Context-Dependent Drivers of Discretionary Debt Decisions: Explaining Willingness to Borrow for Experiential Purchases

Context-Dependent Drivers of Discretionary Debt Decisions: Explaining Willingness to Borrow for... Mental accounting research suggests that consumers prefer borrowing for longer-lasting purchases in order to receive benefits from the purchases as they pay for them. In contrast, two sets of archival data and five lab studies show that consumers are more willing to borrow for experiential versus material purchases, even though experiential purchases tend to have a shorter physical duration. Further, framing the same purchase as more experiential than material increases willingness to borrow. This effect occurs because purchase timing is more important for experiential purchases—a function of consumers’ aversion to missing out on planned consumption. Thus, we moderate the proposed effect by varying whether the borrowing decision impacts planned consumption. Other differences between material and experiential purchases, such as scarcity or expected happiness, cannot similarly explain our results. Moreover, our conceptualization allows us to reconcile the apparent contradiction between the previous and current research by examining the relative impact of purchase-timing importance and payment-benefit duration matching in different contexts (i.e., “purchasing” and “source-of-funding” decisions). experiential purchases, material purchases, consumer borrowing, debt, discretionary purchases, financial decision making, willingness to borrow, mental accounting The average American household spends $12,800 annually on discretionary purchases (Experian Simmons, 2011) and has $7,200 in credit card debt (Picchi 2015). In light of these figures, the current research examines borrowing decisions for discretionary purchases. In particular, we focus on investigating when, why, and for what types of purchases consumers will be more willing to borrow. Mental accounting research suggests that a purchase’s longevity (e.g., how long the purchase will last over time) predicts borrowing preferences (Hirst, Joyce, and Schadewald 1994). People prefer borrowing for purchases that last longer in time because they continue receiving benefits from the purchase over the period of time they pay for it. One classification of purchases that tends to vary systematically by physical longevity is that of experiential and material purchases (Tully, Hershfield, and Meyvis 2015; Van Boven and Gilovich 2003). Since experiential purchases are typically shorter-lived than material purchases, the previous research suggests that people should prefer borrowing for material rather than experiential purchases. In contrast, our research argues that people are typically more willing to borrow for more experiential purchases. Most consumers borrow when they want to make a purchase but do not have adequate funds. Thus, borrowing decisions determine whether or not a purchase can be made. Accordingly, decisions to borrow should be influenced by how important consumers feel it is to make the purchase at the time at which they had planned. We argue that it seems more important to make an experiential purchase when planned, as compared to a material purchase, because consumers are averse to missing out on planned consumption. Experiential (vs. material) purchases, by definition, are made for the primary intention of consumption (vs. retained ownership). Thus, not making an experiential purchase when planned is more likely to impede planned consumption. Hence, consumers are more likely to stick to their plans to buy experiential (vs. material) purchases, even when doing so requires bypassing liquidity constraints and using borrowed funds. Our results suggest that, for most borrowing decisions, purchase-timing importance is a stronger driver than is payment-benefit duration matching, which has been found to drive borrowing decisions in previous research (Hirst et al. 1994; Prelec and Loewenstein 1998). We reconcile our results with those seemingly contradictory results by considering differences across borrowing contexts. We suggest and find that consumers typically borrow when other payment resources are unavailable. However, consumers may at times choose to borrow even when other payment resources are available (e.g., due to attractive financing terms). In the latter borrowing contexts, the borrowing decision determines not whether, but rather how, the purchase is made. Since the purchase will be made regardless of whether one uses borrowed funds, planned consumption is unaffected and thus purchase-timing importance is less likely to affect decisions. Instead, in these contexts, payment-benefit duration matching becomes a more focal consideration and results in greater willingness to borrow for material purchases. We present seven studies that investigate the relationship between material and experiential purchases and willingness to borrow. Archival data and controlled lab studies demonstrate our predicted effects across a range of purchases. Further, we reconcile the current and previous research to provide a greater understanding of when and why purchase-timing importance drives borrowing decisions. THEORETICAL BACKGROUND Consumers’ willingness to borrow can stem from various sources, and we group these sources into three general categories: (1) consumer characteristics, (2) the terms of the loan, and (3) the underlying purchase. In terms of consumer characteristics, research has examined various individual differences and demographic variables that affect willingness to borrow (e.g., age, income, attitude toward credit; Kim and DeVaney 2001; Zhu and Meeks 1994). With respect to loan terms, research has examined how factors such as interest rates, credit limits, and repayment options influence borrowing (Gross and Souleles 2002; Kim and DeVaney 2001; Soman and Cheema 2002). Less research has examined how characteristics of the underlying purchase influence borrowing. We address this gap in the current research. Characteristics of Purchases that Influence Borrowing Seminal work on how purchase characteristics influence borrowing preferences argues that the physical longevity of purchases should be the prominent driver of debt decisions (Hirst et al. 1994). The authors found that when people purchase furniture and a vacation simultaneously, and have the option to pay for one with savings and the other with a loan, they overwhelmingly prefer to use the loan for the furniture. The psychological mechanism driving this effect is that people generally prefer to match the duration of their payments with the duration of the benefits they receive (Auh, Shih, and Yoon 2008; Prelec and Loewenstein 1998). This line of reasoning has implications for one classification of purchases that has received increased attention over the last decade—material versus experiential purchases. Experiential purchases are purchases made for the primary intention of attaining a life experience—living through an event or series of events. Material purchases are made for the primary intention of ownership and possession—getting something that can be retained over time (Van Boven and Gilovich 2003). The distinction between experiential and material purchases is readily recognized by consumers and has substantial consequences for consumer well-being (Carter and Gilovich 2010; Rosenzweig and Gilovich 2012; Van Boven 2005; Van Boven and Gilovich 2003). Importantly, material purchases tend to be longer lasting than experiential purchases (Tully et al. 2015). Given this difference in longevity, previous research on payment-benefit duration matching suggests that people should be more willing to borrow for material rather than experiential purchases. Related research provides additional reasons why people may be more willing to borrow for material versus experiential purchases. For example, consumers predict that material goods provide greater economic value (Pchelin and Howell 2014). Indeed, incurring debt for material goods might seem logical given that material purchases may serve as assets that can be resold if needed (e.g., to repay a debt). While material purchases typically last longer than experiential purchases, we suggest that this difference in longevity may influence borrowing in a different manner than is currently assumed. One way consumers may react to purchase longevity, as suggested by the previous research, is by considering the long-lasting benefits of material purchases (Hirst et al. 1994; Tully et al. 2015). Thinking about the long-lasting benefits of material purchases may prompt considerations regarding payment-benefit duration matching and, in turn, increase willingness to borrow for material purchases. However, another way consumers may react to purchase longevity—which we suggest is more common—is by considering the ephemeral nature of the intended benefits of experiential purchases. Imagine a consumer who has planned to go on a snorkeling tour this weekend. If he does not have the money available, he can either use borrowed funds or forgo making the purchase. If the consumer does not make the purchase this weekend, he could save money to purchase a snorkeling tour in the future. However, to the extent that the consumer has conceptualized the snorkeling tour as the specific event taking place this Saturday (i.e., the planned consumption), the later tour may feel like an entirely different purchase. Thus, the feeling that this exact purchase cannot readily be obtained later may compel the consumer to use debt to fund the purchase to avoid missing out on his planned consumption. In contrast, imagine a consumer who has planned to get snorkeling equipment this weekend but does not have the money available. To the extent that the snorkeling equipment is purchased to be owned and retained over time (material purchase) rather than to gain an experience (experiential purchase), the purchase timing of the snorkeling gear may seem less important. The consumer may therefore be more likely to forgo making the purchase until the snorkeling equipment can be purchased with cash on hand. The above example illustrates why purchase timing is likely to feel more important for experiential purchases. If one comes to regret not buying an experiential purchase, it may seem difficult (or impossible) to reverse that decision and get that originally intended consumption occasion, leading to feelings of inaction regret. In contrast, if one comes to regret not buying a material purchase, it may seem easier to reverse the decision and establish ownership. Previous research alludes to the idea that forgoing a purchase in the present may be more consequential for experiential than for material purchases (Carter and Gilovich 2010; Rosenzweig and Gilovich 2012). For instance, Carter and Gilovich (2010) demonstrate that people view experiential purchases as less comparable to one another relative to material purchases. While this research examined the comparability of purchase options considered simultaneously, the greater difficulty of comparing experiences may extend to intertemporal but otherwise identical alternatives (e.g., comparability of a snorkeling tour at one time vs. another time). To summarize, we propose that making the purchase when planned is more important for experiential (vs. material) purchases, which results in greater willingness to borrow for experiential purchases. More formally, we propose: H1: Consumers are more willing to borrow for experiential (vs. material) purchases when debt is necessary for purchasing. H2: The effect in hypothesis 1 is mediated by purchase-timing importance, which is greater for experiential purchases. A large body of research suggests that experiential purchases result in more long-term happiness (Van Boven and Gilovich 2003). Thus, an alternative explanation for greater purchase-timing importance for experiential purchases is that people perceive these purchases to be better and more important for their well-being. We distinguish between this alternative account and the proposed explanation by identifying a moderating factor: the impact of the borrowing decision on planned consumption. If purchase-timing importance is indeed greater for experiential purchases because consumers are more averse to missing out on planned consumption, then differences in willingness to borrow should be moderated by changes to the relationship between the borrowing decision and planned consumption. More formally, we propose the following: H3: Hypothesis 1 will be moderated by whether consumers believe the decision to borrow impedes planned consumption. Although consumption is typically planned for experiential purchases (consumers know when the snorkeling tour will occur when they buy tickets for it), when people think about an experience more generally (“I want to go snorkeling sometime”), differences in willingness to borrow should be attenuated. Correspondingly, although consumption is not typically planned for material purchases, when it has been planned (“I will use the snorkeling gear this weekend”), purchase timing should feel more important, and differences in willingness to borrow should be attenuated. More formally: H3a: Differences in willingness to borrow will be attenuated when either consumption has not yet been planned for experiential purchases or when consumption has been planned for material purchases. Reconciling the Current and Previous Research Hypothesis 3 also helps to reconcile our proposed effects with those found in the previous research on payment-benefit duration matching (Hirst et al. 1994; Prelec and Loewenstein 1998). To reconcile the current predictions with the predictions in the prior work, we identify two distinct borrowing contexts that differentially affect planned consumption to understand when each factor is more likely to influence borrowing decisions. People may borrow because they do not have the resources to make the purchase otherwise or because they prefer using borrowed funds rather than other available resources (e.g., due to attractive financing). Most consumers borrow when they do not have resources to make the purchase outright.1 Thus, reluctance to borrow results in forgoing the purchase in the present, and it determines whether the purchase is made. The strongest predictor of borrowing decisions should therefore be purchase-timing importance, as it directly relates to the time when a purchase is made. Other factors, such as payment-benefit duration matching, should be secondary considerations. However, when these conditions are not met—such as when a purchase can be funded with debt or another source—the impact of purchase-timing importance on willingness to borrow should decrease. Indeed, the previous research used a joint purchasing context and found that consumers who assume they are buying two purchases (e.g., furniture and a vacation) prefer financing the longer-lasting purchase, and paying outright for the more ephemeral purchase (Hirst et al. 1994). The decision was thus a source-of-funding decision that determined not whether but how the purchase was made. Since purchase timing is unaffected in these situations, people may revert to other decision-making strategies such as matching the purchase’s payments with its utility stream. We would then expect greater willingness to borrow for material rather than experiential purchases. We summarize this proposed reconciliation in hypothesis 3b. H3b: Willingness to borrow will depend on the borrowing context. When consumers face a tradeoff between using debt and forgoing their purchase in the present (i.e., purchasing decisions), they will be more willing to borrow for purchases that are more experiential (vs. material). When consumers face a tradeoff between using debt and an alternative funding source (i.e., source-of-funding decisions), they will be more willing to borrow for purchases that are more material (vs. experiential). Next, we report the results of two archival datasets and five lab studies that demonstrate our proposed effects. Analyses of archival data demonstrate that consumers who buy more experiential purchases have higher credit card balances and pay more financing fees. In a second archival dataset, we find that there are more funded peer-to-peer loans for discretionary experiential versus material purchases. In a more controlled lab setting, we demonstrate greater willingness to borrow for experiential purchases and provide evidence that it is driven by greater purchase-timing importance for experiences. We rule out alternative explanations for this effect by holding constant the focal purchase and showing that manipulating the same purchase to seem more experiential increases willingness to borrow, via increased purchase-timing importance. We further isolate our proposed explanation by demonstrating that when consumption for experiential purchases has not been planned, or when consumption for material purchases has been planned, people are no more likely to borrow for experiential than for material purchases. Finally, we reconcile the results of the current and prior research and demonstrate when to expect payment-benefit duration matching, as compared to purchase-timing importance, to exert a greater influence on borrowing decisions. STUDY 1: ARCHIVAL ANALYSIS OF CREDIT CARD SPENDING As an initial exploration, we studied the relationship between consumer debt and spending on discretionary experiences in archival consumer expenditure data. Specifically, we examined data from the Consumer Expenditure Survey available from the United States Bureau of Labor Statistics (BLS). The Consumer Expenditure Survey provides expenditure data at the consumer unit level, elicited through an interview survey. These data have been used in other research examining discretionary spending on material goods and experiences (Tully et al. 2015). In these data, entertainment purchases are split into three categories: fees and admissions (akin to experiential purchases); TV, radio, and sound equipment (akin to material purchases); and an “other” category for miscellaneous purchases. As in the previous work, we calculated the relative spending on experiential purchases as compared to material purchases using the difference in spending between these categories, as a function of total consumer spending (Tully et al. 2015). That is, we subtracted consumers’ spending in the material category from their spending in the experiential category, as a proportion of their total expenditures. Thus, greater values on this measure reflected greater spending on discretionary experiential purchases relative to discretionary material purchases. We examined the relationship between this measure and consumer debt using two measures of consumers’ debt: (1) the total amount consumers owed on all of their credit cards, and (2) the total amount they paid in financing, late charges, and interest for all their credit cards in the last month. Because the BLS began reporting these credit measures in the second quarter of 2013, we collected and compiled the available quarterly data from the second quarter of 2013 through the fourth quarter of 2014. In support of our hypothesis that consumers are more willing to borrow for experiential versus material purchases, there was a significant positive relationship between relative spending on discretionary experiences and the balance consumers carried on their credit cards (β = 0.03, t(30240) = 5.35, p < .001), as well as the financing interest they paid in the last month (β = 0.03, t(30240) = 4.68, p < .001). The results remain unchanged when we controlled for income (βcredit balance = 0.01, t(30239) = 2.29, p = .022; βfinancing paid = 0.02, t(30239) = 3.64, p < .001). As an additional robustness check, we restricted the dataset to include only the subset of participants who reported having debt and who reported having spent money in at least one of the discretionary spending categories. Among participants reporting a balance on their credit card and who spent money in at least one of the discretionary spending categories, greater relative spending on discretionary experiences significantly and positively predicted higher credit card balances (β = 0.07, t(1833) = 2.80, p = .005). Similarly, among participants who reported paying financing fees and who spent money in at least one of the discretionary spending categories, greater relative spending on discretionary experiences significantly and positively predicted higher financing payments (β = 0.11, t(778) = 3.06, p = .002).2 In sum, using archival data of actual consumer expenditures, we found initial support for the link between discretionary spending on experiences and consumer debt. STUDY 2: ARCHIVAL ANALYSIS OF PEER-TO-PEER LOANS As a secondary examination of the relationship between experiential purchases and willingness to borrow, we sought archival data that could more directly connect the incurrence of debt to the purchase it is used to fund. To this end, we collected real funded loan data from a large, US peer-to-peer lending company that included a description of the purchase for which the loan would be used. In a peer-to-peer lending context, consumers may request and obtain funding for multiple purposes, including both discretionary material and discretionary experiential purchases. Accordingly, peer-to-peer loans, like credit cards, are an ideal debt type for the current research investigation. The dataset included 42,535 loans funded between 2007 and 2011. Each entry included a description of the loan’s intended use, written by the individuals who had requested the loans. A research assistant who was blind to our hypotheses used these descriptions to identify discretionary purchases. The complete dataset of discretionary purchases included 2,571 loans, which represents approximately 6% of the full dataset. To categorize the purchases, we created 69 broad categories to reflect the purchases and asked 30 MTurk workers to rate the categories based on the extent to which they seemed primarily material or primarily experiential (–3 = primarily material, 0 = equally material and experiential, 3 = primarily experiential). In line with predictions, a t-test revealed a positive and significant difference against zero, indicating greater borrowing for more experiential purchases, t(2,570) = 12.60, p < .001. However, the above result could also occur if experiential purchases are rated more extremely than material purchases and are thus being overweighted. Accordingly, we conducted another analysis that allowed us to examine greater instances of experiential versus material loans. We classified each of the 69 categories as material or experiential using a t-test against the midpoint of the scale. The 2,571 discretionary loans were then coded as material (–1) or experiential (+1) based on the category that best fit the description of the loan’s purpose. Out of the 2,571 loans, 158 loans fell into categories that were rated as ambiguous (not different from the scale’s midpoint) and were excluded, leaving a final sample of 2,413 loans (full details on the categories and exclusion criteria for this study are available in the web appendix). To test whether consumers were more willing to borrow for experiential purchases, we conducted a t-test against zero. Replicating the results of the continuous measure, there was a positive and significant difference from zero, t(2412) = 3.04, p = .002, suggesting that consumers took on more peer-to-peer loans for discretionary experiential purchases (53.1%) than for discretionary material purchases (46.9%). Moreover, there were no differences in average loan amount (F < 1), but the interest rates for the loans of more experiential purchases were 1.15% higher than those for more material purchases, F(1, 2411) = 60.81, p < .001. The results of the peer-to-peer loan data were consistent with the results found in the BLS data, providing additional real-world evidence of the proposed phenomenon. Using real funded peer-to-peer loan data, these data explicitly connect the incurrence of debt to its intended purchase and reveal that people took on more peer-to-peer loans to fund the purchase of experiential versus material discretionary purchases. Transitioning from the archival data, we next examined whether consumers’ interest in and willingness to use debt varied as a function of their latent perceptions of their purchases as more material or more experiential. STUDY 3: CORRELATIONAL PEER-TO-PEER LENDING Study 3 was designed to examine naturally occurring relationships between how experiential a purchase seems and consumers’ willingness to borrow. We examined this relationship in the peer-to-peer lending context. We expected participants to be more willing to borrow for purchases that they rated as more experiential (hypothesis 1). Method Five hundred and one MTurk workers (Mage = 33.84, SD = 11.40; 53.9% female) completed the study in exchange for monetary compensation. No subjects were removed from the analysis in this study or in any of the following studies. We designed the stimuli for this study based on a real peer-to-peer lending website that depicted six categories of purchases: debt consolidation, home improvement, auto and vehicle, baby and adoption, small business, and special occasion. Of these categories, special occasion best aligned with discretionary purchases. Thus, aiming for external validity, we asked participants to list and describe one special occasion purchase they were considering buying that cost at least $500. Next, we provided a brief description of peer-to-peer lending and asked participants to indicate whether they had heard of such lending before (yes/no). We then asked participants about two real action steps they might take to fund their purchase with peer-to-peer loans. The first measure read: “How interested would you be in finding out about competitive rates and opportunities to fund your purchase through peer-to-peer lending?” (seven-point scale; 1 = not at all interested, 4 = neutral, 7 = very interested). The second measure read: “How likely are you to visit a peer-to-peer lending website to learn about different funding options for that special purchase today?” (seven-point scale; 1 = not at all likely, 4 = neutral, 7 = very likely). Participants then reviewed a description of material and experiential purchases and were asked to classify their purchase using a seven-point scale (1 = completely material, 7 = completely experiential). Next, participants indicated the month in which they intended to make the purchase and indicated how much the purchase would cost. They also responded to demographic questions: age, gender, and household income. Finally, we asked participants whether they knew the study’s intended purpose. No participant indicated suspicion about the study’s objective nor did any participant guess our research questions or hypotheses. Results and Discussion We first examined the relationship between participants’ ratings of their future purchases and their interest in learning about peer-to-peer lending options for their purchase. As predicted, we found a significant positive correlation, r = .15, p = .001, indicating that the more experiential participants rated their purchase, the more interested they were in learning about peer-to-peer lending rates. These results replicated when we examined the correlation between participants’ purchase ratings and their likelihood of visiting a peer-to-peer lending site on that day, r = .16, p < .001. These results remained unchanged when we adjusted for factors such as cost, age, gender, and income. In sum, this study provided convergent evidence with the peer-to-peer lending archival data, demonstrating a relationship between how experiential a purchase seems and interest in using peer-to-peer lending options to fund the purchase. We conducted another study examining willingness to borrow for people’s planned purchases that manipulated rather than measured the experiential nature of the purchase. In this study, we asked participants to consider a material or experiential purchase they planned to make and measured willingness to finance this purchase with a credit card. Replicating the correlational results in study 1, participants asked to consider an upcoming experiential (vs. material) purchase were more willing to finance the purchase. Further, this study measured a number of potential process measures, finding support for the proposed purchase-timing importance explanation. Other factors, such as the happiness expected from the purchase, could not similarly explain the effect. See supplemental study 1 in appendix A of the web appendix for the complete study and results. STUDY 4: MANIPULATING THE EXPERIENTIAL NATURE OF THE PURCHASE Thus far, we have demonstrated the predicted relationship between experiential purchases and willingness to borrow using archival consumer data and consumers’ self-generated planned purchases. These findings provide external validity, demonstrate the generalizability of the proposed effect across various real purchases, and indicate that the results are not due to specific experimental stimuli. Study 4 was designed to isolate the causal role of purchase type (material vs. experiential) on willingness to borrow by examining whether framing the same purchase as more experiential would increase willingness to borrow. Method We recruited 407 adults (Mage = 33.68, SD = 11.31; 52% male) on MTurk to complete the survey in exchange for monetary payment. The study followed a between-subjects design that manipulated purchase framing—whether participants were led to think about an outdoor grill as more material or more experiential. Participants first indicated whether they owned a grill. Next, they were informed of research showing that people may think about purchases along a continuum ranging from mainly material to mainly experiential. We then introduced the purchase framing manipulation (see appendix B of the web appendix for exact stimuli). Depending on condition, participants were told that many people think about grills as being more experiential (material) and were asked to explain why people might consider outdoor grills to be more experiential (material). Since marketing pitches commonly ask people to imagine what their life would be like with a product, we constructed a marketing pitch for an outdoor grill that included a statement about grill availability that summer, followed by a mental simulation. Specifically, the pitch began: “This summer there are a number of attractive grills available for purchase.” Then, based on condition, participants were asked to imagine having an experience with (experience condition) or owning (material condition) a new outdoor grill and to write about aspects related to those thoughts. After this marketing pitch, we measured purchase-timing importance and three measures related to willingness to borrow. We measured purchase-timing importance using two questions about the extent to which purchase timing was (1) important and (2) relevant for the purchase of the outdoor grill (both questions: 1 = not at all, 7 = very). The primary dependent measure was participants’ interest in using real, currently available financing options to acquire a new outdoor grill. We informed participants that many hardware stores were providing financing options that allow consumers to pay for their purchases over time. We then asked participants to indicate their interest in using such a payment option. Specifically, based on condition, they indicated their interest in using financing that would allow them to have the experience with (experiential condition) or own (material condition) a new outdoor grill (1 = not at all, 7 = extremely). Subsequently, we administered an exploratory behavioral measure of willingness to borrow: participant website visits. Before conducting this study, we compiled a list of popular grill vendors (e.g., Best Buy, Home Depot, Sears) and the current financing offers available from those vendors. We created two Google websites (one for each experimental condition) designed to host that information. The web layout and content was exactly the same on each website. As a real behavioral measure of interest in financing a new outdoor grill, we asked participants whether they would like access to the site at the end of the study (yes/no). Additionally, using Google Analytics, we tracked how many site views each site received. Before completing demographic information, participants completed a manipulation check, rating the extent to which they thought about a grill as more material or more experiential using a seven-point scale (1 = completely material, 7 = completely experiential). Results and Discussion Manipulation Check Participants in the experiential condition (M = 5.09, SD = 1.42) rated an outdoor grill as more experiential in nature than did participants in the material condition (M = 3.03, SD = 1.66), F(1, 405) = 181.49, p < .001, ηp2 = .309. Interest in Borrowing In line with our predictions, there was a significant effect of purchase type on interest in using financing available at hardware stores. Asking participants to think about a grill as an experiential purchase increased their interest in using financing options relative to asking them to think about a grill as a material purchase (Mexperiential = 3.93, SD = 2.13, Mmaterial = 3.44, SD = 2.10), F(1, 405) = 5.36, p = .021, ηp2 = .013. Exploratory Behavioral Measure We next examined the effect of purchase framing on participants’ request to view the financing website. The binary measure of interest in viewing the website was directionally consistent but not significant: 47% of participants in the experiential condition requested access to the site compared to 44% of participants in the material condition, χ2 < 1. There was, however, correlational evidence in support of our hypotheses. There was a significant positive correlation between participants’ self-rated perceptions of an outdoor grill being more experiential (vs. material) and requests to view the website, r = .142, p = .004. Results for actual site visits were similar to the results for the binary website request measure, with 15.5% of participants in the experiential condition visiting the site compared to 12.1% of participants in the material condition, Z = 1.00, p = .32. Purchase-Timing Importance The two measures of purchase-timing importance were combined into a single measure (r = .80, p < .001). Consistent with the results for interest in financing, participants in the experiential condition (M = 5.01, SD = 1.54) rated purchase timing as more important than did participants in the material condition (M = 3.87, SD = 1.88), F(1, 405) = 44.23, p < .001, ηp2 = .098. Mediation Next, we tested whether purchase-timing importance mediated the relationship between purchase type and interest in using hardware store financing. Hayes’s (2013) PROCESS macro with 20,000 bootstrapped samples supported the proposed mediation, as the 95% confidence interval for the indirect pathway did not include zero (.21, .58). We have argued that experiential purchases increase purchase-timing importance because consumers are averse to disruptions to planned consumption. Hence, participants are more willing to use debt when an outdoor grill is framed as more experiential because consumption may otherwise be disrupted. That is, even though an outdoor grill is a durable good that can provide many consumption occasions, participants’ willingness to borrow is greater when it is framed as more experiential because some planned consumption will be otherwise forgone. However, another possibility is that thinking about a grill as more experiential makes the grill more desirable overall (e.g., because it highlights the social value). We begin to distinguish between these two accounts by conducting additional analyses to test hypothesis 3a. Additional Process Evidence In this study, we found greater willingness to borrow for a purchase that we featured in a marketing pitch. Thus, some people likely thought about the grill more generally (“I will use the grill with friends”) as opposed to having planned consumption (“I will use the grill this weekend”). Hence, we used natural variation in participants’ spontaneous consideration of intended consumption occasions to test our prediction. Our theory posits that heightened willingness to borrow for the grill in the experiential condition is driven by participants who spontaneously considered one or more specific consumption occasions they might have with the grill. To directly test this assertion, we revisited participants’ written responses to the marketing pitch for an outdoor grill in which they described either having an experience with the grill or owning the grill. The open-ended responses were coded for whether a time (for consumption or ownership) was specified (e.g., 4th of July, Memorial Day, someone’s birthday; 1 = mentioned, 0 = not mentioned). Only one participant in the material condition specified a time, whereas almost a third of participants (29%) did so in the experiential condition. Thus, we examined willingness to borrow among three groups of participants: (1) those in the material condition, (2) those in the experiential condition who did not specify a consumption time, and (3) those in the experiential condition who specified a consumption time. Two contrasts provided evidence for our proposed explanation. Willingness to borrow was significantly greater among participants in the experiential condition who had specified a time for consumption (M = 4.80, SD = 1.92) compared to those in the material condition (M = 3.44, SD = 2.10), F(1, 404) = 19.31, p < .001, ηp2 = .046. However, there was no difference in willingness to borrow between participants in the material condition and those in the experiential condition who did not specify the time for consumption (M = 3.57, SD = 2.11), F < 1, NS, ηp2 < .001. Overall, in study 4, we induced greater willingness to borrow for the purchase of an outdoor grill by having people think about grills as more experiential rather than material. These results provide greater evidence of a causal relationship between how experiential a purchase seems and people’s willingness to borrow for it. Focusing participants on the experiential aspects of a purchase increased purchase-timing importance and, in turn, willingness to borrow for an outdoor grill. Further, our follow-up analyses provide further evidence that consumers are averse to missing out on planned, but not unplanned, consumption. STUDY 5: DECREASING THE EXTENT TO WHICH BORROWING IMPACTS PLANNED CONSUMPTION FOR EXPERIENTIAL PURCHASES The additional analyses conducted in study 4 provide evidence that the increase in willingness to borrow for experiential purchases depends on having consumption planned. However, it is possible that our effects are not due to the greater perceived impact of changes to planned consumption versus ownership, but rather the mere planning for a purchase. Thus, in study 5, we aimed to disentangle the mere specification of a particular time from the specification of consumption timing in particular. To do so, we asked participants to consider a given experiential or material purchase, and we varied whether people expected to obtain the purchase’s objectives (consumption or ownership) at a specific time or at one out of a number of potential times. We expected participants to be more willing to borrow for experiential versus material purchases when timing was planned. However, when timing was instead flexible, we expected differences in willingness to borrow to be attenuated. Method Participants Participants in the main study were 333 undergraduate students at the University of Southern California who completed the study in exchange for partial course credit (Mage = 20.1, SD = 1.55, 47.9% female). The study followed a 2 (purchase: material vs. experiential) × 2 (timing: planned vs. unplanned) between-subjects design. Procedure Participants were asked to imagine the opportunity to make a given purchase. Half of the participants imagined an experiential purchase (a day trip to Disneyland) and the other half imagined a material purchase (smartwatch). These purchases were determined based on results of a pretest that aimed to equate the purchase types in terms of their perceived desirability, value, uniqueness, and availability to our particular participant sample. See appendix B of the web appendix for complete pretest results. Next, to examine how planning the timing to realize the purchase’s objective influenced willingness to borrow for experiential and material purchases, we told half of the participants to imagine they were planning to make the purchase on the following Saturday (timing planned). We told the remaining half to imagine they were considering a few dates, with one possible date being the following Saturday (timing unplanned). See appendix B of the web appendix for a full description. Participants were then told to imagine they had checked their bank account and did not have enough money for the purchase in their account. They could either wait to save enough money to make the purchase outright, or they could finance the purchase on a credit card with 15% interest. Participants then indicated their likelihood of financing the purchase with their credit card (1 = very unlikely, 7 = very likely). After indicating their borrowing likelihood, participants were asked to categorize the focal purchase as either an experience or a material good. In addition, although we had pretested our stimuli to ensure that the experiential purchase did not seem more unique, we administered four measures assessing the perceived scarcity of the purchases, to ensure that the experiential purchase was not perceived as a scarcer opportunity once timing had been planned. Participants rated the extent to which the purchase they were asked to consider was unique, unusual, rare, and once-in-a-lifetime using a seven-point scale (1 = not at all, 7 = extremely). There were no differences on these measures, so we do not discuss them further. Results and Discussion Manipulation Check Participants’ categorization of their respective purchases varied by condition, χ2 = 190.98, p < .001. Seventy-eight percent of participants considering the material purchase categorized it as a material good, while 96% of participants considering the experiential purchase categorized it as an experience. Willingness to Borrow There was no main effect of timing on willingness to borrow. However, there was a significant main effect of purchase type, such that participants in the experiential condition (M = 3.03, SD = 1.95) were significantly more willing to borrow than were participants in the material purchase condition (M = 2.44, SD = 1.64), F(1, 329) = 8.56, p = .004, ηp2 = .025. However, the main effect of purchase type was qualified by a marginally significant purchase type by timing interaction, F(1, 329) = 2.83, p = .09, ηp2 = .009. Follow-up planned contrasts supported the hypothesized results. Participants were more willing to borrow for the experiential purchase (M = 3.27, SD = 1.96) than for the material purchase (M = 2.33, SD = 1.64) when timing was planned, F(1, 329) = 10.47, p < .001, ηp2 = .031. However, when timing was unplanned, there was no difference in willingness to borrow for the two purchase types (Mexperiential = 2.81, SD = 1.93 vs. Mmaterial = 2.56, SD = 1.81), F < 1, ηp2 = .002. These results are displayed in figure 1. FIGURE 1 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES DEPENDING ON WHETHER TIMING IS PLANNED IN STUDY 5 NOTE.—Error bars indicate standard errors of the mean. FIGURE 1 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES DEPENDING ON WHETHER TIMING IS PLANNED IN STUDY 5 NOTE.—Error bars indicate standard errors of the mean. Study 5 attenuated the greater willingness to borrow for experiential purchases by minimizing the extent to which the decision to borrow affected planned consumption. When consumption was unplanned, willingness to borrow for experiential purchases decreased and differences in willingness to borrow were attenuated. STUDY 6: INCREASING THE EXTENT TO WHICH BORROWING IMPACTS PLANNED CONSUMPTION FOR MATERIAL PURCHASES We designed study 6 to test whether strengthening the link between planned consumption and the decision to borrow could attenuate the greater willingness to borrow for experiential purchases by increasing willingness to borrow for material purchases. To do so, we varied how purchase timing was described. For some participants, purchase timing was described in a standard form: the time when one plans to make the purchase. In these situations, we expected people to infer that purchase timing affects consumption timing for experiential purchases and ownership timing for material purchases, and hence we predicted greater willingness to borrow for experiential purchases. However, for other participants, purchase timing was explicitly linked to planned consumption (i.e., one plans to make the purchase in order to use it at a certain time). In these situations, we predicted that the heightened willingness to borrow for experiential purchases would be attenuated. Method Participants were 405 adults (Mage = 34.8, SD = 12.4; 57% females) on MTurk who completed the survey in exchange for monetary payment. This study followed a 2 × 2 between-subjects design that crossed purchase type (experiential vs. material) and purchase-timing type (standard description vs. consumption-linked description). Participants in the experiential condition imagined purchasing a snorkeling tour, and participants in the material condition imagined purchasing snorkeling equipment. In the standard purchase-timing conditions, participants were simply told to imagine they were making the purchase at the end of the month. Our expectation was that participants would naturally link the experiential purchase (but not the material purchase) to consumption. However, in the consumption-linked purchase-timing conditions, we explicitly linked purchase timing to planned consumption for both purchases. Specifically, we described the experiential purchase as “a snorkeling tour for the summer vacation you are taking at the end of the month,” and the material purchase as “a set of snorkeling equipment to use on the summer vacation you are taking at the end of the month.” Next, participants rated the importance of purchase-timing for their respective purchase (1 = not at all important, 7 = very important). Then they were asked to imagine that, although they had planned to make the purchase, the purchase cost $150, and they did not have the money to pay for it outright. For the dependent measure, they indicated their willingness to finance any part of the purchase (1 = extremely unwilling, 7 = extremely willing). At the study’s end, we collected open-ended responses pertaining to whether participants knew the study’s intended purpose. No participant indicated suspicion about the study’s objective nor did any participant guess our research questions or hypotheses. Results Manipulation Check As intended, participants who read about a snorkeling tour (Mexperiential = 5.88, SD = 1.63) rated their purchase as more experiential (vs. material) than did participants who read about snorkeling equipment (Mmaterial = 4.57, SD = 1.71), F(1, 401) = 61.55, p < .001, ηp2 = .133. Willingness to Borrow There was no main effect of purchase type (F < 1, NS). There was a main effect of timing type, F(1, 401) = 11.02, p = .001, ηp2 = .027. This main effect was qualified by the expected interaction, F(1, 401) = 8.66, p = .003, ηp2 = .021. As expected, in the standard timing condition, participants were more willing to borrow for the experiential (vs. material) purchase, F(1, 401) = 6.32, p = .012, ηp2 = .016. However, in line with our expectations, when timing was specifically linked to consumption, greater willingness to borrow for the experiential purchase was attenuated. In fact, preferences reversed and willingness to borrow was actually marginally greater for material (vs. experiential) purchases, F(1, 401) = 2.75, p = .098, ηp2 = .007. These results are depicted in Figure 2. FIGURE 2 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES BY TIMING TYPE IN STUDY 6 NOTE.—Error bars indicate standard errors of the mean. FIGURE 2 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES BY TIMING TYPE IN STUDY 6 NOTE.—Error bars indicate standard errors of the mean. Purchase-Timing Importance There was a significant main effect of purchase type, F(1, 401) = 4.20, p = .04, ηp2 = .010, and timing type, F(1, 401) = 22.86, p < .001, ηp2 = .054, and these main effects were qualified by a significant purchase type by timing type interaction, F(1, 401) = 4.11, p = .04, ηp2 = .010. Follow-up planned comparisons provided support for the hypothesized pattern of results. In the standard timing conditions, participants indicated greater purchase-timing importance for the experiential (M = 4.81, SD = 1.74) versus material (M = 4.14, SD = 1.80) purchase, F(1, 401) = 8.53, p = .004, ηp2 = .021. However, the difference in purchase-timing importance for the experiential (M = 5.27, SD = 1.64) and material (M = 5.27, SD = 1.46) purchase was attenuated when timing was directly linked to consumption, F < 1. Mediation We have predicted that greater purchase-timing importance drives increased willingness to borrow for more experiential purchases because those purchases are typically linked to planned consumption. However, when material purchases are similarly linked to planned consumption, differences in purchase-timing importance should be minimized and other drivers may better predict borrowing preferences. Thus, we tested for the predicted moderated mediation. Hayes’s (2013) PROCESS macro (model 7) with 20,000 bootstrapped samples supported the proposed moderated mediation, as the 95% confidence interval for the indirect pathway did not include zero (–.35, –.01). In line with predictions, in the standard timing condition, purchase-timing importance explained differences in willingness to borrow (.03, .30). However, in the consumption-linked timing condition, purchase-timing importance did not similarly explain willingness to borrow (–.09, .08). It is important to note that, in the consumption-linked timing condition, purchase-timing importance did not differ between purchase types. Therefore, the observed reversal in willingness to borrow suggests that when purchase timing is no longer a critical difference, other factors (such as payment-benefit duration matching) may become more important. STUDY 7: PURCHASING VERSUS SOURCE-OF-FUNDING DECISIONS We argue that our results differ from those found in the previous research on payment-benefit duration matching because in the current context, borrowing decisions imply purchasing decisions (i.e., whether a purchase is made) as opposed to source-of-funding decisions (i.e., how a purchase is made), which have no impact on consumers’ receipt of the purchase. Study 7 was designed to directly test the predicted differences between these two borrowing contexts. The borrowing decision was framed as a purchasing decision for some participants and a source-of-funding decision for other participants. We predicted that when we framed borrowing as a source-of-funding decision, payment-benefit duration matching would become a more prominent driver and lead to a greater willingness to borrow for material versus experiential purchases. Study 7 included three experimental conditions. In addition to having participants consider their willingness to borrow for either a material or an experiential purchase (purchasing decisions), participants in a third condition imagined both purchases, with only enough cash to pay for one outright (source-of-funding decision). We expected to replicate our earlier results when we framed the borrowing decision as a purchasing decision (either financing the purchase or forgoing the purchase in the present). However, when we framed the decision as a source-of-funding decision (deciding which purchase to finance and which to pay for with cash), we expected participants to be more willing to use debt for the material goods, replicating the findings in the existing literature on payment-benefit duration matching. Method We recruited 756 adults (Mage = 32.1, SD = 10.6; 38.6% females) on MTurk to complete this study in exchange for monetary payment. We asked participants to imagine they had been browsing for purchases they wanted to acquire. Participants in the purchasing decision contexts were told they had found a purchase they really wanted to buy: either a new stereo system (material condition) or two tickets to an upcoming music festival (experiential condition). They were told that the purchase cost $400 and would need to be financed if purchased. In the source-of-funding decision context, participants imagined wanting to buy both the stereo system and the music festival tickets. They were informed that both purchases cost $400. They were then told that they had $400 in savings and would need to finance the other $400 purchase. All participants were told to imagine that they expected to repay the borrowed amount in a couple of months. See appendix B of the web appendix for detailed stimuli. Next, participants indicated their willingness to borrow for the experiential purchase (experiential purchasing condition), material purchase (material purchasing condition), or both (source-of-funding condition) using seven-point scales (1 = not at all willing to borrow for [purchase], 7 = very willing to borrow for [purchase]). In addition, to replicate previous findings (Hirst et al. 1994), participants in the source-of-funding condition made a forced choice, indicating which purchase they would rather use borrowed funds to acquire. Next, all participants indicated the extent to which their financing decision involved considerations about “whether the length of the benefits of the purchase would extend beyond when they finished making payments on the purchase” (seven-point scale: 1 = not at all, 7 = very much) before completing demographic questions. Results and Discussion Willingness to Borrow Consistent with our earlier studies, in the purchasing decision contexts, participants were more willing to borrow for the experiential purchase (M = 3.41, SD = 2.12) than for the material purchase (M = 2.93, SD = 1.90), F(1, 503) = 7.35, p = .007, ηp2 = .014. However, as predicted, the results in the source-of-funding context were consistent with the previous literature. Participants were less willing to borrow for the experiential purchase (M = 2.41, SD = 1.68) than for the material purchase (M = 4.76, SD = 1.97), t(244) = 12.07, p < .001, ηp2 = .374. Further, when forced to decide which purchase they were more likely to finance, 79.1% of participants indicated greater likelihood to borrow for the material purchase. This percentage was significantly greater than 50%, suggesting the effects were not due to mere chance, t(240) = 11.28, p < .001, ηp2 = .346. Additionally, willingness to borrow for the material purchase was significantly lower when borrowing was framed as a purchasing decision than as a source-of-funding decision, F(1, 497) = 112.71, p < .001, ηp2 = .185, whereas willingness to borrow for the experiential purchase was significantly higher when framed as a purchasing decision than as a source-of-funding decision, F(1, 497) = 31.40, p < .001, ηp2 = .059. (See figure 3.) FIGURE 3 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES BY CONDITION IN STUDY 7 NOTE.—Error bars indicate standard errors of the mean. FIGURE 3 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES BY CONDITION IN STUDY 7 NOTE.—Error bars indicate standard errors of the mean. Payment-Benefit Duration Matching As predicted, a planned contrast demonstrated that participants were significantly more likely to consider payment-benefit duration matching when framed as a source-of-funding decision than a purchasing decision for a material or experiential purchase, F(1, 753) = 23.93, p < .001, ηp2 = .031. Thus, we find support for hypothesis 3b. When the decision was framed as a purchasing decision, participants were more willing to borrow for the experiential purchase than for the material purchase. However, when we framed the decision as a source-of-funding decision, we replicated the results of previous research on payment-benefit duration matching and found greater willingness to borrow for the material (and longer-lasting) purchase. These results cannot be explained by differences in participants’ ability to compare and contrast material and experiential purchases in the source-of-funding decision context. For instance, a material (experiential) purchase might highlight the long-lasting (short-lived) nature when presented alongside an experiential (material) purchase. It is also possible that the mere presence of a reference option cued participants to consider borrowing norms or increased the evaluability of the purchase’s longevity. We addressed these possibilities in an additional study that replicates and extends study 7. See supplemental study 2 in appendix A of the web appendix for the complete study. GENERAL DISCUSSION Academic researchers, practitioners, and consumers are increasingly discussing the significance of the experiential-material continuum, highlighting how the purchases affect consumer psychology and well-being (Van Boven and Gilovich 2003). The current work identifies a new consequence of the material-experiential continuum. Using two archival datasets and five lab studies, we show that consumers are more willing to borrow for experiential than for material purchases. This effect emerges across a range of experiential and material purchases—individuals' past and planned purchases, pretested standardized pairs, as well as the same purchase reframed to be more experiential—providing evidence that the effects are not merely an artifact of the chosen stimuli. In addition, we demonstrate that these effects are explained by the greater purchase-timing importance of experiences. We further show that the greater purchase timing for experiences is driven by consumers’ aversion to missing out on planned consumption. Thus, these effects are attenuated when the link between the borrowing decision and planned consumption is weakened for experiential purchases (study 5) and when this link is strengthened for material purchases (study 6). We also considered other plausible factors that might have contributed to these effects. Although prior research has shown that people report greater long-term happiness from experiences rather than material goods, our results cannot be explained by the mere greater desirability of the underlying purchase (e.g., study 4, study 5, and supplemental study 1 in the web appendix). Further, it is important to note that consumers’ greater willingness to borrow for experiential purchases does not imply greater impatience for consuming experiences. That is, delay discounting cannot explain our results. Indeed, prior research by Kumar and colleagues (2016) suggests that, all else equal, people prefer to consume material goods before experiences due to the greater anticipatory pleasure of experiences. Instead, our effects are driven by the importance of making the purchase when planned, which is a function of consumers’ aversion to missing out on planned consumption. The current findings seem to contradict research showing that debt decisions are based on preferences to match a purchase’s duration of payments with its duration of benefits (Hirst et al. 1994; Prelec and Loewenstein 1998). We reconcile this apparent contradiction in study 7. We show that consumers prefer to borrow for experiential versus material purchases when debt decisions are synonymous with purchasing decisions—when the decision determines whether the purchase is made in the present. However, when debt decisions are source-of-funding decisions—when consumers have multiple sources of funds and must decide how to fund a purchase—they prefer borrowing for material versus experiential purchases. Thus, our theoretical framework offers greater understanding of when purchase-timing importance versus payment-benefit duration matching will more strongly influence borrowing decisions, and hence when consumers will be more willing to borrow for experiential or material purchases. Moreover, we contribute to existing research by differentiating these two types of debt decisions (purchasing decisions vs. source-of-funding decisions). Future research may identify other important differences between these different debt decision contexts. In the current work, we demonstrate a causal relationship between experiential purchases and purchase-timing importance. Holding constant the underlying purchase, we demonstrate that framing a purchase as more experiential increases its purchase-timing importance. However, other factors may also contribute to purchase-timing importance, and our findings should help explain when and why those factors may increase willingness to borrow. For example, consumers who consider themselves innovators may care whether they are among the first to own a product. In such instances, planned ownership timing may be as important as planned consumption timing, and purchase-timing importance may increase. Additionally, the scarcity literature offers real-world contexts in which purchase-timing importance may increase for non-consumption-related reasons (Brock 1968; Lynn 1991; Verhallen and Robben 1994). For example, consider cases when the supply of goods outweighs demand (e.g., popular items, restricted production), is limited to a given time window (e.g., limited editions), or is limited to specific geographic locations (e.g., a souvenir from one’s travels). Consistent with prior work showing that restricted purchasing opportunities evoke purchasing urgency (Abendroth and Diehl 2006; Inman and McAlister 1994), the examples discussed above illustrate situations when differences in willingness to borrow for experiential versus material purchases should be attenuated. Future research might further examine such boundary conditions. The current research demonstrates consistent results across a range of debt types including peer-to-peer lending, store financing, and credit card borrowing. However, one should note that some borrowing forms may be more typical for material goods (e.g., personal loans, layaway plans) because those purchases can be used to securitize loans. Future research could examine whether norms for different debt types affect willingness to finance some types of purchases more than others. This work offers a number of managerial insights. The results of study 7 suggest that if a consumer has decided to make the purchase of a material good, they may be more easily persuaded to use an attractive financing offer (vs. cash). Yet the results from our full package of studies suggest that opportunities to finance experiential purchases may be more enticing than otherwise thought. When consumers have no source of funding other than debt, they are more willing to finance experiential versus material purchases. Thus, providing more financing options for experiential purchases may benefit firms as well as customers. In addition, our work suggests ways in which firms may encourage consumers to use financing offers instead of delaying a purchase. For example, focusing consumers’ attention on a particular opportunity for consumption should enhance purchase-timing importance and hence willingness to borrow. Conducting the present investigation in the domain of consumer borrowing is critical to this work given the opposing predictions suggested by the prior borrowing research. A reasonable question that may arise is whether the effects demonstrated in the current work apply more broadly to spending decisions. To this end, we conducted an additional study, finding greater price elasticity for material (vs. experiential) purchases, driven by differences in purchase-timing importance (see supplemental study 3 in appendix A of the web appendix for full study and results). However, we speculate that these effects may not emerge across all types of spending decisions, just as divergent effects emerge in different debt contexts. For example, differences may emerge in situations when people consider purchase likelihood versus willingness to pay. Although the distinction may seem subtle, the two situations may vary on the purchase-timing importance dimension. Willingness-to-pay decisions may be more akin to borrowing decisions because consumers must consider their indifference point: the maximum amount they are willing to spend before it becomes more attractive to forgo the purchase in the present (and potentially wait for a drop in price). Future research would benefit from further investigating the relationship between borrowing and spending decisions. In sum, our package of studies contributes to the literatures on consumer borrowing and material and experiential purchases by uncovering when to expect consumers to be more willing to borrow for experiential or material purchases, detailing what psychological mechanisms drive these effects, and explaining why exactly those mechanisms operate. It is among the first to examine consumer spending decisions for experiential and material purchases, identifies a new dimension (purchase-timing importance) on which experiential and material purchases systematically differ, demonstrates greater willingness to borrow for experiential purchases, and provides a more nuanced understanding of the relative importance of different drivers of consumer borrowing decisions. DATA COLLECTION INFORMATION The first set of archival data was aggregated and analyzed by the first author. The second set of archival data was aggregated and analyzed by both authors with the help of an independent research assistant at Dartmouth College under the supervision of the second author. Both authors jointly managed the online (MTurk) data collection of all studies except study 5, which was collected by the first author at the University of Southern California’s subject pool. Studies 5 and 6 were analyzed by the first author. All others were analyzed by both authors. Study 3 was conducted in summer 2016. Study 4 and 6 were conducted in summer 2016. Study 5 was collected in fall 2015. Study 7 was conducted in spring 2015. The authors thank Adam Alter, Cindy Chan, Shinar Jain, Kevin Keller, Ellie Kyung, Bob Latimer, Debbie MacInnis, Tom Meyvis, Vicki Morwitz, Abby Sussman, and Gulden Ulkumen for comments and assistance at various stages of the project. Part of this research was conducted when Eesha Sharma was a Junior Visiting Faculty Fellow at the UCLA Anderson School of Management, and support from the Morrison Family Center is acknowledged. Footnotes 1 In a pretest examining the circumstances under which people consider using debt to fund discretionary purchases, 83% of participants stated they would borrow when they do not have the money to buy the purchase outright, rather than because they prefer using debt over other available resources (test against 50%, t(99) = 8.74, p < .001). 2 One might expect the effects found in the subset of consumers included in the robustness check to be stronger than the effects found in the entire sample because the former restricted the sample to those who were more inclined to use financing. To examine this possibility, we tested the significance of the difference between the slopes in the full versus restricted samples. Indeed, the effect was significantly stronger in the reduced sample for both dependent measures (credit card balance: t(32,073) = 2.39, p = .017; financing fees paid: t(31,018) = 2.91, p = .004). REFERENCES Abendroth Lisa J., Diehl Kristin ( 2006), “Now or Never: Effects of Limited Purchase Opportunities on Patterns of Regret over Time,” Journal of Consumer Research , 33 3, 342– 51. Google Scholar CrossRef Search ADS   Auh Seigyoung, Shih Eric, Yoon Yeosun ( 2008), “Aligning Benefits with Payments: A Test of the Pattern Alignment Hypothesis,” Journal of Consumer Psychology , 18 4, 292– 303. Google Scholar CrossRef Search ADS   Brock Timothy C. ( 1968), “Implications of Commodity Theory for Value Change,” in Psychological Foundations of Attitudes , ed. Greenwald Anthony G., Brock Timothy C., Ostrom Thomas M., New York: Academic Press, 243– 75. Google Scholar CrossRef Search ADS   Carter Travis J., Gilovich Thomas ( 2010), “The Relative Relativity of Material and Experiential Purchases,” Journal of Personality and Social Psychology , 98 1, 146– 59. Google Scholar CrossRef Search ADS PubMed  Experian Simmons ( 2011), “The 2011 Discretionary Spend Report,” Experian Simmons, http://www.experian.com/assets/simmons-research/white-papers/2011-discretionary-spend-report.pdf, 3. Gross David B., Souleles Nicholas S. ( 2002), “Do Liquidity Constraints and Interest Rates Matter for Consumer Behavior? Evidence from Credit Card Data,” Quarterly Journal of Economics , 117 1, 149– 85. Google Scholar CrossRef Search ADS   Hayes Andrew F. ( 2013), An Introduction to Mediation, Moderation, and Conditional process Analysis: A Regression-based Approach , New York: Guilford Press. Hirst D. Eric, Joyce Edward J., Schadewald Michael S. ( 1994), “Mental Accounting and Outcome Contiguity in Consumer-Borrowing Decisions,” Organizational Behavior and Human Decision Processes , 58 1, 136– 52. Google Scholar CrossRef Search ADS   Inman J. Jeffrey, McAlister Leigh ( 1994), “Do Coupon Expiration Dates Affect Consumer Behavior?” Journal of Marketing Research , 31 3, 423– 8. Google Scholar CrossRef Search ADS   Kim Haejeong, DeVaney Sharon E. ( 2001), “The Determinants of Outstanding Balances among Credit Card Revolvers,” Financial Counseling and Planning , 12 1, 67– 78. Kumar Amit, Gilovich Tom ( 2016), “To Do or to Have, Now or Later: The Preferred Consumption Profiles of Material and Experiential Purchases,” Journal of Consumer Psychology , 26 2, 169– 78. Google Scholar CrossRef Search ADS   Lynn Michael ( 1991), “Scarcity Effects on Value: A Quantitative Review of the Commodity Theory Literature,” Psychology and Marketing , 8 1, 43– 57. Google Scholar CrossRef Search ADS   Pchelin Paulina, Howell Ryan T. ( 2014), “ The Hidden Cost of Value-Seeking: People Do Not Accurately Forecast the Economic Benefits of Experiential Purchases,” The Journal of Positive Psychology , 9 4, 322– 34. Google Scholar CrossRef Search ADS   Picchi ( 2015), “America’s Skyrocketing Credit Card Debt,” CBS MoneyWatch, http://www.cbsnews.com/news/americas-skyrocketing-credit-card-debt/. Prelec Drazen, Loewenstein George ( 1998), “The Red and the Black: Mental Accounting of Savings and Debt,” Marketing Science , 17 1, 4– 28. Google Scholar CrossRef Search ADS   Rosenzweig Emily, Gilovich Thomas ( 2012), “Buyer’s Remorse or Missed Opportunity? Differential Regrets for Material and Experiential Purchases,” Journal of Personality and Social Psychology , 102 2, 215– 23. Google Scholar CrossRef Search ADS PubMed  Soman Dilip, Cheema Amar ( 2002), “The Effect of Credit on Spending Decisions: The Role of the Credit Limit and Credibility,” Marketing Science , 21 1, 32– 53. Google Scholar CrossRef Search ADS   Tully Stephanie M., Hershfield Hal E., Meyvis Tom ( 2015), “Seeking Lasting Enjoyment with Limited Money: Financial Constraints Increase Preference for Material Goods over Experiences,” Journal of Consumer Research , 42 1, 59– 75. Google Scholar CrossRef Search ADS   Van Boven Leaf ( 2005), “Experientialism, Materialism, and the Pursuit of Happiness,” Review of General Psychology , 9 2, 132– 42. Google Scholar CrossRef Search ADS   Van Boven Leaf, Gilovich Thomas ( 2003), “To Do or to Have? That Is the Question,” Journal of Personality and Social Psychology , 85 6, 1193– 202. Google Scholar CrossRef Search ADS PubMed  Verhallen Theo M. M., Robben Henry S. J. ( 1994), “Scarcity and Preference: An Experiment on Unavailability and Product Evaluation,” Journal of Economic Psychology , 15 2, 315– 31. Google Scholar CrossRef Search ADS   Zhu Lillian Y., Meeks Carol B. ( 1994), “Effects of Low Income Families’ Ability and Willingness to Use Consumer Credit on Subsequent Outstanding Credit Balances,” Journal of Consumer Affairs , 28 2, 403– 22. Google Scholar CrossRef Search ADS   © The Author 2017. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Consumer Research Oxford University Press

Context-Dependent Drivers of Discretionary Debt Decisions: Explaining Willingness to Borrow for Experiential Purchases

Loading next page...
 
/lp/ou_press/context-dependent-drivers-of-discretionary-debt-decisions-explaining-7E7rXtM0ct
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0093-5301
eISSN
1537-5277
D.O.I.
10.1093/jcr/ucx078
Publisher site
See Article on Publisher Site

Abstract

Mental accounting research suggests that consumers prefer borrowing for longer-lasting purchases in order to receive benefits from the purchases as they pay for them. In contrast, two sets of archival data and five lab studies show that consumers are more willing to borrow for experiential versus material purchases, even though experiential purchases tend to have a shorter physical duration. Further, framing the same purchase as more experiential than material increases willingness to borrow. This effect occurs because purchase timing is more important for experiential purchases—a function of consumers’ aversion to missing out on planned consumption. Thus, we moderate the proposed effect by varying whether the borrowing decision impacts planned consumption. Other differences between material and experiential purchases, such as scarcity or expected happiness, cannot similarly explain our results. Moreover, our conceptualization allows us to reconcile the apparent contradiction between the previous and current research by examining the relative impact of purchase-timing importance and payment-benefit duration matching in different contexts (i.e., “purchasing” and “source-of-funding” decisions). experiential purchases, material purchases, consumer borrowing, debt, discretionary purchases, financial decision making, willingness to borrow, mental accounting The average American household spends $12,800 annually on discretionary purchases (Experian Simmons, 2011) and has $7,200 in credit card debt (Picchi 2015). In light of these figures, the current research examines borrowing decisions for discretionary purchases. In particular, we focus on investigating when, why, and for what types of purchases consumers will be more willing to borrow. Mental accounting research suggests that a purchase’s longevity (e.g., how long the purchase will last over time) predicts borrowing preferences (Hirst, Joyce, and Schadewald 1994). People prefer borrowing for purchases that last longer in time because they continue receiving benefits from the purchase over the period of time they pay for it. One classification of purchases that tends to vary systematically by physical longevity is that of experiential and material purchases (Tully, Hershfield, and Meyvis 2015; Van Boven and Gilovich 2003). Since experiential purchases are typically shorter-lived than material purchases, the previous research suggests that people should prefer borrowing for material rather than experiential purchases. In contrast, our research argues that people are typically more willing to borrow for more experiential purchases. Most consumers borrow when they want to make a purchase but do not have adequate funds. Thus, borrowing decisions determine whether or not a purchase can be made. Accordingly, decisions to borrow should be influenced by how important consumers feel it is to make the purchase at the time at which they had planned. We argue that it seems more important to make an experiential purchase when planned, as compared to a material purchase, because consumers are averse to missing out on planned consumption. Experiential (vs. material) purchases, by definition, are made for the primary intention of consumption (vs. retained ownership). Thus, not making an experiential purchase when planned is more likely to impede planned consumption. Hence, consumers are more likely to stick to their plans to buy experiential (vs. material) purchases, even when doing so requires bypassing liquidity constraints and using borrowed funds. Our results suggest that, for most borrowing decisions, purchase-timing importance is a stronger driver than is payment-benefit duration matching, which has been found to drive borrowing decisions in previous research (Hirst et al. 1994; Prelec and Loewenstein 1998). We reconcile our results with those seemingly contradictory results by considering differences across borrowing contexts. We suggest and find that consumers typically borrow when other payment resources are unavailable. However, consumers may at times choose to borrow even when other payment resources are available (e.g., due to attractive financing terms). In the latter borrowing contexts, the borrowing decision determines not whether, but rather how, the purchase is made. Since the purchase will be made regardless of whether one uses borrowed funds, planned consumption is unaffected and thus purchase-timing importance is less likely to affect decisions. Instead, in these contexts, payment-benefit duration matching becomes a more focal consideration and results in greater willingness to borrow for material purchases. We present seven studies that investigate the relationship between material and experiential purchases and willingness to borrow. Archival data and controlled lab studies demonstrate our predicted effects across a range of purchases. Further, we reconcile the current and previous research to provide a greater understanding of when and why purchase-timing importance drives borrowing decisions. THEORETICAL BACKGROUND Consumers’ willingness to borrow can stem from various sources, and we group these sources into three general categories: (1) consumer characteristics, (2) the terms of the loan, and (3) the underlying purchase. In terms of consumer characteristics, research has examined various individual differences and demographic variables that affect willingness to borrow (e.g., age, income, attitude toward credit; Kim and DeVaney 2001; Zhu and Meeks 1994). With respect to loan terms, research has examined how factors such as interest rates, credit limits, and repayment options influence borrowing (Gross and Souleles 2002; Kim and DeVaney 2001; Soman and Cheema 2002). Less research has examined how characteristics of the underlying purchase influence borrowing. We address this gap in the current research. Characteristics of Purchases that Influence Borrowing Seminal work on how purchase characteristics influence borrowing preferences argues that the physical longevity of purchases should be the prominent driver of debt decisions (Hirst et al. 1994). The authors found that when people purchase furniture and a vacation simultaneously, and have the option to pay for one with savings and the other with a loan, they overwhelmingly prefer to use the loan for the furniture. The psychological mechanism driving this effect is that people generally prefer to match the duration of their payments with the duration of the benefits they receive (Auh, Shih, and Yoon 2008; Prelec and Loewenstein 1998). This line of reasoning has implications for one classification of purchases that has received increased attention over the last decade—material versus experiential purchases. Experiential purchases are purchases made for the primary intention of attaining a life experience—living through an event or series of events. Material purchases are made for the primary intention of ownership and possession—getting something that can be retained over time (Van Boven and Gilovich 2003). The distinction between experiential and material purchases is readily recognized by consumers and has substantial consequences for consumer well-being (Carter and Gilovich 2010; Rosenzweig and Gilovich 2012; Van Boven 2005; Van Boven and Gilovich 2003). Importantly, material purchases tend to be longer lasting than experiential purchases (Tully et al. 2015). Given this difference in longevity, previous research on payment-benefit duration matching suggests that people should be more willing to borrow for material rather than experiential purchases. Related research provides additional reasons why people may be more willing to borrow for material versus experiential purchases. For example, consumers predict that material goods provide greater economic value (Pchelin and Howell 2014). Indeed, incurring debt for material goods might seem logical given that material purchases may serve as assets that can be resold if needed (e.g., to repay a debt). While material purchases typically last longer than experiential purchases, we suggest that this difference in longevity may influence borrowing in a different manner than is currently assumed. One way consumers may react to purchase longevity, as suggested by the previous research, is by considering the long-lasting benefits of material purchases (Hirst et al. 1994; Tully et al. 2015). Thinking about the long-lasting benefits of material purchases may prompt considerations regarding payment-benefit duration matching and, in turn, increase willingness to borrow for material purchases. However, another way consumers may react to purchase longevity—which we suggest is more common—is by considering the ephemeral nature of the intended benefits of experiential purchases. Imagine a consumer who has planned to go on a snorkeling tour this weekend. If he does not have the money available, he can either use borrowed funds or forgo making the purchase. If the consumer does not make the purchase this weekend, he could save money to purchase a snorkeling tour in the future. However, to the extent that the consumer has conceptualized the snorkeling tour as the specific event taking place this Saturday (i.e., the planned consumption), the later tour may feel like an entirely different purchase. Thus, the feeling that this exact purchase cannot readily be obtained later may compel the consumer to use debt to fund the purchase to avoid missing out on his planned consumption. In contrast, imagine a consumer who has planned to get snorkeling equipment this weekend but does not have the money available. To the extent that the snorkeling equipment is purchased to be owned and retained over time (material purchase) rather than to gain an experience (experiential purchase), the purchase timing of the snorkeling gear may seem less important. The consumer may therefore be more likely to forgo making the purchase until the snorkeling equipment can be purchased with cash on hand. The above example illustrates why purchase timing is likely to feel more important for experiential purchases. If one comes to regret not buying an experiential purchase, it may seem difficult (or impossible) to reverse that decision and get that originally intended consumption occasion, leading to feelings of inaction regret. In contrast, if one comes to regret not buying a material purchase, it may seem easier to reverse the decision and establish ownership. Previous research alludes to the idea that forgoing a purchase in the present may be more consequential for experiential than for material purchases (Carter and Gilovich 2010; Rosenzweig and Gilovich 2012). For instance, Carter and Gilovich (2010) demonstrate that people view experiential purchases as less comparable to one another relative to material purchases. While this research examined the comparability of purchase options considered simultaneously, the greater difficulty of comparing experiences may extend to intertemporal but otherwise identical alternatives (e.g., comparability of a snorkeling tour at one time vs. another time). To summarize, we propose that making the purchase when planned is more important for experiential (vs. material) purchases, which results in greater willingness to borrow for experiential purchases. More formally, we propose: H1: Consumers are more willing to borrow for experiential (vs. material) purchases when debt is necessary for purchasing. H2: The effect in hypothesis 1 is mediated by purchase-timing importance, which is greater for experiential purchases. A large body of research suggests that experiential purchases result in more long-term happiness (Van Boven and Gilovich 2003). Thus, an alternative explanation for greater purchase-timing importance for experiential purchases is that people perceive these purchases to be better and more important for their well-being. We distinguish between this alternative account and the proposed explanation by identifying a moderating factor: the impact of the borrowing decision on planned consumption. If purchase-timing importance is indeed greater for experiential purchases because consumers are more averse to missing out on planned consumption, then differences in willingness to borrow should be moderated by changes to the relationship between the borrowing decision and planned consumption. More formally, we propose the following: H3: Hypothesis 1 will be moderated by whether consumers believe the decision to borrow impedes planned consumption. Although consumption is typically planned for experiential purchases (consumers know when the snorkeling tour will occur when they buy tickets for it), when people think about an experience more generally (“I want to go snorkeling sometime”), differences in willingness to borrow should be attenuated. Correspondingly, although consumption is not typically planned for material purchases, when it has been planned (“I will use the snorkeling gear this weekend”), purchase timing should feel more important, and differences in willingness to borrow should be attenuated. More formally: H3a: Differences in willingness to borrow will be attenuated when either consumption has not yet been planned for experiential purchases or when consumption has been planned for material purchases. Reconciling the Current and Previous Research Hypothesis 3 also helps to reconcile our proposed effects with those found in the previous research on payment-benefit duration matching (Hirst et al. 1994; Prelec and Loewenstein 1998). To reconcile the current predictions with the predictions in the prior work, we identify two distinct borrowing contexts that differentially affect planned consumption to understand when each factor is more likely to influence borrowing decisions. People may borrow because they do not have the resources to make the purchase otherwise or because they prefer using borrowed funds rather than other available resources (e.g., due to attractive financing). Most consumers borrow when they do not have resources to make the purchase outright.1 Thus, reluctance to borrow results in forgoing the purchase in the present, and it determines whether the purchase is made. The strongest predictor of borrowing decisions should therefore be purchase-timing importance, as it directly relates to the time when a purchase is made. Other factors, such as payment-benefit duration matching, should be secondary considerations. However, when these conditions are not met—such as when a purchase can be funded with debt or another source—the impact of purchase-timing importance on willingness to borrow should decrease. Indeed, the previous research used a joint purchasing context and found that consumers who assume they are buying two purchases (e.g., furniture and a vacation) prefer financing the longer-lasting purchase, and paying outright for the more ephemeral purchase (Hirst et al. 1994). The decision was thus a source-of-funding decision that determined not whether but how the purchase was made. Since purchase timing is unaffected in these situations, people may revert to other decision-making strategies such as matching the purchase’s payments with its utility stream. We would then expect greater willingness to borrow for material rather than experiential purchases. We summarize this proposed reconciliation in hypothesis 3b. H3b: Willingness to borrow will depend on the borrowing context. When consumers face a tradeoff between using debt and forgoing their purchase in the present (i.e., purchasing decisions), they will be more willing to borrow for purchases that are more experiential (vs. material). When consumers face a tradeoff between using debt and an alternative funding source (i.e., source-of-funding decisions), they will be more willing to borrow for purchases that are more material (vs. experiential). Next, we report the results of two archival datasets and five lab studies that demonstrate our proposed effects. Analyses of archival data demonstrate that consumers who buy more experiential purchases have higher credit card balances and pay more financing fees. In a second archival dataset, we find that there are more funded peer-to-peer loans for discretionary experiential versus material purchases. In a more controlled lab setting, we demonstrate greater willingness to borrow for experiential purchases and provide evidence that it is driven by greater purchase-timing importance for experiences. We rule out alternative explanations for this effect by holding constant the focal purchase and showing that manipulating the same purchase to seem more experiential increases willingness to borrow, via increased purchase-timing importance. We further isolate our proposed explanation by demonstrating that when consumption for experiential purchases has not been planned, or when consumption for material purchases has been planned, people are no more likely to borrow for experiential than for material purchases. Finally, we reconcile the results of the current and prior research and demonstrate when to expect payment-benefit duration matching, as compared to purchase-timing importance, to exert a greater influence on borrowing decisions. STUDY 1: ARCHIVAL ANALYSIS OF CREDIT CARD SPENDING As an initial exploration, we studied the relationship between consumer debt and spending on discretionary experiences in archival consumer expenditure data. Specifically, we examined data from the Consumer Expenditure Survey available from the United States Bureau of Labor Statistics (BLS). The Consumer Expenditure Survey provides expenditure data at the consumer unit level, elicited through an interview survey. These data have been used in other research examining discretionary spending on material goods and experiences (Tully et al. 2015). In these data, entertainment purchases are split into three categories: fees and admissions (akin to experiential purchases); TV, radio, and sound equipment (akin to material purchases); and an “other” category for miscellaneous purchases. As in the previous work, we calculated the relative spending on experiential purchases as compared to material purchases using the difference in spending between these categories, as a function of total consumer spending (Tully et al. 2015). That is, we subtracted consumers’ spending in the material category from their spending in the experiential category, as a proportion of their total expenditures. Thus, greater values on this measure reflected greater spending on discretionary experiential purchases relative to discretionary material purchases. We examined the relationship between this measure and consumer debt using two measures of consumers’ debt: (1) the total amount consumers owed on all of their credit cards, and (2) the total amount they paid in financing, late charges, and interest for all their credit cards in the last month. Because the BLS began reporting these credit measures in the second quarter of 2013, we collected and compiled the available quarterly data from the second quarter of 2013 through the fourth quarter of 2014. In support of our hypothesis that consumers are more willing to borrow for experiential versus material purchases, there was a significant positive relationship between relative spending on discretionary experiences and the balance consumers carried on their credit cards (β = 0.03, t(30240) = 5.35, p < .001), as well as the financing interest they paid in the last month (β = 0.03, t(30240) = 4.68, p < .001). The results remain unchanged when we controlled for income (βcredit balance = 0.01, t(30239) = 2.29, p = .022; βfinancing paid = 0.02, t(30239) = 3.64, p < .001). As an additional robustness check, we restricted the dataset to include only the subset of participants who reported having debt and who reported having spent money in at least one of the discretionary spending categories. Among participants reporting a balance on their credit card and who spent money in at least one of the discretionary spending categories, greater relative spending on discretionary experiences significantly and positively predicted higher credit card balances (β = 0.07, t(1833) = 2.80, p = .005). Similarly, among participants who reported paying financing fees and who spent money in at least one of the discretionary spending categories, greater relative spending on discretionary experiences significantly and positively predicted higher financing payments (β = 0.11, t(778) = 3.06, p = .002).2 In sum, using archival data of actual consumer expenditures, we found initial support for the link between discretionary spending on experiences and consumer debt. STUDY 2: ARCHIVAL ANALYSIS OF PEER-TO-PEER LOANS As a secondary examination of the relationship between experiential purchases and willingness to borrow, we sought archival data that could more directly connect the incurrence of debt to the purchase it is used to fund. To this end, we collected real funded loan data from a large, US peer-to-peer lending company that included a description of the purchase for which the loan would be used. In a peer-to-peer lending context, consumers may request and obtain funding for multiple purposes, including both discretionary material and discretionary experiential purchases. Accordingly, peer-to-peer loans, like credit cards, are an ideal debt type for the current research investigation. The dataset included 42,535 loans funded between 2007 and 2011. Each entry included a description of the loan’s intended use, written by the individuals who had requested the loans. A research assistant who was blind to our hypotheses used these descriptions to identify discretionary purchases. The complete dataset of discretionary purchases included 2,571 loans, which represents approximately 6% of the full dataset. To categorize the purchases, we created 69 broad categories to reflect the purchases and asked 30 MTurk workers to rate the categories based on the extent to which they seemed primarily material or primarily experiential (–3 = primarily material, 0 = equally material and experiential, 3 = primarily experiential). In line with predictions, a t-test revealed a positive and significant difference against zero, indicating greater borrowing for more experiential purchases, t(2,570) = 12.60, p < .001. However, the above result could also occur if experiential purchases are rated more extremely than material purchases and are thus being overweighted. Accordingly, we conducted another analysis that allowed us to examine greater instances of experiential versus material loans. We classified each of the 69 categories as material or experiential using a t-test against the midpoint of the scale. The 2,571 discretionary loans were then coded as material (–1) or experiential (+1) based on the category that best fit the description of the loan’s purpose. Out of the 2,571 loans, 158 loans fell into categories that were rated as ambiguous (not different from the scale’s midpoint) and were excluded, leaving a final sample of 2,413 loans (full details on the categories and exclusion criteria for this study are available in the web appendix). To test whether consumers were more willing to borrow for experiential purchases, we conducted a t-test against zero. Replicating the results of the continuous measure, there was a positive and significant difference from zero, t(2412) = 3.04, p = .002, suggesting that consumers took on more peer-to-peer loans for discretionary experiential purchases (53.1%) than for discretionary material purchases (46.9%). Moreover, there were no differences in average loan amount (F < 1), but the interest rates for the loans of more experiential purchases were 1.15% higher than those for more material purchases, F(1, 2411) = 60.81, p < .001. The results of the peer-to-peer loan data were consistent with the results found in the BLS data, providing additional real-world evidence of the proposed phenomenon. Using real funded peer-to-peer loan data, these data explicitly connect the incurrence of debt to its intended purchase and reveal that people took on more peer-to-peer loans to fund the purchase of experiential versus material discretionary purchases. Transitioning from the archival data, we next examined whether consumers’ interest in and willingness to use debt varied as a function of their latent perceptions of their purchases as more material or more experiential. STUDY 3: CORRELATIONAL PEER-TO-PEER LENDING Study 3 was designed to examine naturally occurring relationships between how experiential a purchase seems and consumers’ willingness to borrow. We examined this relationship in the peer-to-peer lending context. We expected participants to be more willing to borrow for purchases that they rated as more experiential (hypothesis 1). Method Five hundred and one MTurk workers (Mage = 33.84, SD = 11.40; 53.9% female) completed the study in exchange for monetary compensation. No subjects were removed from the analysis in this study or in any of the following studies. We designed the stimuli for this study based on a real peer-to-peer lending website that depicted six categories of purchases: debt consolidation, home improvement, auto and vehicle, baby and adoption, small business, and special occasion. Of these categories, special occasion best aligned with discretionary purchases. Thus, aiming for external validity, we asked participants to list and describe one special occasion purchase they were considering buying that cost at least $500. Next, we provided a brief description of peer-to-peer lending and asked participants to indicate whether they had heard of such lending before (yes/no). We then asked participants about two real action steps they might take to fund their purchase with peer-to-peer loans. The first measure read: “How interested would you be in finding out about competitive rates and opportunities to fund your purchase through peer-to-peer lending?” (seven-point scale; 1 = not at all interested, 4 = neutral, 7 = very interested). The second measure read: “How likely are you to visit a peer-to-peer lending website to learn about different funding options for that special purchase today?” (seven-point scale; 1 = not at all likely, 4 = neutral, 7 = very likely). Participants then reviewed a description of material and experiential purchases and were asked to classify their purchase using a seven-point scale (1 = completely material, 7 = completely experiential). Next, participants indicated the month in which they intended to make the purchase and indicated how much the purchase would cost. They also responded to demographic questions: age, gender, and household income. Finally, we asked participants whether they knew the study’s intended purpose. No participant indicated suspicion about the study’s objective nor did any participant guess our research questions or hypotheses. Results and Discussion We first examined the relationship between participants’ ratings of their future purchases and their interest in learning about peer-to-peer lending options for their purchase. As predicted, we found a significant positive correlation, r = .15, p = .001, indicating that the more experiential participants rated their purchase, the more interested they were in learning about peer-to-peer lending rates. These results replicated when we examined the correlation between participants’ purchase ratings and their likelihood of visiting a peer-to-peer lending site on that day, r = .16, p < .001. These results remained unchanged when we adjusted for factors such as cost, age, gender, and income. In sum, this study provided convergent evidence with the peer-to-peer lending archival data, demonstrating a relationship between how experiential a purchase seems and interest in using peer-to-peer lending options to fund the purchase. We conducted another study examining willingness to borrow for people’s planned purchases that manipulated rather than measured the experiential nature of the purchase. In this study, we asked participants to consider a material or experiential purchase they planned to make and measured willingness to finance this purchase with a credit card. Replicating the correlational results in study 1, participants asked to consider an upcoming experiential (vs. material) purchase were more willing to finance the purchase. Further, this study measured a number of potential process measures, finding support for the proposed purchase-timing importance explanation. Other factors, such as the happiness expected from the purchase, could not similarly explain the effect. See supplemental study 1 in appendix A of the web appendix for the complete study and results. STUDY 4: MANIPULATING THE EXPERIENTIAL NATURE OF THE PURCHASE Thus far, we have demonstrated the predicted relationship between experiential purchases and willingness to borrow using archival consumer data and consumers’ self-generated planned purchases. These findings provide external validity, demonstrate the generalizability of the proposed effect across various real purchases, and indicate that the results are not due to specific experimental stimuli. Study 4 was designed to isolate the causal role of purchase type (material vs. experiential) on willingness to borrow by examining whether framing the same purchase as more experiential would increase willingness to borrow. Method We recruited 407 adults (Mage = 33.68, SD = 11.31; 52% male) on MTurk to complete the survey in exchange for monetary payment. The study followed a between-subjects design that manipulated purchase framing—whether participants were led to think about an outdoor grill as more material or more experiential. Participants first indicated whether they owned a grill. Next, they were informed of research showing that people may think about purchases along a continuum ranging from mainly material to mainly experiential. We then introduced the purchase framing manipulation (see appendix B of the web appendix for exact stimuli). Depending on condition, participants were told that many people think about grills as being more experiential (material) and were asked to explain why people might consider outdoor grills to be more experiential (material). Since marketing pitches commonly ask people to imagine what their life would be like with a product, we constructed a marketing pitch for an outdoor grill that included a statement about grill availability that summer, followed by a mental simulation. Specifically, the pitch began: “This summer there are a number of attractive grills available for purchase.” Then, based on condition, participants were asked to imagine having an experience with (experience condition) or owning (material condition) a new outdoor grill and to write about aspects related to those thoughts. After this marketing pitch, we measured purchase-timing importance and three measures related to willingness to borrow. We measured purchase-timing importance using two questions about the extent to which purchase timing was (1) important and (2) relevant for the purchase of the outdoor grill (both questions: 1 = not at all, 7 = very). The primary dependent measure was participants’ interest in using real, currently available financing options to acquire a new outdoor grill. We informed participants that many hardware stores were providing financing options that allow consumers to pay for their purchases over time. We then asked participants to indicate their interest in using such a payment option. Specifically, based on condition, they indicated their interest in using financing that would allow them to have the experience with (experiential condition) or own (material condition) a new outdoor grill (1 = not at all, 7 = extremely). Subsequently, we administered an exploratory behavioral measure of willingness to borrow: participant website visits. Before conducting this study, we compiled a list of popular grill vendors (e.g., Best Buy, Home Depot, Sears) and the current financing offers available from those vendors. We created two Google websites (one for each experimental condition) designed to host that information. The web layout and content was exactly the same on each website. As a real behavioral measure of interest in financing a new outdoor grill, we asked participants whether they would like access to the site at the end of the study (yes/no). Additionally, using Google Analytics, we tracked how many site views each site received. Before completing demographic information, participants completed a manipulation check, rating the extent to which they thought about a grill as more material or more experiential using a seven-point scale (1 = completely material, 7 = completely experiential). Results and Discussion Manipulation Check Participants in the experiential condition (M = 5.09, SD = 1.42) rated an outdoor grill as more experiential in nature than did participants in the material condition (M = 3.03, SD = 1.66), F(1, 405) = 181.49, p < .001, ηp2 = .309. Interest in Borrowing In line with our predictions, there was a significant effect of purchase type on interest in using financing available at hardware stores. Asking participants to think about a grill as an experiential purchase increased their interest in using financing options relative to asking them to think about a grill as a material purchase (Mexperiential = 3.93, SD = 2.13, Mmaterial = 3.44, SD = 2.10), F(1, 405) = 5.36, p = .021, ηp2 = .013. Exploratory Behavioral Measure We next examined the effect of purchase framing on participants’ request to view the financing website. The binary measure of interest in viewing the website was directionally consistent but not significant: 47% of participants in the experiential condition requested access to the site compared to 44% of participants in the material condition, χ2 < 1. There was, however, correlational evidence in support of our hypotheses. There was a significant positive correlation between participants’ self-rated perceptions of an outdoor grill being more experiential (vs. material) and requests to view the website, r = .142, p = .004. Results for actual site visits were similar to the results for the binary website request measure, with 15.5% of participants in the experiential condition visiting the site compared to 12.1% of participants in the material condition, Z = 1.00, p = .32. Purchase-Timing Importance The two measures of purchase-timing importance were combined into a single measure (r = .80, p < .001). Consistent with the results for interest in financing, participants in the experiential condition (M = 5.01, SD = 1.54) rated purchase timing as more important than did participants in the material condition (M = 3.87, SD = 1.88), F(1, 405) = 44.23, p < .001, ηp2 = .098. Mediation Next, we tested whether purchase-timing importance mediated the relationship between purchase type and interest in using hardware store financing. Hayes’s (2013) PROCESS macro with 20,000 bootstrapped samples supported the proposed mediation, as the 95% confidence interval for the indirect pathway did not include zero (.21, .58). We have argued that experiential purchases increase purchase-timing importance because consumers are averse to disruptions to planned consumption. Hence, participants are more willing to use debt when an outdoor grill is framed as more experiential because consumption may otherwise be disrupted. That is, even though an outdoor grill is a durable good that can provide many consumption occasions, participants’ willingness to borrow is greater when it is framed as more experiential because some planned consumption will be otherwise forgone. However, another possibility is that thinking about a grill as more experiential makes the grill more desirable overall (e.g., because it highlights the social value). We begin to distinguish between these two accounts by conducting additional analyses to test hypothesis 3a. Additional Process Evidence In this study, we found greater willingness to borrow for a purchase that we featured in a marketing pitch. Thus, some people likely thought about the grill more generally (“I will use the grill with friends”) as opposed to having planned consumption (“I will use the grill this weekend”). Hence, we used natural variation in participants’ spontaneous consideration of intended consumption occasions to test our prediction. Our theory posits that heightened willingness to borrow for the grill in the experiential condition is driven by participants who spontaneously considered one or more specific consumption occasions they might have with the grill. To directly test this assertion, we revisited participants’ written responses to the marketing pitch for an outdoor grill in which they described either having an experience with the grill or owning the grill. The open-ended responses were coded for whether a time (for consumption or ownership) was specified (e.g., 4th of July, Memorial Day, someone’s birthday; 1 = mentioned, 0 = not mentioned). Only one participant in the material condition specified a time, whereas almost a third of participants (29%) did so in the experiential condition. Thus, we examined willingness to borrow among three groups of participants: (1) those in the material condition, (2) those in the experiential condition who did not specify a consumption time, and (3) those in the experiential condition who specified a consumption time. Two contrasts provided evidence for our proposed explanation. Willingness to borrow was significantly greater among participants in the experiential condition who had specified a time for consumption (M = 4.80, SD = 1.92) compared to those in the material condition (M = 3.44, SD = 2.10), F(1, 404) = 19.31, p < .001, ηp2 = .046. However, there was no difference in willingness to borrow between participants in the material condition and those in the experiential condition who did not specify the time for consumption (M = 3.57, SD = 2.11), F < 1, NS, ηp2 < .001. Overall, in study 4, we induced greater willingness to borrow for the purchase of an outdoor grill by having people think about grills as more experiential rather than material. These results provide greater evidence of a causal relationship between how experiential a purchase seems and people’s willingness to borrow for it. Focusing participants on the experiential aspects of a purchase increased purchase-timing importance and, in turn, willingness to borrow for an outdoor grill. Further, our follow-up analyses provide further evidence that consumers are averse to missing out on planned, but not unplanned, consumption. STUDY 5: DECREASING THE EXTENT TO WHICH BORROWING IMPACTS PLANNED CONSUMPTION FOR EXPERIENTIAL PURCHASES The additional analyses conducted in study 4 provide evidence that the increase in willingness to borrow for experiential purchases depends on having consumption planned. However, it is possible that our effects are not due to the greater perceived impact of changes to planned consumption versus ownership, but rather the mere planning for a purchase. Thus, in study 5, we aimed to disentangle the mere specification of a particular time from the specification of consumption timing in particular. To do so, we asked participants to consider a given experiential or material purchase, and we varied whether people expected to obtain the purchase’s objectives (consumption or ownership) at a specific time or at one out of a number of potential times. We expected participants to be more willing to borrow for experiential versus material purchases when timing was planned. However, when timing was instead flexible, we expected differences in willingness to borrow to be attenuated. Method Participants Participants in the main study were 333 undergraduate students at the University of Southern California who completed the study in exchange for partial course credit (Mage = 20.1, SD = 1.55, 47.9% female). The study followed a 2 (purchase: material vs. experiential) × 2 (timing: planned vs. unplanned) between-subjects design. Procedure Participants were asked to imagine the opportunity to make a given purchase. Half of the participants imagined an experiential purchase (a day trip to Disneyland) and the other half imagined a material purchase (smartwatch). These purchases were determined based on results of a pretest that aimed to equate the purchase types in terms of their perceived desirability, value, uniqueness, and availability to our particular participant sample. See appendix B of the web appendix for complete pretest results. Next, to examine how planning the timing to realize the purchase’s objective influenced willingness to borrow for experiential and material purchases, we told half of the participants to imagine they were planning to make the purchase on the following Saturday (timing planned). We told the remaining half to imagine they were considering a few dates, with one possible date being the following Saturday (timing unplanned). See appendix B of the web appendix for a full description. Participants were then told to imagine they had checked their bank account and did not have enough money for the purchase in their account. They could either wait to save enough money to make the purchase outright, or they could finance the purchase on a credit card with 15% interest. Participants then indicated their likelihood of financing the purchase with their credit card (1 = very unlikely, 7 = very likely). After indicating their borrowing likelihood, participants were asked to categorize the focal purchase as either an experience or a material good. In addition, although we had pretested our stimuli to ensure that the experiential purchase did not seem more unique, we administered four measures assessing the perceived scarcity of the purchases, to ensure that the experiential purchase was not perceived as a scarcer opportunity once timing had been planned. Participants rated the extent to which the purchase they were asked to consider was unique, unusual, rare, and once-in-a-lifetime using a seven-point scale (1 = not at all, 7 = extremely). There were no differences on these measures, so we do not discuss them further. Results and Discussion Manipulation Check Participants’ categorization of their respective purchases varied by condition, χ2 = 190.98, p < .001. Seventy-eight percent of participants considering the material purchase categorized it as a material good, while 96% of participants considering the experiential purchase categorized it as an experience. Willingness to Borrow There was no main effect of timing on willingness to borrow. However, there was a significant main effect of purchase type, such that participants in the experiential condition (M = 3.03, SD = 1.95) were significantly more willing to borrow than were participants in the material purchase condition (M = 2.44, SD = 1.64), F(1, 329) = 8.56, p = .004, ηp2 = .025. However, the main effect of purchase type was qualified by a marginally significant purchase type by timing interaction, F(1, 329) = 2.83, p = .09, ηp2 = .009. Follow-up planned contrasts supported the hypothesized results. Participants were more willing to borrow for the experiential purchase (M = 3.27, SD = 1.96) than for the material purchase (M = 2.33, SD = 1.64) when timing was planned, F(1, 329) = 10.47, p < .001, ηp2 = .031. However, when timing was unplanned, there was no difference in willingness to borrow for the two purchase types (Mexperiential = 2.81, SD = 1.93 vs. Mmaterial = 2.56, SD = 1.81), F < 1, ηp2 = .002. These results are displayed in figure 1. FIGURE 1 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES DEPENDING ON WHETHER TIMING IS PLANNED IN STUDY 5 NOTE.—Error bars indicate standard errors of the mean. FIGURE 1 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES DEPENDING ON WHETHER TIMING IS PLANNED IN STUDY 5 NOTE.—Error bars indicate standard errors of the mean. Study 5 attenuated the greater willingness to borrow for experiential purchases by minimizing the extent to which the decision to borrow affected planned consumption. When consumption was unplanned, willingness to borrow for experiential purchases decreased and differences in willingness to borrow were attenuated. STUDY 6: INCREASING THE EXTENT TO WHICH BORROWING IMPACTS PLANNED CONSUMPTION FOR MATERIAL PURCHASES We designed study 6 to test whether strengthening the link between planned consumption and the decision to borrow could attenuate the greater willingness to borrow for experiential purchases by increasing willingness to borrow for material purchases. To do so, we varied how purchase timing was described. For some participants, purchase timing was described in a standard form: the time when one plans to make the purchase. In these situations, we expected people to infer that purchase timing affects consumption timing for experiential purchases and ownership timing for material purchases, and hence we predicted greater willingness to borrow for experiential purchases. However, for other participants, purchase timing was explicitly linked to planned consumption (i.e., one plans to make the purchase in order to use it at a certain time). In these situations, we predicted that the heightened willingness to borrow for experiential purchases would be attenuated. Method Participants were 405 adults (Mage = 34.8, SD = 12.4; 57% females) on MTurk who completed the survey in exchange for monetary payment. This study followed a 2 × 2 between-subjects design that crossed purchase type (experiential vs. material) and purchase-timing type (standard description vs. consumption-linked description). Participants in the experiential condition imagined purchasing a snorkeling tour, and participants in the material condition imagined purchasing snorkeling equipment. In the standard purchase-timing conditions, participants were simply told to imagine they were making the purchase at the end of the month. Our expectation was that participants would naturally link the experiential purchase (but not the material purchase) to consumption. However, in the consumption-linked purchase-timing conditions, we explicitly linked purchase timing to planned consumption for both purchases. Specifically, we described the experiential purchase as “a snorkeling tour for the summer vacation you are taking at the end of the month,” and the material purchase as “a set of snorkeling equipment to use on the summer vacation you are taking at the end of the month.” Next, participants rated the importance of purchase-timing for their respective purchase (1 = not at all important, 7 = very important). Then they were asked to imagine that, although they had planned to make the purchase, the purchase cost $150, and they did not have the money to pay for it outright. For the dependent measure, they indicated their willingness to finance any part of the purchase (1 = extremely unwilling, 7 = extremely willing). At the study’s end, we collected open-ended responses pertaining to whether participants knew the study’s intended purpose. No participant indicated suspicion about the study’s objective nor did any participant guess our research questions or hypotheses. Results Manipulation Check As intended, participants who read about a snorkeling tour (Mexperiential = 5.88, SD = 1.63) rated their purchase as more experiential (vs. material) than did participants who read about snorkeling equipment (Mmaterial = 4.57, SD = 1.71), F(1, 401) = 61.55, p < .001, ηp2 = .133. Willingness to Borrow There was no main effect of purchase type (F < 1, NS). There was a main effect of timing type, F(1, 401) = 11.02, p = .001, ηp2 = .027. This main effect was qualified by the expected interaction, F(1, 401) = 8.66, p = .003, ηp2 = .021. As expected, in the standard timing condition, participants were more willing to borrow for the experiential (vs. material) purchase, F(1, 401) = 6.32, p = .012, ηp2 = .016. However, in line with our expectations, when timing was specifically linked to consumption, greater willingness to borrow for the experiential purchase was attenuated. In fact, preferences reversed and willingness to borrow was actually marginally greater for material (vs. experiential) purchases, F(1, 401) = 2.75, p = .098, ηp2 = .007. These results are depicted in Figure 2. FIGURE 2 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES BY TIMING TYPE IN STUDY 6 NOTE.—Error bars indicate standard errors of the mean. FIGURE 2 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES BY TIMING TYPE IN STUDY 6 NOTE.—Error bars indicate standard errors of the mean. Purchase-Timing Importance There was a significant main effect of purchase type, F(1, 401) = 4.20, p = .04, ηp2 = .010, and timing type, F(1, 401) = 22.86, p < .001, ηp2 = .054, and these main effects were qualified by a significant purchase type by timing type interaction, F(1, 401) = 4.11, p = .04, ηp2 = .010. Follow-up planned comparisons provided support for the hypothesized pattern of results. In the standard timing conditions, participants indicated greater purchase-timing importance for the experiential (M = 4.81, SD = 1.74) versus material (M = 4.14, SD = 1.80) purchase, F(1, 401) = 8.53, p = .004, ηp2 = .021. However, the difference in purchase-timing importance for the experiential (M = 5.27, SD = 1.64) and material (M = 5.27, SD = 1.46) purchase was attenuated when timing was directly linked to consumption, F < 1. Mediation We have predicted that greater purchase-timing importance drives increased willingness to borrow for more experiential purchases because those purchases are typically linked to planned consumption. However, when material purchases are similarly linked to planned consumption, differences in purchase-timing importance should be minimized and other drivers may better predict borrowing preferences. Thus, we tested for the predicted moderated mediation. Hayes’s (2013) PROCESS macro (model 7) with 20,000 bootstrapped samples supported the proposed moderated mediation, as the 95% confidence interval for the indirect pathway did not include zero (–.35, –.01). In line with predictions, in the standard timing condition, purchase-timing importance explained differences in willingness to borrow (.03, .30). However, in the consumption-linked timing condition, purchase-timing importance did not similarly explain willingness to borrow (–.09, .08). It is important to note that, in the consumption-linked timing condition, purchase-timing importance did not differ between purchase types. Therefore, the observed reversal in willingness to borrow suggests that when purchase timing is no longer a critical difference, other factors (such as payment-benefit duration matching) may become more important. STUDY 7: PURCHASING VERSUS SOURCE-OF-FUNDING DECISIONS We argue that our results differ from those found in the previous research on payment-benefit duration matching because in the current context, borrowing decisions imply purchasing decisions (i.e., whether a purchase is made) as opposed to source-of-funding decisions (i.e., how a purchase is made), which have no impact on consumers’ receipt of the purchase. Study 7 was designed to directly test the predicted differences between these two borrowing contexts. The borrowing decision was framed as a purchasing decision for some participants and a source-of-funding decision for other participants. We predicted that when we framed borrowing as a source-of-funding decision, payment-benefit duration matching would become a more prominent driver and lead to a greater willingness to borrow for material versus experiential purchases. Study 7 included three experimental conditions. In addition to having participants consider their willingness to borrow for either a material or an experiential purchase (purchasing decisions), participants in a third condition imagined both purchases, with only enough cash to pay for one outright (source-of-funding decision). We expected to replicate our earlier results when we framed the borrowing decision as a purchasing decision (either financing the purchase or forgoing the purchase in the present). However, when we framed the decision as a source-of-funding decision (deciding which purchase to finance and which to pay for with cash), we expected participants to be more willing to use debt for the material goods, replicating the findings in the existing literature on payment-benefit duration matching. Method We recruited 756 adults (Mage = 32.1, SD = 10.6; 38.6% females) on MTurk to complete this study in exchange for monetary payment. We asked participants to imagine they had been browsing for purchases they wanted to acquire. Participants in the purchasing decision contexts were told they had found a purchase they really wanted to buy: either a new stereo system (material condition) or two tickets to an upcoming music festival (experiential condition). They were told that the purchase cost $400 and would need to be financed if purchased. In the source-of-funding decision context, participants imagined wanting to buy both the stereo system and the music festival tickets. They were informed that both purchases cost $400. They were then told that they had $400 in savings and would need to finance the other $400 purchase. All participants were told to imagine that they expected to repay the borrowed amount in a couple of months. See appendix B of the web appendix for detailed stimuli. Next, participants indicated their willingness to borrow for the experiential purchase (experiential purchasing condition), material purchase (material purchasing condition), or both (source-of-funding condition) using seven-point scales (1 = not at all willing to borrow for [purchase], 7 = very willing to borrow for [purchase]). In addition, to replicate previous findings (Hirst et al. 1994), participants in the source-of-funding condition made a forced choice, indicating which purchase they would rather use borrowed funds to acquire. Next, all participants indicated the extent to which their financing decision involved considerations about “whether the length of the benefits of the purchase would extend beyond when they finished making payments on the purchase” (seven-point scale: 1 = not at all, 7 = very much) before completing demographic questions. Results and Discussion Willingness to Borrow Consistent with our earlier studies, in the purchasing decision contexts, participants were more willing to borrow for the experiential purchase (M = 3.41, SD = 2.12) than for the material purchase (M = 2.93, SD = 1.90), F(1, 503) = 7.35, p = .007, ηp2 = .014. However, as predicted, the results in the source-of-funding context were consistent with the previous literature. Participants were less willing to borrow for the experiential purchase (M = 2.41, SD = 1.68) than for the material purchase (M = 4.76, SD = 1.97), t(244) = 12.07, p < .001, ηp2 = .374. Further, when forced to decide which purchase they were more likely to finance, 79.1% of participants indicated greater likelihood to borrow for the material purchase. This percentage was significantly greater than 50%, suggesting the effects were not due to mere chance, t(240) = 11.28, p < .001, ηp2 = .346. Additionally, willingness to borrow for the material purchase was significantly lower when borrowing was framed as a purchasing decision than as a source-of-funding decision, F(1, 497) = 112.71, p < .001, ηp2 = .185, whereas willingness to borrow for the experiential purchase was significantly higher when framed as a purchasing decision than as a source-of-funding decision, F(1, 497) = 31.40, p < .001, ηp2 = .059. (See figure 3.) FIGURE 3 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES BY CONDITION IN STUDY 7 NOTE.—Error bars indicate standard errors of the mean. FIGURE 3 View largeDownload slide WILLINGNESS TO BORROW FOR EXPERIENTIAL AND MATERIAL PURCHASES BY CONDITION IN STUDY 7 NOTE.—Error bars indicate standard errors of the mean. Payment-Benefit Duration Matching As predicted, a planned contrast demonstrated that participants were significantly more likely to consider payment-benefit duration matching when framed as a source-of-funding decision than a purchasing decision for a material or experiential purchase, F(1, 753) = 23.93, p < .001, ηp2 = .031. Thus, we find support for hypothesis 3b. When the decision was framed as a purchasing decision, participants were more willing to borrow for the experiential purchase than for the material purchase. However, when we framed the decision as a source-of-funding decision, we replicated the results of previous research on payment-benefit duration matching and found greater willingness to borrow for the material (and longer-lasting) purchase. These results cannot be explained by differences in participants’ ability to compare and contrast material and experiential purchases in the source-of-funding decision context. For instance, a material (experiential) purchase might highlight the long-lasting (short-lived) nature when presented alongside an experiential (material) purchase. It is also possible that the mere presence of a reference option cued participants to consider borrowing norms or increased the evaluability of the purchase’s longevity. We addressed these possibilities in an additional study that replicates and extends study 7. See supplemental study 2 in appendix A of the web appendix for the complete study. GENERAL DISCUSSION Academic researchers, practitioners, and consumers are increasingly discussing the significance of the experiential-material continuum, highlighting how the purchases affect consumer psychology and well-being (Van Boven and Gilovich 2003). The current work identifies a new consequence of the material-experiential continuum. Using two archival datasets and five lab studies, we show that consumers are more willing to borrow for experiential than for material purchases. This effect emerges across a range of experiential and material purchases—individuals' past and planned purchases, pretested standardized pairs, as well as the same purchase reframed to be more experiential—providing evidence that the effects are not merely an artifact of the chosen stimuli. In addition, we demonstrate that these effects are explained by the greater purchase-timing importance of experiences. We further show that the greater purchase timing for experiences is driven by consumers’ aversion to missing out on planned consumption. Thus, these effects are attenuated when the link between the borrowing decision and planned consumption is weakened for experiential purchases (study 5) and when this link is strengthened for material purchases (study 6). We also considered other plausible factors that might have contributed to these effects. Although prior research has shown that people report greater long-term happiness from experiences rather than material goods, our results cannot be explained by the mere greater desirability of the underlying purchase (e.g., study 4, study 5, and supplemental study 1 in the web appendix). Further, it is important to note that consumers’ greater willingness to borrow for experiential purchases does not imply greater impatience for consuming experiences. That is, delay discounting cannot explain our results. Indeed, prior research by Kumar and colleagues (2016) suggests that, all else equal, people prefer to consume material goods before experiences due to the greater anticipatory pleasure of experiences. Instead, our effects are driven by the importance of making the purchase when planned, which is a function of consumers’ aversion to missing out on planned consumption. The current findings seem to contradict research showing that debt decisions are based on preferences to match a purchase’s duration of payments with its duration of benefits (Hirst et al. 1994; Prelec and Loewenstein 1998). We reconcile this apparent contradiction in study 7. We show that consumers prefer to borrow for experiential versus material purchases when debt decisions are synonymous with purchasing decisions—when the decision determines whether the purchase is made in the present. However, when debt decisions are source-of-funding decisions—when consumers have multiple sources of funds and must decide how to fund a purchase—they prefer borrowing for material versus experiential purchases. Thus, our theoretical framework offers greater understanding of when purchase-timing importance versus payment-benefit duration matching will more strongly influence borrowing decisions, and hence when consumers will be more willing to borrow for experiential or material purchases. Moreover, we contribute to existing research by differentiating these two types of debt decisions (purchasing decisions vs. source-of-funding decisions). Future research may identify other important differences between these different debt decision contexts. In the current work, we demonstrate a causal relationship between experiential purchases and purchase-timing importance. Holding constant the underlying purchase, we demonstrate that framing a purchase as more experiential increases its purchase-timing importance. However, other factors may also contribute to purchase-timing importance, and our findings should help explain when and why those factors may increase willingness to borrow. For example, consumers who consider themselves innovators may care whether they are among the first to own a product. In such instances, planned ownership timing may be as important as planned consumption timing, and purchase-timing importance may increase. Additionally, the scarcity literature offers real-world contexts in which purchase-timing importance may increase for non-consumption-related reasons (Brock 1968; Lynn 1991; Verhallen and Robben 1994). For example, consider cases when the supply of goods outweighs demand (e.g., popular items, restricted production), is limited to a given time window (e.g., limited editions), or is limited to specific geographic locations (e.g., a souvenir from one’s travels). Consistent with prior work showing that restricted purchasing opportunities evoke purchasing urgency (Abendroth and Diehl 2006; Inman and McAlister 1994), the examples discussed above illustrate situations when differences in willingness to borrow for experiential versus material purchases should be attenuated. Future research might further examine such boundary conditions. The current research demonstrates consistent results across a range of debt types including peer-to-peer lending, store financing, and credit card borrowing. However, one should note that some borrowing forms may be more typical for material goods (e.g., personal loans, layaway plans) because those purchases can be used to securitize loans. Future research could examine whether norms for different debt types affect willingness to finance some types of purchases more than others. This work offers a number of managerial insights. The results of study 7 suggest that if a consumer has decided to make the purchase of a material good, they may be more easily persuaded to use an attractive financing offer (vs. cash). Yet the results from our full package of studies suggest that opportunities to finance experiential purchases may be more enticing than otherwise thought. When consumers have no source of funding other than debt, they are more willing to finance experiential versus material purchases. Thus, providing more financing options for experiential purchases may benefit firms as well as customers. In addition, our work suggests ways in which firms may encourage consumers to use financing offers instead of delaying a purchase. For example, focusing consumers’ attention on a particular opportunity for consumption should enhance purchase-timing importance and hence willingness to borrow. Conducting the present investigation in the domain of consumer borrowing is critical to this work given the opposing predictions suggested by the prior borrowing research. A reasonable question that may arise is whether the effects demonstrated in the current work apply more broadly to spending decisions. To this end, we conducted an additional study, finding greater price elasticity for material (vs. experiential) purchases, driven by differences in purchase-timing importance (see supplemental study 3 in appendix A of the web appendix for full study and results). However, we speculate that these effects may not emerge across all types of spending decisions, just as divergent effects emerge in different debt contexts. For example, differences may emerge in situations when people consider purchase likelihood versus willingness to pay. Although the distinction may seem subtle, the two situations may vary on the purchase-timing importance dimension. Willingness-to-pay decisions may be more akin to borrowing decisions because consumers must consider their indifference point: the maximum amount they are willing to spend before it becomes more attractive to forgo the purchase in the present (and potentially wait for a drop in price). Future research would benefit from further investigating the relationship between borrowing and spending decisions. In sum, our package of studies contributes to the literatures on consumer borrowing and material and experiential purchases by uncovering when to expect consumers to be more willing to borrow for experiential or material purchases, detailing what psychological mechanisms drive these effects, and explaining why exactly those mechanisms operate. It is among the first to examine consumer spending decisions for experiential and material purchases, identifies a new dimension (purchase-timing importance) on which experiential and material purchases systematically differ, demonstrates greater willingness to borrow for experiential purchases, and provides a more nuanced understanding of the relative importance of different drivers of consumer borrowing decisions. DATA COLLECTION INFORMATION The first set of archival data was aggregated and analyzed by the first author. The second set of archival data was aggregated and analyzed by both authors with the help of an independent research assistant at Dartmouth College under the supervision of the second author. Both authors jointly managed the online (MTurk) data collection of all studies except study 5, which was collected by the first author at the University of Southern California’s subject pool. Studies 5 and 6 were analyzed by the first author. All others were analyzed by both authors. Study 3 was conducted in summer 2016. Study 4 and 6 were conducted in summer 2016. Study 5 was collected in fall 2015. Study 7 was conducted in spring 2015. The authors thank Adam Alter, Cindy Chan, Shinar Jain, Kevin Keller, Ellie Kyung, Bob Latimer, Debbie MacInnis, Tom Meyvis, Vicki Morwitz, Abby Sussman, and Gulden Ulkumen for comments and assistance at various stages of the project. Part of this research was conducted when Eesha Sharma was a Junior Visiting Faculty Fellow at the UCLA Anderson School of Management, and support from the Morrison Family Center is acknowledged. Footnotes 1 In a pretest examining the circumstances under which people consider using debt to fund discretionary purchases, 83% of participants stated they would borrow when they do not have the money to buy the purchase outright, rather than because they prefer using debt over other available resources (test against 50%, t(99) = 8.74, p < .001). 2 One might expect the effects found in the subset of consumers included in the robustness check to be stronger than the effects found in the entire sample because the former restricted the sample to those who were more inclined to use financing. To examine this possibility, we tested the significance of the difference between the slopes in the full versus restricted samples. Indeed, the effect was significantly stronger in the reduced sample for both dependent measures (credit card balance: t(32,073) = 2.39, p = .017; financing fees paid: t(31,018) = 2.91, p = .004). REFERENCES Abendroth Lisa J., Diehl Kristin ( 2006), “Now or Never: Effects of Limited Purchase Opportunities on Patterns of Regret over Time,” Journal of Consumer Research , 33 3, 342– 51. Google Scholar CrossRef Search ADS   Auh Seigyoung, Shih Eric, Yoon Yeosun ( 2008), “Aligning Benefits with Payments: A Test of the Pattern Alignment Hypothesis,” Journal of Consumer Psychology , 18 4, 292– 303. Google Scholar CrossRef Search ADS   Brock Timothy C. ( 1968), “Implications of Commodity Theory for Value Change,” in Psychological Foundations of Attitudes , ed. Greenwald Anthony G., Brock Timothy C., Ostrom Thomas M., New York: Academic Press, 243– 75. Google Scholar CrossRef Search ADS   Carter Travis J., Gilovich Thomas ( 2010), “The Relative Relativity of Material and Experiential Purchases,” Journal of Personality and Social Psychology , 98 1, 146– 59. Google Scholar CrossRef Search ADS PubMed  Experian Simmons ( 2011), “The 2011 Discretionary Spend Report,” Experian Simmons, http://www.experian.com/assets/simmons-research/white-papers/2011-discretionary-spend-report.pdf, 3. Gross David B., Souleles Nicholas S. ( 2002), “Do Liquidity Constraints and Interest Rates Matter for Consumer Behavior? Evidence from Credit Card Data,” Quarterly Journal of Economics , 117 1, 149– 85. Google Scholar CrossRef Search ADS   Hayes Andrew F. ( 2013), An Introduction to Mediation, Moderation, and Conditional process Analysis: A Regression-based Approach , New York: Guilford Press. Hirst D. Eric, Joyce Edward J., Schadewald Michael S. ( 1994), “Mental Accounting and Outcome Contiguity in Consumer-Borrowing Decisions,” Organizational Behavior and Human Decision Processes , 58 1, 136– 52. Google Scholar CrossRef Search ADS   Inman J. Jeffrey, McAlister Leigh ( 1994), “Do Coupon Expiration Dates Affect Consumer Behavior?” Journal of Marketing Research , 31 3, 423– 8. Google Scholar CrossRef Search ADS   Kim Haejeong, DeVaney Sharon E. ( 2001), “The Determinants of Outstanding Balances among Credit Card Revolvers,” Financial Counseling and Planning , 12 1, 67– 78. Kumar Amit, Gilovich Tom ( 2016), “To Do or to Have, Now or Later: The Preferred Consumption Profiles of Material and Experiential Purchases,” Journal of Consumer Psychology , 26 2, 169– 78. Google Scholar CrossRef Search ADS   Lynn Michael ( 1991), “Scarcity Effects on Value: A Quantitative Review of the Commodity Theory Literature,” Psychology and Marketing , 8 1, 43– 57. Google Scholar CrossRef Search ADS   Pchelin Paulina, Howell Ryan T. ( 2014), “ The Hidden Cost of Value-Seeking: People Do Not Accurately Forecast the Economic Benefits of Experiential Purchases,” The Journal of Positive Psychology , 9 4, 322– 34. Google Scholar CrossRef Search ADS   Picchi ( 2015), “America’s Skyrocketing Credit Card Debt,” CBS MoneyWatch, http://www.cbsnews.com/news/americas-skyrocketing-credit-card-debt/. Prelec Drazen, Loewenstein George ( 1998), “The Red and the Black: Mental Accounting of Savings and Debt,” Marketing Science , 17 1, 4– 28. Google Scholar CrossRef Search ADS   Rosenzweig Emily, Gilovich Thomas ( 2012), “Buyer’s Remorse or Missed Opportunity? Differential Regrets for Material and Experiential Purchases,” Journal of Personality and Social Psychology , 102 2, 215– 23. Google Scholar CrossRef Search ADS PubMed  Soman Dilip, Cheema Amar ( 2002), “The Effect of Credit on Spending Decisions: The Role of the Credit Limit and Credibility,” Marketing Science , 21 1, 32– 53. Google Scholar CrossRef Search ADS   Tully Stephanie M., Hershfield Hal E., Meyvis Tom ( 2015), “Seeking Lasting Enjoyment with Limited Money: Financial Constraints Increase Preference for Material Goods over Experiences,” Journal of Consumer Research , 42 1, 59– 75. Google Scholar CrossRef Search ADS   Van Boven Leaf ( 2005), “Experientialism, Materialism, and the Pursuit of Happiness,” Review of General Psychology , 9 2, 132– 42. Google Scholar CrossRef Search ADS   Van Boven Leaf, Gilovich Thomas ( 2003), “To Do or to Have? That Is the Question,” Journal of Personality and Social Psychology , 85 6, 1193– 202. Google Scholar CrossRef Search ADS PubMed  Verhallen Theo M. M., Robben Henry S. J. ( 1994), “Scarcity and Preference: An Experiment on Unavailability and Product Evaluation,” Journal of Economic Psychology , 15 2, 315– 31. Google Scholar CrossRef Search ADS   Zhu Lillian Y., Meeks Carol B. ( 1994), “Effects of Low Income Families’ Ability and Willingness to Use Consumer Credit on Subsequent Outstanding Credit Balances,” Journal of Consumer Affairs , 28 2, 403– 22. Google Scholar CrossRef Search ADS   © The Author 2017. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Journal of Consumer ResearchOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off