Production Enjoyment Asymmetrically Impacts Buyers’ Willingness to Pay and Sellers’ Willingness to Charge

Abstract

With the rise of social media and the peer-to-peer economy, sellers can easily tell potential buyers about themselves and their process of producing products and services. This research investigates the influence of a central aspect of the production process that sellers can communicate—their production enjoyment. Buyers are willing to pay a higher price, are more likely to click on ads, and are more likely to choose a product or service when the seller signals that they enjoy producing it. In contrast, sellers are willing to accept lower prices, and actually charge less, for products and services they enjoy producing. Both buyers and sellers make the inference that production enjoyment leads to higher quality products/services, but only buyers rely on this inference when forming their pricing judgments relative to sellers. Nine studies illustrate these effects across a wide variety of products and services, participant samples, and operationalizations of production enjoyment. They show that signals of production enjoyment can influence buyers more than other established signals (e.g., effort) and demonstrate contexts where these effects are more and less likely to occur. These findings offer practical recommendations for both buyers and sellers as well as a variety of theoretical contributions.

In today's marketplace, everyday buyers and sellers can directly connect and conduct business with one another through peer-to-peer networks like Etsy, Upwork, and craft fairs, as well as through social media. Consequently, “buyers” (i.e., those who purchase labor, products, and services) can access more personal information about the production process behind the offerings of their counterpart “sellers” (i.e., those who produce and provide said labor, products, and services; Fuchs et al. 2022). It is well established that buyers evaluate products/services using information about how they are produced (e.g., whether they were handmade, how effortful they were to produce), and this information impacts valuation (Fuchs, Schreier, and Van Osselaer 2015; Kruger et al. 2004). We examine an unstudied marketplace cue that both buyers and sellers use to inform the value of a product or service: the seller's enjoyment of the production process (hereinafter, “production enjoyment”).

We find that buyers respond quite positively to signals of production enjoyment. They infer that products and services that sellers enjoy producing are higher in quality and therefore are willing to pay more for these products and services, even when compared with other signals of quality, such as efficiency and effort. Consistent with our theorizing that perceived quality drives this relationship, we demonstrate in several studies that production enjoyment primarily increases willingness to pay for products and services that require some measure of skill to produce, such as those produced by hand (i.e., not through automation). We then test other moderators (e.g., observability, creativity) in an exploratory analysis.

We also explore how production enjoyment influences the sellers who actually experience it. Sellers are willing to accept a lower price—and do indeed charge less—for the products and services they enjoy producing. Although sellers also generally associate production enjoyment with higher quality, they do not rely on this inference as much in their pricing decisions. Nine primary studies (plus two supplementary ones in the Web Appendix) in both field and lab settings document this asymmetric influence of production enjoyment across a multitude of products and services.

This research carries a range of important contributions for theory. Most notably, enjoyment is integral to the process of creating products and services, yet it has remained understudied as an input into marketplace transactions. We contribute to the growing literature on consumer inference-making (Kirmani and Rao 2000), identifying a novel cue for product quality. Our work thus extends the research on the characteristics of the creation process that influence consumers (Fuchs, Schreier, and Van Osselaer 2015; Kruger et al. 2004; Van Osselaer et al. 2020) and creators (Atakan, Bagozzi, and Yoon 2014; Dahl and Moreau 2007; Norton, Mochon, and Ariely 2012). By showing that the same cue (i.e., production enjoyment) can have opposite effects on the prices buyers are willing to pay versus what sellers are willing to charge, we contribute to the literature on asymmetric transaction effects, such as the influential endowment effect. Often, this work shows that the same cue can have a greater impact on buyers versus sellers (Carmon and Ariely 2000; Morewedge and Giblin 2015; Nayakankuppam and Mishra 2005); however, it rarely demonstrates that the same cue can have the opposite impact on them (cf. Lerner, Small, and Loewenstein 2004).

From a practical standpoint, this research informs managers and producers about a novel attribute that they can signal in the marketplace to increase buyers’ product choice and willingness to pay. By comparing production enjoyment with a variety of other established cues of product quality (e.g., production effort) and identifying several moderators that determine the impact of this signal (e.g., required skill), these studies provide a useful framework for when and how to signal production enjoyment. Furthermore, we demonstrate that buyers and sellers respond to production enjoyment asymmetrically, which suggests to sellers that high production enjoyment might lead them to miss out on available profit by setting prices too low relative to buyers’ willingness to pay. Simultaneously, it suggests to buyers that products/services associated with high production enjoyment may come at a discount.

Theoretical Background

Production Enjoyment

Buyers face uncertainty about how much they should be willing to pay for a good or service, often relying on observable cues to infer an offering's quality (Broniarczyk and Alba 1994; Kirmani and Rao 2000). Typically, buyers use explicit signals, such as brand power; however, for small businesses or peer-to-peer marketplaces, these traditional cues are less available (Caldieraro et al. 2018). Instead, buyers use information about the sellers’ production process to infer quality (Van Osselaer et al. 2020), such as whether the products were handmade (Fuchs, Schreier, and Van Osselaer 2015) or effortful to create (Kruger et al. 2004). We contend that an important but overlooked cue in this vein is the seller's production enjoyment.

We define production enjoyment as the perception of positive feelings, such as pleasure, liking, and fun (Alba and Williams 2013), attributed to the process of producing a good or service. We largely focus on situations where sellers can directly communicate with buyers; however, we ultimately conceptualize production enjoyment as a term that can apply to the vast majority of work settings. To first explore the current usage of production enjoyment as a signal in peer-to-peer marketplaces, we acquired six extant datasets scraped from such marketplaces, namely, Etsy and Upwork. Each dataset displays over 1,000 product or seller profiles written by the seller, in which they could signal production enjoyment. In general, production enjoyment seems to be a lightly used signal, descriptively mentioned in fewer than .1% of profiles on Etsy and about 4% of profiles on Upwork (see Web Appendix A). But are sellers strategically correct in not communicating information about their production enjoyment? As we describe next, revealing high production enjoyment could actually improve buyers’ interest in and willingness to pay for those sellers’ offerings.

Buyers’ Perceptions of Production Enjoyment

Buyers use a variety of cues to infer the quality of a potential purchase and determine their willingness to pay for it (Kirmani and Rao 2000). While some of these high-quality cues are costly to the seller (e.g., warranties, advertising), there are other less costly ones that can be embedded in buyer–seller interactions or promotional materials—such as sellers’ production enjoyment. We propose two reasons that communicating high production enjoyment could serve as a signal of high quality, thus increasing buyers’ willingness to pay.

First, we draw on the influential work on flow theory (Csikszentmihalyi 1990). This research describes the experience of creators entering a “flow” state (aka “being in the zone”), which elicits positive affect as a result of one's relevant skills meeting the upper demands of a task. For example, a musician experiences enjoyment while immersed in producing a piece of high-quality music (Csikszentmihalyi 1990). For buyers who learn that a seller experiences high production enjoyment, it could signal this flow-based association, leading the buyer to conclude that high production enjoyment means a high-quality offering. Research on goals theory supports this notion (Csikszentmihalyi, Abuhamdeh, and Nakamura 2005). People experience positive affect when nearing goal achievement (Carver and Scheier 1999; Freitas, Liberman, and Higgins 2002), meaning that high production enjoyment could similarly imply successful goal attainment in one's product creation (Csikszentmihalyi 1990; Dahl and Moreau 2007). Given this link between enjoyment and skillful achievement of a goal (Teeny et al. 2019), and the notion that affective reactions, such as enjoyment, are seen as highly diagnostic for gaining insights about other people (Andersen 1984), relatively high production enjoyment could signal relatively high quality in the product/service.

Second, and relatedly, if a seller conveys high production enjoyment, it can imply that their labor is motivated intrinsically (Burroughs et al. 2011). Intrinsic motivation (i.e., pursuing something based on inherent enjoyability; Ryan and Deci 2000) has been repeatedly linked to improved quality and performance in work settings (e.g., Bhattacharjee et al. 2014; Sellier and Dahl 2011). Thus, if buyers interpret production enjoyment as a signal of intrinsic motivation, they might infer higher product or service quality.

Based on this theorizing and the robust connection between quality perceptions and willingness to pay (Kirmani and Rao 2000), we propose:

• H₁: Buyers are more interested in and willing to pay more for products and services when sellers appear to have high production enjoyment.

• H₂: The effect in H₁ operates through buyers’ inference that products and services that sellers enjoy producing are of higher quality.

This theorizing leads to predictions about when the effect may not occur. Specifically, in cases where production enjoyment's ability to signal quality is limited, buyers should not show an increase in willingness to pay when sellers express it. One such context is when tasks or jobs are perceived to require little to no skill. Professions (and the tasks within them) vary widely in the amount of skill required to do them well (Deming and Kahn 2018), and the rise of automation has affected the skill required to complete many tasks. In many cases, automation has simplified the production process: tasks once requiring skilled labor (e.g., cooking, copy editing) can now be accomplished with automated machinery and the simple push of a button. We examine contexts where automation has reduced the skill needed to complete a task, along with nonautomated low-skill tasks, and predict that when little skill is required, signaling high production enjoyment has a limited effect on buyers’ quality perceptions.

According to our flow-based theorizing, producers experience enjoyment when their skills match the demands of the task (Csikszentmihalyi 1990). Therefore, if a seller signals high enjoyment of a low-skill (e.g., automated) job, it suggests that their enjoyment may come from other sources besides flow, or if it comes from flow, it means their abilities might be similarly low, mitigating any enjoyment-enhanced perception of the offering's quality. Additionally, for low-skill tasks, such as initiating automated production tasks, the characteristics of the producer completing the task have little impact on product quality. This suggests that sellers should reserve communication about production enjoyment for high-skill jobs or tasks (e.g., for handmade rather than machine-made offerings). Thus, we propose:

• H₃: The effect in H₁ is mitigated for jobs that require low skill, such as automated tasks, because production enjoyment has a limited effect on buyers’ quality perceptions.

In addition to skill, it is feasible that the value of enjoyment as a signal of quality and the extent to which buyers incorporate it into their willingness to pay depends on various practically relevant, job-level characteristics. For example, jobs that are more creative might offer more opportunity for production enjoyment to influence quality, since these types of jobs may better allow for aspects of the creation process to transfer to the product (e.g., Newman and Dhar 2014). Further, enjoyment may be a more powerful signal if buyers can see it for themselves, as buyers tend to incorporate more production-related features when the production is observable (Moran and Ronen 2021), such as with the production of services (vs. products). In our final study, we explore these potential moderators along with two others (the desirability of the job and how highly it pays) in order to provide further, suggestive evidence for managers and sellers in deciding when best to signal production enjoyment.

Sellers’ Perceptions of Production Enjoyment

Between buyers and sellers, marketplace cues can often produce similar inferences (e.g., Monroe and Krishnan 1985). This means sellers (like buyers) likely infer that their high production enjoyment signals a higher quality product/service for similar reasons. But will sellers rely on these quality inferences in a similar way when making their pricing decisions? Sellers vary in their willingness to accept or charge money (i.e., terms we use interchangeably) for their products and services, and years of research show that buyers and sellers approach and evaluate transaction utility in different ways when making such pricing judgments (Carmon and Ariely 2000; Johnson, Häubl, and Keinan 2007; Weaver and Frederick 2012; for a review, see Morewedge and Giblin [2015]). In contrast to buyers, when sellers consider pricing, they incorporate a broad range of factors beyond the product's quality, such as the associations between themselves and the product that result from owning or producing it (Dommer and Swaminathan 2013; Moreau, Bonney, and Herd 2011; Schnurr et al. 2022). Thus, we propose that sellers (unlike buyers) do not rely directly on the higher quality perceptions triggered by production enjoyment when forming their pricing decisions.

When the production process is relatively enjoyable, the process itself is likely to satisfy innate goals (e.g., growth, competence, autonomy; Ryan and Deci 2000) and provide utility in and of itself. For sellers, this contributes to transaction utility, meaning they need less financial compensation to feel as though they benefited from a transaction. From a labor economics perspective, this idea that production enjoyment may provide direct utility to sellers is consistent with Adam Smith's theorizing in The Wealth of Nations (Smith [1776] 1994): Pleasant (vs. unpleasant) jobs require less compensation, because they have a larger pool of available laborers and lower equilibrium wage rates. This suggests that even if sellers draw a similar inference to buyers that high production enjoyment entails a higher quality product or service, they might still discount their pricing for products/services produced with high enjoyment. Taken together, we propose:

• H₄: Sellers charge a lower price for products and services that they enjoy producing.

Overview of Studies

Across nine primary studies and two supplementary studies, we test the effects of production enjoyment on pricing decisions for both buyers and sellers. Studies 1–4 show that signals of production enjoyment enhance buyers’ responses to products and services: High production enjoyment options increased click-through rates (CTRs) in Facebook ads (Studies 1, WA1 in Web Appendix B) and willingness to pay (Studies 2–4b, WA2 in Web Appendix C). Signaling production enjoyment can be more effective than other positive signals (e.g., effort, Study 2; popularity, Study 4a), and the signal boosts willingness to pay through increased quality perceptions (Studies 2, 3, 4b, 7, WA2). Importantly, enjoyment is a more impactful signal of quality for jobs that require more skill (Study 3) and thus is less effective for machine-made products and automated services (Studies 4a–b). The remaining studies examine the effect of production enjoyment on sellers (Study 5) or both buyers and sellers simultaneously (Studies 6–8). These studies show that high production enjoyment asymmetrically leads sellers to accept/charge lower prices, even though buyers would have been willing to pay higher prices. Our final study demonstrates the generalizability of these effects across 100 consumer-facing jobs and suggests further boundary conditions.

Sample sizes or duration in the field for all experiments were determined in advance of data collection, and we report all measures, participant exclusions, and additional analyses either in the main text or supplementary Web Appendix. All experimental data and study materials/procedures are available on the Open Science Framework (https://osf.io/4g9tu/?view_only=4e83770bf10e45a491e0476af7741841).

Study 1: Signaling Production Enjoyment in Facebook Ads

Study 1 examines whether prospective buyers respond positively to knowledge that the seller enjoys producing their product or service. If so, a key practical implication is that production enjoyment can be an effective signal that sellers can intentionally communicate to buyers. To test this, we conduct a field study through Facebook to examine how consumers respond to advertising that mentions (vs. does not mention) high production enjoyment.

Method

We targeted Facebook users in the United States who were over the age of 18 and who listed “small business owner(s)” in their interests, employers, or behaviors. Over four days in August 2020, 30,468 unique Facebook users were randomly exposed to one of two ads in a two-cell (production enjoyment: high vs. control) between-subjects field experiment.

We created two ads for a search engine optimization (SEO) specialist (see Web Appendix D) and used Facebook's A/B testing feature to randomly present one of the two ads to our target audience. In the control condition, the ad read, “Need SEO Help? I have a lot of experience.” In the production enjoyment condition, the ad additionally read, “I really enjoy SEO.” We observed the number of ad clicks as a proportion of ad exposures (i.e., the CTR).

Results and Discussion

Results

Over the course of the experiment, the ads were displayed a total of 38,099 times (n_control = 20,519, n_enjoyment = 17,580), and more users clicked on the enjoyment ad than the control ad (n_control = 147, n_enjoyment = 177). A chi-square test revealed that the ad mentioning enjoyment was significantly more effective at generating clicks than the control ad (CTR_control = .72%, CTR_enjoyment = 1.01%; χ²(1) = 9.47, p = .002, Cramér's V = .016; see Web Appendix D for alternate specification with consistent results). These percentages, although low, are consistent with prior studies (e.g., Tucker 2014). Facebook provided additional information regarding the cost per click (CPC) of each ad (CPC_control = $.48, CPC_enjoyment = $.40) and, consistent with our conclusions, identified the production enjoyment condition as the winner of the A/B test.

Discussion

The results of this study show that consumers have a positive response to ads that signal the seller's production enjoyment. Specifically, the ad that mentioned that the seller enjoys SEO yielded a 40% higher CTR and a 17% lower CPC than the otherwise identical ad that did not mention production enjoyment. We conceptually replicated this result in another Facebook study (Study WA1; N = 40,866) using a different domain and showed consistent results, suggesting that sellers who enjoy producing their products and services should readily signal production enjoyment. Subsequent studies address the alternative explanation that the enjoyment condition may have been better received because it presented participants with two pieces (vs. one) of positive information.

Study 2: Willingness to Pay and Quality Perceptions for Enjoyment Versus Effort

Next, we provide evidence that high production enjoyment (communicated via realistic ads in a laboratory experiment) increases buyers’ willingness to pay, because buyers infer that it signals a high-quality offering. This study further compares the strength of the enjoyment signal to that of effort, which has been shown to increase quality perceptions (Kruger et al. 2004). We also examine the role of other inferences consumers can make from production enjoyment (e.g., the extent to which a product contains love, how authentic it is, how unique it is) to determine their influence relative to inferences about high quality.

Method

Participants

A total of 605 participants from CloudResearch completed the study. Nine participants were excluded for failing the instructional manipulation check (IMC; Oppenheimer, Meyvis, and Davidenko 2009). The final sample includes 596 participants (M_age = 40.43 years, SD = 12.59; 52% female, 46% male, 2% prefer not to say/other).

Procedure

This study uses a 2 (production enjoyment: high vs. control) × 2 (effort: high vs. control) between-subjects design. Participants imagined hiring a plumber and saw one of four possible realistic Instagram ads for a possible plumber (see Web Appendix E). Each ad emphasized their enjoyment of plumbing (“Why not try a plumber who loves the work? I enjoy repairing pipes and fixing leaks more than any other plumber that I’ve met”), the effort they put into it (“Why not try a plumber who works really hard? I work harder at repairing pipes and fixing leaks than any other plumber that I’ve met”), a combination of the two, or neither. Participants then indicated their willingness to pay for the plumber (“Compared to other plumbers, how much would you be willing to pay for this person's plumbing work?” 1 = “Much less than other plumbers,” and 7 = “Much more than other plumbers”) and perceived quality (“How good do you think the plumbing work will be?” 1 = “Much worse than other plumbers,” and 7 = “Much better than other plumbers”). Participants were also asked about alternative mediators, including how much the plumber's work contains love, is done by hand, is unique, and is authentic (see Web Appendix E). Participants then responded to two manipulation checks asking how much the ad said that the plumber enjoys their job and works hard at their job (1 = “Not at all,” and 7 = “Very much”). Finally, participants indicated their age and gender and responded to an IMC asking them to complete a simple math problem.

Results and Discussion

Manipulation checks

For the enjoyment manipulation check, as intended, participants who were (high-enjoyment condition) versus were not (control-enjoyment condition) exposed to the enjoyment manipulation indicated that the ad mentioned enjoyment (M_{high enjoyment/control effort} = 6.58, SD = .93; M_{control enjoyment/control effort} = 3.40, SD = 1.89; M_{high enjoyment/high effort} = 6.35, SD = .99; M_{control enjoyment/high effort} = 2.86, SD = 1.97; F_enjoyment(1, 592) = 711.82, p < .001, ${\eta }_{\mathrm{p}}^2$  = .55). There was also a significant main effect of effort (F_effort(1, 592) = 9.47, p = .002, ${\eta }_{\mathrm{p}}^2$  = .02), such that mentions of effort decreased perceptions of enjoyment. How might this affect our primary results? If enjoyment indeed increases willingness to pay as hypothesized, the lower enjoyment arising from higher effort may artificially deflate the possible benefits of effort on willingness to pay. However, less perceived enjoyment in the high-effort conditions cannot account for a potential main effect of enjoyment on willingness to pay, which is our focal hypothesis in the current study. Thus, the unexpected main effect of effort on the enjoyment manipulation check neither contributes to nor detracts from our predicted main effect of enjoyment. There was a nonsignificant interaction (F_interaction(1, 592) = 1.56, p = .212, ${\eta }_{\mathrm{p}}^2$  < .01). For the effort manipulation check, participants who were (vs. were not) exposed to the effort manipulation indicated that the ad mentioned effort (M_{high enjoyment/control effort} = 3.87, SD = 2.09; M_{control enjoyment/control effort} = 3.99, SD = 1.82; M_{high enjoyment/high effort} = 6.32, SD = 1.17; M_{control enjoyment/high effort} = 6.43, SD = 1.10; F_effort(1, 592) = 347.15, p < .001, ${\eta }_{\mathrm{p}}^2$  = .37). There was no main effect of enjoyment nor an interaction (Fs < 1).

Willingness to pay

An ANOVA using the willingness to pay measure revealed no main effect of effort and a significant main effect of enjoyment qualified by a significant interaction (M_{high enjoyment/control effort} = 4.57, SD = 1.14; M_{control enjoyment/control effort} = 3.80, SD = 1.07; M_{high enjoyment/high effort} = 4.23, SD = 1.01; M_{control enjoyment/high effort} = 3.91, SD = 1.09; F_enjoyment(1, 592) = 37.94, p < .001, ${\eta }_{\mathrm{p}}^2$  = .06; F_effort(1, 592) = 1.81, p = .18, ${\eta }_{\mathrm{p}}^2$  < .01; F_interaction(1, 592) = 6.39, p = .012, ${\eta }_{\mathrm{p}}^2$  = .01). Compared with ads that did not mention enjoyment, mentioning enjoyment increased willingness to pay. The interaction revealed that the positive effect of enjoyment was strongest when effort was not mentioned (F_{simple effect}(1, 592) = 37.49, p < .001, ${\eta }_{\mathrm{p}}^2$  = .06) and weaker, but still significant, when effort was also mentioned (F_{simple effect}(1, 592) = 6.64, p = .010, ${\eta }_{\mathrm{p}}^2$  = .01).

Perceived quality

An ANOVA using the perceived quality measure revealed a significant main effect of enjoyment and a marginal main effect of effort, qualified by a significant interaction (M_{high enjoyment/control effort} = 5.15, SD = 1.12; M_{control enjoyment/control effort} = 4.30, SD = 1.07; M_{high enjoyment/high effort} = 4.72, SD = 1.17; M_{control enjoyment/high effort} = 4.37, SD = 1.16; F_enjoyment(1, 592) = 41.75, p < .001, ${\eta }_{\mathrm{p}}^2$  = .07; F_effort(1, 592) = 3.74, p = .054, ${\eta }_{\mathrm{p}}^2$  < .01; F_interaction(1, 592) = 7.33, p = .007, ${\eta }_{\mathrm{p}}^2$  = .01). Compared with ads that did not mention enjoyment, mentioning enjoyment increased quality perceptions. Further investigation of the interaction revealed that the positive effect of enjoyment was strongest when effort was not mentioned (F_{simple effect}(1, 592) = 41.76, p < .001, ${\eta }_{\mathrm{p}}^2$  = .07), and weaker, but still significant, when effort was mentioned (F_{simple effect}(1, 592) = 7.09, p = .008, ${\eta }_{\mathrm{p}}^2$  = .01).

Mediation

First, we consider whether high quality explains the relationship between production enjoyment and willingness to pay through a mediation analysis using 5,000 bootstrapped samples (Hayes 2022, Model 4). Indeed, perceived quality statistically mediated the relationship between production enjoyment and buyers’ willingness to pay (b = .21, SE = .03; 95% CI: [.15, .28]). Next, we added the alternative mediators in parallel to quality, and quality remained significant (b = .16, SE = .03; 95% CI: [.11, .22]). Authenticity (b = .05, SE = .02; 95% CI: [.02, .09]) and uniqueness (b = .05, SE = .02; 95% CI: [.02, .08]) also mediated enjoyment's effect on willingness to pay. See Web Appendix E for full results.

A test of moderated mediation (Model 7) revealed that quality was a significant mediator whether effort was mentioned or not, though the indirect effect was stronger when effort was not mentioned (index of moderated mediation: b = −.18, SE = .07; 95% CI: [−.31, −.05]). The index of moderated mediation for quality remained significant when the alternative mediators were entered in parallel (b = −.14, SE = .05; 95% CI: [−.25, −.04]), while authenticity (b = −.03, SD = .02; 95% CI: [−.08, −.00]) and uniqueness (b = −.03, SE = .02; 95% CI: [−.06, −.00]) exhibited the same pattern of results (i.e., significant index of moderated mediation). However, across studies, only quality consistently emerges as a robust mediator. See Web Appendix E for full results.

Discussion

Production enjoyment once again proved an effective way to increase buyers’ willingness to pay, and this was mediated by increased perceptions of quality. Production enjoyment increased willingness to pay and quality perceptions more than an effort signal, which has been shown in other contexts to increase willingness to pay (Kruger et al. 2004). Finally, we note that the indirect effect through perceptions of quality held while simultaneously entered alongside multiple, alternative mediators.

Study 3: Production Enjoyment and Skill

Since production enjoyment affects willingness to pay through quality perceptions, we expect that the effect will be attenuated in contexts where enjoyment's ability to signal quality is limited. Flow theory suggests that production enjoyment is more likely to occur when the producer's skill level matches the demands of the task (Csikszentmihalyi 1990), and achieving such a state is only impressive when the task requires skill. Thus, for production tasks that do not require skill, enjoyment should not be a useful signal of quality, and buyers’ willingness to pay for the product should not increase. In Study 3, we test these predictions by manipulating whether a bartender verbally signals production enjoyment for a complex cocktail versus a simple mixed drink. We expect that production enjoyment will increase quality perceptions of and willingness to pay for the complex (vs. simple) drink.

Method

Participants

We recruited 600 CloudResearch workers, and 601 completed the study. Excluding the 6 participants who failed the IMC left a final sample of 595 participants (M_age = 41.81 years, SD = 12.52; 51.3% female, 47.7% male, 1.0% prefer not to say/other).

Procedure

This study uses a 2 (production enjoyment: high vs. control) × 2 (skill: low vs. high) between-subjects design. Participants imagined they were at a bar ordering a drink, but they were not sure what to order. The bartender recommends a drink that was described as a “simple whisky on the rocks” that “only has two ingredients” (low-skill condition) or “an old fashioned cocktail” that is “complicated to make” (high-skill condition). In the high-enjoyment condition, the bartender mentioned that they “really enjoy” making the drink because it is either simple or complex. In the control-enjoyment condition, the bartender mentioned that they can make the drink, and “it's tasty,” because it is either simple or complex. Participants then indicated their willingness to pay for the bartender's drink compared with other bartenders and the quality of the bartender's drink compared with other bartenders (both on scales where 1 = “Definitely less than others,” and 7 = “Definitely more than others”). Participants then responded to manipulation checks about the skill and enjoyment manipulations (see Web Appendix F for full measures). Finally, participants indicated their age and gender and responded to an IMC asking them to complete a simple math problem.

Results and Discussion

Manipulation check

For the enjoyment manipulation check, results revealed that participants in the high-enjoyment conditions indicated that the bartender enjoys making the drink more than participants in the control-enjoyment conditions (M_{high enjoyment/high skill} = 6.33, SD = .97; M_{control enjoyment/high skill} = 4.54, SD = 1.50; M_{high enjoyment/low skill} = 5.89, SD = 1.28; M_{control enjoyment/low skill} = 4.45, SD = 1.25; F_enjoyment(1, 591) = 243.67, p < .001, ${\eta }_{\mathrm{p}}^2$  = .29). There was also a main effect of skill and a marginally significant interaction (F_skill(1, 591) = 6.40, p = .012, ${\eta }_{\mathrm{p}}^2$  = .01; F_interaction(1, 591) = 2.73, p = .099, ${\eta }_{\mathrm{p}}^2$  = .01). The enjoyment manipulation was significant within both the high-skill (F_{simple effect}(1, 591) = 146.77, p < .001, ${\eta }_{\mathrm{p}}^2$  = .20) and low-skill (F_{simple effect}(1, 591) = 98.91, p < .001, ${\eta }_{\mathrm{p}}^2$  = .14) conditions. However, we do note that the main effect of skill on the enjoyment manipulation check may increase the potential main effect of skill on willingness to pay. That is, skill may signal enjoyment, which in turn can boost willingness to pay; however, it is unlikely that this affects our primary prediction in this study (i.e., the interaction showing that enjoyment primarily affects willingness to pay for tasks that require high [vs. low] skill).

Regarding the skill manipulation check, participants in the high-skill conditions indicated that their drink required more skill to make than participants in the low-skill conditions (M_{high enjoyment/high skill} = 4.82, SD = 1.47; M_{control enjoyment/high skill} = 4.46, SD = 1.66; M_{high enjoyment/low skill} = 1.85, SD = 1.26; M_{control enjoyment/low skill} = 2.05, SD = 1.29; F_enjoyment(1, 591) = 529.76, p < .001, ${\eta }_{\mathrm{p}}^2$  = .47). There was no significant main effect of enjoyment (F < 1), but there was a significant interaction (F_interaction(1, 591) = 5.66, p = .018, ${\eta }_{\mathrm{p}}^2$  = .01). That is, the skill manipulation was slightly stronger within the enjoyment condition (F_{simple effect}(1, 591) = 323.03, p < .001, ${\eta }_{\mathrm{p}}^2$  = .35) but still strong and highly significant within the control condition (F_{simple effect}(1, 591) = 212.58, p < .001, ${\eta }_{\mathrm{p}}^2$  = .27).

Willingness to pay

An ANOVA using the willingness to pay measure revealed a significant main effect of enjoyment, a significant main effect of skill, and the predicted significant interaction, which is displayed in Figure 1 (M_{high enjoyment/high skill} = 4.43, SD = 1.06; M_{control enjoyment/high skill} = 3.90, SD = 1.34; M_{high enjoyment/low skill} = 3.46, SD = 1.24; M_{control enjoyment/low skill} = 3.53, SD = 1.19; F_enjoyment(1, 591) = 5.30, p = .022, ${\eta }_{\mathrm{p}}^2$  = .01; F_skill(1, 591) = 45.31, p < .001, ${\eta }_{\mathrm{p}}^2$  = .07; F_interaction(1, 591) = 8.81, p = .003, ${\eta }_{\mathrm{p}}^2$  = .02). An analysis of the simple effects revealed that enjoyment served as a positive signal for willingness to pay in the high-skill condition (F_{simple effect}(1, 591) = 13.69, p < .001, ${\eta }_{\mathrm{p}}^2$  = .02), but not in the low-skill condition (F < 1).

Figure 1.

High Enjoyment Only Increases Willingness to Pay for High-Skill Jobs.

Perceived quality

An ANOVA on the perceived quality measure similarly revealed a significant main effect of enjoyment, a significant main effect of skill, and the predicted significant interaction (M_{high enjoyment/high skill} = 5.08, SD = 1.18; M_{control enjoyment/high skill} = 4.08, SD = 1.40; M_{high enjoyment/low skill} = 4.22, SD = 1.16; M_{control enjoyment/low skill} = 3.88, SD = 1.12; F_enjoyment(1, 591) = 44.77, p < .001, ${\eta }_{\mathrm{p}}^2$  = .07; F_skill(1, 591) = 28.37, p < .001, ${\eta }_{\mathrm{p}}^2$  = .05; F_interaction(1, 591) = 10.96, p < .001, ${\eta }_{\mathrm{p}}^2$  = .02). Although enjoyment increased perceptions of quality in both the high-skill condition (F_{simple effect}(1, 591) = 49.27, p < .001, ${\eta }_{\mathrm{p}}^2$  = .08) and the low-skill condition (F_{simple effect}(1, 591) = 5.80, p = .016, ${\eta }_{\mathrm{p}}^2$  = .01), the effect was significantly stronger in the high-skill condition.

Mediation

A moderated mediation analysis using 5,000 bootstrapped samples (Hayes 2022, Model 7) revealed that perceived quality differentially mediated the relationship between production enjoyment and buyers’ willingness to pay as a function of skill (index of moderated mediation: b = .23, SE = .07; 95% CI: [.09, .37]). Specifically, perceived quality had a significantly stronger indirect effect in the high-skill condition (b = .34, SE = .05; 95% CI: [.24, .45]) than the low-skill condition (b = .12, SE = .05; 95% CI: [.03, .21], see Web Appendix F).

Discussion

This study demonstrates an additional ecologically valid way to signal production enjoyment to customers (i.e., verbally) and provides further evidence that production enjoyment increases willingness to pay via increased perceptions of quality. Finally, it also shows that production enjoyment increases willingness to pay more for production tasks that require higher levels of skill. This moderation analysis provides practically relevant insight for producers about when this signal works best and is consistent with our theorizing around perceived quality. However, we do acknowledge that the slightly inconsistent strength of the manipulations across conditions may have contributed to the results. We further explore this moderator in Studies 4a and 4b.

Studies 4a and 4b: Production Enjoyment for Handmade and Machine-Made Products

The prior study demonstrated that production enjoyment has a greater impact on willingness to pay for tasks that require relatively more (vs. less) skill, because quality inferences are greater in these contexts. As an additional way to operationalize skill level, we next turn to an increasingly relevant factor in producing products and services: automation. Recent developments in technology have led to a proliferation of machine-automating tools, which lower the skill required in a variety of production settings. While not all automated jobs require less skill, automation has a tendency to reduce the skill needed in production compared with a handmade alternative. Further, consistent with our theoretical framework, machine-made products leave little room for characteristics of the producer to affect the quality of the product (Fuchs, Schreier, and Van Osselaer 2015). In our next set of studies, we vary whether sellers express production enjoyment for handmade versus automated, machine-made offerings.

In Study 4a, we conduct a field experiment at a U.S. university along with a process-focused posttest. In them, we compare the impact of production enjoyment on actual product choice versus another positive signal commonly used in marketing, product popularity (Cialdini and Goldstein 2004), when applied to a handmade versus machine-made product (brownies). In Study 4b, we conduct a conceptual replication of these findings in a laboratory setting to extend the results to a service domain (dog washing). We also include additional process measures and controls (see Web Appendices G and H for full details). Across both studies, we expect the effect of production enjoyment to be attenuated in the low-skill (i.e., automated) condition relative to the higher-skill (i.e., handmade) one.

Study 4a Method

Participants

We collected as many participants as we could during our allotted 2.5-hour window at a popular student fair. In total, 313 people stopped at our table, which advertised receiving a free brownie in exchange for participation in a brief survey (M_age = 18.99 years, SD = 1.77, 1 missing; 60.1% female, 37.0% male, 2.9% nonbinary). To provide additional process evidence, we also conducted a posttest to support our theorizing (CloudResearch; N = 183; M_age = 39.30 years, SD = 10.35; 44.3% female, 55.2% male, .5% prefer not to say).

Procedure

We reserved a table at a popular student fair on campus (∼3,000 attendees per day), where research assistants (trained for consistency and unaware of the hypothesis) set up a large sign along with other materials indicating that they were conducting market research for a local bakery. Passersby were informed that if they indicated which of the two ostensibly different brownies they were willing to pay more for, they would receive it for free. Next to each brownie (Brownie A and Brownie B), there was a handwritten sign that provided more information about it, and it was used to employ a 2 (signal: production enjoyment vs. popularity; within-subjects) × 2 (skill: high vs. low; between-subjects) mixed design.

As the within-subjects signal manipulation, each of the two signs included a quote from the “Head Pastry Chef.” The quote about Brownie A signaled high production enjoyment (“I really enjoy making this version”), while the quote about Brownie B signaled high popularity (“It's a very popular item”). To manipulate the between-subjects skill condition (i.e., handmade vs. machine-made), the research assistants changed versions of the two signs every 25 participants. In the high-skill condition, the quotes explicitly described each brownie as handmade, and in the low-skill condition, they explicitly described each brownie as machine-made (full stimuli details in Web Appendix G). After participants read the quotes, they reported which brownie they would “pay more for” in a store along with their age and gender on a slip of paper. They then received their chosen brownie.

Study 4a Results

Results

First, we tested whether there was a significant difference in the distribution of participants’ choices in the high-skill (handmade) versus low-skill (machine-made) conditions, where we found one (χ²(1) = 9.33, p = .002, Cramér's V = .173). In the high-skill condition, 67.1% of participants chose the brownie high in production enjoyment (vs. 32.9% who chose the popular brownie). A difference in proportions test revealed this difference was significant compared with chance (z = 3.23, p = .001, Cohen's h = .349). In contrast, in the low-skill condition, participants were equally likely to choose the brownie high in production enjoyment (50%) and the popular brownie (50%).

Posttest

In our posttest, we provided participants with a detailed, scenario-based description of the setting and design of our focal study. Rather than assessing a choice between brownies, however, we asked them to evaluate the brownies’ perceived quality (1 = “Brownie A is definitely higher quality,” 3 = “Brownies A and B are equal in quality,” and 5 = “Brownie B is definitely higher quality”). In the high-skill (handmade) brownie condition, a one-sample t-test against the midpoint (3, representing equal quality perceptions) revealed that participants rated the brownie with a production enjoyment (vs. popularity) signal as higher in quality (M = 2.66, SD = 1.20; t(92) = −2.76, p = .007, d = .29). In contrast, in the low-skill (machine-made) condition, participants rated the brownies as equal in quality (M = 3.01, SD = 1.09; t < 1). These findings suggest that, in the field study, a production enjoyment (vs. popularity) signal increased perceived quality in the handmade condition, but not the machine-made condition, and this ultimately drove the observed differences in choice share (see Web Appendix G for more methodological details and results).

Study 4b Method

Study 4b provides further support for the moderating role of the job's skill level (operationalized as handmade vs. automated) on the effects of production enjoyment. We recruited dog owners via CloudResearch, received 401 responses, and eliminated three participants for failing the IMC (n = 398; M_age = 39.65 years, SD = 12.23, 52.0% female, 46.5% male, 1.5% other/prefer not to say). Participants viewed realistic components of a website for a dog washing service, including its landing page, “About” page, and employee profiles. The focal employee profile stressed either production enjoyment or efficiency in dog washing, and was located next to a filler profile. We further manipulated whether the dog washing was done by hand (high skill) versus by a state-of-the-art, automated machine (low skill), resulting in a 2 (signal: enjoyment vs. efficiency) × 2 (skill: low vs. high) between-subjects design. Participants reported their willingness to pay (1 = “$30,” and 11 = “$50”), quality perceptions relative to another dog washer (1 = “Much worse,” and 7 = “Much better”), and the alternative process measures described in Study 2 plus perceived effort. Additional methodological details are reported in Web Appendix H.

Study 4b Results

An ANOVA on willingness to pay revealed a main effect of production enjoyment, a main effect of skill, and the predicted interaction, replicating Study 4a (M_{enjoyment/high skill} = 6.88, SD = 2.15; M_{efficiency/high skill} = 5.44, SD = 2.13; M_{enjoyment/low skill} = 5.29, SD = 2.42; M_{efficiency/low skill} = 5.66, SD = 2.38; F_signal(1, 393) = 5.57, p = .019, ${\eta }_{\mathrm{p}}^2$  = .01; F_skill(1, 393) = 8.98, p = .003, ${\eta }_{\mathrm{p}}^2$  = .02; F_interaction(1, 393) = 15.67, p < .001, ${\eta }_{\mathrm{p}}^2$  = .04). Although enjoyment (vs. efficiency) increased willingness to pay in the high-skill (handmade) condition (F_{simple effect}(1, 393) = 20.96, p < .001, ${\eta }_{\mathrm{p}}^2$  = .05), it had no effect in the low-skill (automated) condition (F_{simple effect}(1, 393) = 1.22, p = .270, ${\eta }_{\mathrm{p}}^2$  < .01). A test of moderated mediation through quality perceptions using PROCESS Model 7 (Hayes 2022) supported the conclusion that quality perceptions mediated the effect of production enjoyment on willingness to pay only in the high-skill service (index of moderated mediation: b = .66, SE = .17, 95% CI: [.34, 1.00]). Furthermore, when treating quality perceptions as a parallel mediator alongside the alternative mediators, quality's indirect effect remained significantly stronger in the high- versus low-skill conditions (index of moderated mediation: b = .60, SE = .15, 95% CI: [.30, .91]). Additionally, no alternative mediator (love, authenticity, uniqueness, handmade, or effort) was able to explain this pattern of results (i.e., no other measure produced the same moderated mediation). See Web Appendix H for full results.

Studies 4a and 4b Discussion

Across a field study, posttest, and conceptual replication, we found consistent evidence that the positive effect of production enjoyment on willingness to pay, operating through quality perceptions, is attenuated for low-skill (i.e., automated) jobs. For jobs that do not require much skill, production enjoyment no longer signals higher quality, which diminishes its effect on willingness to pay. We note that, in Study 4b, our effects arise relative to a conservative control condition, which may have additional benefits for willingness to pay and quality perceptions compared with baseline (i.e., efficiency, which is beneficial for both the dog and dog owner). In Study 8, we return to the moderating role of skill level. Next, however, we examine how production enjoyment influences the pricing judgments of sellers.

Study 5: Actual Sellers’ Pricing Strategies

Having established the influence of production enjoyment on buyers, we turn our attention to sellers and test the proposition that they charge less for tasks high (vs. low) in production enjoyment. As a first study, we use a sample of real sellers from the online platform Upwork, which connects freelance sellers of services (e.g., graphic design, market research) with prospective buyers. We focus our comparison between high and low enjoyment (rather than high vs. neutral enjoyment) on sellers’ pricing strategies to match the experience of sellers who can experience the full range of production enjoyment.

Method

Participants

We recruited English-speaking participants in the United States by posting an actual job opportunity within the “data entry” category. The job description mentioned that they should only sign up if they do multiple different tasks and charge different prices for different tasks. We had 50 available job openings for a two-cell within-subjects design. A total of 48 participants both accepted the job posting and completed the survey. Excluding those who did not follow instructions (e.g., not writing names of tasks) left a final sample of 44 participants (M_age = 30.09 years, SD = 8.74; 72.7% female, 22.8% male, 4.5% nonbinary).

Procedure

Participants first named two tasks that they do that require similar levels of skill: one that they enjoy more than other Upworkers (high enjoyment) and one that they enjoy less than other Upworkers (low enjoyment). Participants then indicated how much the average Upworker would charge per hour for each of the two tasks. We used these average prices as the midpoints of the willingness to charge scales that sellers subsequently used to indicate their minimum willingness to charge for their high-enjoyment and low-enjoyment tasks. Each scale point deviated by 10% from the average price. For example, for a $20 task, participants were presented with a nine-point scale labeled in increments of 10% of that $20 (i.e., $2), so 1 = “The minimum I would accept is $12 per hour,” 5 = “The minimum I would accept is $20 per hour,” and 9 = “The minimum I would accept is $28 per hour.” Participants answered this question for each of the two tasks and concluded by indicating their age and gender.

Results and Discussion

Results

A paired-samples t-test revealed that sellers charge significantly less for tasks that they enjoy more than others (M_{high enjoyment} = 3.55, SD = 2.14; M_{low enjoyment} = 4.75, SD = 2.45; t(43) = 2.56, p = .014, d = .39). More specifically, participants were willing to accept prices 14.5% less than average for these high-enjoyment tasks. Notably, the low- and high-enjoyment tasks did not differ in the price that participants expected the average Upworker to charge (M_{high enjoyment} = $28.45, SD = $44.37; M_{low enjoyment} = $22.71, SD = $16.86; t < 1).

Discussion

The results of this study demonstrate that real sellers who provide various services online use production enjoyment as an input for their pricing decisions: They charge less for tasks that are relatively more enjoyable for them to perform (vs. less enjoyable). While this study provides ecologically valid evidence of sellers' pricing strategies, it does have a small sample and does not compare these effects relative to matched buyers. Thus, in the next study, we recruit a larger sample of participants and examine both buyers and sellers.

Study 6: Comparing Actual Sellers and Yoked Buyers

This study examines both buyers and sellers within the same study to provide evidence of the pricing asymmetry and does so across a broad range of actual jobs. Using a yoked design, sellers (i.e., participants who have full-time jobs creating products or performing services) are matched with potential buyers (i.e., participants who consider purchasing the offerings of those sellers). This study carries more ecological validity for sellers, who are reflecting on their actual jobs (vs. buyers, who respond to a hypothetical question). We compare tasks the sellers enjoy doing with a control condition involving tasks the seller routinely does and expect a buyer/seller asymmetry such that buyers demonstrate higher willingness to pay for the tasks that sellers enjoy, while sellers demonstrate a lower willingness to accept for the tasks that they enjoy. We also predict that both buyers and sellers assume that high production enjoyment is a signal of quality; however, we expect that sellers do not rely on this inference as much in their pricing decisions.

Method

Participants

For the seller-side sample, we specifically recruited participants who were fully employed outside of survey completion jobs (N = 201; CloudResearch). We excluded 9 participants for failing the IMC and 27 who provided unusable responses (i.e., charging below minimum wage, n = 4; same task mentioned twice, n = 8; no response, n = 2; survey taking as a task, n = 1, otherwise unusable responses [e.g., typically enjoyable tasks that cannot feasibly be charged for such as “my hours,” “Seeing the children actually enjoy what they’re learning about,” “helping others in the department,” etc.], n = 12). We made these exclusions prior to soliciting the buyer-side sample. We further removed 14 participants who did not match with a buyer due to buyer-side randomization. The final seller-side sample included 150 responses (M_age = 40.00 years, SD = 11.61; 50.7% male, 47.3% female, 2.0% other), which were later yoked to a buyer-side sample. For the buyer-side sample, we received 201 CloudResearch participants. We removed 11 participants for failing the IMC, and in cases where multiple buyers saw the same seller's response, we eliminated the second buyer's data (n = 40). A total of 150 buyer-side responses remained (M_age = 41.81 years, SD = 13.59; 57.3% male, 41.3% female, 1.3% other).

Seller-side procedure

Participants with full-time jobs identified the name of their job/profession and how much, in U.S. dollars, someone at their level with this job would “typically earn” per hour. Participants named two tasks core to their job: one task that they enjoy more than most others with this job and a control task (i.e., a routine task they do a lot). The enjoyable (vs. control) task was described with an average of four more characters (M_enjoy = 27.67, SD = 24.76; M_control = 23.49, SD = 19.57; t(149) = 2.34, p = .027, d = .18). Results remain consistent when including the lengths of the responses as covariates.

Next, participants were asked to indicate the minimum amount of money they would need to be paid per hour to do each of the two tasks on a nine-point scale. We used a similar scale from Study 5, where the midpoint showed the “typical” hourly price that they had inputted earlier, and each scale point varied this price by ±10%. We then asked participants to indicate how good they were at each task on a seven-point scale anchored by “Much worse than other people” and “Much better than other people.” Participants then indicated their age and gender and responded to an IMC asking them to perform a simple math calculation.

Buyer-side procedure

Each buyer-side participant saw a response from a randomly selected seller and was asked to imagine that they needed to hire someone with this seller's job. These buyer-side participants were then shown the two tasks and which one the seller indicated as enjoyable versus routine. The buyer-side participants were then asked to consider the maximum amount they would pay this person to do each individual task on the identical, yoked nine-point scale from seller-side participants, such that the midpoint was the “typical earning” amount as indicated by the yoked, seller-side participant and each scale point varied by ±10%. Participants then completed the same quality, demographic, and IMC measures.

Results and Discussion

Buying and selling prices

A repeated-measures ANOVA revealed a nonsignificant main effect of our between-subjects factor of marketplace role (buyer vs. seller; F < 1) and a nonsignificant main effect of our within-subjects factor of task (F(1, 298) = 3.23, p = .073, ${\eta }_{\mathrm{p}}^2$  = .01). Importantly, as Figure 2 illustrates, results revealed the predicted interaction (M_seller/enjoy = 4.66, SD = 1.74; M_{seller/control} = 5.18, SD = 1.18; M_buyer/enjoy = 5.45, SD = 1.95; M_{buyer/control} = 4.58, SD = 2.00; F(1, 298) = 51.73, p < .001, ${\eta }_{\mathrm{p}}^2$  = .15). Decomposing this interaction revealed support for our hypotheses: Buyers were willing to pay more for tasks high in production enjoyment (vs. control; t(149) = 6.42, p < .001, d = .52), while sellers were willing to accept less for them (vs. control; t(149) = −3.78, p < .001, d = .31). To quantify these results, buyers were willing to pay about 9% more for the task that sellers enjoy doing (vs. control), while sellers were willing to accept 5% less for those tasks. These and all subsequent results hold when not eliminating any respondents for imperfect pairing (see Web Appendix I).

Figure 2.

Production Enjoyment Asymmetrically Influences Willingness to Accept/Pay.

Perceived quality

A repeated-measures ANOVA revealed that both buyers and sellers infer high quality from enjoyment (F(1, 298) = 60.03, p < .001, ${\eta }_{\mathrm{p}}^2$  = .17). There was no main effect of marketplace role (F(1, 298) = 1.26, p = .263, ${\eta }_{\mathrm{p}}^2$  < .01), and there was a nonsignificant interaction (M_seller/enjoy = 5.84, SD = .94; M_{seller/control} = 5.45, SD = 1.17; M_buyer/enjoy = 5.83, SD = 1.05; M_{buyer/control} = 5.23, SD = 1.08; F(1, 298) = 2.95, p = .087, ${\eta }_{\mathrm{p}}^2$  = .01). Simple effects revealed that enjoyment is a signal of quality among both buyers and sellers (t_buyers(149) = 7.00, p < .001, d = .57; t_sellers(149) = 4.09, p < .001, d = .33).

Mediation

To conduct within-subject mediation within each marketplace role, we used MEMORE (Montoya and Hayes 2017, Model 1, 5,000 Monte Carlo samples). That is, we tested whether differences in quality perceptions (between the enjoyable and routine task) mediated differences in willingness to pay for buyers and willingness to accept for sellers. Among buyers, differences in quality perceptions stemming from high production enjoyment did mediate differences in willingness to pay for the enjoyment versus control job (b = .20, SE = .08; 95% CI: [.041, .369]); however, this was not the case for sellers (b = −.07, SE = .05; 95% CI: [−.179, .021]). See Web Appendix I for additional results.

Discussion

This study asked sellers to report their willingness to accept prices for tasks within their actual jobs that they enjoy versus perform routinely, which potential buyers then evaluated. Buyers were willing to pay more for tasks that sellers enjoy, whereas sellers were willing to accept less money for tasks they enjoy. In line with our expectations, both buyers and sellers perceived higher quality in the work that was enjoyed; however, only buyers relied on this information in making their pricing judgments. A conceptual replication of this study (Study WA2; N = 1,201) provides additional support for these observed asymmetries when manipulating production enjoyment (high vs. neutral vs. low). That is, Study WA2 shows the direction of the effect of high and low production enjoyment, relative to neutral, on pricing decisions and perceived quality for both buyers and sellers. For sellers, it shows that high production enjoyment leads to lower selling prices than neutral production enjoyment, and neutral production enjoyment leads to lower selling prices than low production enjoyment for sellers. The reverse pattern of effects was true for buyers (i.e., high production enjoyment led to greater buying prices than neutral production enjoyment, and neutral production enjoyment led to greater buying prices than low production enjoyment).

Study 7: Helping Sellers Make Better Pricing Forecasts

One key takeaway from the previous two studies is that sellers may be undercharging for products and services they enjoy producing. This may occur in part because sellers undervalue the weight that buyers place on signals of quality when deciding their willingness to pay. If true, reminding sellers of buyers’ interest in high-quality products and services may increase sellers’ willingness to charge for their high production enjoyment. In other words, sellers associate production enjoyment with high quality, but do not rely on this inference when making pricing decisions (Study 6). By emphasizing the association between high quality and buyers’ willingness to pay, we expect sellers to charge more for high production enjoyment. At the same time, this reminder would have little influence on buyers’ willingness to pay, because they already rely on inferences about quality stemming from production enjoyment in their pricing judgments. We test this possibility in a widely used peer-to-peer network, Amazon Mechanical Turk (MTurk). Here, MTurk workers “sell” the service of completing surveys, while academic researchers “buy” this service. We predict that the asymmetric responses to production enjoyment on pricing decisions will be attenuated when MTurkers (sellers) are reminded of the value that academic researchers (buyers) place on receiving high-quality data, thus prompting MTurkers to increase their willingness to charge for surveys they enjoy.

Method

Seller-side participants

A total of 105 MTurk workers participated in this study (M_age = 40.37 years, SD = 11.60; 47.6% female, 52.4% male). While this study included an IMC (a simple math calculation, which three participants failed), because the buyer-side data (described subsequently) did not include an IMC, we retained all seller-side participants to maintain equivalent exclusion procedures. The results excluding these three participants are consistent (see Web Appendix J).

Seller-side procedure

Participants were told that they would be answering questions about the work that they do for academic researchers on MTurk. Participants were randomly divided into two conditions: quality emphasis versus control.

In the control condition, sellers saw the following prompt: “Some people hate completing surveys for research, and some people really enjoy doing it.” In the quality emphasis condition, sellers additionally read a short prompt emphasizing that “researchers are willing to pay higher prices for survey-takers who provide better, more valuable data” (full prompt in Web Appendix J). All participants then reported their willingness to charge for surveys (1 = “I would definitely charge more if I hated completing surveys,” and 7 = “I would definitely charge more if I really enjoyed completing surveys”). We acknowledge that this measure may not be in line with how MTurk workers make pricing decisions (e.g., accept vs. skip, depending on pay); however, we wanted to keep the language consistent when we later asked it of buyers. Participants finished the study by providing their age and gender and completing the IMC.

Buyer-side participants

A total of 103 academic researchers who all use MTurk to pay individuals to complete surveys participated in this study. Sixty-eight participants were recruited to participate in person by two authors of this research at the 2020 Society for Personality and Social Psychology conference. Demographic information was not recorded for these participants. An additional 38 participants were recruited through an email solicitation sent to academic researchers (M_age = 34.63 years, SD = 5.77; 40% female, 60% male).

Buyer-side procedure

Participants in the buyer condition (i.e., academic researchers) were randomly exposed to one of two conditions (control vs. quality emphasis), where the prompts were identical to those for sellers. They then answered a parallel question to seller-side participants about willingness to pay as a function of production enjoyment (1 = “I would definitely be willing to pay someone more if they hated completing surveys,” and 7 = “I would definitely be willing to pay someone more if they really enjoyed completing surveys”). Participants responding online indicated their age and gender.

Results and Discussion

Results

First, we sought to understand whether the results are consistent with our primary predictions (that buyers are willing to pay more for enjoyment [H₁] while sellers are willing to charge less for enjoyment [H₄]). To that end, we ran two one sample t-tests against the midpoint of the scale (i.e., 4), focusing on buyers and sellers separately while collapsing across the quality emphasis manipulation. The midpoint of the scale reflects a neutral point where enjoyment does not affect pricing. Thus, we predict that buyers would be willing to pay more when sellers enjoyed (vs. did not enjoy) their work, as evidenced by results significantly above the midpoint of the scale. We find support for this prediction (M_buyers = 4.39, SD = 1.60; t(102) = 2.47, p = .015, d = .24). Among sellers, we predict the opposite: sellers would charge less when they experience production enjoyment, as evidenced by results significantly below the midpoint of the scale. We also find support for this prediction (M_sellers = 3.02, SD = 1.92; t(104) = −5.25, p < .001, d = .51).

Next, we examine the role of our quality emphasis manipulation. We again use our willingness to pay measure as the dependent variable, where higher numbers indicate that production enjoyment increases willingness to pay for buyers (researchers) and willingness to charge for sellers (MTurkers). An ANOVA using marketplace role (buyer vs. seller) and prompt (control vs. quality emphasis) revealed a significant main effect of marketplace role such that higher production enjoyment led to higher willingness to pay for buyers versus willingness to charge for sellers (F(1, 204) = 33.00, p < .001, ${\eta }_{\mathrm{p}}^2$  = .14), a significant main effect of prompt (F(1, 204) = 5.09, p = .025, ${\eta }_{\mathrm{p}}^2$  = .02), and a significant interaction (F(1, 204) = 6.32, p = .013, ${\eta }_{\mathrm{p}}^2$  = .03; M_{seller/control} = 2.44, SD = 1.58; M_{seller/quality emphasis} = 3.58, SD = 2.06; M_{buyer/control} = 4.42, SD = 1.47; M_{buyer/quality emphasis} = 4.36, SD = 1.72). In the control condition, there was a large difference in how buyers and sellers respond to production enjoyment (F_{simple effect}(1, 204) = 33.45, p < .001, ${\eta }_{\mathrm{p}}^2$  = .14), while in the quality emphasis condition, in line with expectations, this difference was significantly reduced (F_{simple effect}(1, 204) = 5.32, p = .022, ${\eta }_{\mathrm{p}}^2$  = .03). Put differently, emphasizing the importance of quality to sellers (MTurkers) increased their willingness to charge for high (vs. low) production enjoyment as theorized (F_{simple effect}(1, 204) = 11.50, p < .001, ${\eta }_{\mathrm{p}}^2$  = .05), while buyers (researchers) were unaffected by the manipulation (F_{simple effect} < 1).

Discussion

Using a real peer-to-peer marketplace and sampling actual buyers and sellers within it, this study further supports a buyer/seller asymmetry in pricing that stems from the two parties’ responses to production enjoyment signals. At baseline, buyers (i.e., researchers) were willing to pay more when sellers (i.e., MTurkers) enjoy producing a service, while these sellers were willing to charge less when they enjoy doing so. However, when sellers were reminded of the influence of quality in buyers’ pricing decisions, they relied more on production enjoyment's signal of quality in their willingness to charge, and their pricing decisions more closely aligned with buyers’ willingness to pay. Although this study does have limitations (e.g., the specific language of the prompts/questions; the potential difficulty for sellers to connect “charging for a task” to the reality of accepting a task at the listed payment), they cannot explain the focal interaction.

Study 8: The Influence of Enjoyment on Buyers and Sellers Across 100 Jobs

In our final study, we seek to generalize the asymmetry in buyers’ and sellers’ pricing judgments across a wide variety of jobs. To do this, we derived a list of 100 of the most common consumer-facing jobs, from accountants to woodworkers, and treat these jobs as our unit of analysis, examining how high versus low production enjoyment within them affects buyers’ and sellers’ pricing decisions. Additionally, we use this study to explore moderators of the buyer/seller asymmetry. In particular, we reexamine skill level (Studies 3 and 4) now for both buyers and sellers. We also explore other possible moderators that provide managerial insight on the types of jobs where buyers are most likely to exhibit a higher willingness to pay for production enjoyment and sellers are most likely to exhibit a lower willingness to charge. These include the degree to which the job involves creativity, is observable, is oriented around products or services, is desirable, and is well-paid.

Method

Stimuli selection

We created our list of the 100 most common consumer-facing jobs by first scraping a list of all jobs listed on the U.S. Bureau of Labor Statistics (2020) website (747 jobs; bls.gov). We then asked a research assistant unaware of the hypotheses to remove jobs where individual consumers would be very unlikely to directly pay sellers for producing the product or service (e.g., astronaut, diplomat; 146 jobs remaining). We further refined this list by asking MTurk respondents (N = 182) to evaluate how likely a consumer is to buy products or services from someone with each job, and we selected the 100 jobs that scored highest on this measure. These various jobs involved the production of a diverse array of products/services (e.g., chef, florist, housekeeper, massage therapist; see Web Appendix K for procedural details).

Participants

We first recruited 304 MTurk workers through Positly, a participant recruitment platform. After removing those who failed the IMC, we had a final sample of 294 participants (M_age = 40.60 years, SD = 12.17; 45.9% female, 53.8% male, .3% missing/other). To examine possible moderators, we recruited an additional 897 MTurk workers through Positly, removed those who failed the same IMC, and examined a final sample of 829 participants (M_age = 41.48 years, SD = 12.26; 49.6% female, 49.6% male, .8% other).

Main study procedure

Participants were randomly assigned to a marketplace condition: buyer versus seller. Those in the buyer condition imagined that they needed to hire someone with a particular job (randomly selected from the list of 100 jobs) and then responded to the following willingness-to-pay question: “In which case would you pay more, if the person really didn’t enjoy doing the work you hired them for, or if they really did enjoy doing the work you would hire them for?” (in increments of 5%, 1 = “I would pay 20% more for the job if they didn’t enjoy the work,” 5 = “The amount of money I would pay would be the same,” and 9 = “I would pay 20% more for the job if they enjoyed the work”).

Participants in the seller condition imagined that they had a particular job (randomly selected from the list of 100 jobs) and were available for hire. They answered an analogous willingness-to-charge question: “In which case would you charge more, if you really didn’t enjoy doing the work you would be hired for, or if you really did enjoy doing the work you would be hired for?” (on a scale with the same 5% increments, 1 = “I would charge 20% more for the job if I didn’t enjoy the work,” 5 = “The amount of money I would charge would be the same,” 9 = “I would charge 20% more for the job if I enjoyed the work”). All participants repeated this procedure for 40 randomly selected jobs, then they responded to an IMC asking them to perform a simple math calculation. Positly, the participant recruitment platform, provided participants’ age and gender.

Moderators procedure

Each participant answered one of the following questions about 40 randomly selected jobs: (1) “How much skill does it take to perform this job effectively?” (1 = “not much skill at all,” and 7 = “a lot of skill”), (2) “How much creativity is involved in doing this job?” (1 = “not much creativity involved at all,” and 7 = “a lot of creativity involved”), (3) “To what extent can customers directly observe the person while they are performing this job?” (1 = “cannot be observed at all,” and 7 = “can very much be observed”), (4) “To what extent does this job involve creating a product versus performing a service?” (1 = “entirely involves creating a product,” and 7 = “entirely involves performing a service”), (5) “How desirable is it to have this job?” (1 = “not at all desirable,” and 7 = “very desirable”), and (6) “On average, would you describe this job as low paying or high paying?” (1 = “very low paying,” and 7 = “very high paying”). Finally, participants answered the same IMC question, and reported their age and gender.

Results and Discussion

Main study results

Following prior research (Kruger 1999; Kruger and Dunning 1999), we conducted the main analysis at the job-level (vs. participant-level). That is, by aggregating participants’ ratings of each job (an average of 50.83 participants [SD = 5.50] rated each job), we computed average willingness to pay and willingness to charge for each individual job. This allowed us to then treat each job as a single data point, where our final data set for this main analysis consisted of 100 rows (one for each job) and two columns, one for mean willingness to pay and one for mean willingness to charge for each job as a function of production enjoyment. We then used a paired-samples t-test to compare the influence of production enjoyment on buyers’ willingness to pay versus sellers’ willingness to charge.

From this analysis, we replicated the predicted asymmetry: Buyers were willing to pay more for high production enjoyment (relative to low enjoyment), while sellers were willing to charge less for high production enjoyment (relative to low enjoyment; M_buyer = 5.68, SD = .30; M_seller = 4.03, SD = .49; t(99) = 41.39, p < .001, d = 4.14). Among buyers, a one-sample t-test against the midpoint of the scale (i.e., 5, representing no differences in willingness to pay based on production enjoyment) revealed that buyers were willing to pay a premium for high production enjoyment (3.38% more; t(99) = 188.58, p < .001, d = 2.24). Among sellers, the same analysis revealed that they would instead charge less for high production enjoyment (4.87% less; t(99) = 82.89, p < .001, d = 2.01). An additional participant-level analysis directly compared pricing judgments of buyers and sellers within each job using a series of independent samples t-tests. These analyses revealed significant differences between how much buyers would be willing to pay versus sellers would willing to charge for high (vs. low) production enjoyment for 99 of the 100 jobs, and a marginal difference for the one remaining job (see Web Appendix K for the results for each job).

Moderator results

Again, we first aggregated job-level responses by averaging raters’ responses to each question within each job (an average of 55.2 participants [SD = 6.06] rated each question for each job). We then conducted two separate multiple linear regressions. The first regression assessed how each of the six aspects of the jobs affect buyers’ tendency to pay for high production enjoyment. Replicating Studies 3–4b, buyers were willing to pay more for high production enjoyment when that production is associated with a more skillful job (b = .110, SE = .039, t(93) = 2.81, p = .006). For the exploratory moderators, buyers were willing to pay more for high production enjoyment when that production involves more creativity (b = .131, SE = .025, t(93) = 5.27, p < .001), is more observable (b = .103, SE = .024, t(93) = 4.34, p < .001), is more service-based (vs. product-based, b = .058, SE = .023, t(93) = 2.46, p = .016), is part of a more desirable job (b = .129, SE = .037, t(93) = 3.53, p = .001), and is associated with a lower-paying job (b = −.096, SE = .033, t(93) = −2.95, p = .004).

The second regression assessed how these six aspects affect sellers’ tendency to charge for production enjoyment. Here, sellers were willing to charge less for high production enjoyment when that production requires more skill (b = −.121, SE = .057, t(93) = −2.12, p = .036), involves less creativity (b = .085, SE = .036, t(93) = 2.36, p = .021), and is part of a less desirable job (b = .450, SE = .053, t(93) = 8.50, p = .001). The three remaining moderators (observability, product vs. service, payment) were not significant (ts < 1).

Discussion

Across 100 consumer-facing jobs—including jobs frequently represented in peer-to-peer marketplaces like Upwork and Freelancer—buyers were willing to pay more when the seller enjoys the work, while sellers charged less when they enjoy the work. Although the degree of seller control over setting prices may vary across the jobs examined, these results highlight the striking generalizability of the results across many types of jobs.

From our moderation analyses, first, we further emphasized the role of the job's skill level on buyer-side effects (i.e., production enjoyment increases buyers’ willingness to pay more for high-skill jobs), while also showing how it affects sellers (i.e., it decreases sellers’ willingness to charge for high-skill jobs). Notably, this suggests that high-skill jobs may be a domain where sellers are particularly unlikely to take advantage of production enjoyment as a signal. We also provided some initial evidence for other, managerially relevant moderators. Specifically, production enjoyment increases buyers’ willingness to pay more for highly observable, creative, service-oriented, desirable, and low-paying jobs. For sellers, production enjoyment decreased their willingness to charge more for less creative and desirable jobs.

General Discussion

In the current work, we document an asymmetry in buying and selling prices driven by an often fundamental aspect of the seller's experience: production enjoyment. We present nine studies that explore the distinct influence of production enjoyment on both buyers and sellers. We show that buyers respond positively to signals of production enjoyment (even compared with other positive signals of quality) across a variety of response types (e.g., willingness to pay and product choice; H₁), because high production enjoyment serves as a signal of high quality (Studies 2, 3, 4b, and 6; H₂₎. Sellers, however, accept and charge lower prices following production enjoyment (H₄). This buyer/seller asymmetry is robust across a selection of the 100 most common consumer-facing jobs (Study 8) and a wide range of participants’ own jobs (Studies 5–6), including both products and services.

Several of these studies compare the signal of high production enjoyment with other signals that sellers use to signal quality (e.g., effort, popularity; Studies 2, 4a–b). In our operationalizations, production enjoyment emerges as more effective than these (often costlier) signals. We further showed the unique effect of production enjoyment on buyers’ willingness to pay through quality inferences, which remained significant when also including the effect of alternative explanations (e.g., love, authenticity, effort, uniqueness; Studies 2, 4a–b, and WA2).

The findings also shed light on when signaling high production enjoyment is more versus less likely to attract buyers. That is, production enjoyment boosts buyers’ willingness to pay for products and services that require relatively high levels of skill (Studies 3, 4, and 8; H₃), meaning that sellers should use other signals of quality for jobs perceived to be low in skill, such as those that employ machine automation. Further results from our final study suggest that production enjoyment may have the greatest impact on buyers’ willingness to pay when that production is more observable, creative, and low-paying, and involves more desirable, service-based jobs.

Theoretical Implications

The current research establishes production enjoyment as a novel cue that differentially affects buyers and sellers. This contributes to and connects disparate streams of literature that examine (1) the role of enjoyment in consumer behavior, (2) buyer/seller asymmetries in pricing, and (3) how cues, especially those relating to the creation process, and subsequent inferences guide judgments and decision-making.

Enjoyment is a key construct influencing both consumers and creators (Alba and Williams 2013; Csikszentmihalyi 1990). While the majority of prior research has examined either the antecedents to enjoyment (Alba and Williams 2013; Dahl and Moreau 2007; Sellier and Dahl 2011) or enjoyment's consequences for consumers’ goal pursuit (Custers and Aarts 2010; Woolley and Fishbach 2016), we build most directly on research that examines enjoyment as a central variable in the creation process (Atakan, Bagozzi, and Yoon 2014; Dahl and Moreau 2007; Sellier and Dahl 2011). The current research highlights its asymmetric effect on buyers versus sellers, offering a nuanced understanding of the consequences of production enjoyment for both sides of the marketplace.

By exploring this buyer/seller asymmetry, these results also build on previous research on buyer/seller differences, most notably the endowment effect (Carmon and Ariely 2000; Morewedge and Giblin 2015; Nayakankuppam and Mishra 2005). Our results contribute to research about how marketplace roles influence the type of information that most readily influences pricing judgments (e.g., Carmon and Ariely 2000; Johnson, Häubl, and Keinan 2007), demonstrating that buyers and sellers both infer quality from production enjoyment, but this only mediates willingness to pay (vs. charge). We show a rare occasion where buyers and sellers do not simply weigh information about the product or service differently; they actually have opposite responses to production enjoyment. Our findings therefore suggest that production enjoyment likely moderates the endowment effect for sellers who produce a product/service, such that high production enjoyment could attenuate this classic effect.

Finally, we build on research on consumers’ inference-making in judgments of value, and specifically on variables associated with the production of products and services. For example, consumers value goods that were made by hand (Fuchs, Schreier, and Van Osselaer 2015), that were particularly effortful or time-consuming to create (Buell and Norton 2011; Kruger et al. 2004), that were user-designed (Nishikawa et al. 2017), and that were made in the original factory (Newman and Dhar 2014). Our findings suggest that production enjoyment may serve as an important moderator of these effects. For example, handmade products from identifiable sellers (Van Osselaer et al. 2020) might be particularly valuable if the seller enjoyed the production process. Our findings similarly suggest that buyers might be less likely to respond positively to information about high production effort (Buell and Norton 2011; Kruger et al. 2004) when that effort implies a lack of production enjoyment. There is less research on how production-related cues such as these impact sellers, and we highlight the potential disconnect between buyers’ and sellers’ responses to characteristics of the production experience.

Practical Implications

The current research illuminates a novel factor that drives buyers’ willingness to pay, and in doing so, offers guidance for sellers seeking to maximize their profits. First, awareness of this pricing asymmetry may benefit sellers. If sellers charge more for low (vs. high) production enjoyment, they may be missing out on potential profits by undervaluing, relative to buyers’ perceptions, the products/services they enjoy producing/providing. Based on our findings, sellers might reconsider their willingness to charge for different products/services. This knowledge can also benefit buyers, who can recognize production enjoyment as a signal of a potential discount, because sellers charge less when they enjoy the production process.

Second, this research points to an important tool that sellers can leverage given buyers’ inference-making tendencies, particularly in peer-to-peer marketplaces or other vertically integrated markets where a single actor is responsible for both creating and selling the item. Such industries (Etsy, OnlyFans, Fiverr, etc.) are becoming increasingly popular and are estimated to be worth $355 billion by 2025 (PricewaterhouseCoopers 2015). Here, as well as on social media more generally, sellers have control about how to present themselves and their production process to prospective buyers. If sellers mention high production enjoyment in their profiles and marketing, buyers may be more interested in their products and services. Thus, we shed light on Shakespeare's ([1590–1592] 18871590,1592,1887undefined, 1.1.40) line in The Taming of the Shrew that “No profit grows where is no pleasure ta’en.”

While costlier signals are often more credible (Kirmani and Rao 2000), we find that production enjoyment is interpreted as a credible signal of product quality despite its seemingly low cost to sellers. However, signaling production enjoyment can be quite “costly” in different ways. First, it is dependent on how the message reaches consumers. For example, mentioning production enjoyment during customer interactions or on one's Etsy page may not be very costly, but mentioning production enjoyment via an advertising campaign can become quite expensive for sellers. Second, additional “costs” may arise if consumers feel misled by the seller's claims and/or a product's quality does not live up to expectations (Kopalle and Lehmann 2006). Third, in cases where a seller's work is very easily observable, signaling production enjoyment elsewhere can carry costs in terms of requiring inauthentic, overt signals of enjoyment (i.e., emotional labor; Grandey 2000). However, if production enjoyment is genuine, observability can benefit sellers. Study 8 supports this assertion, where we find that production enjoyment has a bigger impact on willingness to pay for more observable jobs. In summary, a better understanding of the asymmetries that arise as buyers and sellers evaluate production enjoyment can help sellers make more informed decisions about the cues they send and when and why to use production enjoyment as a cue.

Future Directions

Future research should investigate potential commonalities and distinctions between enjoyment and related variables such as intrinsic motivation (Bhattacharjee et al. 2014; Woolley and Fishbach 2016), passion in the workplace (Kim et al. 2020), and effort (Kruger et al. 2004), and to consider how these variables might operate differently for buyers and sellers. In our work, we showed that inferred quality drives buyers’ increased willingness to pay for production enjoyment; however, this is not the case for sellers. We speculate that sellers directly incorporate the affective utility of enjoyment into their pricing decisions, reducing their willingness to charge. Since pricing is a key component of an effective marketing mix, we recommend that future research consider the process by which sellers incorporate enjoyment and other related variables (e.g., passion, effort) into their transaction utility and pricing decisions. Additionally, while we differentiate the effect of production enjoyment through high quality from perceptions that the product “contains love,” colloquially stating that “I love my work” simultaneously reflects a high amount of production enjoyment. Future work could better illuminate when signals like this drive one inference (e.g., love) or the other (e.g., enjoyment).

Future research can also explore when information about production enjoyment works best as a messaging strategy. In our studies, we showed that production enjoyment positively affects buyers when used both in explicit advertising (Studies 1 and 2) as well as messages directly from the provider (Studies 3 and 4). However, it might be particularly influential when communicated both directly and indirectly (e.g., through nonverbal cues), as production enjoyment seems to have a bigger impact on willingness to pay when the seller's work is observable (Study 8). Future research is also required to further test the other exploratory moderators we observed in Study 8. The finding that buyers respond best to production enjoyment for creative jobs is quite consistent with our theorizing about the role of skill and quality in the effect, but more research is needed to determine why production enjoyment also works best as a signal for desirable, low-paying, and service-based jobs. These may all be domains where the skills of the producer vary more, or have more impact on product quality, but these moderators may also hint at other mechanisms.

In Study WA1 (Web Appendix B), we document the results of an additional field study run on Facebook showing that production enjoyment does not depend on whether a single person versus a collection of people signal production enjoyment. However, these effects may depend on other related variables. For example, the effects may be attenuated for nonidentified creators (Small and Loewenstein 2003) and for looser (vs. tighter) groups of creators (Smith, Faro, and Burson 2013). Finally, future research could examine the role of persuasion knowledge (Friestad and Wright 1994) in these effects. Is production enjoyment a stronger cue than others we tested because it is less likely to trigger consumers’ persuasion knowledge? Does general skepticism limit the effectiveness of production enjoyment as a cue in the same way as other frequently used cues in advertising (e.g., “I have the lowest prices”)? Regardless of the answers to those questions, we would like to emphasize to readers how much we enjoyed conducting the current research.

Supplemental Material

sj-pdf-1-jmx-10.1177_00222429241257913 - Supplemental material for Production Enjoyment Asymmetrically Impacts Buyers’ Willingness to Pay and Sellers’ Willingness to Charge

Supplemental material, sj-pdf-1-jmx-10.1177_00222429241257913 for Production Enjoyment Asymmetrically Impacts Buyers’ Willingness to Pay and Sellers’ Willingness to Charge by Anna Paley, Robert W. Smith, Jacob D. Teeny and Daniel M. Zane in Journal of Marketing