Service Provider Salience: When Guilt Undermines Consumer Willingness to Buy Time

=1592459 Epp, A. M., & Velagaleti, S. R. (2014). Outsourcing parenthood? How families manage care assemblages using paid commercial services. Journal of Consumer Research, 41(4), 911–935. DOI: https://doi.org/10. 1086/677892 Fassiotto, M., Li, J., Maldonado, Y., & Kothary, N. (2018). Female surgeons as counter stereotype: The impact of gender perceptions on trainee evaluations of physician faculty. Journal of Surgical Education, 75(5), 1140–1148. DOI: https://doi.org/10.1016/j. jsurg.2018.01.011 Flynn, F. J., & Schaumberg, R. L. (2012). When feeling bad leads to feeling good: Guilt-proneness and affective organizational commitment. Journal of Applied Psychology, 97(1), 124. DOI: https://doi.org/10.1037/ a0024166 Furnham, A. (1982). The Protestant work ethic and attitudes towards unemployment. Journal of Occupational and Organizational Psychology, 55(4), 277–285. DOI: https://doi.org/10.1111/j.2044-8325.1982.tb0 0101.x Furnham, A. (1984). Personality and values. Personality and individual differences, 5(4), 483–485. DOI: https://doi.org/10.1016/0191-8869(84)90018-7 Gino, F. (2017). Uber Shows How Not to Apply Behavioral Economics. Harvard Business Review. Retrieved March 29, 2019 from https://hbr.org/2017/04/ uber-shows-how-not-to-apply-behavioral-economics Gleason, M. E., Iida, M., Bolger, N., & Shrout, P. E. (2003). Daily supportive equity in close relationships. Personality and Social Psychology Bulletin, 29(8), 1036– 1045. DOI: https://doi.org/10.1177/014616720325 3473 Grant, A. M. (2008). Does intrinsic motivation fuel the prosocial fire? Motivational synergy in predicting persistence, performance, and productivity. Journal of applied psychology, 93(1), 48. DOI: https://doi. org/10.1037/0021-9010.93.1.48 Greenberg, J. (1977). The Protestant work ethic and reactions to negative performance evaluations on a laboratory task. Journal of Applied Psychology, 62(6), 682. DOI: https://doi.org/10.1037/0021-9010.62.6. 682 Hershfield, H. E., Mogliner, C., & Barnea, U. (2016). People Who Choose Time Over Money Are Happier. Social Psychological and Personality Science, 7(7), 697–706. DOI: https://doi.org/10.1177/1948550616649239 Hochschild, A. R. (2012). The outsourced self: Intimate life in market times. New York: Metropolitan Books. Hsee, C. K., Zhang, J., Cai, C. F., & Zhang, S. (2013). Overearning. Psychological Science, 24(6), 852–859. DOI: https://doi.org/10.1177/0956797612464785 Jeffrey, S. A. (2009). Justifiability and the motivational power of tangible noncash incentives. Human Performance, 22(2), 143–155. DOI: https://doi.org/ 10.1080/08959280902743659 Kogut, T., & Ritov, I. (2005). The “identified victim” effect: An individual group or just a single individual. Journal of Behavioral Decision Making, 18(3), 157–167. DOI: https://doi.org/10.1002/bdm.492 Lang, A., & Yegiyan, N. S. (2008). Understanding the interactive effects of emotional appeal and claim strength in health messages. Journal of Broadcasting & Electronic Media, 52(3), 432–447. DOI: https://doi. org/10.1080/08838150802205629 Lee, A. Y., & Aaker, J. L. (2004). Bringing the frame into focus: the influence of regulatory fit on processing fluency and persuasion. Journal of personality and social psychology, 86(2), 205. DOI: https://doi. org/10.1037/0022-3514.86.2.205 Lee-Yoon, A. J., & Whillans, A. V. (2018). Making seconds count: When valuing time promotes subjective wellbeing. Current Opinion in Psychology, 26, 54–57. DOI: https://doi.org/10.1016/j.copsyc.2018.05.002 Mogilner, C. (2010). The pursuit of happiness: Time, money, and social connection. Psychological Science, 21(9), 1348–1354. DOI: https://doi.org/10. 1177/0956797610380696 Mogilner, C., Whillans, A., & Norton, M. I. (2018). Time, Money, and Subjective Well­being: Handbook of Well­ being. Edited by Diener, Oishi, & Tay. Mosher, D. L. (1980). Guilt. Encyclopaedia of clinical assessment, RH Woody editions. Oishi, S. (2010). The psychology of residential mobility: Implications for the self, social relationships, and wellbeing. Perspectives on Psychological Science, 5(1), 5–21. DOI: https://doi.org/10.1177/1745691609356781 Pfeffer, J., & DeVoe, S. E. (2009). Economic evaluation: The effect of money and economics on attitudes about volunteering. Journal of Economic Psychology, 30(3), 500–508. DOI: https://doi.org/10.1016/j. joep.2008.08.006 Rao, T. S. S., & Indla, V. (2010). Work, family or personal life: Why not all three? Indian Journal of Psychiatry, 52(4), 295–297. DOI: https://doi. org/10.4103/0019-5545.74301 Roxburgh, S. (2004). ‘There Just Aren’t Enough Hours in the Day’: The Mental Health Consequences of Time Pressure. Journal of health and social behavior, 45(2), 115–131. DOI: https://doi.org/10.1177/002214650 404500201 Sandstrom, G. M., & Dunn, E. W. (2014). Social interactions and well-being: The surprising power of weak ties. Personality and Social Psychology Bulletin, 40(7), 910–922. DOI: https://doi.org/10. 1177/0146167214529799 Shah, A. K., Mullainathan, S., & Shafir, E. (2012). Some consequences of having too little. Science, D ow naded rom http://online.ucpress.edu/le-pdf/6/1/28/442008/252-4582-2-pb.pdf by gest on 05 D ecem er 2021 Whillans et al: Outsourcing and Guilt Art. 28, page 19 of 19 338(6107), 682–685. DOI: https://doi.org/10.1126/ science.1222426 Shockley, K. M., Smith, C. R., & Knudsen, E. A. (2017). The Impact of Work–Life Balance on Employee Retention. In H. W. Goldstein, E. D. Pulakos, J. Passmore & C. Semedo (eds.), Selection and Employee Retention, New York: John Wiley & Sons. DOI: https:// doi.org/10.1002/9781118972472.ch24 Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2011). False-positive psychology: Undisclosed flexibility in data collection and analysis allows presenting anything as significant. Psychological science, 22(11), 1359–1366. DOI: https://doi.org/10.1177/0956797611417632 Small, D. A., & Loewenstein, G. (2003). Helping a victim or helping the victim: Altruism and identifiability. Journal of Risk and Uncertainty, 26(1), 5–16. DOI: https://doi.org/10.1023/A:1022299422219 Tangney, J. P., Miller, R. S., Flicker, L., & Barlow, D. H. (1996). Are shame, guilt and embarrassment distinct emotions? Journal of Personality and Social Psychology, 70, 1256–1269. DOI: https://doi.org/ 10.1037/0022-3514.70.6.1256 Walsh, B. (2011). Today’s smart choice: Don’t own. Share. Times Magazine, Retrieved April 7, 2019 from http://content.time.com/time/specials/packages/ article/0,28804,2059521_205 9717,00.html Wheeler, C. S., Petty, R. E., & Bizer, G. Y. (2005). Selfschema matching and attitude change: Situational and dispositional determinants of message elaboration. Journal of Consumer Research, 31(4), 787–797. DOI: https://doi.org/10.1086/426613 Whillans, A. V. (2019). Time for happiness: Why the pursuit of money isn’t bringing you joy–and what will. Harvard Business Review. Retrieved March 29, 2019 from https:// hbr.org/cover-story/2019/01/time-for-happiness Whillans, A. V., Dunn, E. W., & Norton, M. (2018). Overcoming barriers to time-saving: Reminders of future business encourage consumers to buy time. Social Influence, 13, 117–124. DOI: https://doi.org/10. 1080/15534510.2018.1453866 Whillans, A. V., Dunn, E. W., Smeets, P., Bekkers, R., & Norton, M. I. (2017). Buying time promotes happiness. Proceedings of the National Academy of Sciences, 201706541. DOI: https://doi.org/10.1073/ pnas.1706541114 Whillans, A. V., Weidman, A. C., & Dunn, E. W. (2016). Valuing time over money is associated with greater happiness. Social Psychological and Personality Science, 7(3), 213–222. DOI: https://doi. org/10.1177/1948550615623842 White, K., & Peloza, J. (2009). Self-benefit versus otherbenefit marketing appeals: Their effectiveness in generating charitable support. Journal of Marketing, 73(4), 109–124. DOI: https://doi.org/10.1509/ jmkg.73.4.109 Zauberman, G., Kim, K. B., Malkoc, S. A., & Bettman, J. R. (2009). Discounting time and time discounting: Subjective time perception and intertemporal preferences. Journal of Marketing Research, 46(4), 543– 556. DOI: https://doi.org/10.1509/jmkr.46.4.543 Zauberman, G., & Lynch, J. G. (2005). Resource slack and propensity to discount delayed investments of time versus money. Journal of Experimental Psychology: General, 134(1), 23. DOI: https://doi. org/10.1037/0096-3445.134.1.23 Zhao, X., Lynch, J. G., Jr., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of Consumer Research, 37(2), 197–206. DOI: https://doi.org/10. 1086/651257 How to cite this article: Whillans A. V., Lee-Yoon, A. J., & Dunn, E. W. (2020). Service Provider Salience: When Guilt Undermines Consumer Willingness to Buy Time. Collabra: Psychology, 6(1): 28. DOI: https://doi.org/10.1525/collabra.252 Senior Editor: Don Moore Editor: Oleg Urminsky Submitted: 08 April 2019 Accepted: 31 May 2020 Published: 30 June 2020 Copyright: © 2020 The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. See http://creativecommons.org/licenses/by/4.0/. Collabra: Psychology is a peer-reviewed open access journal published by University of California Press. OPEN ACCESS Peer review comments The author(s) of this paper chose the Open Review option, and the peer review comments can be downloaded at: http://doi.org/10.1525/collabra.252.pr D ow naded rom http://online.ucpress.edu/le-pdf/6/1/28/442008/252-4582-2-pb.pdf by gest on 05 D ecem er 2021

In 2011, Time Magazine rated the sharing economy as one of the top 10 ideas that would change the world (Walsh, 2011). Today, the possibility of outsourcing just about anything from grocery shopping to dog walking, to standing in line for the latest iPhone is only a few clicks away. Companies such as TaskRabbit and Hello Alfred enable customers to outsource nearly any household chore by connecting people who need tasks done with people who have time to do them. With the growing popularity of the sharing economy, it has never been easier for consumers to outsource dreaded tasks to others. Despite the rise of the sharing economy, little is known about factors that predict whether individuals decide to 'buy time. ' This question is increasingly important in light of the fact that most working adults feel increasingly pressed for time. Large-scale survey data suggests that 80% of working adults in the US feel "time poor"-like they have too many things to do and not enough time to do them (Whillans, 2019). Employees who feel overwhelmed with the demands of work and life report lower happiness, are less productive, and more prone to depression and anxiety (Rao & Indla, 2010, Roxburgh, 2004, Shockley, Smith & Knudson, 2017). Public health researchers have even highlighted time poverty as one of the most important emerging social trends underlying rising rates of obesity (Banwell, Hinde, Dixon & Sibthorpe, 2005).

Benefits of buying time
As a potential solution to rising rates of time poverty, research provides evidence that time-saving services can reduce time stress and promote happiness (Whillans, Dunn, Smeets, Bekkers & Norton, 2017). In surveys conducted with working adults from Canada, Denmark, the US, and the Netherlands (N = 6,613), respondents who spent money on time-saving services, such as by outsourcing disliked tasks to others, reported greater life satisfaction and positive mood (Whillans et al., 2017). These effects held across the income spectrum: even respondents with very little discretionary income reported greater well-being when they spent money on time-saving purchases. In a related experiment, participants reported greater positive mood after spending $40 on a timesaving purchase (e.g., grocery delivery) as compared to after spending $40 on a material purchase for themselves, in part because time-saving purchases protected people from the negative impact of time-stress on happiness (Whillans et al., 2017).
Consistent with these findings, there has been a trend for organizations rewarding employees with time-saving services. As part of a recent initiative, Stanford University conducted a small pilot study where they rewarded busy doctors with vouchers for time-saving services such as grocery delivery and house cleaning services (Fassiotto, Maldonado, & Kothary, 2018). Doctors who received vouchers for time-saving services reported greater work-life balance, received more awards, and were less likely to quit as compared to matched controls. These results suggest that organizations might benefit from both offering and encouraging employees to redeem for time-saving rewards.
Despite the psychological and productivity benefits of time-saving services, people often fail to spend their money on time-saving services even when they can afford to do so. In a nationally representative sample of employed Americans living in the US (N = 1,265), only 17% of respondents spent money on time-saving services; yet, 99% of respondents could think of a disliked task that they wanted to outsource such as cooking, cleaning, or shopping (Whillans et al., 2017). An obvious explanation for these findings is that respondents could not afford to make these purchases. Yet, even among the over 800 millionaires surveyed as part of this research, only 52% reported spending any money to outsource their disliked tasks to others.
Similar patterns emerge in the workplace. When employees are provided with the choice between material goods and time-saving services, employees often select material goods or gift cards. In a recent study conducted with 77 US companies that offered time-saving rewards to employees, only 3.2% of employees redeemed reward points for these services, whereas 67% of employees redeemed reward points for material goods (In this study, 18% of employees redeemed for experiences and 13% of employees donated their reward points to charity; Whillans, Dunn & Norton, 2018). These findings point to the possibility that psychological barriers prevent individuals from selecting time-saving services in their daily life.
Consistent with this proposition, recent research sheds light on the psychological factors that might prevent people from spending money on time-saving services (see Lee-Yoon & Whillans, 2018;Mogilner, Whillans & Norton, 2018 for related reviews). For example, one welldocumented reason that people do not spend money on time-saving purchases is because they often fail to realize that they will be as busy in the future as they are in the present moment ('future time slack'; Zauberman & Lynch, 2005;Zauberman, Kim, Malkoc, & Bettman, 2009). Following these findings, people are more likely to make time-saving purchases when they are prompted to see their future as more similar to the busy present (Whillans, Dunn & Norton, 2018). These results point to the role of 'future time slack' in explaining why people often fail to make time-saving purchases.

Guilt as an Unexplored Barrier to Buying Time
Building on this research, we examine another potential barrier to making time-saving purchases: feelings of guilt. Initial qualitative research supports the role of guilt in major outsourcing decisions, such as couples' decisions about whether or not to outsource childcare (Epp & Velagaleti, 2014). Relatedly, we recently conducted a short pilot study with N = 153 employed Americans. In this pilot study, respondents who were familiar with TaskRabbit and Amazon, and had both services available to them, were asked to make a hypothetical choice between receiving an $80 Amazon gift card or a TaskRabbit gift card from their employer. Replicating previous research, nearly all of our respondents selected the Amazon Gift Card (99%). Most central to our argument, a majority of respondents (52%) stated that the primary reason that they chose the Amazon gift card was because they felt guilty about burdening a service provider with their household chores. This data is available through the Open Science Framework (OSF; https://osf.io/vp2kx/).
Following these results, we propose that time-saving purchases are most likely to promote feelings of guilt when these purchases involve burdening other people with daily tasks. Of course, the strength of this feature varies within purchase category: in the real world, time-saving purchases vary in the extent to which they highlight the effort of the service provider. For example, housecleaning services often involve directly outsourcing to another person, whereas purchasing pre-cooked meals less directly involves the effort of another person. People themselves may also vary in the extent to which they see the effort of time-saving purchases as a benefit (vs. burden) to the service provider.
To begin to unpack these issues, we turned to the social support literature. Research in social psychology shows that receiving social support from close romantic partners is most beneficial when support is invisible. That is, people benefit most when their partner provides social support, but they themselves do not take notice. In a seminal study, individuals faced with a major stressor (e.g., increased workload) adjusted more successfully when their partners reported providing social support, but they themselves did not report receiving social support (Bolger, Zuckerman & Kessler, 2000). Furthermore, experimental research shows that supportive behaviors are most effective when these behaviors are accomplished without the recipients' awareness (Bolger & Amarel, 2007). The receipt of visible social support can create feelings of indebtedness, and most critically to our argument, can increase feelings of guilt (Bolger & Amarel, 2007;Gleason, Iida, Bolger, & Shrout, 2003).
Further indirect evidence for the idea that people might fail to buy time because they feel guilty about burdening the service provider comes from research on the identifiable victim effect. This research shows that people experience greater emotional distress when faced with the suffering of an identifiable person vs. a broad group (Kogut & Ritov, 2005;Small & Loewenstein, 2003). People exhibit greater concern about a single individual in need of medical help than about an unidentified group of people in need of help.
Extrapolating from these findings, we propose that people may feel elevated discomfort about buying their way out of daily tasks when doing so entails thinking about service providers. These feelings of discomfort should emerge more to the extent that individuals believe they are inflicting suffering on someone else. For example, people may experience more discomfort from purchasing a housecleaning service vs. a pre-cooked meal that only involves reheating the meal-both purchases might save about the same amount of time, but only the housecleaning involves thinking about delegating a daily unpleasant task to a salient other person.
The experience of discomfort that people may experience when outsourcing disliked tasks to others is likely to manifest as feelings of guilt. Feelings of guilt are a painful affective experience that encompass remorse, selfblame, and regret (Mosher, 1980). People who feel guilt are often reacting to a specific situation where they feel personally responsible for negative outcomes inflicted on others (Tangney, Miller, Flicker & Barlow, 1996). People feel heightened emotional distress when faced with identifiable targets (Small & Loewenstein, 2003). People also experience guilt when asking other people for assistance in daily life (Bolger & Amarel, 2007). Building on these findings, we predicted that the feelings of guilt that arise when faced with making time-saving purchases that involve salient service providers could undermine individuals' willingness to spend money on these purchases. We also predicted that the guilt that arises from considering time-saving purchases involving salient service providers could suppress the happiness that people experience from buying time (Figure 1). Across studies, we define salience as whether the individual thought about outsourcing to a service provider (or did not).
Although our hypotheses are grounded in the social support literature, there may be critical differences between receiving support through the market economy vs. from close social relationships. In market contexts, support providers receive payment for their assistance, which could be perceived as a benefit (rather than a burden) to the service provider. When people know that their actions benefit a recipient, they report increased happiness (Aknin, Dunn, Whillans, Grant & Norton, 2013) and greater engagement (Grant, 2010). Thus, to the extent that consumers view employment as a positive experience, these perceived benefits could, in turn, mitigate the negative feelings of guilt-even when the service provider is salient. This theorizing results in two possible predictions. First, it is possible that when the service provider is salient, consumers could feel more guilt and report reduced intentions to buy time. Second, it is possible that when the service provider is salient, consumers could be reminded about the benefits of service provision (e.g., compensation), reducing their guilt and increasing their purchase intentions.
We explore these possibilities by examining whether and how the salience of the service provider shapes consumers' feelings of guilt and their intentions to buy time (Figure 1). 1

Overview of Studies
To examine the role of guilt in predicting happiness and intentions to make time-saving purchases, we conducted two large-scale survey studies (Study 1a and 1b). In these studies, we examined whether time-saving purchases that involved a salient service provider were associated with greater guilt, undermining individuals intentions to spend money on these purchases and the happiness that individuals expected to receive (N = 1,337). For people who thought about a service provider (vs. did not), we examined whether believing time-saving services were a benefit (vs. burden) to the provider lowered feelings of guilt.
To rule out potential confounds that might result from the spontaneous recollection of time-saving purchases, we then designed a novel lab paradigm to examine the causal role of service provider salience on feelings of guilt. We used this lab paradigm to examine whether knowing that a task is being completed by another individual would provoke feelings of guilt, undermining willingness to spend discretionary income to have more free time (Study 2).
To provide causal evidence that perceptions of the benefit (vs. burden) of time-saving purchases for the service provider would mitigate feelings of guilt, we conducted an additional experiment (Study 3). In Study 3, we adapted advertisements from the website of a sharing economy company, Hello Alfred, which offers timesaving services for people to purchase at home and for employees to redeem at work. We randomly assigned some participants to view advertisements that focused on the benefits of the purchase for the service provider. Other participants viewed ads focused on the benefits of the purchase for the consumer. Following from the results of our correlational and experimental studies, we predicted that ads highlighting the benefits of the purchase for the service provider would mitigate feelings of guilt.
To examine the relationship between guilt, happiness, and intentions to buy time, we triangulated across multiple methods. We utilized large-scale survey data, developed a new lab paradigm, and conducted an experiment with a diverse sample of working adults. The detailed characteristics of participants from each study are presented in Table 1. Across studies, we follow the standards proposed by Simmons, Nelson, and Simonsohn (2011): we reported all exclusions, every measure given, every condition tested, and the stopping rule for each study.
In light of recent calls for Open Science, we undertook several measures to maximize the transparency and replicability of our studies. First, we conducted a wellpowered replication (Study 1b) that confirmed the correlational results of Study 1a. Second, we pre-registered our hypothesis for Study 1b and Study 2 through OSF, and provided a detailed account of all the procedures and measures as well as supplementary materials in the SOM. Lastly, we also made our datasets and syntax available on OSF (https://osf.io/rqjq9/).

Overview of Studies 1a&b
In Study 1a, we conducted a brief survey to examine whether service provider salience was associated with greater feelings of guilt. We then examined whether feelings of guilt were associated with lower intentions to spend money to outsource disliked tasks and lower anticipated happiness from these purchases. In Study 1b, we sought to replicate the results of Study 1a using a larger, representative sample of employed Americans living in the US.

Participants
In Study 1a, we targeted a sample of approximately 300 respondents. A power analysis showed that a sample size of 200 gave us 80% power to detect an effect size of d = 0.20. In Study 1a, we succeeded in recruiting 309 working adults over the age of 19 through the professional survey company Qualtrics (51% female; Mdage = 35-44 years old).
See Table 2a for the correlation table of all variables assessed in Study 1a.

Outsourcing
To help respondents think about spending money on a time-saving purchase, we asked respondents to write down a task that they disliked doing and that they would like to buy themselves out of. Respondents replied to the following question: "If you could buy your way out of completing one task that you dislike doing, what task would this be (e.g., household chores, shopping)?" (Whillans et al., 2017). Next, we asked respondents to describe in 1-2 sentences how they would arrange to buy themselves out of this task.

Mood
Respondents reported how much guilt they felt when thinking about buying themselves out of this task (0 = Not Guilty, 100 = Guilty). Respondents also reported how happy they felt when thinking about buying themselves out of this task (0 = Not Happy, 100 = Happy). We adapted these measures from previous studies using single-item measures to assess in-the-moment mood (e.g., Anik, Aknin, Norton, Dunn, & Quoidbach, 2013;Killingsworth & Gilbert, 2008).

Service Provider Salience
We then asked respondents to report whether they thought about the person who would complete this task on their behalf. Respondents indicated whether they "thought about the person(people) who would have to complete this task on their behalf" or whether they "did not think about the person(people) who would have to complete this task on their behalf." This measure allowed us to determine whether respondents spontaneously   considered a time-saving purchase that involved outsourcing to another person (i.e., a salient service provider) or whether respondents did not spontaneously consider a purchase that involved outsourcing to another person (i.e., a non-salient service provider).

Buying Time Intentions
To capture intentions to buy time, respondents reported their agreement with three items assessing willingness to spend money to buy themselves out of this task on a scale from -5 = Strongly Disagree, 0 = Neither Agree or Disagree, 5 = Strongly Agree. E.g., "I plan to buy myself out of this task in the near future" (α = 0.96; Grant, 2008).

Burden vs. Benefit
Respondents who thought about a salient service provider were also asked about the perceived benefit (vs. burden) that they were causing this person. Respondents were asked, "To what extent did you think that you would be burdening vs. benefiting this person(people)?" from -50 = Burdening, 0 = Neither, +50 = Benefitting. We only asked this question to respondents who reported thinking about a salient service provider.

Covariates
Respondents reported their age, gender, experience with buying time (i.e., whether they spent money to buy time), whether they were married, how many of their children were currently living with them, as well as their annual income. We selected these covariates based on previous research examining time-use and well-being (e.g., Hershfield, Mogilner, Barnea, 2016;Mogilner, 2010;Whillans, Weidman & Dunn, 2016;Whillans et al., 2017).

Moderators
In this study, we also examined whether any of our results were moderated by relevant individual differences including working class background, gender, and protestant work ethic. We did not observe consistent, reliable moderating effects of these demographic characteristics across studies. To promote readability, these additional analyses from Studies 1a-Study 3 are reported in the Supplemental Online Material (SOM).

Results Overview
We first examined whether respondents who spontaneously considered the other person who would have to complete the disliked task on their behalf reported greater anticipated guilt. We then examined whether higher feelings of guilt undermined respondents' purchase intentions and their anticipated happiness with buying themselves out of a disliked task.

Salient Service Provider
In this study, 68% of respondents spontaneously considered a buying time opportunity that involved a salient service provider; whereas 32% of respondents considered a buying time opportunity that did not. People who reported making purchases from a salient service provider reported wanting to make purchases such as "hiring a maid service," "hiring someone to clean the windows," and "hiring a professional landscaper." People who reported on purchases in the absence of identifying a service provider wanted to make purchases such as "paying to outsource my grocery shopping," and "hiring a lawn mowing service." Descriptively, the purchases that people considered when thinking about service providers vs. not were similar in content; yet, the critical difference was that respondents explicitly reported that they thought about the service provider(s) who would have to complete the disliked task on their behalf.

Happiness
We then examined the extent to which service provider salience influenced anticipated happiness via guilt. First, we examined whether there was a total effect of service provider salience on happiness. In this model, there was no total effect of salience on anticipated happiness, total effect = -2.03 (3.25), p = 0.533, 95% CI [-8.42, 4.37]. Next, we examined the joint regression of the predictor (salient service provider) and mediator (guilt) when predicting the DV (happiness). In this model, controlling for guilt, there was no effect of salience on anticipated happiness, We then examined our conceptual model by assessing whether the extent to which individuals experienced guilt when thinking about salient service providers undermined their purchase intentions. Consistent with this hypothesis, there was a significant indirect effect, -0.23 (0.11), 95% CI [-0.47, -0.05]. Respondents who thought about the service provider were less likely to intend to buy themselves out of a disliked task if they felt guilty about the decision. These results held controlling for our predetermined set of covariates, indirect effect = -0.23 (0.10), 95% CI [-0.45, -0.05]. In this mediation, the presence of a positive direct effect of salience on purchase intentions suggests that part of the service-provider salience and intention relationship is unexplained by guilt in our model. For example, thinking about a specific service provider could help some consumers follow through to make time-saving purchases. More work is needed to assess alternative mediators that match the positive sign of the direct effect (Zhao, Lynch & Chen, 2010).

Outsourcing as a Benefit vs. Burden
We then restricted our analyses to people who spontaneously considered a salient service provider (N = 211). With this restricted sample, we examined whether respondents who viewed outsourcing their task as a benefit (vs. burden) to the service provider would report lower feelings of guilt, greater anticipated happiness, and greater intentions to buy time. Consistent with these hypotheses, respondents who reported that outsourcing the task was a benefit (vs. burden) for the service provider felt less guilt r(211) = -0.19, p = .006, greater anticipated happiness, r(210) = 0.32, p < .001, and greater intentions to spend money to outsource their disliked task in the near future, r(211) = 0.15, p = .035.
Conducting indirect effect analyses, respondents who reported that outsourcing was a benefit (vs. burden) to the service provider reported lower feelings of guilt, which in turn was associated with greater anticipated happiness, indirect effect = 0.04 (0.03), 95% CI [0.01, 0.10] and greater intentions to spend money to outsource their disliked task in the near future, indirect effect = 0.02 (0.01), 95% CI [0.002, 0.04]. These results held controlling for covariates.

Study 1b Participants and Procedure
In Study 1b, we conducted a preregistered replication of Study 1a by recruiting a nationally representative sample of employed Americans living in the US (N = 1,028) through the survey company Qualtrics (47.6% women; see Table 1 for demographics). Participants completed the identical items from Study 1a, including individual difference measures (gender, social class, and protestant work ethic; see SOM). See Table 2b for a correlation table of all variables examined. Following the results of Study 1a, we pre-registered three central hypotheses. The preregistration is available through the OSF (https://osf.io/ gtb8e/), along with the data and materials. 2 We predicted that respondents who spontaneously reported time-saving purchases that involved a salient service provider would report greater guilt when thinking about buying themselves out of a disliked task. We predicted that these greater feelings of guilt would in turn undermine anticipated happiness and intentions to make a time-saving purchase.
To account for potential individual differences in respondents' previous experiences with outsourcing, we pre-registered that we would control for whether respondents reported that their parents typically spent money to outsource disliked tasks while they were growing up, and for respondents' recent experiences with buying time. We conducted all of the analyses below controlling for our pre-registered set of covariates (age, gender, number of kids living at home, marital status, personal income, and outsourcing experience). We also pre-registered hypotheses about moderators (gender, social class, and protestant work ethic; see SOM).

Salient Service Provider
In this study, 64% of respondents spontaneously considered a buying time opportunity that involved another person; 36% of respondents spontaneously considered a buying time opportunity that did not involve another person.

Guilt
As predicted in our pre-registration, respondents who thought about a service provider who would complete the task on their behalf reported greater feelings of guilt (M = 41.98, SD = 32.67) as compared to people who did not (M = 28.98, SD = 30.29), t(814.15) = 6.42, p < .001, 95% CI [9.03, 16.98]. Reporting these results in regression, respondents who thought about the service provider who would complete the disliked task on their behalf reported greater feelings of guilt when thinking about buying time as compared to respondents who did not, B = 13.00 (2.07), β = 0.19, p < .001, 95% CI of B [8.94, 17.06]. These results held controlling for the pre-registered covariates described above, B = 8.80 (2.28), β = 0.13, p < .001, 95% CI of B [4.33, 13.26].

Happiness
We then examined whether the extent to which service provider salience increased guilt in turn undermined anticipated happiness. Consistent with Study 1a, there was   Respondents who thought about the service provider reported lower anticipated happiness from buying themselves out of a disliked task, via increased feelings of guilt. These results held for covariates, indirect effect = -1.08 (0.37), 95% CI [-1.87, -0.43]. This indirect-only mediation provided partial support for our proposed conceptual model.

Intentions
Consistent with Study 1a, there was a positive total effect of service provider salience on purchase intentions, 1.87 (0.22), p < .001, 95% CI [1.45, 2.30]. Thus, on average, thinking about the service provider was linked to greater purchase intentions. Next, we examined the joint regression of the predictor (salient service provider) and mediator (guilt) when predicting the DV (intentions). In this model, controlling for guilt, there was a significant positive effect of salience on purchase intentions, 1.80 (0.22), p < 0.001, 95% CI [1.37, 2.24]. Controlling for salience, there was a significant negative effect of guilt on purchase intentions, -0.01 (0.003), p = .017, 95% CI [-0.01, -0.001]. As with Study 1a, the presence of a direct effect suggests that there could be alternative mechanisms other than guilt that has been omitted. More research is needed to understand mechanisms underlying this positive effect.
We then examined our conceptual model by assessing whether the extent to which individuals experienced guilt when thinking about service salient providers undermined their purchase intentions. Consistent with this hypothesis, there was a significant indirect effect, -0.01 (0.05), 95% CI [-0.10, -0.08]. Respondents who thought about the service provider expressed lower intentions to buy themselves out of a disliked task in future, via increased feelings of guilt. These results held controlling for our predetermined set of covariates, indirect effect = -0.06 (0.04), 95% CI [-0.14, -0.01]. As with Study 1a, the presence of a direct effect suggests that there could be alternative omitted mediators. Further research is needed to find potential omitted variables that match the positive sign of our direct effect.

Benefit vs. Burden
As per our pre-registered analysis plan, we also examined whether respondents who thought about a timesaving purchase that involved outsourcing to a salient service provider would report lower guilt if they viewed outsourcing the task as a benefit (vs. a burden) to the service provider. Consistent with Study 1a, respondents who reported that outsourcing the task was a benefit (vs. burden) for the service provider reported lower feelings of guilt, r(655) = -0.11, p = .004, greater anticipated happiness, r(655) = 0.29, p < .001, and greater intentions to outsource their disliked task in the near future, r(649) = 0.19, p < .001.

Discussion of Studies 1a&b
Across Studies 1a and 1b, we examined the role of guilt in predicting respondents' anticipated happiness and their intentions to spend money on time-saving purchases. Across both studies, respondents who spontaneously thought about the person who would have to complete the task on their behalf (salient service provider) reported significantly higher feelings of guilt as compared to respondents who did not spontaneously consider the person or people who would have to complete the task on their behalf (nonsalient service provider). Across studies, greater feelings of guilt undermined respondents' anticipated happiness. Guilt also undermined respondents' intentions to spend money to outsource their disliked tasks in the future. However, we did not see a significant direct effect of guilt on outsourcing intentions, suggesting that there might be two contrasting effects of service provider salience on intentions. First, we saw a negative effect of salience on intentions via heightened guilt, but there could also be a positive effect of salience on intentions via some other mechanism that we did not explore in the current study. Taken together, Studies 1a and 1b provide convincing initial evidence to support our theoretical framework, showing that there is a strong inter-relationship between salience, perceived benefits, guilt, and anticipated happiness.
Critically, the results of these studies could not be explained by previous experiences with spending money to outsource disliked tasks or demographic factors such as income. We did not find reliable evidence that these results were moderated by individual differences such as gender, social status, or protestant work ethic (see SOM). In Study 1b, we found initial evidence that women reported lower intentions to spend money to outsource their disliked tasks, in part because they believed these tasks would be more of a burden (vs. benefit) for the service provider (results reported in the SOM). However, we did not replicate these findings in our subsequent studies, therefore these results should be interpreted with caution. In contrast to our predictions, people of higher social status (vs. lower social status) felt greater guilt when thinking about outsourcing a disliked task. Once again, however, these results did not replicate in subsequent studies and should be interpreted with caution (results reported in SOM).
Across Studies 1a and 1b, respondents' feelings of guilt were moderated by how much the respondent felt as if they were benefitting (vs. burdening) the service provider. These results are consistent with research on social support (Bolger & Amarel, 2007) and the identifiable victim effect (Kogut & Ritov, 2005). Not everyone who thinks about outsourcing to another other person experiences guilt when doing so, as some people think about the opportunity to outsource as a benefit vs. burden.
Overall, the results of Study 1a and 1b suggest that people spontaneously think about the involvement of other people when making time-saving purchases. Although we found some evidence that the perceptions of others' involvement positively predicted purchase intentions, these perceptions also had a cascade of psychologically detrimental effects, including greater feelings of guilt, lower anticipated happiness, and lower intentions to buy time.
Of course, because of the correlational nature of these data, it is not possible to rule out alternative explanations. It is possible that those who are more committed to outsourcing are more likely to have a specific person in mind. Therefore, in these studies, the positive relationship between salience and intentions could reflect an effect in the opposite direction, whereby intentions affect salience. People who think about purchases made by salient service providers might also think about tasks that are more unpleasant or that they are more able to complete themselves, which could help to explain why people who recalled purchases that involve salient service providers experienced greater feelings of guilt. In Study 2, to rule out these possible confounds, we developed a lab paradigm to explore the causal relationship between service provider salience, time-saving purchases, and guilt. We examined whether thinking about a salient service provider (vs. not) increased guilt and decreased people's intentions to buy time.

Development of a Buying Time Paradigm
In developing this paradigm, our goal was to mirror realworld dilemmas where people must first decide whether to spend discretionary income to have more free time. Similar lab paradigms have been developed to study the psychological consequences of resource scarcity (Shah, Mullainathan, & Shafir, 2012) and the fundamental tendency for people to earn more than they can consume (Hsee, Zhang, Cai, & Zhang, 2013).
In the first part of this online paradigm, participants are asked to complete an e-task in which they are asked to copy letter strings containing the letter "e" into a new browser window (see also Greenberg, 1977). This task is meant to mimic the boring, mundane chores that chip away at free time in daily life. Participants are provided with monetary compensation for completing this initial e-task that they are not expecting to receive at the outset of the study, providing participants with income that they can spend at their discretion.
Participants are then faced with a choice: they can continue to complete the boring "e-task" for another 30 minutes, or they can pay their earned income to buy themselves out of this task (leaving them with 30 minutes of free time). By administering this study online instead of in the lab, participants can have 30 minutes to spend at their discretion, mirroring the real-world situation of spending money to have more free time.

Study 2 Participants
Three hundred and fifty students from a large public university participated in this experiment in exchange for course credit (74% female, M = 20.24, SD = 2.50). We pre-registered this study through the OSF (https://osf.io/ h2n4t/).

Paradigm
In this study, we wanted to ensure that the decision of whether to keep the cash or to outsource were equally desirable. Thus, we conducted a pilot study to ensure that the decision to keep the cash was perceived by participants as equally pleasing as the decision to exchange the cash for 30 minutes of free time. In this pilot study (N = 106), we asked students to complete five minutes of the counting e's task. We then asked students to report the minimum payment that they would be satisfied receiving in order to continue completing the e-task for the next 30 minutes. On average, participants reported that they would like to receive approx. $7.00 to continue completing the e-task (M = $6.35, SD = $3.20). Based on the results of this pilot study, we were able to set our reward amount to $7 to ensure that the cash option was equally as satisfying as the free time option for most participants on average.

Protocol
All participants were recruited online and earned $7 to complete the initial e-task for 5 minutes. Participants were then provided with the opportunity to trade the $7 for 30 minutes of free time, where they could presumably do anything they wanted since the study was administered online. In the non-salient service provider condition, students were presented with the choice to receive 30 minutes of free time or to receive $7 for continuing the e-task. They could "exchange the cash for 30 minutes of free time" or "keep the cash and complete the counting e's task for the next 30 minutes." In the salient service provider condition, students were presented with the choice to receive 30 minutes of free-time only if they outsourced the e-task to another student who would complete the task on their behalf. Specifically, students were presented with the choice to "receive 30 minutes of free time by having another student complete the e-task on your behalf" or to "keep the cash and complete the counting e's task for the next 30 minutes." As a result, Study 2 provides a further test of the implications of whether the service provider was salient or not. In the salient service provider condition, participants were asked to think about the 'student who would complete the task on their behalf.' In the control condition, participants were not asked to think about the person who would complete the task on their behalf; they were simply told that the task would be completed. Participants in the experimental condition were prompted to think about the service provider, whereas participants in the control condition were not. Thus, this study provides a test of the question of whether service provider 'salience' shapes time-saving purchase intentions and feelings of guilt.

Measures
Immediately before making the decision about whether to keep the cash or to give up the cash to have 30 minutes of free time, students reported how guilty and how happy they felt about exchanging the cash for 30 minutes of free time. Consistent with Study 1b, participants reported their responses on a scale from 0-100. 3

Pre-Registered Prediction
In Study 2, our key dependent variable was the choice students made to outsource the e-task or to keep the $7 and continue to complete the e-task themselves. We predicted that students assigned to the salient service provider condition would be less likely to outsource the disliked task. We also predicted that students assigned to the salient service provider condition would feel greater feelings of guilt as compared to students who were randomly assigned to the non-salient service provider condition. Consistent with the results of our correlational studies, we predicted that students who were randomly assigned to the salient service provider condition would experience greater guilt, undermining their willingness to choose free time.

Main Effects Percentage Who Outsourced
In Study 2, 60% of students decided to keep doing the task, whereas 40% of students exchanged their $7 for 30 minutes of free-time.

Guilt
Consistent with our pre-registered predictions, participants who were randomly assigned to salient service provider condition reported that choosing the free time would cause them significantly greater feelings of guilt (M = 38.90, SD = 30.52) as compared to participants who were randomly assigned to the non-salient service provider condition (M = 29.74, SD = 30.58), t(348) = 2.73, p = .007, 95% CI [2.56,15.76].

Choice
To see how many participants in each condition chose to outsource, we conducted a binary logistic regression. A significantly lower number of participants in the service-provider salience condition chose to outsource. Specifically, 44.6% of participants assigned to the control condition outsourced their disliked tasks. In the serviceprovider salience condition, 31.9% outsourced the task, χ 2 (1) = 5.57, p = .018. Next, we examined whether there was a direct effect of service provider salience on participants' choice to outsource. Most importantly, and consistent with our pre-registered analyses, we examined whether there was an indirect effect of service provider salience on the choice to outsource. We found a significant indirect effect, -0.16 (0.07), 95% CI [-0.31, -0.04]. To the extent that participants assigned to the salient (vs. non-salient) service provider condition reported greater feelings of guilt, they were also less likely to exchange their $7 payment for 30 minutes of free time.

Direct Effect on Happiness
On an exploratory basis, we examined whether condition assignment shaped anticipated happiness. Participants assigned to the salient service provider condition reported that choosing the free time would cause them to feel significantly lower happiness (M = 38.91, SD = 25.75) as compared to participants who were assigned to the nonsalient service provider condition (M = 51.51, SD = 27.21), t(322) = -4.04, p < .001, 95% CI [-18.33, -6.32]. In this study, there was no significant indirect effect of service provider salience on happiness via guilt, indirect effect = -.27 (0.35), 95% CI [-1.10, 0.37], meaning that service providers salience was a direct predictor of lower anticipated happiness, regardless of how much guilt participants experienced.

Study 2 Discussion
In Study 2, we used a novel lab paradigm to examine whether outsourcing to a salient service provider increased guilt and undermined participants' willingness to buy time. As predicted in our pre-registered analyses, service provider salient time-saving purchases resulted in greater guilt as compared to non-salient purchases. In turn, these feelings of guilt undermined respondents' willingness to give up discretionary income to have more free time. In Studies 1a-2, we observed a consistent pattern: timesaving purchases involving a salient service provider increased feelings of guilt. These feelings of guilt, in turn, undermined respondents' anticipated happiness and their intentions to spend money to buy time.
If guilt is the critical factor in the causal model that we proposed in Figure 1, we should be able to manipulate feelings of guilt by focusing on the benefits vs. burden of time-saving purchases for the service provider. This hypothesis is consistent with the results that we observed in Studies 1a and 1b, where respondents who perceived timesaving purchases as a benefit (vs. burden) to the service provider reported lower feelings of guilt. We conducted a study highlighting the benefits of task completion for the service provider (Study 3). In Study 3, we assigned participants to view advertisements from a popular US sharing economy company that focused on the benefits of the company for the service provider or the consumer.
We predicted that participants assigned to view advertisements that focused on the benefits to the service-provider (vs. consumer) would experience lower guilt, greater happiness, and greater intentions to use the service. In testing this question, we provide a valuable test of a practical intervention strategy implied by our theoretical model. If consumers feel guilty when thinking about the person or people who will be completing their tasks, then emphasizing the benefits of task-completion for the service provider(s) should mitigate these feelings of guilt. If true, this framing strategy could be a useful intervention that sharing economies could use to reduce consumers' guilt and improve their satisfaction.

Study 3 Overview
A diverse sample of three-hundred and ninety participants from Boston, MA completed three unrelated studies for a $10 payment (48% female; Md age = 28.00). Participants were randomly assigned to view one of two ads for a popular outsourcing company, Hello Alfred. This company offers a variety of services, including in-home services like dry-cleaning, grocery shopping, and home cleaning. These tasks are certainly disliked by people -in a nationally representative sample of Americans (N = 1,265) -80% of respondents reported thinking about outsourcing household cleaning tasks (Whillans et al., 2017).
Participants were randomly assigned to the consumerbenefits condition or the provider-benefits condition. In the consumer-benefits condition, participants read an ad from Hello Alfred that focused on the benefits of using the service for the consumer. In the provider-benefits condition, participants read an ad from Hello Alfred that focused on the benefits of using the service for the service provider. As a manipulation check, after viewing this ad, participants were asked whether they thought that they were benefitting or burdening the service provider. Participants were then asked how guilty and how happy they would feel using this service, as well as their intentions to use this service in the near future.

Appeals
In this study, we examined two appeals that were already in use by Hello Alfred. We selected these appeals because they clearly highlighted (1) the benefits of the service to the service provider, or (2) the benefits of the service to the consumer. These appeals depicted a photo of the consumer or a service provider and included a brief description of the benefits of the advertisements for each group (See Table 3). We constrained the ethnicity of the targets featured in the ads and we gender matched the targets and participants to minimize variability because our studies were not adequately powered to test for these differences.
With a separate sample of participants that we recruited from Amazon's Mechanical Turk (N = 385; 52% female), we pre-tested these appeals to ensure that they did not differ on any key dimension that could impact our results. In particular, we pre-tested these appeals for differences in how high in social status the consumers/providers appeared. We also pre-tested these appeals for differences in basic message characteristics, including how positive and negative the ads were (Lang & Yegiyan, 2008), how easy the ads were to process (Lee & Aaker, 2004;White & Peloza, 2009), and how involved and attentive participants felt when reading the ads (Wheeler, Petty & Bizer, 2005).
The ads that highlighted the benefits to the service provider and the ads that highlighted the benefits to the consumer did not differ on any of these dimensions (Tables 4a and 4b). Thus, the appeals used in Study 3 did not differ on characteristics (such as positivity, fluency, or clarity) that could have accounted for our results.

Anticipated Mood
After reading the ads for Hello Alfred, participants completed the two mood questions from Studies 1a and 1b. Participants reported how much guilt they felt when thinking about using the service (0 = Not Guilty, 100 = Guilty) and they reported how happy they felt when thinking about using the service (0 = Not Happy, 100 = Happy).

Buying Time Intentions
To capture consumer intentions, participants reported their agreement with three-items assessing willingness to spend money to use this service in the near future from -5 = Strongly Disagree, 0 = Neither Agree or Disagree, 5 = Strongly Agree. E.g., "I plan to use this service in the near future" (α = 0.98; Grant, 2008). Participants were told that "at the conclusion of this study, your ID will be entered into a draw. If you win this draw you can choose one of two prizes: a $100 voucher for Hello Alfred or $50 cash." We examined participants' decision of whether they would like to "receive the $50 cash prize" or whether they would like to "receive the $100 voucher for Hello Alfred" as a behavioral measure assessing interest in using the service. Given that cash is typically perceived as more valuable when compared to non-cash rewards Table 4b: Advertisements used that focus on customer vs. service provider benefits (male).  (Jeffrey, 2009), we doubled the value of the time-saving voucher as compared to the cash.

Manipulation Check
Participants who were randomly assigned to view the advertisement that emphasized the benefits to the service provider reported that the purchase was more likely to benefit the service provider (M = 18.26, SD = 25.26) as compared to participants who were randomly assigned to view the ads that emphasized the benefits to the consumer (M = 11.23, SD = 28.37), t(386.89) = 2.60, p = .010, 95% CI [1.71,12.34]. These results suggest that the manipulation successfully encouraged participants to think about the benefits of using the service for the service provider (vs. to consider the burden of the service).

Guilt
As predicted, participants who were randomly assigned to view the ad that emphasized the benefits to the service provider reported significantly lower feelings of guilt when thinking about using the service (M = 24.64, SD = 27.32) as compared to participants who were randomly assigned to view the advertisement that emphasized the benefits to the consumer (M = 31.55, SD = 29.93), t(393) = 2.40, p = .017, 95% CI [-12.58, -1.23].

Happiness
In this study, there was no main effect of condition assignment on anticipated happiness, t(394) = 0.86, p = .392, 95% CI [-2.88, 7.35]. However, consistent with the results of Study 1a and 1b, to the extent participants in the service provider-benefits condition reported lower guilt, they also reported greater anticipated happiness with using the service, indirect effect = 1.82 (0.84), 95% CI [-0.32, 3.65]. 4

Intentions
In this study, there was no direct effect of condition assignment on intentions to make a time-saving purchase in the near future, t(394) = 0.161, 95% CI [-0.47, 0.79]. However, consistent with Study 1a and 1b, to the extent that participants in the provider-benefits condition reported lower levels of guilt, participants also reported greater intentions to use the service in the near future, indirect effect = 0.22 (0.09), 95% CI [0.03, 0.41].

Purchase Behavior
Because only 3% of respondents chose the voucher vs. the cash prize in the lottery, we did not analyze this outcome measure; chi-square analyses require an expected value of at least 5 per cell/condition to yield valid results (MacDonald, 2014). Instead of a chi-square analysis, we conducted a Monte Carlo permutation test to examine whether there was a significant effect of treatment on participants' decision to outsource. This exact test indicated that there was no significant difference between treatment groups under the null hypothesis distribution, p = 0.201. The low base-rates of choosing the time-saving service in the lottery presented in Study 3 highlights the fact that many other factors might limit people's intention to use these services beyond those studied in the paper. Indeed, people often believe they will have more time in the future than they do in the present, and express a limited desire to give up a concrete resource (money) to have more of an abstract resource (i.e., time; see Dunn, Whillans, Aknin & Norton, 2019 for a comprehensive review of the barriers to 'buying time'). Future research should further explore how these barriers might also cause guilt and contribute to the relatively low rates of outsourcing we observed across studies.
It is worth noting that we sought to prioritize ecological validity. As a result, the benefits that were highlighted across ads in Study 3 differed across versions. Some of the ads featured benefits that highlighted the pleasant nature of providing the services (e.g., service providers acquiring meaning from having a happy client), while other ads highlighted the positive impact for the provider's skills (e.g., service providers acquiring skills from helping clients). While we were interested in the broader question of whether highlighting the benefits (regardless of what these benefits were) would shape how consumers felt about the services, future research should explore which types of benefits are most effective. For example, previous research has shown that prosocial benefits (like highlighting how the service has a positive impact for people's family) might be more effective than self-focused benefits (like highlighting how the service might have a positive impact in terms of developing one's skills; Feiler, Tost, & Grant, 2009). More research should explore these and related possibilities.
While Study 3 shows that emphasizing the benefits for the providers through a single advertisement exposure significantly reduced consumer's feelings of guilt, a more prolonged, stronger interventions might be necessary to shift an average consumer's perceptions to motivate usage of time-saving services. More research is needed to explore such interventions.

General Discussion
Across two large-scale surveys, including a nationally representative sample of working adults, respondents who thought about the person who would complete a disliked task on their behalf reported greater feelings of guilt, in turn undermining their anticipated happiness and their purchase intentions (Studies 1a and 1b). In these studies, it is important to note that we observed significant positive direct effects of service-provider salience on intentions to purchase time-saving services. The presence of a direct effect suggests that there could be one or more omitted mediators in our model. For example, people might be more likely to outsource when they can think about the specific person or people they would want to outsource to. Future research should explore these and other, related possibilities.
In an experiment using a novel lab paradigm, simply telling participants that another person would complete the task on their behalf increased guilt and undermined participants' willingness to spend money to have additional free time (Study 2). On an exploratory basis, in Study 2, we also examined whether thinking about a salient service provider decreased happiness via guilt. Inconsistent with Studies 1a and 1b, we did not see a significant indirect effect of service provider salience on happiness in this study. In Study 3, we found a simple method of mitigating consumer's guilt: highlighting the benefits of the timesaving purchase for the service provider. In Study 3, emphasizing the benefits for providers (vs. consumers) did not have a direct effect on purchase intentions, yet it increased purchase intentions through lowering consumers' feelings of guilt. However, more research is needed to understand when saliency creates guilt, and when guilt undermines purchase intentions.
Overall, our evidence for the effect of salience on intentions is mixed across studies, such that salience had a positive effect on intentions in Studies 1a and 1b and a negative effect in Study 2. Therefore, we should note that our studies support the conclusion that service provider salience undermines intentions, only to the extent that salience increases guilt.
Across studies, we observed mixed evidence for the effects of service provider salience on anticipated happiness. Our framework proposed that service provider salience would impact guilt, which in turn would impact anticipated happiness. Although Studies 1a & 1b supported this prediction, Study 2 found only a direct effect of service provider salience on happiness and no indirect effect, meaning that service provider salience was a direct predictor of lower anticipated happiness from outsourcing, regardless of how much guilt participants experienced.

Theoretical Implications
This research adds to a growing body of research documenting the benefits and barriers of spending money in ways that promote happiness (Whillans et al., 2017;Dunn, Whillans, Aknin & Norton, 2019). Critically, this research points to an underexamined barrier that can prevent people from utilizing time-saving services: feelings of guilt that arise from outsourcing to a salient service provider. It is worth noting that these findings contrast with research showing that operational transparencydisclosing information about the company's practices, policies, and decision-making processes-promotes customer satisfaction (Buell & Norton, 2011). In doing so, this research adds nuance to the idea that operational transparency always enhances customer satisfaction. When the effort that is expended on a customers' behalf involves another person (e.g., a housecleaner), companies might want to consider downplaying the effort that this person or group of people is expending or highlight the benefits of using the service for this person or group of people to mitigate consumers' feelings of guilt.
Although the findings documented here diverge from research on operational transparency, these findings are consistent with research in social psychology showing that people who receive social support from their romantic partners benefit most when this social support goes unnoticed (Bolger, Zuckerman & Kessler, 2000). The current research, therefore, suggests that a central idea from the social support literature-invisible social support-can be used to understand the psychology of the support purchased through the market economy. In doing so, these findings provide a theoretical contribution to the literature by connecting the seemingly disparate research on social support in close relationships with emerging research on the benefits of social support that is purchased through the sharing economy.
Drawing on these novel connections, further research should examine when there might be divergent effects of the support that is received through close relationships and the support that is purchased through the market economy. Previous social support research suggests that people feel as if they are burdening the support provider, which in turn can promote greater feelings of guilt (Bolger & Amarel, 2007). In the current research, we find evidence that this typical pattern can be reversed in the context of the market economy. Across studies, when individuals believed that their transaction had benefits for the service provider, they no longer felt as guilty about paying others to complete disliked tasks on their behalf.
Taking this insight further, researchers could examine how these effects unfold over time. In the social support literature, the more often people receive visible social support from their partners, the worse they typically feel (Bolger, Zuckerman & Kessler, 2000). In the context of support purchased through the market economy, the reverse might be true: Repeated interactions with a service provider could foster feelings of closeness and trust (Sandstrom & Dunn, 2014) or a heightened recognition of the benefits of the task for the service provider (e.g., in terms of financial resources or feelings of purpose; Grant, 2008). More research is needed to further document the similarities and differences between the psychology of receiving social support via close others vs. through the market economy.

Practical Implications
Our findings also support the idea that individuals are sensitive to the treatment of the service providers employed by the sharing economy. When sharing economy companies treat employees well, companies are likely to benefit from advertising the benefits of the service for their employees. In contrast, when sharing economy companies do not treat their employees well, these companies are likely to drive away business.
More broadly, our findings, and the results from Study 3, suggest that individuals are responsive to the welfare of service providers employed by sharing economy companies. While some service providers, prioritize the well-being of employees, others fail to prioritize the wellbeing of their workers (Gino, 2017). Our results suggest that companies should think carefully about the benefits that they provide their employees. By offering benefits to service providers and highlighting these benefits to consumers, sharing economy companies could promote the well-being of employees, consumers, and increase the number of individuals who use their services, increasing profit.

Limitations and Future Directions
Despite this promising experiment, the current research provides limited behavioral evidence in support of our core hypotheses. Although we have preliminary results showing that promoting the benefits of task provision for the service provider (vs. consumer) increases purchase intentions, we are not able to observe whether these intentions result in actual purchase decisions outside of a laboratory context. Thus, further research should investigate the influence of service provider saliency on actual purchase behavior over time.
Another limitation of the current paper is that we only tested the moderating role of perceived benefits (vs. burdens) among individuals already thinking about service provider(s). Future studies should, therefore, examine how consumers' general perceptions of time-saving purchases (in the absence of thinking about a salient service provider) shapes guilt, satisfaction, and consumption intentions. A related limitation is that we do not have a study that provides a direct comparison between people who thought about the benefit vs. burdens for the service provider and the benefit vs. burdens of time-saving services more broadly. Although the scope of our paper focused on individual service providers and guilt, future research should also examine why people don't spontaneously think about the benefits of time-saving services for consumers. Exploring why people fail to think about service provider benefits could allow researchers to identify the specific interventions that would help consumers recognize the benefits (to the extent they exist). For example, if people are simply mispredicting that the task will be unenjoyable to another person, companies could implement ads that highlight to consumers that many service providers enjoy their work. Therefore, investigating when and why people think about the benefit or burden of service providers could lead to the development of more effective advertisements, increasing people's intentions to buy time.
In our research, we find that a critical source of guilt is the perceived burden placed on the service provider. These findings are consistent with research on the identifiable victim effect showing people are more emotionally responsive to the hardship of others when they know details about them (e.g., Dickert & Slovic, 2009;Cryder, Loewenstein, & Scheines, 2013). However, there are likely several other causes of guilt, including a predisposition for guilt-proneness (Flynn & Schaumberg, 2012), perceived moral obligation to complete one's household chores or the perception that buying time is a luxury expense. Purchase characteristics-such as perceived intimacycould also influence guilt (see Hochschild 2012 for a similar argument). Cultural acceptability could also play a moderating role, such that if it is more socially acceptable to buy time then time-saving services would not evoke guilt and would be more likely to promote happiness. Future research should further delineate which of these personality and situational factors are the most important contributors to feelings of guilt.
Another possible moderator is feeling of self-indulgence or personal inadequacy. This possibility is in line with our examination of protestant work ethic (PWE) in Studies 1a and 1b in the SOM. On an exploratory basis, we examined whether personal characteristics, such as PWE, shaped whether exposure to salient service providers undermined purchase intentions. People who score higher on protestant work ethic might experience greater feelings of guilt when faced with the opportunity to outsource their disliked tasks because they are more likely to believe that hard-work signals competence (e.g., Furnham, 1982;Furnham, 1984;Greenberg, 1977). Consistent with this reasoning, we found that respondents who scored higher in PWE reported greater feelings of guilt when thinking about buying themselves out of this disliked task. However, those who scored higher in PWE were alsoon average-more likely to spend money to buy time; therefore, these divergent findings require additional research to disentangle.
To further explore the role of PWE, we examined whether these results differed across income levels. We found a significant interaction-wealthier people who scored higher in PWE felt greater guilt when outsourcing to a salient (vs. non-salient) service provider ( Figure S1 in SOM). Yet, there was also a significant interaction between PWE and wealth on purchase intentions, such that wealthier people reported greater purchase intentions when they were high in PWE ( Figure S2 in SOM). These divergent results suggest a complex interplay between PWE, income, and guilt-such that under high PWE beliefs-wealthier people are both more likely to outsource to salient service providers and are more likely to experience guilt from doing so, which could undermine their actual enjoyment of their resulting free time. More research is needed to replicate and substantiate these patterns of results.
Since PWE was a significant predictor of guilt and intentions to buy time, we controlled for PWE in all of our analyses and confirmed that all of our reported results remained significant. Overall, these results provide evidence that service provider salience predicts guilt above and beyond alternative constructs (i.e., PWE) that could otherwise explain when and how individuals experience guilt, lower anticipated satisfaction from buying time, and reduced intentions to buy time. However, there could still be other potential explanations for this phenomenon that are outside the scope of our current work. For example, people who under-recognize the value of time may feel guilty from making time-saving purchases because they feel as if time-saving purchases are a waste of money (see Pfeffer & DeVoe, 2009 for a similar argument). Moreover, people's broader beliefs about the benefit vs. burdens of sharing economy companies could be an alternative source of guilt-particularly if consumers feel that these sharing economy companies are unethical or treat their workers unfairly (see Study 2; Akinola, Martin, & Philips, 2018 for a similar argument). Future research should explore these additional pathways linking salient identification of service providers with enhanced feelings of guilt, lower anticipated happiness, and consumer intentions.
Given that the popularity of the sharing economy is on the rise, understanding when the benefits of time-saving purchases are likely to emerge is increasingly important. In 2015, the Pew Research Centre found that seven-in-ten