To Tip, or Not to Tip?

The transformation of tipping during the Covid-19 pandemic through the lens of Practice Theory and Intersectionality

By: Kevin Shum & Jeff Pabst

Introduction

The Covid-19 pandemic affected many different aspects of society and caused serious social and economic ramifications across the globe. One of the hardest hit industries, especially during the periods of strict lockdowns, was the food service industry. Due to health and safety concerns amidst strict social distancing and sanitation guidelines, many restaurants in the US shifted their businesses to focus on takeout and adopted new point-of-sale (POS) systems, such as the one designed by Square, to digitize and automate parts of the transaction and tipping experience. In this paper, we apply Practice Theory and Intersectionality to uncover new understandings of how the performance, skills, and meaning of tipping with Square POS systems changed during the pandemic. First, we will introduce the two theories and describe the components of each that will be used as part of our analysis. Next, we will introduce the Square POS system and discuss how our combined lens will help us analyze the transformation in further detail. Finally, we will present our analysis and discuss the results, culminating in key findings, insights, and future research questions.

Theoretical Basis

In this section, we will discuss our theoretical approach and explain the core concepts that we will use in our analysis of tipping with the Square POS systems.

Practice Theory Expanding our analysis beyond individual needs or interactions, Practice Theory focuses on how shared activities, like tipping, shape our social structures as they are re-performed and stabilized across people over time [10]. Practices can be broken down into three main elements: Motivation, Skills & Knowhow, Embodied Performances, and Tools & Materials. In order to understand broader societal trends or behavior, we can leverage the reflexive nature of Practice Theory to understand how changes in activity or social structures can shape each other or break down over time through a “crisis of routine” [6], such as Covid-19, to create new emergent practices.

Practice Theory Terms	Definitions
Motivation, Skills & Knowhow	The knowledge and capabilities required and the reasons that motivate a subject to perform the practice.
Embodied Performances	Routinized physical activity involved in carrying out the practice.
Tools and Materials	The necessary artifacts appropriated by the subject to perform a practice.

Intersectionality Intersectionality offers a historical and critical lens to unpack how the practice of tipping impacts the lived experiences of customers and servers in complex ways [4, 20]. Design is not a neutral practice, and inequities are not the product of a single factor in isolation. People encounter overlapping forms of discrimination based on intersecting identities — including gender, age, race, class, socioeconomic status, geography, sexuality, citizenship, religion, familial status, disabilities, and abilities. Intersectionality helps us better embrace the margins and identify various forms of oppression and privilege that are encoded in and performed through our designs.

Product Our case study focuses on tipping with the Square POS system for takeout orders in the US. Each Square POS device handles customer transactions using an interactive touchscreen interface with compatibility for credit card or tap-to-pay payment methods and functionality for automatic tip selection and receipt options. These systems are typically found at the front kiosk or service counters in restaurants and were the primary means of paying for takeout orders during the pandemic.

Left: An example of an interactive Square POS system with tap-to-pay functionality (Source: Square) Right: An example of a tipping screen with three default percentages and options for custom and no tip options below (Source: TouchBistro).

Combined Theoretical Approach While intersectionality is a broad and critical framework for critiquing the impacts of tipping and the Square POS system on certain communities, Practice Theory grounds our intersectional analysis on specific and concrete elements of tipping that create our social structures. Intersectionality compliments Practice Theory by offering a lens to unpack how the practice of tipping creates margins that overlap based on different combinations of identities. Combined, we can better understand how the crisis of routine during the pandemic may have exacerbated marginalizations as the practice is reshaped. With this new lens, we can then untangle hidden and complex inequalities perpetuated by tipping before and after the pandemic, critique the consequences of our adoption of the Square POS systems, and generate new research questions around reducing harms for similar technologies.

Analysis

Within each of the three main components of a practice, we will apply our combined theoretical approach, drawing from elements of both Practice Theory and Intersectionality to uncover new understandings of how tipping has been impacted by the pandemic.

Motivation, Skills & Knowhow - Combined Analysis While tipping is largely pervasive across the US, motivation for tipping can vary widely across contexts. Prior to the pandemic, tips were not generally expected for takeout orders, and were reserved for rewarding good customer service as a bonus. Differential wage laws, however, may motivate customers to tip more to compensate for service workers’ wages, particularly in areas with lower minimum wages and/or higher costs of living. For example, Washington State and California currently have the highest tipped minimum wage at $12/hour, while 18 other states follow the federal tipped minimum wage set at a mere $2.13/hour [18]. With the adoption of Square POS systems that accept tip payments before the service is received, the platform further “silences the tip’s feedback about customer satisfaction” and “requires the customer to make a moral choice” around compensating for labor upfront [2]. Since the outbreak of Covid-19, the societal pressure of tipping has further evolved into compensating essential workers putting their health and lives at risk and keeping local restaurants in business, even without a full-service sit-down dining experience. During the initial lockdowns in Spring of 2020, Square POS data showed average tip percentages increased by almost 2.5% in the US, with another spike during the third wave of the pandemic [1]. Notably, tipping also transformed into fundraising efforts and an existential rallying cry in support for AAPI community members and allies, especially in the midst of increasing racially-motivated, anti-Asian hate crimes and attacks throughout the pandemic [19].

Average tip percentages temporarily increased for restaurants using Square POS machines during the start of the Covid-19 pandemic and the third wave [1].

In terms of skills and knowhow, Square’s new UX flow has reshaped the tipping practice with unique consequences. Instead of manually doing the math themselves — which often meant pulling out a calculator or using round numbers as opposed to strict percentages and decimal points — customers now have access to pre-calculated tip amounts via a set of three default options, such as a 15%, 20%, and 25% tip, to select from with a single tap to expedite the calculation. These shortcuts are presented on the POS system before they complete the transaction, and intentionally appear larger and more prominent than the “no tip” or “custom tip amount” options below. Once customers select a tip option, there is no “undo” button if a mistake was made. As opposed to an alternative flow that defaults to no tips and asks customers to opt-in to providing a tip, this opt-out flow increases friction and reduces customers’ agency by privileging certain tip defaults without prior consent and avenues to backtrack. This UX design can ultimately be interpreted as an intentional and manipulative “dark pattern” to increase tipping rates from customers and benefit the servers and restaurant owners.

Lastly, because the server often is present and watching while the customer completes the tipping flow on the POS, customers now feel the pressure of “guilt tipping” into tipping a higher amount, or tipping at all in situations that they wouldn’t normally otherwise, as a result of feeling socially judged. As a result, Square has reshaped the meanings and motivations of the tipping practice by creating unique UX flows that constrain user agency and behavior.

Embodied Performances - Combined Analysis The pandemic and Square converged together to reconfigure the physicality of our interaction between customers and employees during the practice of tipping. Previously, we typically exchanged cash, cards, and receipts back and forth. We struggled with calculating the tip manually by hand or dropped change in tip jars next to the cashier. During the pandemic, servers and customers were required to follow health and safety guidelines, like masking and social distancing. Physical barriers created new boundaries that made it more difficult to communicate and hand things off to one another. Customers relied on “tap-to-pay” methods on card readers to further reduce contact, but needed to wait until the server and Square card reader were ready to accept the payment. The rest of the transaction was entirely carried out using our stubby fingers to navigate around Square UI to sign and select the tip and receipt options. People even learned to adapt to heightened sanitation measures like disinfecting the screens after each use and using hand sanitizer before and after each transaction.

Intersectionality, we can unpack how our tipping behavior disproportionately impacts women and single mothers, BIPOC, and LGBTQ communities who are already overrepresented¹ in the restaurant industry. Studies show that female, BIPOC, and LGBTQ workers are tipped less than their male, white, straight, and cisgender counterparts in general [3, 11, 12, 15]. Servers who identify within more than one community are more likely to rely on tips for their income and are at higher risk of overlapping marginalization [8]. Thus, our decisions on how much we tip on Square POS machines can be influenced by our implicit biases against perceived embodied axes of identities, privileging some groups while harming others.

We can also look at how embodied performances impact interpersonal interactions in a tipping practice, particularly for servers in their role as front-line workers. Given that the restaurant industry has the highest rates of reported harassment [18], servers face increased risk of intersectional marginalization within the inherent power imbalance of depending on customer satisfaction and tips for their livelihood. Tipping exposes tipped workers to higher rates of sexual harassment from customers than their non-tipped counterparts by almost 25% [14]. Further, women, BIPOC, and LGBTQ servers are more more likely to report sexual harassment than their male, white, straight, and cisgender counterparts — 45% more likely for women [17] and 15% more likely for LGBTQ servers [12]. The act of tipping opens up higher rates of unprovoked harassment from customers that increase for certain intersectional identities.

The pandemic has only exacerbated power imbalances and rates of harassment further for marginalized communities who also risked their health and lives to serve customers. Women and BIPOC women reported even higher rates of harassment [15]. In a study, 43% of women reported harassment around Covid-related safety measures, including a customer telling a server to “take off your mask so I can know how much to tip you” [5, 13]. Black workers who enforced Covid protocols received 11% fewer tips than their white colleagues [15]. Members of the AAPI community suffered from additional harassment and physical attacks driven by racism and xenophobia, including increased marginalization and risks for AAPI women, nonbinary, immigrants, LGBT, and elder individuals based on their overlapping identities [19]. Thus, the embodied elements of tipping unfairly fuels disproportionate discrimination and marginalization that perpetuates life-threatening harms against some of our most vulnerable communities.

Tools and Materials - Combined Analysis Square POS systems, with the pandemic as an enabler, replaced traditional tipping methods with digital tools and fewer physical artifacts. Instead of pens, physical receipts, and cash, we transitioned to interactive touchscreen terminals, digital receipts, credit cards, and tap-to-pay methods. With Square, the footprint for tipping also extends into the global infrastructure that stores and processes the transaction data in the cloud and sends customers their receipts via texts and emails. Servers and restaurant owners also now have the ability to manage their tipping data and options within the Square platform.

The shift in payment methods, however, is one that unfortunately carries with it significant discrmination against multiple communities. While the primary and recommended means of payment with Square is a credit card, not everyone has access to or is able to obtain one. For example, the percentage of people with credit cards is lower among Black Americans and Hispanics than it is for their white counterparts [9]. BIPOC and LGBTQ communities have also been found to have lower credit scores than other groups [9], which can both limit their ability to obtain a credit card and discourage them from using ones that they have. Square also encourages tap-to-pay payment methods with smartphones, but when it comes to smartphone ownership, those who are 65+, earn less than $30K/year, and live in rural areas have significantly lower ownership rates [16]. This illustrates the disproportional disadvantage of these payment methods for those who identify as female, BIPOC, lower income, and LGBTQ in different combinations that Square exacerbates further.

Lastly, interacting with the POS machines requires a certain level of digital literacy and motor/visual skills to navigate through the transaction in ways that perpetuates ableism as yet another layer of oppression. The Square POS systems are often fixed at one height or location, font style, color palette, and UI. Studies have shown that touchscreens present major accessibility issues for the elderly, as well as people with certain medical conditions or professions that involve working with their hands [7]. This can result in frustrating experiences for those customers, which may affect their satisfaction with the interactive UX and the amount they choose to tip or cause them to avoid transacting with establishments that rely on Square. Those with visual impairments can be further marginalized by the lack of tactile accessibility options on the Square touchscreens that are missing by default. With an intersectional lens, we have found that the shift in materiality within the tipping practice can create complex and compounding harms in our communities that we may have otherwise neglected.

Discussion

By combining Practice Theory and Intersectionality, we gain a deeper understanding of not only how the activity of tipping changed due to effects of the Covid-19 pandemic, but how those changes disproportionately affected already marginalized communities. Practice Theory uncovers how the Square POS systems impacts each element of the tipping practice by digitizing and automating the process, such as reducing customer agency by biasing them towards predetermined defaults, increasing the social and moral pressure of guilt tipping, and forcing wage compensation before the service is received. Intersectionality shifts the focus from the specific changes in the activity to how those changes disadvantage significant segments of our population. From servers who are at higher risks of harassment and discrimination based on their overlapping identities, to customers who are more likely to be excluded from the platform due to affordability and accessibility concerns, Square enables, encodes, and perpetuates existing power imbalances within the tipping practice through its platform and fuels new inequities through its features. With more data, we can drill into other marginalized communities and understand further disparities between identities like citizenship, religion, and language abilities and their own intersectional dynamics.

However, there are some limits when it comes to applying our approach. Despite its intent to explain general phenomenon and societal influences, the intersectional perspective has the potential to generalize the experiences of marginalized communities depending on what data is collected and how it is used. Many other personal and specific contextual factors that can impact someone’s tipping experience warrant future research. For example, Critical Race Theory can also help us understand how the origins of tipping has ties to slavery in the US [5] and unpack how the practice and the platform normalizes and perpetuates racism onto the design of the Square POS systems.

To better understand the agency that tipping tools and artifacts have on the practice and on creating social structures, sociomaterial theory can better highlight the entanglement between both the agencies of technology and culture in shaping each other. Posthumanism can also offer insights into nonhuman actors to understand the sustainability of the POS devices which may break down over time and the introduction of new global infrastructure to power the digital platform.

Future Research

This combined theoretical approach offers powerful capability which could be applied to other technology design problems relating to practices around consumer purchasing behavior and workflows. We can apply our analyses to other POS systems including ones that ask for charitable donations, online payment, billing, and subscription platforms, and financial, healthcare, and insurance applications. In each case, we can consider the social behavior around each example as a practice that is enabled by tools, skills, and performances that shape our institutions and perpetuate power imbalances and marginalization by design. By centering on the lived experiences of those most harmed, we can then design interventions to reverse discriminatory tools and practices. However, we may also need to rely on Critical Race Theory to uncover legacies of racism and exclusion in our political and economic institutions and Sociomateriality to analyze the agency tools themselves have on perpetuating harms.

Looking to the future, we can use the combined approach to generate new speculative research questions around uncovering hidden forms of discrimination in other transactional practices as we enter a new era of robots, automation, and intelligence. As we depend on more complex and invisible non-human actors to fully mediate our practices, behaviors, and life decisions, we can investigate what types of practices we codify into our technology as we build fully-automated systems that don’t require human-to-human interaction like with robots, autonomous transportation, and the metaverse. How might we rethink and reshape routines and practices into ones that are more equitable, accessible, and inclusive for marginalized and low resource communities who have traditionally been left out? However, we will need to draw from posthumanism and sociomaterial theory to complement our approach in order to account for the agency of novel non-human actors.

Conclusion

In this study, we applied elements of Practice Theory and Intersectionality to uncover new understandings of how the practice of tipping with Square POS systems changed during the Covid-19 pandemic. Our analysis highlighted the shift in physicality, materiality, and meaning, and how these transformations adversely impact already marginalized communities for both customers and servers. With a combined lens, we find that Square perpetuates harmful legacies of tipping and encodes marginalization within its infrastructure and platform by default.

Although our combined theoretical approach may not be adequate for explaining rich subjective experiences, it provided unique insights in this study and shows strong promise for many other design and research problems in which changes in technology have the opportunity to reinforce institutional discrimination, particulary within certain combinations of communities and identies. In the future, careful consideration should be given to understanding not only who new technology can benefit, but who it might disadvantage.

References

[1] Christopher Adams. 2022. Average Tip Percentages by Month. Flourish. Retrieved August 20, 2022 from https://public.flourish.studio/visualisation/8763501/.

[2] Jonathan Bean and Melanie Wallendorf. 2017. Tipping the scale. interactions 24, 5 (September - October 2017), 22–23. https://doi-org.offcampus.lib.washington.edu/10.1145/3125395.

[3] Zachary W. Brewster and Christine Mallinson. ‘Racial Differences in Restaurant Tipping: A Labour Process Perspective’. The Service Industries Journal 29, no. 8 (August 2009): 1053–75. https://doi.org/10.1080/02642060902764343.

[4] Kimberle Crenshaw. 1989. ‘Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics’, n.d., 31.

[5] Sarah E. Dempsey. 2021. Racialized and Gendered Constructions of the “Ideal Server”: Contesting Historical Occupational Discourses of Restaurant Service. Frontiers in Sustainable Food Systems 5:727473. https://doi.org/10.3389/fsufs.2021.727473.

[6] Kristen Dew, Class Lecture, Topic: “HCDE 501 Week 6: From interactions to practices”. Department of Human Centered Design and Engineering, University of Washington, Seattle, WA, July 19, 2022.

[7] Gabe Fender. 2022. Touch screens don’t work for everyone. Retrieved August 20, 2022 from https://www.gabefender.com/writing/touch-screens-dont-work-for-everyone.

[8] Laurel E. Fletcher. ‘Working Below the Line: How the Subminimum Wage for Tipped Restaurant Workers Violates International Human Rights Standards’. SSRN Electronic Journal, 2015. https://doi.org/10.2139/ssrn.2758842.

[9] Tamara E. Holmes. 2021. Credit card race, age, gender and income statistics. Retrieved August 20, 2022 from https://www.creditcards.com/statistics/race-age-gender-statistics/.

[10] Kari Kuutti and Liam J. Bannon. 2014. The turn to practice in HCI: towards a research agenda. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’14). Association for Computing Machinery, New York, NY, USA, 3543–3552. https://doi.org/10.1145/2556288.2557111.

[11] Michael Lynn and Zachary W. Brewster. ‘Racial and Ethnic Differences in Tipping: The Role of Perceived Descriptive and Injunctive Tipping Norms’. Cornell Hospitality Quarterly 56, no. 1 (February 2015): 68–79. https://doi.org/10.1177/1938965514553466.

[12] One Fair Wage. 2022. No Pride In Subminimum Wages: The Harsh Realities Of Being An LGBTQ Restaurant Worker. The UC Berkeley Food Labor Research Center. https://onefairwage.site/wp-content/uploads/2022/06/OFW_LGBTQ_RestaurantWorkers.pdf.

[13] One Fair Wage. 2020. Take Off Your Mask So I Know How Much to Tip You. Service Workers’ Experience of Health and Harassment During Covid-19. The UC Berkeley Food Labor Research Center. https://onefairwage.site/wp-content/uploads/2020/12/OFW_Covid_WorkerExp_MA_4.pdf.

[14] One Fair Wage. 2022. The Tipping Point: How the Subminimum Wage Keeps Incomes Low and Harassment High. The UC Berkeley Food Labor Research Center. https://onefairwage.site/wp-content/uploads/2021/03/OFW_TheTippingPoint_3-1.pdf.

[15] One Fair Wage. 2022. Unlivable: Increased Sexual Harassment and Wage Theft Continue to Drive Women, Women of Color, and Single Mothers Out of the Service Sector. The UC Berkeley Food Labor Research Center. https://onefairwage.site/wp-content/uploads/2022/04/OFW_Unlivable.pdf.

[16] Pew Research Center. 2021. Mobile Fact Sheet. Retrieved August 20, 2022 from https://www.pewresearch.org/internet/fact-sheet/mobile/.

[17] Restaurant Opportunities Centers United (ROC-United). 2014. ‘The Glass Floor: Sexual Harassment in the Restaurant Industry’, New York: ROC-United. http://hdl.handle.net/20.500.11990/609.

[18] Jacqueline Ross and John Welsh. ‘Understanding the “Tipped Minimum Wage”: Critical Directions for US Policy Research’. Social Policy and Society 20, no. 2 (April 2021): 192–210. https://doi.org/10.1017/S1474746420000214.

[19] Stop AAPI Hate. 2022. Two Years and Thousands of Voices What Community-Generated Data Tells Us About Anti-AAPI Hate. Retrieved August 20, 2022 from https://stopaapihate.org/year-2-report.

[20] Neha Kumar and Naveena Karusala. 2019. Intersectional computing. interactions 26, 2 (March - April 2019), 50–54. https://doi.org/10.1145/3305360.

---

1. In the US, around 67% of tipped restaurant workers are female, 30% of them are mothers, and 54% of them are single; around 45% of restaurant workers are BIPOC; and 18% of restaurant workers identified as LGBTQ, nearly double the national LGBTQ identification rate [8, 12].↩︎

---

---