Labor advocates are seeking to abolish restaurant tipping, even at fast-casual restaurants such as Applebee’s, Olive Garden, and TGI Friday’s. Tipping is supposedly a repugnant practice because it dehumanizes servers who “must grovel for our change” to make their living. It also supposedly perpetuates poverty among servers, “enshrines” racial and gender discrimination, and encourages sexual harassment. Moreover, tipping does not improve customer service, or so the abolitionists say.

Tipping abolitionists seek to replace tipping with a “living wage” of maybe $15 or $18. They think servers would be ecstatic with the change in their compensation from tip-dependent income that the abolitionists consider low and variable, to non-tip-dependent income that the abolitionists believe would be consistently higher. Yet my research indicates the case for the universal abolition of tipping is misguided on both factual and theoretical grounds, at least for fast-casual restaurants with table service—a sector that especially concerns tipping abolitionists.

Better income? / I investigated the economics of tipping last fall by surveying 40 servers in fast-casual restaurants in Orange County, CA. (To be included in the survey, restaurants had to offer on their menus a cheeseburger-fries combination priced between $8 and $12, and beer and wine.) When asked, most servers had a fairly firm idea of their tip income. Some consulted a smartphone app on which they recorded their daily tips and computed their average hourly total income, which they gave me.

I posed a question to each privately, “What hourly wage would cause you to voluntarily give up your current compensation package, including the state’s minimum wage and your tip income?” The responses ranged from $18 to $50 an hour. The median no-tipped wage was $30 an hour. Some 62 percent of the servers demanded an hourly wage of at least $30 to replace their current compensation. To put that in perspective, if two married servers work full-time and earn $30 an hour, they would have an annual household income of $124,800, putting them in the top 17 percent of households by income (and yet they would still be entitled to the California minimum wage, which supposedly targets only low-income workers).

This response is apparently not unique to Orange County. In the spring of 2016, I similarly interviewed 10 servers in piedmont North Carolina. They had a median response of $26 an hour, which is actually higher than the median hourly wage demanded in Orange County after adjusting for the difference in the cost of living.

Tipping abolitionists might be surprised to learn that all servers surveyed chortled at the suggested replacement of their tip incomes with a “living wage” of $15 an hour. Most servers responded with comments of the essence, “How stupid can these people be?”

Better service? / Studies on the effect of tipping on service, and vice versa, suggest a positive but very weak tie between tipping and service quality. However, these studies are replete with problems, not the least of which is that one prominent study used “subjects” who were community college students. The students were asked to assess the “quality” of their restaurant service (on a scale of 1 to 5) over some undefined period of time as the students ate at different restaurants at different times of the day. The students were asked to self-report their tips and service quality, which means their subjective assessments of service quality could vary widely across students and through time. The researchers found that servers’ tip income rose only 2 percent for a 1‑point increase in reported service quality on the 5‑point scale—hardly a reliable finding.

To examine this issue, I asked the servers I interviewed if the service they provided affected their tips. All strongly agreed it did. Indeed, servers said that if they raised their service level from a “3” (average service) to a “4” (above-average), their average tip percentage (not total tips) would rise by over 25 percent. If they elevated their service from “4” to “5” (excellent service), their average percentage tip would rise another 25 percent, which means that an increase in service level from average to excellent would raise their average percentage tips by 57 percent. All servers strongly agreed that overall service quality would drop precipitously if their tip income were replaced with a fixed hourly wage, especially for “loud,” “obnoxious,” and “arrogant” customers, as well as customers with unruly and messy children.

Particular knowledge / Tipping abolitionists portray tipping as a loss to everyone: servers, owners, and customers. They don’t understand that an array of restaurants face a serious management problem that economists call the “principal/​agent problem.” Managers and owners may know much about the restaurant business, but they don’t know all that their servers know about particular customers, especially “regulars”—such as their names, what they typically order, and how they want to be treated (whether they want to talk or be left alone). Many customers want some control over the service they receive, and tips give them that control. Servers want to be rewarded for their extra effort. Owners want servers to use their local knowledge to increase sales and repeat visits. One way of doing this is to make servers de facto commissioned salespeople, making sure that servers are partial residual claimants on the added service they provide and the resulting additional sales and profits the restaurants garner.

Granted, several restaurant chains are currently experimenting with no-tipping policies, combining them with higher worker wages and menu prices. While such policy changes may work for some restaurants in fast-food markets (whose products are standardized and whose workers can be easily monitored by managers), they are likely to be counterproductive for many other restaurants in fast-casual and fine-dining market segments (where service must, to one degree or another, be tailored for customers’ needs).

In 2015, the owners of two pricey San Francisco restaurants, Bar Agricole and Trou Normand, instituted a 10-month trial of a no-tipping policy, higher base wage, and higher menu prices to finance the new wage. The restaurants returned to tipping in 2016 because server morale crashed and turnover skyrocketed to 70 percent during the trial period. Why? The company had lowered servers’ de facto hourly wage from a range of $35–$45 to $20–$35.

Likewise, national fast-casual seafood chain Joe’s Crab Shack tried a no-tipping policy with higher prices at a handful of its locations in late 2015, but returned to tipping in spring 2016. Management gave two reasons for the return: they would have to raise worker wages—and thus menu prices—far higher than they had expected in order to satisfy workers’ demand for a no-tip wage, and they found their customers wanted the return of tipping.

Nevertheless, no-tipping policies could spread in the near future as more states and localities increase their mandated minimum benefits and hourly wages (now scheduled to go to $15 an hour in California and New York) and remove “tipped wage” exemptions. Many restaurant executives will begin to see server tips as a revenue stream they can tap to offset some of their higher mandated labor cost increases, implementing no-tipping policies in exchange for menu price increases.

Tipping abolitionists may be surprised to find that some of the most ardent opponents of tipping abolition are servers and their customers. One North Carolina server volunteered: “I made $60,000 in tips last year, reported $40,000—and had a before-tax income of $80,000! That’s why I quit my teaching job.” And customers will likely suffer impaired service as the tipping incentive disappears.