Restaurant groups: The data professor got Boulder restaurant wages wrong

This commentary is by the Colorado Restaurant Association, EatDenver, Boulder Chamber, Visit Boulder and the Downtown Boulder Partnership board of directors. The organizations represent restaurants, tourism businesses and employers in Boulder and across Colorado. 

Read more Federal complaint accuses Boulder Valley School District of failing Jewish student amid years of antisemitic bullying

Professor Brian Keegan opens his May 10 column in Boulder Reporting Lab by accusing Boulder City Councilmember Matt Benjamin of spreading a myth about what restaurant workers earn. That’s a serious charge, especially when Keegan’s own analysis contains a methodological flaw that he acknowledges but then proceeds to ignore.

The column was written amid an ongoing Boulder City Council debate over how restaurant workers who receive tips should be paid.

Councilmembers are considering changes to the city’s tip credit, which allows restaurants to count a portion of employees’ tips toward meeting minimum wage requirements. A central focus of Keegan’s column was disputing estimates that many restaurant workers earn about $40 an hour in total compensation and arguing that the figure was being used to justify slowing planned wage increases for tipped workers.

We represent Boulder restaurants and a broad cross-section of businesses invested in the vitality of our local economy. Every restaurant operates differently, finding its own way through the often-impossible puzzle of making an independent restaurant work. Some use the tip credit; some don’t because forgoing it lets restaurants legally pool tips with kitchen staff and narrows the wage gap between the front and back of the house. Some restaurants don’t accept tips at all. Some opt for service charges. If you don’t know how all that works, welcome to the mayhem that is the restaurant industry.

But whatever path restaurants take, they share one goal while navigating the constant balance between math, regulation and the people those things affect: to pay their entire staffs —including cooks, dishwashers, servers and bartenders — as well as they possibly can, and still eke out the 3-5% margin that keeps an independent restaurant alive. That’s how narrow the fine line is between staying open and closing.

Keegan admits his hourly math is wrong, then uses it anyway.

Keegan derives his central $22-$23 per hour figure by taking annual Census earnings for “food service workers” and dividing by 2,080 hours, the equivalent of a 40-hour, year-round work week. He then uses that figure to declare that Benjamin’s $40-per-hour estimate is “wrong by nearly half.”

But here is what Keegan writes in the very same paragraph: “Both figures assume full-time, year-round work, which skews the numbers upward compared with the part-time, slow-Tuesday reality that defines much of restaurant work.” He flags his own math as inflated and then publishes it as the rebuttal anyway. Apply a more honest assumption of 25 to 30 hours per week, typical for a mixed workforce of full- and part-time restaurant employees, and that same $46,305 in annual earnings translates to roughly $29 to $35 per hour. The gap Keegan claims exists is substantially a product of math he himself admits is skewed.

The ACS “food service workers” category compounds the problem further. It includes cafeteria employees, fast food counter workers, school lunch staff, hospital kitchen workers and dishwashers, the majority of whom earn no tips at all. It is not a measure of what tipped servers and bartenders earn at full-service restaurants, which is the only population this policy debate is about.

Keegan is not correcting a myth; he’s building a bigger one.

The workers who need help most are being ignored.

This raises a question the policy debate consistently sidesteps: Why are policymakers focused on the highest-paid workers in restaurants — tipped servers and bartenders — when their back-of-house counterparts, many of them from the underserved communities this policy claims to champion, are the ones earning wages closest to the minimum?

The hard truth, in our view, is that mandatory tipped-wage increases do not primarily benefit the workers most in need. Higher fixed-labor costs on the front of the house make it harder, not easier, for operators to raise wages in the kitchen.

The real myth: A higher wage floor raises what workers take home.

A 2026 Denver restaurant industry report compared Denver, whose tipped minimum wage of $16.27 per hour is the highest of any major American city, to Austin, Dallas and Houston, where tipped workers earn $2.13 per hour.

Read more Anonymous Reddit attack rattles Boulder CU Regent primary race

The finding: Servers and bartenders in all four cities earn essentially the same total hourly compensation. Denver’s dramatically higher base wage does not produce higher take-home pay for servers. It produces higher fixed-labor costs for operators, costs that crowd out back-of-house wages and must be covered before a single table is seated, regardless of how slow a Tuesday turns out to be.

The gap between what a Denver operator pays and what an Austin operator pays doesn’t go to workers. It disappears into structural overhead. With Boulder’s full minimum wage scheduled to jump by 8% to $18.17 in 2027, Denver is not a hypothetical; it is a preview for Boulder.

One Denver server, quoted in recent reporting, put it plainly: “Even with the raises, I don’t feel like I’m walking away with more money.” Nationally, full-service tip rates fell to 19.1% in mid-2025, the lowest Toast POS has recorded in seven years. The workers this policy intends to protect are being squeezed from both ends.

Restaurants can’t fix Boulder’s cost of living. Neither can this policy.

Restaurant owners pay Boulder rents too. They watch their teams navigate the same impossible housing math. But raising one input cost doesn’t solve the underlying problem; it merely shifts the burden onto operators, onto untipped back-of-house workers who see no direct benefit, and ultimately onto the front-of-house workers themselves when their hours are cut or restaurants close. Driving up a single factor in an already broken affordability equation is attacking a symptom. The root causes of housing costs, commercial rents, the structural squeeze on everyone in a service economy in a city this expensive, don’t move. Only the restaurants do.

Boulder’s full-service restaurant employment has already declined 3.4% from 2023 to 2025. The number of full-service restaurants here is down 4% in two years, all before the 2027 wage increase has taken effect. To argue there is no crisis, Keegan cites establishment counts through 2023, stopping precisely the year before conditions deteriorated.

Three cities tried this and ultimately reversed course.

Massachusetts voters rejected full tipped-wage elimination by a vote of 64–36 in November 2024. Washington, D.C., was forced to roll back its tip-credit elimination in emergency legislation in 2025 after record business closures and workers testified that service charges were cutting their take-home pay. In March 2026, the Chicago City Council voted 30–18 to freeze its One Fair Wage ordinance, with restaurant owners and servers appearing together at the same press conference to call for the pause because workers said their hours and earnings were falling.

Restaurants know their teams. This isn’t corporate capitalism.

Nobody opened a restaurant in Boulder to get rich, and nobody opened one to stick it to their employees.

Over 90% of the restaurants downtown are independently owned. These aren’t regional managers and labor-cost spreadsheets. These are people who live here, whose kids go to school here, who are personally on the hook for the lease and the equipment and the payroll every two weeks. Boulder has over 400 eating and drinking establishments serving a city of 108,000 people. That didn’t happen because some algorithm identified us as an underserved market. It happened because we kept betting on this place and on each other. And this dining scene is a genuine part of why Boulder is Boulder. Residents here have a quality of life that most cities would trade for in a heartbeat, and the food and drink community is a big piece of that.

The people running our restaurants know exactly who works for them. They know who’s saving for a car, who just had a baby, who needs Tuesdays off to deal with something at home.

The question restaurants are asking Boulder’s City Council is not whether local workers deserve more. The question is whether this policy actually delivers more to them, or whether it accelerates restaurant closures, reduces hours, and leaves workers with no better total compensation and fewer places to work.

Professor Keegan accused Councilmember Benjamin of building policy on a myth. The myth worth examining is the one Keegan half-admitted in his own column: that you can divide annual earnings by 40 hours, wave away the caveat, and call the result definitive. Boulder’s hospitality workers, front of house and back of house, deserve a policy debate built on honest numbers. Restaurants are still waiting for one.

Read more 🏭 Xcel advances Boulder coal ash cleanup. Questions remain.