Why does the suggested tip only ever go up?
The suggested tip keeps climbing because a tipping norm behaves like a ratchet: it has a mechanism that pushes it up and almost nothing that pulls it back down. The standard US restaurant tip went from 10% in the first half of the 1900s, to 15% by the 1980s, to 20% today — and the minimum the screen suggests keeps creeping higher still. Each new average quietly becomes the next baseline. The number rises, sticks, and rises again.
What makes 2026 strange is that the climb continues into open revolt. Popmenu’s 2026 consumer study found 78% of Americans say tipping practices have “become ridiculous,” and 44% report tipping less than they did a year ago. Yet in the same survey, 74% noticed restaurants raising the suggested minimums on their screens. The backlash and the ratchet are running at the same time.
Sources: Tim Sablik, “Tipping: From Scourge of Democracy to American Ritual,” Econ Focus, Federal Reserve Bank of Richmond (2024); Popmenu, “Consumer Tipping Study” (2026)
What is the tipping ratchet?
The tipping ratchet is the self-reinforcing loop that turns this year’s average tip into next year’s expectation. In 2025, Dartmouth’s Laurens Debo and Tel Aviv University’s Ran Snitkovsky published the first formal model of it in Management Science. They built tipping from two motives the research has long documented: the urge to show appreciation, and the pressure to conform to the going rate. Run those two motives forward over many rounds and the math converges on a rising norm.
In their model, the “recommended tip” is simply the average tip in the market — the figure printed at the bottom of the check or pre-loaded on the screen. Each period, people tip, the new average is computed, and that average becomes the recommendation for the next period. Two groups emerge at equilibrium: those who genuinely value the service and tip generously, and the rest who tip “just to avoid negative feelings.” The generous ones do the lifting.
Appreciators pull the conformists upward. The conformists never pull them back down.
That single asymmetry is the whole ratchet. Generous tippers raise the average; everyone else follows the average to avoid social pain; the new average becomes the new floor. There is no equal and opposite force — no group whose discomfort drags the norm lower — so the equilibrium only travels one way.
Tip above the norm to signal genuine gratitude. Their generosity lifts the average — and the average is the recommendation everyone else inherits.
Tip at the norm to avoid the discomfort of standing out. They follow the line wherever it has moved, which locks each new high in place.
Source: Debo & Snitkovsky, “A Modeling Framework for Tipping in the Presence of a Social Norm,” Management Science (2025)
How does a number become the new normal?
A number becomes the norm through a feedback loop with no thermostat. You don’t tip what the service was worth — the research finds tips barely track service quality at all. You tip what you think other people tip, because tipping below the perceived rate carries a real psychological cost. And most people don’t actually know the rate: Pew Research found only about a third of Americans feel it’s easy to know how much to tip. Unsure, they default to the figure on offer — which is the rising average fed back to them — and matching it is exactly what re-sets the rate for the next person in line.
“If everybody tips 20 percent, I don't want to be the person who tips 5 percent, unless the service really stinks.
Laurens Debo, Tuck School of Business, Dartmouth
Debo and Snitkovsky show that when people are highly sensitive to that social pressure, the loop overshoots into what they call a “tipping war” — an “unnecessarily aggressive” equilibrium where the norm climbs past anything service quality could justify. The suggested figures on the screen aren’t describing what’s fair. They’re nudging the average up, and the average is what gets recommended back to you next time. The mechanism that turned 10% into 15% and then 20% is the same one straining to lift the norm again.
Sources: Debo & Snitkovsky, “A Modeling Framework for Tipping in the Presence of a Social Norm,” Management Science (2025); Drew DeSilver & Jordan Lippert, “Tipping Culture in America,” Pew Research Center (2023)
Wasn’t tipping once un-American?
It was — which proves the ratchet had a beginning and isn’t a law of nature. For the first half of the 20th century, tipping in the US was, in the Richmond Fed’s words, “rare and reviled.” Reformers called it a “scourge of democracy,” an aristocratic habit that turned free citizens into servants chasing coins. The opposition was organized and serious — and for a while, it was winning.
Six states ban tipping outright — Arkansas, Iowa, Mississippi, South Carolina, Tennessee, and Washington pass laws criminalizing the giving or soliciting of tips, with fines and, in South Carolina, even jail time.
The Itching Palm. Social activist William R. Scott’s anti-tipping book frames gratuities as fundamentally un-American: “Every tip given in the United States is a blow at our experiment in democracy.”
The bans collapse. The laws prove ineffective, are largely ignored, and are repealed or struck down. Tipping survives — and the ratchet starts its long climb.
10% becomes 15%. The customary tip of the early century quietly resets upward; by the 1980s, 15% is the new baseline.
15% becomes 20% — and screens raise the minimum. Digital prompts accelerate the climb, and tips spread into counter service, pickup, and delivery.
The history matters because it kills the excuse that 20% is simply “what tipping is.” A century ago, the right number was zero, and six states wrote that into law. The figure has only ever traveled in one direction since — not because service got better, but because the ratchet only turns one way. Economist Ofer Azar, surveying the field in the Journal of Economic Perspectives, notes that tipping persists as a “welfare-increasing and sustainable social norm” precisely because it’s voluntary and self-enforcing — no law required.
Sources: Tim Sablik, “Tipping: From Scourge of Democracy to American Ritual,” Econ Focus, Federal Reserve Bank of Richmond (2024); Ofer H. Azar, “The Economics of Tipping,” Journal of Economic Perspectives (2020)
If everyone hates it, why doesn’t the norm fall?
Because the backlash hits behavior, not the suggestion — and the two move independently. In Popmenu’s 2026 data, realized tipping is fraying everywhere: 44% of people tip less than last year, the share leaving 20%+ for servers slid to 41% (from 45% the prior fall), and the rate of pickup orders carrying a digital tip fell from 78% in 2022 to 62% in 2026. But the suggested figures kept climbing — 74% of diners watched restaurants raise the minimums on the screen. People are pushing back with their wallets while the anchor ratchets up over their heads.
The split that explains 2026: a tip norm has two layers — the expectation (the number the screen recommends) and the compliance (what people actually leave). The ratchet lives in the expectation, where appreciators and rising defaults only ever push up. Fatigue lives in the compliance, where tired consumers shave a few points off. The expectation can keep rising even as compliance cracks — which is exactly the contradiction the 2026 numbers describe.
This is why “everyone’s fed up” never translates into a lower norm. There’s no organized appreciator-equivalent on the downside — no group whose discomfort with over-tipping is strong enough to drag the recommendation lower. The most a fatigued tipper does is quietly under-comply, which the model treats as noise around a norm that’s still anchored high. To actually reset the ratchet, you’d need a coordinated shift, not a million private grumbles. (For the consumer-frustration side of this story, see iPad tipping fatigue.)
Source: Popmenu, “Consumer Tipping Study” (2026)
Does tipping more actually buy better service?
No — and that’s the quiet punchline of the ratchet. People believe they tip to reward good service and punish bad service, but decades of data show tips barely move with service quality. Debo and Snitkovsky’s model corroborates it: the social norm is so strong that even after poor service, most people still tip near the going rate to avoid the discomfort of stiffing someone. The norm, not the service, sets the number.
Popmenu’s 2026 survey makes the same point from the consumer side: 66% of people have tipped a worker despite receiving poor service. If tips were a performance review, that figure would be near zero. So what is the rising tip actually doing? In the model’s terms, it’s a transfer that firms rely on to keep workers when labor is mobile — closer to a pricing tool than a reward.
“Tipping is in essence a form of price skimming. It allows firms to extract rents from customers that can be used to retain workers.
Laurens Debo, Tuck School of Business, Dartmouth
That reframing is the reason the ratchet feels so frustrating. If higher tips bought better service, the climb would at least buy something. Instead, the model finds tipping is “a rather weak incentive” for service effort — restaurants would do better to train or pay their staff directly. The number goes up; the service it supposedly rewards mostly doesn’t.
Sources: Debo & Snitkovsky, “A Modeling Framework for Tipping in the Presence of a Social Norm,” Management Science (2025); Popmenu, “Consumer Tipping Study” (2026)
What did the 2026 backlash actually change?
It split the screen from the wallet. The suggested tip kept ratcheting up while the tip people actually leave drifted down across nearly every category. The table below lines up the two halves of Popmenu’s 2026 data — the expectation rising on the left, compliance falling on the right — so the contradiction is impossible to miss.
| The suggestion (ratcheting up) | The behavior (fraying down) |
|---|---|
| 74% noticed restaurants raising suggested minimums | 44% tip less than they did a year ago |
| Restaurants raised the minimum suggested tip on screens (e.g. from 10% to 15%) | 20%+ tips for servers fell to 41% (from 45%) |
| Prompts now appear at pickup, counters, kiosks | Pickup digital-tip rate fell from 78% (2022) to 62% |
| 59% still feel pressured to tip at a screen | Coffee-shop tipping dropped to 39% (from 46%) |
Read together, the numbers say the backlash dented compliance but never touched the anchor. That’s the ratchet’s signature: pressure builds, behavior gives a little, and the recommended number stays put — or rises — waiting for the next round of appreciators to lift it again.
Source: Popmenu, “Consumer Tipping Study” (2026)
How does the ratchet show up when a group splits the bill?
At a group table, the ratchet runs in miniature and the stakes scale with the total. One person says “let’s just do 20%,” or someone taps the highest figure the screen suggests on a $400 check, and that number becomes the table’s norm for the night. Nobody wants to be the friend arguing the tip down, so the highest comfortable number wins — the same appreciator-pulls-conformists dynamic, compressed into one dinner.
Then the split makes it worse. When the group divides the bill evenly, the person who ordered a salad and water pays the same tip dollars as the person who had two cocktails and the steak. The tip rides on top of an already-uneven bill, so an even split quietly taxes the light orderers to subsidize the heavy ones. The norm sets the percentage; the method decides who actually carries it. (For whether an even split is fair in the first place, see is splitting the bill evenly fair?)
The compounding problem: a ratcheted-up tip percentage applied through an even split is two unfairnesses stacked. You can’t easily fix the percentage — that’s a social norm — but you can fix the split, so the tip lands proportionally on whoever ran up the bill that produced it.
How does splitty split the tip without judging it?
splitty doesn’t fight the norm — it makes sure whatever tip your table lands on gets divided fairly. Scan the receipt, assign each item to whoever shared it, and splitty distributes the tip proportionally to each person’s share instead of in equal slices. The person with the $14 cocktail carries more of the tip than the person who had water, because the tip follows what each person actually ordered.
The ratchet is a social norm; no app reverses it. What splitty fixes is the part you can control — making sure the tip you already paid is shared by what each person ordered, so the rising number at least lands on the right people. Settle tonight’s receipt and move on; the norm fight can wait for the next table.
FAQ
Frequently asked questions about rising tip norms
Research-backed answers to the questions behind the ever-climbing suggested tip.
01 Why does the suggested tip keep increasing?
Because a tipping norm is a ratchet. A 2025 Management Science model by Debo and Snitkovsky shows that generous 'appreciator' tippers pull the average up, everyone else conforms to the new average to avoid social discomfort, and that average becomes the recommendation for next time. There's no equal force pulling it down, so the standard climbed from 10% in the early 1900s to 20% today.
02 Are Americans tipping less in 2026?
Yes, in behavior but not in expectation. Popmenu's 2026 study found 44% tip less than a year earlier, and the pickup-order digital-tip rate fell from 78% in 2022 to 62% in 2026. But 74% also noticed restaurants raising the suggested minimums on screens — so compliance is fraying while the recommended number keeps rising.
03 Does tipping more actually improve service?
Barely. Tips track service quality very weakly, and the Debo–Snitkovsky model corroborates it: the social norm is strong enough that 66% of people tip even after poor service. The researchers describe tipping as 'a rather weak incentive' for service effort and closer to 'price skimming' that helps firms retain workers.
04 Why does the minimum suggested tip on the screen keep rising?
Preset percentages raise the perceived norm, and the norm is what the model says people match. The screen recommendation is essentially the average tip fed back to you, so a higher suggested minimum nudges the average — and the next recommendation — upward. In Popmenu's 2026 study, 74% of diners noticed restaurants raising the minimum suggested tip on digital screens. For the choice-architecture mechanics of these prompts, see our guide to counter service tipping.
05 Will tipping norms ever go back down?
Not easily. The ratchet only has an upward force; private frustration shows up as under-tipping, which the model treats as noise around a norm still anchored high. A century ago tipping was reviled and even banned in six states, which proves it can change — but reversing it would take a coordinated shift, not scattered grumbling.
