Covering shifts with overtime because a station is short? We advertise your openings to people already on Facebook.
Book a demoEmployer guideRead at source, 9 September 2026
Restaurant Labor Costs in 2026: The Industry Spends 28.9% of Sales on Wages, and That Share Has Been Falling for Six Years
Every article about restaurant labor costs quotes a rule of thumb — 25 to 35 percent of sales — and almost none of them says where it comes from. So we worked it out from the federal data. Multiply the number of people employed in food services and drinking places by their average weekly hours and average hourly earnings, and divide by what the industry sold, and the wage bill came to 28.8 percent of sales in 2025 and 28.9 percent in the first half of 2026, down from 32.3 percent in 2017. That is wages only, so your own figure will be higher once payroll taxes and benefits are in it. The striking part is the direction: the average hourly wage in this industry is up 56 percent since 2017, and the share of the check it consumes has still fallen, because sales per employee rose faster. Meanwhile the one cost that has not improved is churn — the sector loses 4.1 percent of its people to voluntary quits every month, roughly twice the all-industry rate.
What the industry actually spends on wages
There is no federal statistic called “restaurant labor cost percentage.” There are two statistics that, divided into each other, produce one. BLS publishes how many people the industry employs, how many hours they average in a week and what they average an hour. Census publishes what the industry sold. The first three multiplied together are a wage bill; the wage bill over the sales is a share.
| Year | People employed | Average hourly earnings | Average weekly hours | Wage bill | Industry sales | Wages as a share of sales |
|---|---|---|---|---|---|---|
| 2017 | 11.72m | $14.01 | 25.6 | $218.7bn | $678.1bn | 32.3% |
| 2019 | 12.07m | $15.13 | 25.6 | $242.5bn | $757.3bn | 32.0% |
| 2021 | 10.68m | $17.02 | 25.8 | $244.2bn | $825.7bn | 29.6% |
| 2023 | 12.16m | $19.67 | 25.1 | $312.3bn | $1,066.1bn | 29.3% |
| 2024 | 12.22m | $20.53 | 25.0 | $326.5bn | $1,121.3bn | 29.1% |
| 2025 | 12.27m | $21.23 | 25.1 | $339.7bn | $1,181.1bn | 28.8% |
| 2026 (Jan–Jun) | 12.30m | $21.86 | 25.0 | $175.0bn | $604.6bn | 28.9% |
Employment, hours and earnings are annual averages of the monthly BLS series for food services and drinking places, not seasonally adjusted. Sales are the sum of the monthly Census series for the same industry. Both read 9 September 2026. Using the seasonally adjusted sales series instead moves every share by less than two-tenths of a point.
Why your number is higher than 28.9 percent
This is a wages-only figure: it counts what lands on the pay stub and nothing else. Your labor line almost certainly also carries the employer half of payroll taxes, unemployment insurance, workers’ compensation and whatever you spend on benefits, and those together commonly add a fifth to a third on top of wages. It is also an industry aggregate that includes bars, caterers and every quick-service chain with a drive-through, so it averages together businesses with very different service models. Treat 28.9 percent as the floor the whole industry sits on, not as a target for your restaurant.
The wage went up 56 percent. The share went down.
The dominant story about this industry for the last five years has been that labor costs are crushing it. The wage half of that is true and then some: average hourly earnings in food services went from $14.01 in 2017 to $21.86 in the first seven months of 2026, a rise of 56 percent. Over the same stretch the wage share of sales fell by three and a half points. Both things are facts, and reconciling them is the whole point of the page.
The reconciliation is sales per employee, which rose from about $57,900 to $96,300 between 2017 and 2025 — 66 percent against the wage’s 52 percent over the same years. Some of that is menu prices. Some of it is a genuinely smaller crew doing more covers: the industry is running on 25.0 average weekly hours against 25.6 in 2017, and it employed only about 200,000 more people in 2025 than it did in 2019 while selling 56 percent more. However the mix breaks down, the arithmetic says the same thing. Wages did not outrun revenue. They chased it and lost ground.
If your own labor percentage has gone the other way over those years, that is worth knowing precisely because it is not what the industry did. It usually means one of three things: your prices moved less than the market’s, your covers per labor hour fell, or you are paying for the same shift more than once.
The line that did not improve
Paying for the same shift more than once is what churn is, and it is the cost that never appears as its own row on a P&L. Accommodation and food services runs the highest quit rate in the American economy and it has barely moved in a decade of pay rises.
| Year | Monthly quit rate, accommodation and food services | Monthly quit rate, all industries | Quits in the sector that year |
|---|---|---|---|
| 2019 | 4.9% | 2.3% | 8.28m |
| 2021 | 5.8% | 2.7% | 8.48m |
| 2022 | 5.8% | 2.8% | 9.45m |
| 2023 | 5.0% | 2.4% | 8.47m |
| 2024 | 4.2% | 2.1% | 7.03m |
| 2025 | 4.2% | 2.0% | 7.25m |
| 2026 (Jan–Jul) | 4.1% | 2.0% | 4.12m |
Source: BLS Job Openings and Labor Turnover Survey, accommodation and food services (NAICS 72) and total nonfarm, seasonally adjusted monthly rates averaged over each year, read 9 September 2026. Quits counts are the sum of the monthly levels.
Two things follow. The first is that the sector’s quit rate has come down since 2022 and is now below where it sat in 2019, which is the opposite of the “nobody wants to work” framing. The second is that even at 4.1 percent a month it is still twice the rate of the economy as a whole, and a restaurant that keeps a hundred people on the schedule is replacing roughly four of them every month before it has grown by a single cover. The cost of that sits in three places on your P&L and none of them is labeled turnover: overtime paid to cover the gap, the training hours a new starter absorbs, and the sales you did not make because the section was closed. We work through what a refill actually costs on our page about cost per hire in restaurants, and the turnover rates themselves are on restaurant turnover rate.
Work out your own number
Put a week of your own figures in and this gives you the wages-only share on the same basis as the federal number above, plus the fully loaded share once your taxes and benefits are counted. The comparison line is the one that matters: it tells you whether you are above or below what the whole industry runs at.
Restaurant labor cost calculator
One typical week. Hours are all paid hours, hourly and salaried together.
—
The comparison is the 28.9 percent industry wage share for January to June 2026, computed above from BLS and Census data. It is wages only, which is why the fully loaded figure is shown separately rather than compared against it.
What filling the shift costs
Here is where we have to be careful about our own data. Boostpoint runs social recruitment advertising, and our 2026 Social Job Advertising Benchmark covers 891 campaigns — but restaurant campaigns are a small enough slice of it that we do not publish a restaurant cost per applicant, and we would rather say that than dress an adjacent number up as one. The nearest honest reference points are the customer service and administrative role family, which ran a $2.71 median cost per applicant at a 25 percent apply rate, and event-driven campaigns built around a dated hiring day, which ran a $8.02 median at 21 percent across 78 campaigns. Restaurant roles behave more like the first than like a nursing campaign, and a hiring-day structure is the one most operators actually want.
The reason any of that belongs on a page about labor cost is the arithmetic in the section above. If wages are already the smaller share of sales they have been in a decade, the money left on the table is not in the wage rate. It is in the shifts you covered with overtime, the covers you did not turn because a station was short, and the four people in a hundred walking out the door every month. Those are recruiting-speed problems, and speed is the one thing advertising can actually buy.
Frequently asked questions
What percentage of sales should restaurant labor cost be?
Across the whole industry, wages came to 28.8 percent of sales in 2025 and 28.9 percent in the first half of 2026, computed from BLS employment, hours and earnings for food services and drinking places against Census sales for the same industry. That is wages only. Once payroll taxes, workers’ compensation and benefits are added, most operators land higher, which is where the familiar 30 to 35 percent rules of thumb come from.
How do you calculate restaurant labor cost percentage?
Divide total labor cost for a period by total sales in the same period and multiply by 100. The decision that matters is what goes in the numerator: wages alone gives you a number comparable to the 28.9 percent industry figure, while wages plus payroll taxes, benefits and workers’ compensation gives you the fully loaded figure your accountant uses. Compare like with like or the benchmark is meaningless.
Are restaurant labor costs rising?
Wages are, sharply: average hourly earnings in food services rose 56 percent between 2017 and 2026, from $14.01 to $21.86. Labor cost as a share of sales is not: it fell from 32.3 percent in 2017 to 28.9 percent in the first half of 2026, because sales per employee rose faster than pay did.
What is the average hourly wage in restaurants?
The BLS average hourly earnings figure for all employees in food services and drinking places was $21.86 across January to July 2026, against $21.23 for 2025 and $14.01 for 2017. Average weekly hours have drifted slightly down over the same period, from 25.6 to 25.0.
How high is restaurant turnover?
Accommodation and food services averaged a 4.1 percent monthly quit rate across January to July 2026, against 2.0 percent for all industries. That is down from a 5.8 percent peak in 2021 and 2022 and slightly below the 2019 level, but it is still roughly double the economy-wide rate. The sector recorded 7.25 million quits in 2025.
Does cutting labor hours lower labor cost percentage?
Only if sales hold. The percentage is a ratio, and the industry lowered it over the last six years mainly by raising the denominator rather than by cutting the numerator. If cutting hours costs you covers, closes a station or slows service enough to shrink the check, the ratio can get worse while the payroll gets smaller.
What is a good sales per labor hour figure?
The industry-wide figure implied by the federal data is roughly $74 of sales per paid hour in 2025, calculated from about $1,181bn of sales against 12.27 million employees at 25.1 hours a week. Service models vary enormously around that, so it is more useful as a direction of travel for your own restaurant than as a target.
What does it cost to hire restaurant staff?
We do not publish a restaurant-specific cost per applicant because restaurant campaigns are a small share of our benchmark. The nearest reference points from our 2026 Social Job Advertising Benchmark are the customer service and administrative family at a $2.71 median cost per applicant and event-driven hiring-day campaigns at an $8.02 median. The full cost-per-hire ledger, including the interview and training time that dwarfs the advertising, is on our restaurant cost-per-hire page.
The wage line is not your problem. The unfilled shift is.
Bring your weekly sales, your paid hours and your turnover. We will show you what applicant flow costs in your market, and what the shifts you are covering twice are costing you now.
Book a Demo