Restaurant Staff Turnover: The Real Numbers, and What They Cost

Restaurant turnover is one of the most quoted and least sourced statistics in hospitality. The figures below come from national statistical agencies, with the periods stated. Where no official figure exists, we say so rather than repeating a number from a vendor blog.

1. The US numbers, and what they actually say

The Bureau of Labor Statistics publishes quits rates by industry through JOLTS. Annual averages of the monthly rate, not seasonally adjusted:

Three honest readings of that table:

• Accommodation and food services remains the highest-churn major industry in the United States, at roughly 2.1 times the all-industry rate.
• A 4.2% monthly quits rate annualises to roughly 50% of the workforce leaving voluntarily in a year — and that is quits only, before dismissals and layoffs.
• It has fallen sharply from the 5.8% peak of 2021–22. If you have been told the labour shortage is getting worse, the data does not support it. "Structurally high but well off the peak" is what these numbers say.

2. Singapore

Singapore's Ministry of Manpower publishes an average monthly resignation rate by industry, covering private establishments with 25 or more employees plus the public sector.

Same shape as the US: elevated relative to the national rate — about 1.8 times — and cooling. The internal split is instructive. Within Singapore F&B, rank-and-file staff resign at 2.8% a month against 1.2% for professionals and managers. Seniority is the strongest predictor of who stays.

3. The UK, and why you should distrust the figure you have seen

There is no official UK statistic for hospitality staff turnover. The ONS does not publish industry turnover rates, and the House of Commons Library briefing on hospitality employment (February 2026) contains no turnover, vacancy-rate or recruitment metric. Figures circulating in the UK trade press trace back to private consultancies rather than official statistics.

The closest official UK number is a vacancy count: ONS recorded 70,000 vacancies in accommodation and food service activities in July 2026, down from 75,000 in December 2025. That is a stock of open roles, not a turnover rate, and it should not be presented as one.

We mention this because it is a good test of any hospitality statistic you are shown: ask which agency published it and for which period. If the answer is a consultancy report with no sample disclosed, treat it as marketing.

4. Japan: the supply side

Turnover is only half the problem. The other half is how easily the person is replaced, and in Japan that has been hard for years.

The Ministry of Health, Labour and Welfare's job-openings-to-applicants ratio for food preparation workers (飲食物調理従事者, including part-time):

All occupations in Japan stood at 1.05 in July 2026, so cooks sit at roughly twice the economy-wide ratio — about 218 openings per 100 applicants. Excluding part-time, the ratio is 1.83 against 1.16 for all occupations.

Note the classification change: the old category 「調理の職業」 was discontinued, and 「飲食物調理従事者」 is the current equivalent. Searching for the old label returns nothing current, which is why some published comparisons stop abruptly around 2022.

As with the other two markets, the direction is loosening — 2.90 down to 2.18 — but the level is still structurally tight. A chef in Japan has options.

5. What churn costs a Japanese restaurant

There is no citable restaurant-industry cost-per-hire statistic that we could verify, in any country. The closest authoritative figure is Japanese and all-industry: Recruit's 就職白書2020 put average cost per hire at ¥1,033,000 for a mid-career hire in financial year 2019, up from ¥830,000 the year before. Recruit stopped publishing the metric after that edition, so that six-year-old figure is the most recent authoritative one available.

Rather than trusting a borrowed benchmark, build your own from things you can count:

• Advertising or agency fee for the replacement
• Management hours spent screening and interviewing, at a real hourly cost
• Overtime paid to cover the gap
• Training time from senior staff, which is also lost production
• The productivity ramp — weeks before the new person works at full speed
• For an international hire: visa fees, levies, flights, and any relocation support

The last line is why the economics of a sponsored chef are different in kind. A UK three-year large-sponsor package runs to roughly £9,200 in government fees alone. An Australian four-year nomination carries A$7,200 in Skilling Australians Fund levy. A Singapore S Pass costs S$650 a month in levy, S$15,600 over two years. Lose that person at month five and none of it comes back.

6. The relocated chef is a different risk

A local hire who leaves has usually found a better offer nearby. A relocated chef who leaves has usually decided the country did not work out — and that decision is driven by things outside the kitchen.

The recurring reasons, in the order they tend to come up:

1. Isolation. No one to speak Japanese with, long hours, no social life. This is the quiet number one.
2. The first month went badly. Housing, banking, phone, residence registration — a chef who spends three weeks fighting bureaucracy alone starts the job already tired of the place.
3. The job was not what was described. Usually menu authority, or volume, or the guest mix.
4. Ingredients. A chef recruited to cook Japanese food who cannot get the fish will stop believing in the job.
5. Family. A spouse who cannot work or a child who cannot settle ends the posting regardless of how good the job is. Worth noting that Malaysia's Employment Pass Category III began permitting dependants with the June 2026 policy revision, which removes one version of this problem.

7. What actually reduces it

• Hire more senior. The Singapore split — 2.8% for rank-and-file against 1.2% for professionals and managers — is the clearest evidence in this article. Senior people stay.
• Fix the first thirty days. Meet them at the airport. Have accommodation ready. Go to the bank with them. This is cheap and it has an outsized effect.
• Write the role down accurately, including the bad parts. A chef who was told about the 200 covers and came anyway will not leave over the 200 covers.
• Put a second Japanese speaker in the building if you possibly can.
• Make the path visible. Renewal, permanent residence, a promotion, or an agreed end date. Ambiguity is what people leave.
• Publish the schedule properly. Days off and closing time are the terms most often left vague and most often cited when people go.

Figures are as published by the agencies named, for the periods stated. Where no official statistic exists we have said so rather than substituting an unsourced one.