Key takeaways
- ACRA recorded 3,074 cessations of Food & Beverage (SSIC 56) business entities in 2025, against 4,103 formations. Cessations have risen every year since 2020.
- For restaurants (SSIC 56111), there were 853 cessations for every 1,228 formations in 2025: about 69 per 100, up from about 50 per 100 in 2017.
- A cessation is a business entity leaving the register, not an outlet closing its doors. No official series counts outlet closures.
- F&B sales volumes are still below 2019 levels, and F&B sales fell 1.9% year on year in July 2026 (SingStat).
- The number of licensed food shops kept rising, from 18,035 in 2017 to 24,359 in 2025 (SFA via SingStat). The industry is churning rather than shrinking.
What counts as a “closure” in official data
Singapore has no official count of restaurant outlets that close. The nearest measure is ACRA’s count of business entity cessations, published by SingStat and broken down by industry using the Singapore Standard Industrial Classification (SSIC). Before you quote any closure figure, keep these points in mind:
- An entity is not an outlet. One company can run ten outlets, and closing three of them does not show up as a cessation. Equally, a sole proprietor who deregisters a business that never traded does count.
- The industry code is self-declared. The SSIC code is chosen at registration. A company classed as a restaurant may have pivoted to catering or a central kitchen.
- Cessation includes strike-offs. ACRA’s footnote says that since September 2023 it has “significantly increased the number of ACRA-initiated striking-off actions”. Dormant companies being removed from the register push the count up.
- December 2024 is distorted. Counts that month are lower because of a Bizfile system migration and ACRA’s periodic removal exercise.
So treat cessation counts as a measure of churn in the register, not as a list of shuttered dining rooms. Media counts of closures that don’t name an official table cannot be checked, so we have left them out.
F&B business formations and cessations, 2017 to 2025
The table below uses ACRA’s annual series for SSIC 56 (Food & Beverage Service Activities), from SingStat tables M085851 (formation) and M085861 (cessation), last updated 9 July 2026. The last two columns are our own arithmetic.
| Year | Formations | Cessations | Net change | Cessations per 100 formations |
|---|---|---|---|---|
| 2017 | 3,298 | 2,115 | +1,183 | 64 |
| 2018 | 3,341 | 2,132 | +1,209 | 64 |
| 2019 | 2,985 | 2,358 | +627 | 79 |
| 2020 | 3,285 | 2,034 | +1,251 | 62 |
| 2021 | 3,934 | 2,457 | +1,477 | 62 |
| 2022 | 3,335 | 2,746 | +589 | 82 |
| 2023 | 3,669 | 2,754 | +915 | 75 |
| 2024 | 3,795 | 3,048 | +747 | 80 |
| 2025 | 4,103 | 3,074 | +1,029 | 75 |
Two points stand out. Cessations have risen every year since 2020 and passed 3,000 for the first time in 2024. But formations also reached a new high of 4,103 in 2025, so the number of F&B entities entering the register still exceeded the number leaving it in every year shown. For scale, across all industries ACRA recorded 78,146 formations and 60,445 cessations in 2025. F&B made up about 5% of each.
Restaurants, cafés and kiosks: the 2025 breakdown
The detailed SSIC codes show where churn is highest. Figures are for 2025, from the same ACRA tables. The ratio column is our arithmetic.
| SSIC | Activity | Formations | Cessations | Cessations per 100 formations |
|---|---|---|---|---|
| 56111 | Restaurants | 1,228 | 853 | 69 |
| 56112 | Cafés | 405 | 324 | 80 |
| 56121 | Fast food outlets | 58 | 50 | 86 |
| 56123 | Food and drink kiosks mainly for takeaway and delivery | 444 | 225 | 51 |
| 56200 | Food caterers | 193 | 183 | 95 |
| Other 56 | Other food and beverage services | 1,775 | 1,439 | 81 |
For restaurants the trend is clear. Cessations rose from 453 in 2017 to 861 in 2024 and 853 in 2025, while formations peaked at 1,343 in 2023 and have eased since. That takes the ratio from roughly 50 cessations per 100 formations in 2017 to about 69 in 2025. Takeaway and delivery kiosks are the fastest-growing code: formations rose from 247 in 2017 to 444 in 2025.
2026 so far: the monthly figures
ACRA also publishes monthly counts (M085831 and M085841, updated 7 September 2026). Totals for January to August are our sums.
| SSIC 56, January to August | 2025 | 2026 |
|---|---|---|
| Formations | 2,741 | 2,985 |
| Cessations | 2,007 | 2,354 |
| Highest single month of cessations | 616 (February) | 603 (March) |
Monthly cessations are lumpy because strike-offs are processed in batches. In 2025 they ranged from 8 in January to 616 in February, so compare periods of several months, not single months. On that basis, both formations and cessations in the first eight months of 2026 were higher than a year earlier.
What sales data shows: the F&B Services Index
SingStat’s Food & Beverage Services Index measures sales at restaurants, fast food outlets, caterers, cafés, and food courts and other eating places. Individual stalls in hawker centres, food courts, coffee shops and canteens are not covered. The annual index in chained volume terms (M602241, 2025 = 100) strips out price changes:
| Year | All F&B services | Restaurants |
|---|---|---|
| 2019 | 113.4 | 130.5 |
| 2020 | 81.2 | 86.0 |
| 2022 | 100.1 | 109.1 |
| 2023 | 102.8 | 108.3 |
| 2024 | 101.4 | 104.7 |
| 2025 | 100.0 | 100.0 |
In volume terms, restaurant sales in 2025 were well below 2019 and have fallen each year since 2022. The latest SingStat release (7 September 2026) reports that F&B sales at current prices fell 1.9% year on year in July 2026, after a 2.3% fall in June. Restaurants were down 0.3%, cafés 6.4%, and food courts and other eating places 6.6%. Fast food outlets rose 4.6%. More entities chasing flat or falling volumes is consistent with rising cessations. It does not prove that falling sales caused any particular closure.
Labour: vacancies and retrenchments in F&B
MOM’s quarterly tables, published through SingStat (job vacancies and retrenchments, updated 22 September 2026), give a labour-market view of the same sector.
| F&B services | 2Q 2025 | 1Q 2026 | 2Q 2026 |
|---|---|---|---|
| Job vacancies (end of quarter) | 4,100 | 3,900 | 3,400 |
| Retrenched employees | 30 | 150 | 170 |
F&B retrenchments were 10 to 70 a quarter through 2025, then rose to 150 and 170 in the first two quarters of 2026. Vacancies fell from 5,000 in 2Q 2024 to 3,400 in 2Q 2026. Fewer openings and more retrenchments suggest operators are cutting headcount. MOM does not say how many of these retrenchments came from outlets closing.
The other side: licensed food shops keep rising
SFA’s count of licensed food establishments (M890531, end of year) shows that the stock of premises has grown despite the churn:
| End of year | Food shops | Food stalls |
|---|---|---|
| 2017 | 18,035 | 14,567 |
| 2021 | 21,073 | 14,455 |
| 2023 | 22,747 | 14,178 |
| 2025 | 24,359 | 13,893 |
Licensed food shops rose by about 35% between 2017 and 2025, while food stalls edged down. The market is getting more crowded, which helps explain why the businesses that do close are often those that could not hold their share of a flat pool of spending.
Why restaurants close: the common operating causes
No official source records why a business ceases, so this part is operating experience, not statistics. The same four pressures come up again and again:
- Rent and occupancy cost. Rent, service charges and any turnover rent that grow faster than sales. See our guide to restaurant occupancy cost for how to calculate your ratio.
- Labour. Wages, levies and the cost of being short-staffed. Our restaurant labour cost guide covers how to measure it.
- Food cost. Ingredient prices and portion drift that nobody measures until the margin has gone.
- Footfall. A mall repositioning, a new competitor next door, or a shift in customers to delivery and takeaway.
Most closures involve more than one of these, and cash usually runs out before the owner has a clear picture. Our restaurant losing money checklist is a structured way to find the leak, and how to reduce restaurant costs covers the fixes.
Early-warning signs owners can track
You can’t control the national numbers, but you can see your own trouble early if you track the right figures every week:
- Same-period sales. Compare this week with the same week last year, not just with last week.
- Covers and average spend. Falling covers with a stable average spend points to footfall. Stable covers with falling spend points to menu or pricing.
- Food cost percentage against target. A gap between theoretical and actual cost usually means waste, over-portioning or theft.
- Labour hours per 100 covers. Rosters that don’t flex with demand are a common silent cost.
- Occupancy cost as a share of sales. Recalculate it monthly, not just when the lease renews.
- Cash runway. The number of weeks of fixed costs you can cover from cash in hand.
A POS with proper reporting makes most of these a two-minute check. ChaChaCha, powered by AppsPOS, includes sales reports, recipe-level stock deduction and recipe costing per menu item, and an ERP view for multi-outlet head offices. That lets you watch food cost and sales trends outlet by outlet. If you want to see how that would look for your business, talk to us.
