Hawker vs Restaurant Spending in Singapore: When Consumers Trade Up and When They Won't
A participant in one of our sessions, an engineer in his forties, told a story about a plate of fishball noodles. His usual stall had raised the price by 50 cents, and he had stopped going for about three weeks (he had counted), "on principle". In the same conversation, maybe ten minutes later, he described a birthday dinner for his wife where the bill came to several hundred dollars, and he said, without any irony at all, that it was "worth it, very reasonable". The room didn't find this strange. Everyone had a version of it. I think it is the single most useful thing to understand about food spending in Singapore, and it is the reason price-led thinking keeps getting F&B strategy wrong.
Singaporeans behave as if they carry separate pockets of money for different kinds of eating, and each pocket follows its own rules about what counts as expensive. Behavioural economists would call it mental accounting. Our participants just call it "hawker money" and "makan-out money". This post looks at what the spending data shows, how those pockets actually work, and when a consumer will move money from one to the other (and when nothing will persuade them to).
What the spending data shows
The long view helps here. The SingStat Household Expenditure Survey 2023 found households spending $966 a month on eating out, up from $810 in 2017/18, and the survey attributes the increase mainly to restaurants, cafes, and pubs. Spending there rose from $296 to $403 a month. Spending at hawker centres, food courts, coffee shops, and canteens rose from $437 to $491. I did the arithmetic twice, because it surprised me. Restaurant spending grew by roughly a third, hawker spending by about 12 per cent (my arithmetic, not SingStat's), and hawker-type venues still took the largest share of eating-out money at 50.8 per cent. So Singapore did two things at once. It kept the hawker centre as its daily kitchen, and it put its extra money into occasions.
Then the price pressure arrived at the cheap end. Hawker food prices rose 6.1 per cent in 2023, the highest since 2008, according to figures reported by Mothership from a parliamentary exchange. The Ministry of Sustainability and the Environment's written reply on hawker food prices explained why, citing an NEA survey in which raw materials made up 56 per cent of stallholders' operating costs and manpower another 20 per cent. A hawker has very little room to absorb that. The customer, it seems, has even less patience for being asked to.
And now the picture for 2026 looks strained across the board. SingStat's F&B services index for July 2026 shows overall sales down 1.9 per cent year on year, with food courts and other eating places down 6.6 per cent, cafes down 6.4 per cent, restaurants down only 0.3 per cent, and fast food outlets up 4.6 per cent. YouGov's 2025 dining out survey found 32 per cent of Singaporeans eating out less than a year earlier, and 89 per cent feeling that restaurant prices had risen. Closures have followed. ACRA figures reported by VnExpress put F&B closures at 2,101 in the first seven months of 2026, up 25.1 per cent, against 2,594 new registrations.
If people were simply trading down, wouldn't the cheapest venues be the ones holding up?
Why the cheapest meals feel the most expensive
They aren't holding up, at least not in that index, and I suspect the reason is that the hawker wallet is the one people watch most closely. A daily meal gets repeated five or ten times a week (sometimes twice a day). Its price becomes an anchor, a number people know by heart, so a 50-cent rise is noticed instantly and remembered as a small betrayal. A restaurant main has no such anchor. Nobody knows what a steak "should" cost, which is exactly why a restaurant can raise prices more quietly than a chicken rice stall ever could.
There is a moral dimension too (and Singaporeans feel it strongly). Hawker food carries civic meaning. When hawker culture was inscribed on UNESCO's intangible heritage list in December 2020, NEA described hawker centres as "community dining rooms". Participants talk about affordable hawker food almost as a public right, something closer to a bus fare than a purchase. The government's own policy tools lean the same way. At Socially-conscious Enterprise Hawker Centres, NEA requires operators to propose initiatives to keep food affordable, and a value meal at every stall is the current practice. So a price rise at the hawker centre gets judged as fairness. A price rise at a restaurant gets judged as value. Those are different courts, with different juries.
The Three Food Wallets
The first draft on our whiteboard was a price line, cheap meals on one side and expensive meals on the other, and the meal-sorting exercise kept scrambling it. Actually, the split runs by purpose, and the same venue can sit in more than one wallet depending on why you are there. I've started calling these the Three Food Wallets, though I'm still testing whether a fourth (something like "convenience money", for delivery) deserves its own slot.
The Three Food Wallets
Fuel money
The weekday lunch and the tired dinner. Price is anchored and watched, and a small increase feels personal.
Coffee shop economy rice, office food courtTreat money
A small reward that needs no excuse. Flexible, spent on a queue-worthy stall as readily as a cafe.
Satay night out, bubble tea, a famous stallOccasion money
Birthdays, clients, parents, anniversaries. Price matters less than whether the place makes the moment feel right.
Family restaurant booking, celebration dinnerFuel money is where the pressure lands. It is spent on autopilot (few people can recall Tuesday's lunch by Thursday), and people defend it fiercely because it covers most of their meals. Treat money is the interesting one, because it moves between venue types without friction. A Friday satay dinner at Lau Pa Sat (a cast-iron market completed in 1894, with its Satay Street running since 1996) comes out of treat money, not fuel money, even though it is technically hawker food. So does a craft beer and a plate of wings at Timbre+ One North, which describes itself as a food park with "restaurant-grade food and casual eats at affordable prices" and live music every evening. Those hybrid venues probably earn their keep by letting people spend treat money in a setting that still feels like hawker value.
Occasion money barely behaves like money at all (at least, not the way a budget spreadsheet would expect). It is spent on behalf of a relationship. A participant will book the restaurant her mother likes, whatever it costs, and then eat economy rice for three days afterwards to balance the ledger in her head. I think this explains the July 2026 index better than a simple downturn story. Restaurants held flat while food courts fell, possibly because occasions are protected and fuel meals are where people cut, cook at home, or switch to a fast food value set.
When consumers trade up, and when they won't
For operators, the practical question is whether a given diner will move money from one wallet into your venue. Here is how the pattern tends to break down across our sessions.
| Situation | Trades up? | What decides it |
|---|---|---|
| Weekday lunch near the office | Rarely | A known price anchor and a 45-minute break leave no room for persuasion |
| Friday evening with colleagues | Often | Treat money opens up when the week is over and the group shares the cost |
| Family birthday or visiting parents | Almost always | The occasion is the product, and choosing cheaply can read as disrespect |
| A heritage or famous hawker stall | Yes, within limits | People pay more and queue longer, while still expecting "hawker price" logic |
| A new restaurant with a promotion | Once | A discount buys a trial, and a return visit needs an occasion to attach to |
That last row is where many openings go wrong. YouGov found that among Singaporeans cutting back, 71 per cent were choosing cheaper restaurants and 60 per cent were using discounts and coupons. A promotion gets you into someone's treat budget for one visit. It does not move you into their occasion budget, and it certainly does not make you fuel. We see the same trial-without-return pattern in Instagram-driven restaurant discovery, where a viral dish fills the tables for a month and then the room goes quiet. It is also why testing an F&B concept before signing the lease should ask which wallet the concept will be paid from, not only whether people like the food.
Why asking about price gets the wrong answer
If you ask diners what they would pay for a meal, they answer from fuel money, because that is the number they know. So surveys tend to tell operators their prices are too high, and then the same respondents turn up at a $40-a-head restaurant for a birthday. Or rather, that is how it looks from the survey data. They aren't lying about price. They are answering a different question from the one the operator meant to ask. We have written before about why Singapore consumer research keeps misleading brands, and food spending might be the cleanest example.
What works better is reconstructing real meals. In focus groups we ask people to sort a week of their own meals into piles, then explain each pile, which surfaces the wallet logic without anyone naming it. For households, in-depth interviews matter more, because the person who books the occasion dinner is often not the person eating economy rice to balance it. And diaries (we lean on mobile ethnography for this) capture the tired Tuesday dinner that nobody remembers in a group discussion.
"Tell me about the last time you paid much more than usual for a meal and felt fine about it. Now tell me about the last time a stall charged you 50 cents more and you minded." The distance between those two stories shows you exactly where the wallets are divided, and what your venue would need to be to cross that line.
The same logic reaches well beyond food. Skincare shoppers run a nearly identical split between everyday products and indulgences, which we mapped in when Singaporeans splurge and when they save. And delivery sits awkwardly across all three wallets, as our work on food delivery behaviour in Singapore suggests. Operators who want to see how this plays out in real projects can look at our anonymised case studies.
Price the wallet as well as the plate
A Singapore diner is not one customer with one budget. She is at least three, and they don't talk to each other much. The MTI's February 2026 reply on F&B closures described the sector as competitive because of "rapidly shifting consumer preferences", and that is true, though I'd add that the preferences shift less than the wallets they get paid from. An operator who knows which wallet funds their table can price, promote, and design for it. One who doesn't will keep discounting into the wrong pocket. I'm fairly sure the three wallets aren't the final word on this, and delivery may still force a fourth. But the engineer who boycotted his noodle stall over 50 cents, and then called a birthday bill reasonable, was not being inconsistent. He was being very precise, just about something other than price.
What F&B operators ask about hawker and restaurant spending
How much do Singapore households spend at hawker centres compared with restaurants?
Why do Singaporeans react so strongly to hawker price increases?
When will Singapore consumers pay more to eat at a restaurant?
What is mental accounting in food spending?
How can F&B brands research dining spending behaviour in Singapore?
Finding out which food wallet your diners pay from, and what moves them between them
Singapore diners judge a hawker price rise on fairness and a restaurant bill on the occasion, so a pricing or promotion plan built on average spend misses the decision entirely. We run focus groups, household interviews, and food diaries that reconstruct real meals, so you can see which budget funds your concept and what would earn a return visit.
Request a quote →
