Follow the Plate · D.8

The Plate Tracker

Twenty-one indicators that would let anyone follow the plate year by year, with what Singapore publishes today and how to close each gap. Feeds the yearly Plate Ledger.

Singapore publishes 3 of 21 regularly, 11 in part and 7 not at all.

The diner

The diner: what it measures, today's source and how to close the gap
Indicator	Status	What it measures	Today's source	To close the gap
Hawker food prices	Published	Hawker-food CPI and its components	SingStat, monthly	Keep; add a split by type of landlord
The plate price by town and landlord	Partial	Price of a fixed basket of 10 dishes	IPS Makan Index 2.0 (prices of late 2022, published 2023; one-off)	Repeat yearly with NEA and HDB data
What households spend	Published	Hawker-type share of food spend, by income	HES, every five years	Keep

The stall

The stall: what it measures, today's source and how to close the gap
Indicator	Status	What it measures	Today's source	To close the gap
Cost shares	Partial	Ingredients, manpower, rent, utilities as % of costs	NEA's yearly survey, quoted by MSE for 2022, 2023 and 2024; not published	Publish NEA's survey yearly; a voluntary panel of about 300 stalls for what it misses, seeded by the Year 0 stall baseline: orders a day, plates an order and dollars an order, by dish type (§6.1)
Hawker income and hours	Partial	Net income, hours and days closed, as distributions: quartiles, the mean and the bottom tenth, by type of stall and of centre	One median: full-time own-account hawkers, 2021, about $1,470 a month (MOM, in the ACPW report, 2022, read from a chart); MSE's "second decile" (2022)	From the same panel and CPF self-employed records; published as distributions, not medians
Health and cover	Missing	Days off a year; days lost to illness and injury; MediSave status; insurance held; the share over 65 with no successor in view	None	One block of questions in the stall panel; MediSave status from CPF, in aggregate
Platform and payment take	Partial	Delivery share and commission; e-payment share and fees	Industry sources only	Panel, plus Enterprise Singapore's SGQR data
Flows: entries, exits and transfers	Partial	Entries; exits within three years; time to let a vacant stall; successors in place; closures by reason; HSS transfers; by type of centre	Ad hoc parliamentary replies	Publish quarterly
Stalls at risk, and the Regulars' List	Missing	Stalls whose hawkers say no successor is in view and that they mean to stop within three years, or that a closure notice or a lease ending flags: national and per-centre totals, never naming a stall; and stalls on the Regulars' List, by tier, Bronze to Platinum, by count of regulars or share of weekly orders, with orders a week and the regulars' share of orders by centre	None: MCCY confirmed on 7 Oct 2026 that the Government keeps no list of hawker trades at risk	The Seed Bank's yearly count, from the end of Year 1 (§3.6)

The landlord

The landlord: what it measures, today's source and how to close the gap
Indicator	Status	What it measures	Today's source	To close the gap
NEA rents and bids	Partial	Rent and bid distributions (quartiles and the mean, not only the median); the subsidised share; the newcomer's path under the November 2024 renewal rule	Ad hoc parliamentary replies	Publish quarterly by centre, as distributions
HDB coffeeshop rents	Partial	Tender rents; stall rents and fees	Tender results; stall data collected since Jan 2026	Publish stall-rent medians by town
Private coffeeshop and food-court rents	Missing	Stall rents, turnover shares, O&M fees	Listings and company reports	Voluntary disclosure by Heritage Anchor partners
Discount and support pass-through	Partial	Share reaching stallholders	HDB clawback rule (no data published)	Publish compliance rate
Floor costs by type of centre	Partial	Each charge on its own line, a month, before GST: rent, conservancy, table cleaning, dishwashing, point-of-sale fees, fines and supplier space, at NEA, Town Council and operator-run centres	MSE's 2025 medians for rent, conservancy and table cleaning (8 Sep 2026): together 1.6 times the rent at operator-run centres and 1.5 times at comparable NEA centres (derived); dishwashing, point-of-sale and other fees not published	Publish every charge yearly, as distributions by type of centre

The supplier

The supplier: what it measures, today's source and how to close the gap
Indicator	Status	What it measures	Today's source	To close the gap
Imported food prices	Published	Import price index, food and meat	SingStat, monthly	Keep
Wholesale prices of hawker staples	Missing	Weekly prices for about 20 staples	None published	SFA weekly price sheet
Bought-in cooking	Missing	Share of stalls using central-kitchen bases	None	One question in the stall panel
The margin spread	Partial	Retail and wholesale prices of hawker staples against the food import price index, quarterly	Retail food CPI and import prices: SingStat, monthly; wholesale: none	Publish the spread each quarter (B.5.3)

The public purse

The public purse: what it measures, today's source and how to close the gap
Indicator	Status	What it measures	Today's source	To close the gap
Support reaching hawkers	Partial	Rent support, grants, voucher redemptions at hawker stalls	Scattered replies and releases	One yearly table
Rent collected as landlord	Missing	NEA and HDB rents from cooked-food stalls	Not published	One yearly line
Hawkers' CPF position	Missing	MediSave compliance, Workfare take-up, CPF balances of hawker SEPs	Not published for hawkers	CPF aggregate, by licence type

The Plate Ledger: one page a year

From these indicators, a single yearly page: where a typical plate's dollar went, by type of landlord, with the change on last year. It would be produced by the plan's independent evaluator (the Plan's Part VI), as Method, Levers and Data, F.6 already provides, from published series, aggregated administrative data and a voluntary stall panel, and released with the Singapore Hawkers' Celebration and Awards. Its first stall-level numbers come from the Year 0 baseline (the Plan, §6.1): a two-week till count at the pilot centres' consenting stalls, of orders a day, plates an order and dollars an order, by dish type (D.4.2). It reports distributions, not medians: quartiles (a quarter of a ranked list) and the mean of rent, fees, income and hours, by type of stall and of centre, with the bottom tenth, because a median hides both the pioneer on a subsidised rent and the newcomer still paying his bid. Beside them it counts flows, entries and exits, and health, the days lost to illness, which a median of incomes cannot show. It respects the plan's rule against data mandates and surveillance (the Plan, §1.5): no stall is required to report, no individual stall's figures are published, and the panel is paid for its time.

As of 2026-10-29.

This is a working paper for public discussion. Every number in it is an invitation to improve it. Corrections and better data are welcome.