§2.10Market Transparency

INDEPENDENTLY PUBLISHED

Every quarter the independent evaluator (Part VI) – never the programme itself – publishes the range of takings by dish category, ingredient price indices, typical cleaning and utility costs, and how often stalls change hands, so the market's data and the institution's report card stay in separate hands. Aggregate data only; no individual stall identifiable; voluntary revenue reporting compensated at $25 per quarter. The yearly at-risk count of §3.6 keeps the same rule: national and per-centre totals only, from signals hawkers choose to give. Sunlight, not surveillance.

Follow the Plate (D.8) specifies what to publish: twenty-one indicators across diners, stalls, landlords, suppliers and the public purse, of which Singapore publishes three regularly today, and a one-page yearly Plate Ledger showing where a typical plate's dollar goes, by type of landlord (Follow the Plate, D.8). Most of the missing figures are already held by NEA, HDB or CPF.

Prices are already counted; costs are not. The IPS Makan Index, Prof Ivan Png's indices built from NEA tender bids, CASE's Price Kaki and NEA's own published top bids already track what diners pay and what stalls fetch (the Record, A.1.2). Publishing them works: MTI (Ministry of Trade and Industry) found that one more stall selling the same dish goes with prices about 8.4% lower, and its 2015 list of measures included an online platform to publish the prices and locations of affordable food (MTI, 2015). The Plate Ledger licenses or partners for that price data rather than collecting it again. Its contribution is the other side of the plate: ingredients, rent and wages. No one publishes them regularly; only one-off medians and cost shares exist: MTI's cost shares of 2015, the shares from NEA's yearly cost survey that MSE has quoted for 2022, 2023 and 2024, and medians for rent, service charges and table cleaning given in a reply to Parliament (MTI, 2015; NEA, 2022; MSE, 4 Aug and 8 Sep 2026). The Government has said as much. Answering Victor Lye of the People's Action Party on 7 October 2026, Dr Tan See Leng, answering for MTI, said that "comparative data on the prices of these cooked food items across types of food establishments are not available" (Official Report). That is the gap the Ledger's site-cost and price lines fill.

One line for the question readers ask most. Are the middlemen taking more? The Plate Ledger answers it every quarter with a margin-spread indicator, which sets the retail and wholesale prices of hawker staples against the price of imported food: if the gap widens, the middlemen are taking more (Follow the Plate, D.8). So far the evidence says no. From 2019 to 2025 retail food prices rose 18.5% and food import prices 20.9% (SingStat; our arithmetic), so the Singapore end of the chain charged about the same share of a bigger bill. If that ever changes, the spread will show it first, and the plan will not need to guess. A second new line prices the stall's site itself: rent, conservancy, table cleaning and dishwashing a month, before GST, by type of centre (Follow the Plate, D.5.4, D.8). In 2025 rent, conservancy and table cleaning alone came to a median of about $2,750 at operator-run centres against $2,170 at comparable centres (MSE, 8 Sep 2026; our sum), and MSE warns that the two are not directly comparable. Tracked each year, that gap is measured rather than argued about. MSE published three charges in September 2026: rent, conservancy and table cleaning, which together come to 1.6 times the rent at operator-run centres and 1.5 times elsewhere (our arithmetic). It did not publish dishwashing, point-of-sale or the other fees asked about. Hawkers and their advocates say those can take the total to double the rent (§2.2). The Ledger records each charge separately, so that the argument can be settled with numbers.

What we count, and why not the median

The record hides the trade's trouble in medians. The median rent has held near $1,250 for a decade while the subsidised share fell and newcomers paid their bids (§2.1, §2.2). The median age has been about 60 since 2018, across a trade of veterans and newcomers who entered at a median of 46 (§1.1). The one official income figure is a median for hawkers who work alone, and the measures of success the agencies report are occupancy, satisfaction and enrolment. So the Ledger publishes the whole spread, not one figure: for rent, fees, income and hours, the lowest quarter, the middle, the top quarter, the average and the poorest tenth, by stall type and centre type. It publishes movement: stalls opened, stalls closed within three years, how long a vacant stall takes to let, how many hand-overs have a successor in place, and every closure by its stated reason. And it publishes health: days closed a year for illness or injury, whether MediSave is paid up, cover held, and the share of stallholders over 65 with no successor in view (§2.11.2). The Year 0 starting count is published as a spread from the start (§6.1), and the Regulars' List adds orders a week, the regulars' share of them and the spread of grades, by centre (§3.6). Each line is set out in Follow the Plate, D.8, and none is ever published by stall.

Its first stall-level numbers. The Ledger should start from what stalls actually sell, not from the model's guess. In Year 0 the listening teams run a two-week till count at every consenting stall at the pilot centres: orders a day, plates an order, dollars an order, and the share of orders from people the hawker names as his regulars, by dish type (§6.1). It is published only by dish type and centre, never by stall. Its first use is to set the Regulars' List's thresholds (§3.6); its second is to give the Ledger its first stall-level line, against which every later year is read (Follow the Plate, D.8).