§7.1Reading the Ledgers

Every year's ledger comes in two parts, the way a young company keeps its books. A, the operating budget, is the institution's own running cost – the team, the office, the evaluator, the Hawker Council and the national platforms; B, the programme budget, is the money that reaches hawkers, shown centre by centre. To be sure, running costs are not waste, and no Fellowship pays itself out; still, a donor should be able to see how much of each dollar lands at a stall, and so should a hawker.

Each part carries a 10% contingency, a reserve for costs no one can foresee, on its gross cost. Programme revenue is deducted from B, and the net funding requirement is A + B. Revenue lines are deliberately conservative – Relief-Corps co-pays, masterclasses and non-exclusive licensing – because a plan that renounces exclusive IP control must not lean on licensing fantasies to balance its books. The ledgers cost the central plan: the two Lighthouses, Chinatown Complex and Geylang Serai, from Year 1, and Tekka from Year 2, each with the full loop (§6.3). No centre is costed as an average. Each is costed from its own shape: how many of its stalls are heritage-worthy, and how many stalls it has. Heritage-worthy is a stand-in measure – the stalls the MICHELIN Guide, the Hawker Heritage Awards and the heritage records already name, with an allowance for what they miss – because no one publishes the age of a centre's stallholders (the Budget Book, B.9.6). The measure sizes the budget, and only the budget. It counts the stalls that have already been noticed, so it tilts: at Tekka the heritage-worthy count is 7.6% of the stalls, against 11.9% at Chinatown Complex and 12.7% at Geylang Serai (derived). So from Year 1 a Fellowship is offered where a hawker wants to teach and the Registry's open test finds his dish fragile (§3.6), never for fame alone, and each year the Seed Bank publishes who received them, by centre, cuisine and community. For each heritage-worthy stall the ledgers fund one Master Fellowship over the centre's first three years, with two apprentices and one documented dish, and a funded first stall for about 30% of the apprentices. The cover, the relief cooks and the circle leads follow the stall count: about 70 in 120 stallholders are 55 or older (assumption: only the median, 60, is published), and a centre has about ten relief cooks for every 100 stalls. Each centre's gifts come in the year before it opens: Year 0 at the Lighthouses, the second half of Year 1 at Tekka. Each centre's listening sessions pay a $150 fee to about 1.25 participants for every cooked-food stall (150 participants at the 120-stall centre the ledgers were first drawn for): 283 at Chinatown Complex, 79 at Geylang Serai and 149 at Tekka. The Taste Council, the repository, the Hawker Council, the Buying Circles' price board and the core team stay national and shared, with one pilot-centre coordinator; so does the Regulars' List, open at all 123 government centres from Year 1 (§3.6). Every assumption is listed in the Budget Book (B.9). Neither part pays for the hawker centre itself: the centres' repairs and upgrading run through repairs and redecoration and the Hawker Centres Upgrading Programme 2.0, and a stall's own duct and equipment are the hawker's, not this programme's (§2.9). Nor do the ledgers carry the table-cleaning fee subsidy, which is the Government's (§2.2, §7.7).

An earlier version gave every centre a flat number of Fellowships. That was about right in total – 38 Master Fellowships, 18 at the one Lighthouse and 10 at each later centre, against the 44 the heritage count asks for – but in the wrong places: nine short at Chinatown Complex, two too many at Geylang Serai and one at Tekka. Yet money is not the real limit. Willing masters are: NEA's Hawkers Succession Scheme completed six handovers between January 2022 and September 2025 (MSE, 22 Sep 2025). So the Fellowship line is a ceiling, not a target, and what a centre does not spend on it returns to the endowment.

The four-year total is $19,683,808: $14,253,702 of it reaches hawkers directly (72.4%), and $5,430,106 runs the institution, about $113,000 a month. Let me be plain about the change: that is $4,693,468 more than version 1.2's $14,990,340, a rise of 31.3%. Most of it, about $3.31M with the contingency, is the apprentices, now paid the Progressive Wage for a stall assistant with employer CPF for all 18 months of the cycle (§3.3). The Fellowships add about $0.91M: their CPF, at the 12.5% rate for Fellows over 60, and a two-year stewardship after the teaching year, so that no Fellow's pay stops dead when the teaching ends (§4.2). The relief cook adds about $0.31M: his day rate now follows the Progressive Wage for a cook, and his day is now a lawful one, at most 12 hours with the 4 beyond the normal 8 paid as overtime (§2.7). The Regulars' List, with a booklet and map for every centre, adds about $0.06M. The starting count, a two-week till count at every consenting stall at the three centres in the year before each opens, adds about $0.10M (§6.1). I would rather pay those wages and print the bill than keep the old total. The Buying Circles' part-time role, price board and circle leads' honoraria cost $206,580 of that; their savings reach hawkers at the till, not through B, so they do not show in the share. The Regulars' List costs $416,570 of it: a coordinator, 800 listing visits over three years, the decals, the Registry pages, the booklets and maps, and the listing. None of it reaches hawkers as money, so it sits in A, and it lowers the share by about 1.6 points. Hawker Council fees and directors' stall-visit vouchers are one lump sum; the Council's tenth seat, held by FMAS, and the board's Government director are paid for within it (§3.8). The two Lighthouses alone would cost $16,224,542 (§7.8), and I would rather print the difference than hide it. For the price of a large building refurbishment, Singapore gets insurance for 238 of its oldest hawkers at three centres, three pools of relief cooks, forty-four Master Fellowships, eighty-eight apprenticeships, twenty-six funded first stalls, a Regulars' List of about 800 stalls across the country – and a blueprint tested at three centres rather than one.

• YR 0$906,598
• YR 1$4,629,249
• YR 2$6,706,257
• YR 3$7,441,704

A fourth or a fifth centre is an add-on, not part of the central plan. Old Airport Road and Maxwell are the examples, each costed at its own size and opening in Year 2, and their cost is stated here rather than left for later:

Each pilot centre over Years 0–3, costed from its own shape: the year before it opens, then its first years. Its programme budget is set apart from the operating costs it adds:

Chinatown Complex's own staff, evaluator and Taste Council tastings sit in the core team, as the first Lighthouse's did, so most of its money shows as B; at 226 stalls it adds one more centre staff. Elsewhere a centre adds 1.5 staff for every 120 stalls, rounded down to the nearest half and never fewer than 1.5, plus the evaluator's fieldwork, Taste Council tastings and community events. Every centre, Chinatown Complex too, adds its starting count in the year before it opens (§6.1). Geylang Serai is small: 63 cooked-food stalls, so its operating costs weigh more in its share. Fellowships, apprentices and dishes follow the planning count; the cover, the relief cooks, the circle leads and the listening fees follow the stall count. From Year 4, once its planning count has been offered, each centre keeps its stall-driven lines and funds new Fellowships at a maintenance rate fed by the yearly at-risk count (§3.6, §7.8). A further centre not yet named, with 100 stalls – the median, by stall count – and 10 heritage-worthy stalls, would add $3,636,004.