C.8.1The retirement match: costed, and left for later

• OPTIONAL · NOT YET AN ASK

The Protection Floor (the Plan, §2.11) proposes a retirement match for stallholders: Government matches the hawker's own CPF (Central Provident Fund, Singapore's compulsory savings for retirement, health and housing) top-ups, up to $1,200 a year, as a platform now shares the CPF cost for a younger rider (C.9.1, C.9.2). That share is paid for riders born in or after 1995, and for the roughly 27% of older riders who opted in (MOM, 12 Jan 2026); every rider has work-injury cover. The match is for stallholders under 55. From 55, the Matched Retirement Savings Scheme already matches cash top-ups to the Retirement Account, up to $2,000 a year, for those who pass its income and home tests. Paying twice makes no sense, so there the plan helps hawkers claim it. This plan costs the match but does not yet ask for it. The base case is a match for every stallholder under 55 in a public centre, NEA- and operator-run (about 6,900 stalls in all): up to about $4.1M a year. That is a ceiling: it assumes every one of them takes the match up in full. The stretch case matches every stallholder under 55, coffeeshops and food courts included, at 70% take-up (the share of those eligible who join), and adds shared hospital-cash cover for stallholders of every age: about $7.9M a year.

Three sources outside the Budget could meet part of the cost: gifts set aside for the match (half of the continuing gifts the programme itself does not need); 40% of Heritage Plate licence income (the Plan, §5.5); and the tender rent HDB (Housing & Development Board, Singapore's public-housing landlord) gives up when coffeeshop operators bid to fund their stallholders' retirement (the Budget Book, B.7), which is HDB's contribution. The table adds a fourth column, an idea rather than an ask: a Government match on licence income in the mark's first years (Method, Levers and Data, F.9). That would itself be a Government cost. The endowment cannot carry the match on its own (Method, Levers and Data, F.9). The table shows what would be left.

Central assumptions; nominal dollars. No match is paid in Years 1–3, not even at the Lighthouses: the match starts only if and when it is adopted, rolling out nationally from Year 4 and reaching full take-up in Year 7. "Gifts set aside" is half of the continuing gifts the programme does not need, earmarked (set aside for one use) for the match instead of the corpus (the endowment's invested capital). At three centres, on the heritage logic from Year 4 (the Budget Book, B.8), the programme spends about two-thirds of its continuing gifts, so about $0.55M a year is set aside. That has a price: with half set aside the corpus ends Year 20 at about 101% of the $40M target in today's dollars, against about 121% with none and about 86% with all of it. Half is about as far as the earmark can go without spending the endowment's future on today's match. Money not needed in a year carries forward: Year 3's $84K helps pay Year 4, and the base case runs short from Year 5. HDB's contribution phases in as coffeeshops are re-tendered (a quarter in Year 4, all by Year 7). The tender rent HDB gives up pays for coffeeshop stallholders, so it counts only in the stretch case, and only its under-55 part (half of it, about $0.84M a year at full roll-out) counts towards the match; operators' payments for older stallholders go into their own savings. A Government match on licence income in the Heritage Plate's first three years is an idea, not an ask. Without HDB's contribution, the stretch shortfall over Years 4–10 rises to about $37.0M; the base is unchanged. Licence income is net of a third of licensees failing each year (assumption; the Budget Book, B.9.5), which is why the sources outside the Budget now meet only part of the base, about 40% of it. The decision is left for later. The scenarios behind this table are set out in the Budget Book, B.9.5; the match should be decided on that evidence, not ahead of it.