Keystone is a three-rung financing-recovery ladder for buyers S P Setia's panel banks have already rejected. This case sets out what Setia puts in, the numbers behind the Phase 1 gate, the Rung 2 bridge economics, and the risks, so the pilot can be evaluated on evidence rather than narrative.
1. What Setia puts in
- Data. The rejected-lead list for the pilot scope (100 historical leads plus 50 live rejections as they occur), the matching unsold affordable-tier inventory list, and confirmation of current panel-bank relationships.
- Units, Phase 2 only. Up to 3 completed, unsold units allocated to a dedicated Islamic SPV for the rent-to-own bridge. Setia decides which units. This step is contingent on clearing the Phase 1 gate and on the named legal opinion commissioned in week 1.
- A sponsor. One named contact on Setia's Tech and Innovation Council to evaluate results and authorise Phase 2.
- Capital: zero. Phase 1 is packaging work against an existing government guarantee and existing bank programmes; Setia holds no financing risk. Phase 2 capital and credit risk sit in the SPV, not on Setia's balance sheet.
- Cost: programme co-funded. The pilot is designed to run under Cradle's BIG 2.0 programme, Venture Client track, where the pilot contract is co-funded by the programme's matching grant (up to RM2m per corporate). The developer's share follows the programme's matching terms, not a Univa price list; Univa's Phase 1 effort is carried inside that contract.
2. Inputs and assumptions
| Input | Value | Status | Source |
|---|---|---|---|
| Housing loan approval rate | 39.2% (Jan-Apr 2026), down from 41% (2025), 42% (2024) | Verified | BNM data via The Malaysian Reserve, 29 Jun 2026 |
| Developers facing end-financing difficulty | 83% | Verified | Rehda Property Industry Survey H2 2025 via Malay Mail, 13 Mar 2026 |
| Launch take-up rate, H2 2025 | 21%, down from 38% | Verified | Malay Mail, 13 Mar 2026 |
| National residential overhang, Q1 2026 | 32,801 units, RM16.37bn | Verified | NAPIC via The Malaysian Reserve, 29 Jun 2026 |
| Rejection rate, RM500,001 to RM700,000 band | 31% to 45% | Verified | Malay Mail, 13 Mar 2026 |
| SJKP guarantee capacity | RM20bn, up to 80,000 first-home buyers | Verified | Budget 2026 coverage, tabled Oct 2025 |
| SJKP-MADANI financing cap | up to 120%, cap RM360,000 | Verified | sjkp.com.my/en/hcgs/hcgs-madani |
| SJKP standard-tier financing cap | up to 110%, cap RM500,000 | Verified | sjkp.com.my/en/hcgs/hcgs-madani |
| SJKP DSR ceiling | 65% on gross income | Verified | sjkp.com.my/en/hcgs/hcgs-madani |
| S P Setia FY2025 sales | RM5.11bn | Verified | EdgeProp, 2 Mar 2026 |
| BIG 2.0 matching grant | up to RM2m per corporate | Verified | MOSTI press statement, 29 Jan 2026 |
| Phase 1 conversion gate | 20% of fed leads close a cash booking | Assumption (a target, not a measured result) | Keystone proposal v2 |
| Affordable-tier average unit price, low case | RM330,000 | Assumption (demo persona reference price) | Internal |
| Affordable-tier average unit price, high case | RM450,000 | Assumption (illustrative, upper affordable band) | Internal |
| Annual holding cost | 3% of unsold value | Assumption (industry rule of thumb, not Setia-specific) | Internal |
| Rung 2 bridge terms (Halim profile) | rent RM1,900/month, 50% credited, graduation after 56 on-time months, RM53,200 credited | Verified (Keystone engine deterministic output) | Internal, config.ts / sjkp.ts |
3. Scenario table: recovered cash sales
Per 100 rejected affordable-tier leads fed through Rung 1.
| Recovery rate | Units recovered | Cash sales at RM330,000 average | Cash sales at RM450,000 average |
|---|---|---|---|
| 10% | 10 | RM3.3m | RM4.5m |
| 20% (Phase 1 gate) | 20 | RM6.6m | RM9.0m |
| 30% | 30 | RM9.9m | RM13.5m |
20% is the gate the proposal sets for a Phase 2 go decision; 10% and 30% bound a conservative case and an upside case. All three rows are scenario math on assumed price and conversion, not a forecast or a measured result.
4. Holding-cost view
Assumption: annual holding cost equals 3% of unsold value (financing carry, maintenance, quit rent and assessment, opportunity cost). This is an industry rule of thumb, not a Setia-specific figure, and it is used here only to illustrate the avoided-cost side of a faster sale.
For 20 completed affordable-tier units at RM330,000 average (RM6.6m total unsold value), annual holding cost runs to about RM198,000. Moving those 20 units to sale 6 months earlier avoids roughly half a year of that carry, about RM99,000. This avoided-cost figure sits alongside, not inside, the recovered cash-sale value in Section 3: one is the revenue the ladder generates, the other is the cost it removes by shortening days-to-sale.
5. Rung 2 bridge economics on 3 units
Modelled on Halim's verified engine output, replicated across the Phase 2 allocation cap of 3 units: rent RM1,900 a month per unit, 50% credited toward equity, graduation after 56 on-time months with RM53,200 accumulated credit per unit. Across 3 units that is about RM159,600 in accumulated buyer credit by the point each plan graduates. Using Halim's RM520,000 unit price as the reference, graduation converts to roughly RM1.56m in eventual full-price sale value once each unit is handed over and sold under a normal SPA.
The capital and the credit risk sit in the Islamic SPV throughout the 56-month term; none of it touches Setia's balance sheet, and every plan carries a deposit and a guarantor. Note: 3 units is the Phase 2 cap set out in the proposal, not a committed or forecast figure, and Halim's rent and price are one persona's terms used to illustrate the mechanism, not a portfolio average.
6. Commercial terms
[AHNAF: commercial terms decision, e.g. success fee per recovered sale, platform fee, or grant-funded pilot then licence]
Illustrative structure only, not a commitment: Phase 1 runs as a BIG 2.0 Venture Client pilot contract co-funded by the matching grant, so Setia's exposure during the 90 days is the programme's matching share, not a separate Univa fee. Post-gate, one plausible model is a success fee payable only on a closed cash booking (for example, a fixed fee per recovered sale, set well below the value the developer recovers), plus an optional platform or licence fee for continued use of the dossier engine once the pilot ends. Final terms need Ahnaf's decision before this section goes into any external submission.
7. Risks and mitigations
| Risk | Mitigation |
|---|---|
| Bank underwriting discretion still declines packaged files despite SJKP eligibility. | Routing across 5 participating banks; a 10-file validation experiment walked to an SJKP-specialist officer before scaling; dossiers built to the guarantee's own published criteria. |
| Rung 2 carries legal exposure under the HDA or Moneylenders Act lines. | One named legal opinion commissioned in week 1 gates Phase 2 entirely; Rung 2 is scoped strictly to completed post-CCC stock sold under a normal SPA. |
| Setia's rejected-lead or inventory data is incomplete, stale, or the rejection reasons are not SJKP-recoverable profiles. | Week 1 price-band fit analysis sizes the addressable wedge honestly against the SJKP caps before any recovery target is locked. |
| AI document parsing introduces an incorrect figure into a bank-facing dossier. | Deterministic cross-checks against extracted figures, an anti-hallucination contract unit-tested in both languages, human-in-the-loop operator review before any submission. |
| Conversion falls short of the 20% gate; the gate itself is a target, not a measured result. | Phase-gated structure: no Phase 2 commitment, no unit allocation, and no SPV until Phase 1 clears the gate. Setia's downside if it does not is data and time, not capital. |
8. The gate: what proves it in 90 days
Phase 1's KPI is cash recovered and bookings closed, measured against 100 historical rejected leads plus 50 live rejections fed through the engine over 90 days. Week 1 delivers the price-band fit analysis, sizing the addressable wedge against the SJKP caps before any target is locked. The gate to Phase 2 is one number: at least 20% of fed leads convert to a closed, cash booking.
That number has to move before an SPV exists, before any unit is allocated, or before the legal opinion becomes operative for Phase 2. Everything upstream of the gate, data intake, dossier generation, bank submission, is packaging work Univa carries inside the pilot contract. Everything downstream of the gate, unit allocation and capital, is a decision Setia makes with 90 days of evidence in hand, not before it.