The PropKaki Curve™
The PropKaki Curve is Singapore's standing quarterly measure of HDB lease decay: the median transacted price per square foot of HDB resale flats in five remaining-lease bands, computed from every resale registered with the Housing & Development Board. This page is the metric's permanent definition — its formula, rules and version history. Current values live on the HDB Lease Decay Tracker.
Definition
For each completed calendar quarter, the PropKaki Curve is the set of five median transacted prices per square foot of HDB resale flats, grouped by remaining lease at the point of sale: 90 or more years, 80–89, 70–79, 60–69, and under 60 years. The curve's shape is the market's observed price on remaining lease; the percentage gap between the 90+ band and the under-60 band is the curve's headline reading. Two companion series are published per band: the median 4-room resale price (a flat-type control) and the transaction count (a liquidity reading).
Computation
psf(txn) = resale price ÷ (floor area sqm × 10.7639)
Curve(q, b) = median(psf of all transactions in quarter q, band b)
Gap(q) = 1 − Curve(q, under60) ÷ Curve(q, 90plus)
Banding is by completed years — a flat with 89 years 11 months of lease counts as 89 and sits in the 80–89 band. The in-progress quarter is always excluded. Any band-quarter with fewer than 10 transactions is suppressed rather than published, including its transaction count. No estimates, forecasts, seasonal adjustments or valuations are applied at any step.
Data sources
The sole input is the Housing & Development Board's resale transaction dataset published on data.gov.sg, which records every resale registered with HDB, including its resale price, floor area and remaining lease. The granular dataset begins in January 2017, so the Curve's history starts there. PropKaki re-ingests the dataset automatically every week; every published figure states the exact quarter it refers to.
Known limitations
The bands control for size (per-square-foot pricing) or for flat type (the 4-room companion series) but not for location, floor or condition. Because newer flats cluster in non-mature towns while the oldest sit in mature central ones, the raw gaps likely understate true like-for-like lease decay. Band medians reflect whichever flats actually transacted in the quarter; they are market context, not valuations of any individual flat.
Version history
| Version | Date | Change |
|---|---|---|
| 1.0 | 12 June 2026 | First published definition. Five bands, k≥10 suppression, completed-years banding, quarterly medians from the full HDB resale record (the underlying band series have been published on the tracker since June 2026). |
Any future change to the bands, suppression threshold or computation will be recorded here with a new version number, and the change will be announced on the market data changelog. Numbers already published are never silently restated.
Citing the PropKaki Curve
Preferred citation:
The PropKaki Curve™ name and methodology are published by Straits Intelligence Pte. Ltd. (UEN 202615220E), operator of PropKaki. The underlying transaction data belongs to the Housing & Development Board. The computed series are free to reuse with attribution to "PropKaki (propkaki.sg)" under CC BY 4.0.
Current Curve values, charts and per-band history are on the HDB Lease Decay Tracker. Free JSON and CSV endpoints are listed in its "Use this data" section.