Three categories that must stay separate
A headline "supply" number that sums all three is not a forecast of future stock — it is an upper bound built on the assumption that every plan completes. That assumption has never held in any property market.
- Completed units: built and delivered. This is existing stock and it is the only category that definitely exists.
- Under construction: committed, with capital deployed and a reasonable expectation of delivery, though timing can slip.
- Announced or registered but not started: a plan. Some will be delivered, some will be delayed substantially, and some will not proceed.
Why expected delivery dates move
Expected delivery dates are planning information. Construction timelines shift for reasons ranging from contractor availability to funding to approvals, and slippage is normal rather than exceptional across the industry.
The practical implication is that a pipeline concentrated in a single future year should be read as indicative of pressure in that general period, not as a precise delivery schedule. Analysis that depends on a specific quarter of delivery is analysis built on the least reliable part of the dataset.
How to read supply usefully
- Compare pipeline against existing stock in the same community — that ratio, not the absolute number, indicates relative supply pressure.
- Separate by construction status and weight committed supply more heavily than announced supply.
- Look at the delivery distribution across years, not just the total.
- Match the pipeline to the segment you care about: units expected in a community tell you little if they are a different property type or size from your interest.
- Read supply together with recorded transaction and rental activity, since the question is always about supply relative to demand.
Coverage limitations we will not paper over
Registered project coverage can be incomplete. Where it is, total future stock will be higher than the figures shown. For this reason supply figures are best treated as a well-evidenced floor rather than a complete census of what is coming.
We also do not publish oversupply risk indicators unless a verified methodology supports them. Labelling a community "oversupplied" is a strong analytical claim, and it requires a defensible method relating pipeline to absorption — not a large pipeline number and an adjective.