The Capital Radar, Issue 006
Every figure below comes from the public record as it stood on 30 August 2026, with its source attached to the entry it came from. Where we have not verified something against a primary source, we say so rather than rounding the doubt away.
This issue is about the shape of a number, not its size. It also explains why our headline number went down this week.
In brief
- 24 projects across the Gulf, South Asia and South East Asia are recorded as seeking finance, for a combined USD 18.72 billion.
- Eight of those 24 account for USD 15.6 billion, or 83% of the total. The headline is a handful of gigawatt-scale programmes.
- We withdrew 12 records this week because we could not stand behind their sourcing. That removed USD 560 million from the total.
We took twelve records down
Before the analysis, the correction, because it affects every number in it.
An audit of the entries we had not re-checked in ninety days found twelve that could not be verified. Ten carried a placeholder developer rather than a named counterparty. One named a real utility but no public document supported the capital figure attached to it. One conflated two emirates in its own title.
None of them should have been public. They are now withdrawn, and the totals in this issue are calculated without them. Projects seeking finance moved from 36 to 24, and the recorded total from USD 19.28bn to USD 18.72bn.
Both numbers are smaller. Both are now true. A database whose only claim is that every record names its source cannot keep twelve that do not.
The headline is eight projects
A regional capital figure invites a conclusion about a region. This one mostly describes eight decisions.
The largest entries are Terra Solar in the Philippines (3,500 MW, USD 3.9bn), the Khavda Renewable Energy Park in India (5,000 MW, USD 3.0bn), Dholera Solar Park in India (5,000 MW, USD 2.5bn), and Duqm Green Hydrogen Solar in Oman (1,300 MW, USD 1.5bn). Add Kuwait's Shagaya Phase III entries and the total passes USD 13bn while the list is still a quarter of the way down.
Anyone sizing a market from the aggregate is sizing those programmes. Most are already in construction, already sponsored, and not looking for the reader.
Where the count is, and where it is not
The pattern reverses if you count projects instead of dollars. Ten of the twenty-four are seeking under USD 150 million: rooftop and C&I portfolios, a utility solar programme in Sri Lanka, efficiency projects in India and Indonesia.
That band is 42% of the projects and 4% of the money. It is also the band where a single funder can take a whole position, which is a different proposition from joining a syndicate on a multi-gigawatt park.
We are not saying that band is better. We are saying the aggregate hides it.
The instrument, where it is stated
Of the 24, the recorded capital type is blended for 22, debt for 1 and equity for 1.
Read that with care, and note it moved sharply this week. Several of the withdrawn records carried debt, equity and grant labels that no source supported. What remains is dominated by development-bank and sovereign programmes, where blended is the standard framing. It reflects which institutions publish their financing plans, not what private sponsors across the region are actually seeking.
What we have verified, and what we have not
Ten of the 24 have been checked against a primary source. Fourteen have not, and thirteen of the twenty-four carry a direct source link.
The verified share rose from 28% to 42% this week, and none of that came from new verification work. It rose because we removed unverified records. That is worth stating plainly, because a ratio that improves through subtraction is not the same as one that improves through effort.
Unverified does not mean wrong. It means nobody has yet put eyes on an original document, and until that happens the entry carries the flag. Every record shows its own status, its source, and the date it was last checked, so a reader can decide what to lean on rather than taking our word for the lot.
Method
Entries are compiled from named public sources. Each carries a link and a check date. Nothing is marked verified without a primary-source check. Figures are never estimated from capacity or inferred from comparable projects. Where a figure is not published, the field stays empty.
Contact details are not published. Introductions are requested through the record and the parties deal with each other directly.
Corrections are welcome and applied to the public record, with the change visible. This issue is one.
Information only. Not financial advice. We are not a broker and do not hold funds or securities.