BlackEdge's Real-Estate Book: How Deep Is The Cushion After Senior Debt
BlackEdge presents about NIS 1.2 billion of collateral against roughly NIS 692 million of secured balances, but the report explicitly says that figure is shown without safety haircuts and without taking senior debt into account. This follow-up isolates what the LTV tables, collateral mix and lien ranking actually tell us about the depth of the cushion.
What This Follow-Up Is Isolating
The main article framed the core debate around BlackEdge: the book expanded, funding improved, but margin has not yet recovered. Inside that bigger picture, the real-estate book is both the main growth engine and the main foundation for the collateral story. This continuation narrows the question: how much of that cushion is really left for BlackEdge once the debt sitting ahead of it gets paid first.
The short answer is that the report gives solid reason to take the collateral seriously, but not to read the gross collateral value as a net protection layer. The reason is written explicitly. When the company shows about NIS 1.2 billion of collateral against roughly NIS 692 million of secured balances, and when it separately shows about NIS 878 million of collateral against roughly NIS 485 million of loans in the real-estate snapshot, it adds the same qualification: both figures are shown without safety haircuts and without taking senior debt into account. That is not a technical footnote. It is the entire point.
The implication is straightforward. The report does say there is far more pledged asset value than secured debt. It does not say how much of that gap is actually available to BlackEdge after senior lenders are repaid, after haircuts, and after the time and friction needed to realize the collateral.
This Is Not A Uniform Mortgage Book
The sub-segment table already complicates the headline. Roughly NIS 425.9 million sits in entrepreneurial real estate, about NIS 109.3 million in income-producing real estate, about NIS 25.5 million in land-acquisition financing, and about NIS 1.6 million in other. But the more important split is by debt ranking: about NIS 376.5 million is classified as senior debt and about NIS 185.8 million as subordinated debt.
That already changes the way the cushion should be read. On one hand, roughly two-thirds of the real-estate exposure in the table is senior debt. That matters, because the book is clearly not built mostly as mezzanine risk. On the other hand, about one-third still sits as subordinated debt, and there the question is not only what the asset is worth, but how many layers of capital stand ahead of the company on the same asset.
The collateral-type table breaks the story even further away from a simple mortgage read. The reported real-estate book is not backed only by property mortgages. It includes NIS 193.2 million secured by land, NIS 82.6 million by surplus-account pledges, NIS 82.4 million by income-producing property, NIS 16.6 million by residential apartments, NIS 1.5 million by commercial property, and NIS 102.3 million by listed securities.
That matters because collateral type determines not only headline value but also realization mechanics, realization speed, and the gap between a credit event and actual cash recovery. Note 5 does say that part of the book is backed by hard collateral that can be realized within a reasonable period, but it immediately conditions that on the type of collateral. So even the report itself does not invite the reader to treat surplus pledges, land, income-producing property, and marketable securities as if they were one identical cushion.
Why The LTV Table Looks Tougher Than The Credit Policy
There is one point here that is especially easy to miss, and it is probably the sharpest insight in the whole collateral story. In the credit policy, the company sets LTV ceilings that sound relatively conservative: up to 75% for first-lien real-estate collateral, up to 70% in subordinated real-estate financing for land-equity completion and surplus pledges, and up to 85% in other second-lien real-estate transactions.
If you stop there, you would expect the disclosed LTV distribution to sit mostly below 75%, or at worst below 85%. In practice, the reported table looks different: only 13.9% sits up to 50%, only 0.9% sits between 60.1% and 70%, 53.9% sits in the 70.1% to 80% band, another 13.1% in the 80.1% to 90% band, and another 18.3% in the 90.1% to 100% band. There is no exposure at all in the 50.1% to 60% band and none above 100%.
At first glance that almost looks like a contradiction to policy. But the footnote under the table resolves it. For subordinated debt, the reported LTV is calculated using total debt on the asset, including liens not in the company’s favor, divided by asset value. In other words, the table is not measuring only BlackEdge’s own loan against asset value. It is measuring the full leverage stack on the asset. That distinction is critical.
This leads to the key conclusion. The disclosed LTV level tells the reader how much cushion exists under the entire asset-level capital stack, not only whether BlackEdge itself stayed within its own underwriting cap. So the 18.3% sitting in the 90.1% to 100% band is not automatically proof of loose internal policy. It is proof that in part of the deals where the company sits in a subordinated position, the residual cushion behind it is much thinner than the headline gross-collateral numbers suggest.
Another striking point is where the mass of the book actually sits. Nearly 78.4% of the entire LTV distribution lies in two project structures: projects under construction backed by surplus plus a first mortgage, and planning-stage projects or land backed by surplus plus a second mortgage. Only about 18.1% sits in the category of land, existing assets, and income-producing assets with a first-lien charge. That means a very large part of the cushion is tied to projects, surplus accounts, and capital-stack position, not just to a clean first mortgage on a standing, income-producing asset.
Why The Approval Ladder Matters To The Collateral Story
The report adds one more useful clue. It does not rely only on formal LTV rules. It also discloses a fairly tight approval ladder. Exposure up to NIS 10 million can be approved at the business-line level subject to a CRO opinion. Between NIS 10 million and NIS 20 million requires CEO approval. Above NIS 20 million and up to 5% of the credit book requires joint approval by the CEO and chairman, and above 5% of the credit book requires board approval. On top of that, a deal that materially misses at least two threshold conditions must be escalated one level higher.
That is not administrative detail. It means the company itself treats the protection layer as something that cannot be judged only through a single LTV number. Deal quality, threshold compliance, and policy exceptions all trigger governance escalation. In a book where a meaningful share already sits at relatively high LTV bands and in subordinated structures, that is a soft protection layer, but still an important one.
What Still Cannot Be Calculated
The issue is not a lack of disclosure. The issue is that disclosure remains gross and partial. The report gives several slices of the same story, but not one full bridge linking transaction type, collateral type, lien ranking, the debt sitting ahead of the company, and the haircuts it actually applies.
| What is disclosed | What is still missing | Why that matters |
|---|---|---|
| Gross collateral value against secured debt | Senior debt balance for each deal or each sub-book | Without that, you cannot calculate what is left for BlackEdge after the lender ahead of it |
| Senior-versus-subordinated split | A deal-by-deal bridge from debt ranking to collateral type | Not every subordinated loan sits on the same collateral and not every senior loan enjoys the same protection |
| A detailed LTV table | Haircuts and realization assumptions | Gross asset value is not recovery value |
| A general statement that hard collateral can be realized within a reasonable period | Timing by collateral class | A paper cushion and a liquid cushion are not the same thing |
That is exactly why an intuitive reading can go wrong. You cannot take a gross 1.7x or 1.8x collateral ratio and automatically treat it as an equity cushion. Anyone holding the table without asking how much debt sits ahead of the company is only reading half the story.
Bottom Line
BlackEdge’s real-estate book looks stronger when you focus on the existence of collateral, and less strong when you ask what is left after the senior layer. That is not an argument against the collateral itself. Quite the opposite. The report shows that the book is built with far more protection than a generic non-bank credit label would imply. But it also shows that this protection is heterogeneous, that a meaningful part of it sits in project structures involving surplus pledges and second liens, and that for subordinated debt the reported LTV is measured against the full debt stack on the asset, not just against BlackEdge’s own tranche.
So the right reading is not “there is NIS 878 million of collateral against NIS 485 million of loans, therefore everything is covered.” The right reading is different: BlackEdge does have a real protection layer, but its economic depth depends on each deal’s capital structure, on collateral type, and on how much debt sits ahead of the company. Until the report gives a full bridge across those layers, the collateral cushion should be read as a gross cushion, not a net one.
Disclosure: Deep TASE analyses are general informational, research, and commentary content only. They do not constitute investment advice, investment marketing, a recommendation, or an offer to buy, sell, or hold any security, and are not tailored to any reader's personal circumstances.
The author, site owner, or related parties may hold, buy, sell, or otherwise trade securities or financial instruments related to the companies discussed, before or after publication, without prior notice and without any obligation to update the analysis. Publication of an analysis should not be read as a statement that any position does or does not exist.
The analysis may contain errors, omissions, or information that changes after publication. Readers should review official filings and primary sources before making decisions.