Erech Finance: How Much of the Legacy Book Can Still Turn Into Cash
Erech Finance still reports a NIS 49.2 million gross legacy book, but after a NIS 38.7 million provision only NIS 10.5 million remains net. Even that residue sits almost entirely in impaired debt, long-dated files, and legal collection proceedings, so the real issue is no longer book size but recovery quality and speed.
What Is Actually Left in the Book
The main article argued that Erech Finance is no longer an active lender, but a public shell trying to buy time until it can bring in a new engine. This follow-up isolates the narrower question that sits inside that broader thesis: how much of the old book can still turn into real cash.
The short answer is much harsher than the balance-sheet headline. The legacy book still stands at NIS 49.21 million gross, but after a NIS 38.716 million credit-loss provision only NIS 10.494 million remains net. Even that remaining amount does not sit inside a reasonably performing book. It sits almost entirely in impaired debt, mostly in exposures older than 12 months, inside a portfolio that has already moved deep into legal collection.
That is the key economic point. The NIS 10.5 million is not a nearly ready cash box. It is an accounting estimate of what may still remain after file-by-file analysis of collateral, guarantees, and legal proceedings. Anyone looking only at the gross book misses the heart of the problem. Anyone looking only at the net book misses the heart of the risk.
| Layer | December 31, 2025 figure | What it really means |
|---|---|---|
| Gross debt book | NIS 49.21 million | The size of the portfolio before recognizing how much has already deteriorated |
| Credit-loss provision | NIS 38.716 million | Roughly 78.68% of the book has already been cut away |
| Net book | NIS 10.494 million | What remains as an economic estimate, not as committed cash |
| Impaired debt | NIS 48.738 million gross, NIS 10.032 million net | Almost all of the value left still sits in troubled files |
| Debts over 12 months | NIS 48.454 million gross, NIS 9.869 million net | Most of the remaining book is old, slow, and heavier to realize |
| Top ten customers | NIS 33.353 million gross, NIS 6.547 million net | Recovery remains concentrated even after provisions |
| Legal proceedings | More than 60 files | This is no longer routine collection but an enforcement workload |
That chart is the right starting point. The issue here is no longer NIS 49.2 million in headline book size, but the NIS 10.494 million left on the books after provisions. And even that should not be read as a near-term collection forecast. It is simply the amount left after a very deep haircut.
The credit-risk note makes the number even more fragile. The overwhelming majority of the debts carry specific rather than general provisions. In other words, management and its legal advisers assessed major files one by one, estimated what may still be realized through proceedings or collateral, and booked the gap as a provision. The 1.5% general provision is left only for a very small cleaner tail. So the net book is not a broad portfolio supported by portfolio statistics. It is a collection of case-level judgments on files that have already been cut aggressively.
This is where the internal migration of the book becomes clear. At the end of 2024, NIS 3.875 million net still sat in the category of debts with increased credit risk since initial recognition. By the end of 2025, that bucket had almost disappeared, leaving only NIS 14 thousand. At the same time, NIS 10.032 million out of NIS 10.494 million net, about 95.6%, already sits in impaired debt. In plain terms, the middle layer has almost vanished. What is left is no longer the economics of shaky borrowers that may normalize. It is the economics of files the company already classifies as impaired.
That also explains why the NIS 10.5 million no longer behaves like the residual book of a going lending business. Only NIS 448 thousand of the net book is still classified as debt with no increase in credit risk, and another NIS 14 thousand sits in the increased-risk bucket. Almost everything left is the residue of a portfolio that has already crossed into enforcement.
Where the Residual Value Still Looks Recoverable
If there is still real economics inside the old book, the way to find it is not through the headline exposure, but through collateral structure and concentration. That is where the portfolio stops deteriorating evenly. Some layers look very weak, while others still carry most of the net value left.
On a gross basis, the book is driven mainly by personal guarantees, NIS 36.712 million, or about 74.6% of the portfolio. But after provisions the picture changes sharply. The net amount left in that layer falls to only NIS 5.027 million, or about 13.7% of the original exposure. By contrast, loans backed by real estate and equipment account for only NIS 8.004 million gross, about 16.3% of the book, but NIS 4.474 million net, about 42.6% of the total net book. That is a major point. Almost half of what is still left economically in the book no longer sits in the broad layer of personal guarantees. It sits in the harder-collateral slice.
The implication is straightforward. The real economic question is no longer how big the book is, but how much can actually be realized from the secured layer. If collections reach the files backed by real estate and equipment, the legacy book can still produce some cash. If even that layer proves slower, more expensive, or more partial than expected, the remaining net book may turn out to be generous.
Concentration also survives the haircut. The top ten customers hold NIS 33.353 million gross, about 67.8% of the total book, and still account for NIS 6.547 million net after provisions, about 62.4% of the total net book. So even after management has already cut the accounting value of the portfolio, most of what remains still sits in a handful of debtors.
The note on material customers sharpens that point further. Two material customers alone, Customer A and Customer C, account together for NIS 4.223 million net. That is about 40.2% of the total net book. So reading the NIS 10.5 million as a broad and diversified recovery pool is too soft. In practice, the remaining portfolio looks much more like a few large cases plus a tail than like a wide base of smaller paying borrowers.
Why the Recovery Process Will Stay Slow and Heavy
The biggest bottleneck inside the legacy book is not only credit quality. It is time. The company states clearly that all customers are already past due and that collection proceedings are now under way. That means the maturity table is no longer a normal repayment schedule. It is a way to understand how much of the remaining book is tied to old and long-dated files.
The numbers are close to one-directional. NIS 48.454 million out of the gross book, about 98.5%, sits in debts over 12 months. Even after provisions, NIS 9.869 million out of NIS 10.494 million net, about 94%, remains in that same long-dated layer. So the old book is no longer built on a stream of near-term repayments. It is built on patience, proceedings, and the ability to turn collateral and settlements into cash over time.
The sector mix does not make the job easier. Construction and real estate together account for 65% of the gross book. That does not by itself prove lower recoverability across those files, but it does mean that a large part of the portfolio is concentrated in two similar sectors, so any continued collection weakness there would hit a large share of the book. Once the company adds that more than 60 files are already in legal proceedings, the picture becomes that of a book where recovery is no longer only a question of borrower quality. It is also a question of legal burden and time.
It also matters how the company describes its own process. It does not rely on one broad percentage across the book. It reviews customer financials, collateral value, guarantor quality, legal developments, and credit reports. That is exactly why the residual value is so heavy. The old book may still turn into cash, but it is likely to do so through a small number of important cases, in stages, and with much more friction than the NIS 10.5 million headline might suggest at first glance.
Conclusion
Erech Finance’s old debt book has not gone to zero, but it no longer looks like NIS 49.2 million waiting to be collected. It looks like NIS 10.494 million of accounting value left after a very deep haircut, and almost all of that residue sits in impaired debt, long-dated files, and high concentration.
That is the core point of this follow-up. The real question is no longer whether value remains in the book, but how much of the remaining value is actually accessible and at what pace. As of the end of 2025, the more plausible source of future cash sits mainly in the real-estate-and-equipment-backed layer and in a few large files that survived the provisioning cycle. On the other side, any disappointment in those files would hit a large part of the net book directly, because there is barely any middle layer left to absorb the damage.
That also sharpens the conclusion of the main article. The legacy book may still buy the shell some time, but it no longer looks like a broad enough funding base to carry the structure on its own. If collections in the secured files move forward, some usable interim cash may still come out of it. If not, even the NIS 10.5 million may prove to be less an extractable value pool and more an accounting stopping point inside a book that has become almost entirely a recovery project.
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.