Bull Trade 2025: The Loan Book Is Shrinking, but the Real Pressure Has Shifted to the License and the Debt

Getting to Know the Company

At first glance Bull Trade still looks like a small non-bank lender. That is no longer the right way to read it. By the end of 2025 this is not a credit company looking for growth, new customers, or book expansion. Since July 2023 it has stopped extending new credit, and since September 2024, under the debt arrangement, it has been constrained both legally and operationally to collections, occasional approved reschedulings, and management of the legacy book. The central question is therefore no longer how much finance income is still being booked, but how much real cash can still be extracted from the portfolio before the pressure fully shifts to creditors and the regulator.

What is still working is the stripped-down operating layer. During 2025 the company carried out five mandatory early redemptions of Series B bonds totaling NIS 26.15 million par value, cut selling, general and administrative expenses to NIS 6.24 million from NIS 7.93 million in 2024, and also reduced office costs after moving in December 2025 to the supervising accountant's office, at roughly NIS 10 thousand per month versus roughly NIS 34 thousand before. This is not a business recovery, but it does show that the company is being run as a workout shell that is trying to burn less cash on the way.

What is not clean is almost everything else. The gross loan book fell to NIS 130.4 million from NIS 172.0 million, but the credit-loss allowance rose to NIS 99.2 million, leaving a net book of only NIS 31.1 million. Some 95% of the portfolio is already classified as Stage 3, and 95% is also more than 181 days past due. Year-end cash stood at just NIS 3.3 million, and equity moved from a positive NIS 35.9 million at the end of 2024 to a negative NIS 5.1 million at the end of 2025. Put differently, the portfolio is still large in nominal terms, but the company's own margin of safety has almost disappeared.

This is exactly where a superficial read can go wrong. The company still reports positive working capital of NIS 12.8 million, and it still holds meaningful collateral. But that working capital is mostly distressed receivables rather than cash, and the collateral note explicitly says the disclosed values do not include forced-sale haircuts, taxes, levies, or expenses, and may include floating liens that do not automatically turn every shekel of stated value into a shekel of recovery. The active bottleneck at Bull Trade is therefore not "profitability" in the usual sense, but the pace and quality of recovery, against a bond liability that keeps compounding and a license now challenged by the breach of the minimum-equity requirement.

The table below frames the company correctly. This is no longer a credit platform in growth mode. It is a collection book inside a listed shell.

Events and Triggers

Cash That Actually Reached Creditors in 2025

The first trigger is not growth but cash transfers to creditors. During 2025 the company completed five mandatory early redemptions of Series B bonds totaling NIS 26.15 million par value, in January, March, April, August, and December. That matters for two reasons. On one hand, it proves the company was still able to convert part of the legacy book into real cash and move it upstream. On the other hand, it also shows that whatever value is created in the book does not remain at the company level for long. Under the arrangement it is quickly pulled toward bondholders. Even when cash comes in, it does not necessarily enlarge the margin of safety for equity holders.

Control Has Moved From Growth to Recovery Mechanics

The second trigger is structural. The debt arrangement no longer looks like a temporary framework. It looks like the company's actual operating model. At completion, 80% of the fully diluted share capital was transferred to bondholders, a supervising accountant with oversight and joint-signature powers was appointed, quarterly expenses were capped at NIS 1.5 million, and the company undertook not to take new loans, not to distribute dividends, not to pledge additional assets, and not to extend new credit. In January 2026 Series B bondholders also appointed a representative, Mr. Pikovski, to review and advance proposals to acquire the company's activity or its shares. That is a clear sign that the conversation has shifted from "managing the loan book" to "choosing a realization path."

The important point is that if a proposal to buy the activity or the shares appears, decision-making no longer sits primarily with the legacy equity layer. Any value, if created, will first be examined through the lens of Series B bondholders and the arrangement terms. That is a critical distinction between accounting value or operating value and value that is actually accessible to ordinary shareholders.

December Bought Time, March Changed the Tone

The third trigger is the move from temporary relief to potential enforcement. In December 2025 bondholders approved a specific waiver of an immediate-acceleration trigger related to collection targets through June 30, 2026, pushing the next possible date for that trigger to September 30, 2026. That buys time, but it does not solve the problem. In fact, the company itself already says it does not expect to meet the collection target on that date.

In March 2026 the tone moved up another level. On March 11 a consultation meeting of Series B bondholders discussed a request by a holder of more than 5% of the series to initiate legal proceedings for the appointment of a receiver over all company assets in a liquidation path, alongside a proposal to ask the court to appoint CPA Aliza Sharon as both receiver and trustee. Even without a final decision, the fact that this discussion took place means the market is no longer dealing only with collection management. It is also dealing with an enforcement track.

The License Has Moved From a Regulatory Footnote to an Operating Threat

The fourth trigger, and probably the heaviest one, is the license. The annual report itself states that as of December 31, 2025 the company did not meet the NIS 4 million minimum-equity requirement for an expanded credit license. On March 12, 2026 the company also informed the Capital Market Authority that, based on work on the 2025 financial statements, it appears equity is below the required threshold and the supervisor may suspend or revoke the license after giving the company an opportunity to be heard.

This risk matters because the company no longer relies on the ability to resume credit growth. The license therefore matters not only for future lending, but also for the ability to keep managing an orderly runoff, approve isolated reschedulings when needed, and preserve some optionality if a purchase proposal or some other framework appears. If the licensing layer breaks, even the recovery path can become more rigid.

Efficiency, Profitability and Competition

Bull Trade's key metric is no longer competitive advantage versus other lenders. The company is not competing for new customers, so the real question is what happens to each shekel still trapped in the legacy book, and how much of it survives after finance costs, provisions, and collection expenses.

2024 Was an Accounting Profit, 2025 Brought the Story Back to Earth

This is the most important point in reading the numbers. In 2024 the company reported profit before tax of NIS 26.9 million, but almost all of that came from an accounting gain of NIS 68.6 million on the debt arrangement. In 2025, without that one-off gain, the company swung back to a pre-tax loss of NIS 40.9 million. Anyone treating 2024 as a genuine turnaround is therefore missing the fact that the healing then was mainly accounting, not operating and not cash-based.

Finance income fell in 2025 to NIS 6.7 million from NIS 7.5 million in 2024 and NIS 27.7 million in 2023. Finance expenses, meanwhile, rose to NIS 25.15 million. The company explains that the increase reflects the post-arrangement accounting treatment of the bonds, and the notes state explicitly that during 2025 the company booked finance expenses on the bonds of NIS 25.06 million at an effective annual rate of 104.56%, even though the contractual rate is 15%. That is the core of the story. The arrangement that created an accounting gain in 2024 created, in 2025, an effective interest burden that now weighs on the income statement.

The meaning is that 2025 was not just a weak year. It was a year that removed an illusion. If 2024 created the impression that the debt arrangement materially improved the economics of the company, 2025 shows that this improvement did not make the model healthier. It merely changed the shape of the debt and the way the debt cost now flows back into the accounts.

The Book Is Deteriorating Faster Than It Is Shrinking

The decline in the loan book may look positive on the surface, but the quality of what remains matters more than the shrinkage itself. Gross receivables fell to NIS 130.4 million from NIS 172.0 million, yet the credit-loss allowance increased to NIS 99.2 million from NIS 95.2 million. As a result, the net carrying value collapsed to NIS 31.1 million from NIS 76.8 million.

The average allowance ratio jumped to 76.12% of the book, versus 55.35% a year earlier and 39.87% in 2023. This is not just extra conservatism. It is a real change in risk composition. According to the report, 95% of the portfolio is already in Stage 3, meaning impaired loans. The aging schedule shows 95% of the book more than 181 days past due, while only 1% is not past due at all. The company itself says that around 100 out of roughly 105 customers with an outstanding balance are in arrears. This is no longer a loan book with some troubled pockets. It is a troubled book almost throughout, with only a few pockets still looking collectible.

There was no real sign of improvement in the fourth quarter either. Quarterly finance income fell to NIS 1.16 million, finance expense reached NIS 7.75 million, and the pre-tax loss for the quarter was NIS 12.0 million. In other words, 2025 did not finish with stabilization. It finished with another indication that time is not necessarily improving book quality.

Collateral Helps Framing the Book, but It Does Not Create Immunity

This is where overly optimistic readings can creep in. In the portfolio of real-estate secured loans and other tangible collateral, gross exposure stood at NIS 61.5 million against collateral value of NIS 91.5 million. On paper that looks comfortable. Even some of the largest borrowers appear to have relatively low loan-to-value ratios, and the largest single borrower accounts for 19.46% of the portfolio, operates in real estate, carries collateral value of NIS 10 million, and shows an LTV of 39.41%.

But the report itself cools that reading. The note explains that collateral values are shown without quick-sale discounts, without taxes, levies, and expenses, and may include floating liens that are not specific to a single exposure or whose surplus cannot be transferred to other loans. In other words, collateral is a starting point for analysis, not an immediate recovery estimate. In the deferred-check discounting portfolio the picture is even harsher: gross exposure of NIS 68.9 million against only NIS 6.5 million of collateral.

Concentration has not disappeared either. The three largest customers still account for 21.57% of the portfolio at the end of 2025. On the other hand, finance-income concentration from the three largest customers fell to 10.91% from 34.47% in 2024. That may look like healthier diversification, but here the more plausible reading is that finance income itself has simply collapsed and a larger part of the book has stopped generating ordinary current income. The decline in income concentration is therefore not necessarily a quality improvement.

Cash Flow, Debt and Capital Structure

Here the right frame is all-in cash flexibility. The story is not the theoretical cash-generation capacity of a going, growing lender. It is how much cash remains after all actual uses of cash under an arrangement that pulls most of it upstream to creditors.

Cash, Not Working Capital

At the end of 2025 the company had NIS 3.3 million of cash and cash equivalents plus a restricted deposit of NIS 2.9 million. Cash flow from operating activity was negative NIS 0.2 million, financing cash flow was negative NIS 2.0 million, and total cash declined by NIS 2.19 million during the year. That is a non-obvious point. Despite a gross receivables book still above NIS 130 million, and despite meaningful mandatory redemptions to bondholders, reported operating cash flow has already turned negative.

Working capital was positive, NIS 12.8 million, but that number can mislead. Of the NIS 29.3 million of current assets, only NIS 3.3 million is cash. The rest is tied to customer credit, a restricted deposit, other receivables, and tax receivables. That means the company is not sitting on a liquid cushion. It is sitting on claims whose realization depends on a book that has already moved overwhelmingly into Stage 3.

Debt Is Still the Story, Even If the Accounts Look Calmer

At year-end 2025 Series B bonds stood at NIS 80.27 million par value. Their carrying value in the financial statements was NIS 45.3 million, accrued interest was roughly NIS 25.1 million, and their exchange value was NIS 24.1 million. The gap between the carrying amount and market value is an important external warning signal. The bond market is pricing the liability at a much lower level than the accounting carrying value. That is not proof of an extreme outcome, but it is a reminder that creditors themselves are not reading the numbers with the same optimism the balance sheet may sometimes allow.

The payment structure is also very tight. Principal is due in a single bullet payment on June 30, 2027, but in practice the company is required to make mandatory early redemptions every quarter-end, or whenever NIS 4 million accumulates in the cash box, leaving only NIS 1.0 million to NIS 1.25 million for the next quarter's expenses. The implication is that each successful collection is rapidly pulled out of the company and moved to the bondholders.

Another crucial point is that the company no longer has bank lines or other non-bank funding sources. According to the report, Series B is the only financing source left. There is no quiet backstop behind the scenes. If collections fall short, there is almost no alternative liability layer to bridge the gap.

Outlook

Before getting into the forward numbers, four non-obvious findings should be fixed in place:

• The 2024 profit did not heal the company. It came mainly from a one-off accounting gain of NIS 68.6 million on the debt arrangement, while 2025 already shows what happens once the new debt structure begins to flow through heavy finance expense.
• Positive working capital is not the same as cash flexibility. The company has NIS 12.8 million of positive working capital, but only NIS 3.3 million of cash, while most current assets are already troubled receivables.
• The waiver is not a solution. Bondholders waived the immediate-acceleration trigger only through June 30, 2026, while the company itself already estimates it will miss the September 30, 2026 collection target.
• The license issue has moved from a theoretical regulatory risk to an execution constraint. Once equity fell below the minimum threshold, even the orderly runoff path became more fragile.

2026 Is a Collection Test Year, Not a Stabilization Year

The company's own forecast for 2026 and 2027 is very explicit. For 2026 it expects to open with NIS 3.3 million of cash, generate NIS 20.1 million of operating cash flow, and pay NIS 22.39 million on Series B, leaving only NIS 1.0 million of year-end cash. For 2027 the picture gets thinner still: NIS 1.0 million opening cash, NIS 4.67 million operating inflow, NIS 5.67 million bond payments, and a closing cash balance of zero.

This is not a stabilization forecast. It is a walk-the-line forecast. And there is another yellow flag here. The same forecasting framework already proved optimistic in 2025. In the 2024 annual report the company forecast 2025 operating cash flow of NIS 37.77 million, but actual operating cash flow came in at NIS 29.76 million. The company explains the miss by saying the estimate partly relied on arrangements that were not fully signed, and because some customers who do sign arrangements later fail to honor them. That is a very material admission. If the 2025 forecast missed by roughly NIS 8 million, the market has little reason to assign a full premium to the NIS 20.1 million operating inflow projected for 2026.

What Has to Happen for the Thesis to Hold

First, actual collections in 2026 need to get at least close to the range implied by the company's own forecast. Without that, the September 2026 discussion quickly moves from a theoretical covenant-style issue to a real confrontation with creditors.

Second, the company needs to stabilize the licensing layer. If the Capital Market Authority takes a strict line, even the orderly management of the runoff can become more complicated. True, the company is no longer relying on the ability to extend new credit, but it still needs a valid regulated framework to run a financial platform, approve isolated restructurings, and preserve optionality if a purchase proposal or some other framework emerges.

Third, Series B bondholders need to prefer continued managed runoff over a more aggressive realization path. The appointment of a representative to review proposals on one side, and the discussion of a possible receivership process on the other, show that both routes are currently open. This is exactly the kind of setup where market interpretation can change even without a sharp move in the income statement.

What a First Read Can Miss

A first read may focus on the fact that the book is still large, that collateral exists, and that mandatory redemptions were actually carried out. All of that is true. But the more important point is that almost the entire system is now working for Series B rather than for the equity layer. Once 80% of the fully diluted share capital has moved to bondholders, once excess cash is swept upward through the mandatory-redemption mechanism, and once the bond market values the debt well below its accounting carrying amount, it is not enough to ask whether value exists in the book. The question is who can actually access that value.

That is why 2026 looks like a decision year. Not a growth year, not a genuine rebuilding year, and not even a stabilization year. It is a collection, regulatory, and creditor test year, where each of those three forces can determine whether the company continues on a controlled path or shifts into a much sharper one.

Risks

License Risk

This is the most direct risk. The minimum-equity requirement for the expanded license is NIS 4 million, and the company ended 2025 with negative equity of NIS 5.1 million. So this is not just broad regulatory tightening. It is a breach of an existing license condition. Even if the supervisor does not act immediately, the mere existence of the breach changes the company's operating framework and the confidence envelope around it.

Collection Risk and Debt Acceleration

The company itself estimates it will miss the September 30, 2026 collection target and will not be able to repay Series B in full by 2027. Once the specific waiver expires on June 30, 2026, any further collection weakness can become a practical basis for a renewed confrontation with creditors. The March 2026 discussion of a possible receivership route shows that this is not a remote tail scenario.

Risk of Over-Reading the Collateral

The report provides collateral numbers, but it also explains why they should not be read as clean liquidation value. There are no quick-sale haircuts, no deduction of expenses or taxes, and in some cases floating liens are included. A meaningful part of the company's exposure is also tied to construction and infrastructure, and the company itself warns that the security environment and labor shortages in those sectors may hurt collection ability and gradually pressure asset values.

Legal and Market Risk

A motion to certify a class action is pending against the company and certain officers, with an estimated range of NIS 7 million to NIS 9.5 million. The company says insurance coverage exceeds the claim amount and that the matter is in mediation, so this is not the main existential risk, but it adds another burden to the management of the runoff period.

The share has also been on the TASE preservation list since July 22, 2024. Even without taking a view on valuation, that is an actionability constraint for shareholders. A company with negative equity, a preservation-list share, a bond market already pricing a weak recovery, and a license challenged by an equity shortfall is not a case where "paper value" is enough by itself.

---

Conclusions

The Bull Trade story is simpler than the tables initially suggest. There is still a sizable receivables portfolio, there is still some collection activity, and expense discipline has improved. But the pressure point is no longer whether the company can return to being a growing non-bank lender. It is whether what remains in the book is enough to serve Series B, preserve an orderly recovery path, and survive the license test.

What supports the thesis today is that the company is still extracting cash from the book, cutting overhead, and managing a relatively orderly runoff. What blocks a cleaner read is that equity has been wiped out, most of the portfolio is already Stage 3, and the company itself admits the next collection target in September 2026 will probably be missed. In the near to medium term, the market is therefore likely to react less to changes in finance income and more to signals from the regulator and the bondholders.

Current thesis: Bull Trade is no longer a credit story. It is a recovery story, centered on how much value remains after creditors and the regulator take first claim on both time and cash.

What changed: In 2024 it was still possible to argue that the debt arrangement had bought time. In 2025 it became clear that the time bought rested on a one-off accounting gain, while the real pressure migrated to equity, licensing, and collections.

Counter-thesis: The company may already have recognized most of the losses, and because the book is now smaller and more tightly controlled, actual recovery could still end up better than the market currently assumes.

What could change the market read: A genuine positive surprise would have to come from collections beating plan or from a structured solution that reduces the license risk. A negative surprise would come if the receivership discussion turns operational or if the collection target begins to fail faster than expected.

Why this matters: Once a lender turns into a workout vehicle, recovery quality matters more than book size, and access to value matters more than the value stated on paper.

Over the next two to four quarters the company needs to show that actual collections are not breaking down, that the license remains on a controlled path, and that Series B bondholders still prefer managed runoff over a more aggressive process. If those three things do not happen together, the thesis does not merely weaken. It shifts from "how much can still be collected" to "through which legal and structural path will realization now happen."