Isracard: How Much Capital Headroom Really Remains Before The ESH Deal

Starting point

The main article’s core argument was that capital had become Isracard’s active bottleneck. This follow-up isolates only that thread: after the special dividend, after the VAT hit, and after another year of credit growth, how much capital room is actually left before ESH enters the picture.

The first number that matters here is not normalized profit and not market cap. It is a year-end CET1 ratio of 10.7%, against an internal target of 9.75% and a regulatory minimum of 8.0%. Based on CET1 capital of NIS 3.011 billion and risk-weighted assets of NIS 28.248 billion, the room above the internal target is only about NIS 257 million. Above the regulatory floor, the room is about NIS 751 million. That is still compliance. It is no longer a wide cushion.

What matters is that this number already includes everything that hurt in 2025. It is already after the NIS 1.2466 billion special dividend paid in August 2025, after the additional VAT-related charge of about NIS 290 million before tax and NIS 223 million after tax, and after an 11.3% increase in risk-weighted assets to NIS 28.248 billion. In other words, this is not a theoretical cushion. It is the cushion that remained after the year had already consumed much of the comfort.

Four points frame the picture:

• 2025 was more a denominator squeeze than a numerator collapse. CET1 after deductions slipped only from NIS 3.025 billion to NIS 3.011 billion, while risk-weighted assets jumped from NIS 25.375 billion to NIS 28.248 billion.
• Even the internal bar was lowered. Midroog states that in November 2025 the board revised the internal CET1 target to 9.75% from 10.0%, so part of today’s room comes from a lower hurdle, not only from a stronger cushion.
• The rating signal already says this is not viewed as excess capital. Midroog kept the rating but moved the outlook to negative because of capital erosion, mainly after the foreign-VAT court ruling.
• The ESH memorandum itself looks like a deliberate attempt to protect capital through deal structure, not like a transaction launched from obvious surplus capital.

That table matters because the question is no longer whether Isracard is meeting the rules. It is. The question is how much freedom is left above those rules, and whether that freedom is enough for more credit growth, a dividend policy reset to up to 40% of semiannual net profit, and another strategic move.

What the rating report is really saying about 2026

Midroog’s March 17, 2026 follow-up report adds the critical forward layer here. Its message is straightforward: the cushion is still adequate relative to regulation, but it is no longer wide relative to the rating.

In its base case, Midroog assumes 18% to 20% annual credit-book growth, a CET1 ratio in the 10.6% to 10.7% range, and dividends from ongoing profits during the forecast period. That matters because the “comfortable” case is already carrying three simultaneous sources of pressure: more risk-weighted assets, resumed distributions, and only moderate profitability. Anyone reading year-end 2025 as a return to a clear comfort zone is ignoring the fact that Midroog’s own base case does not rebuild the capital cushion back above 11%.

There is also an important distinction to keep clean. The new liquidity policy, requiring committed unused facilities of at least 135% against debt maturities over the next 24 months, addresses liquidity. It does not solve the capital question. That is helpful for creditors, but it does not create new CET1.

This is also where the key conclusion starts to form: ESH is landing on a company that is already planned forward with a relatively narrow capital cushion, not on a company sitting on obvious surplus capital waiting to be deployed. The fact that Midroog is underwriting 10.6% to 10.7% CET1, not a rebuild back to a visibly stronger level, means Isracard enters 2026 needing capital discipline, not just strategic ambition.

What the ESH memorandum is really asking from the balance sheet

The non-binding memorandum signed on March 17, 2026 is structured in a way that tells the story by itself. The bank leg is paid mainly in Isracard shares, not in cash. That is not a technical detail. It is the core economic signal.

Under the memorandum, Isracard would issue shares to the bank’s shareholders at NIS 15.96 per share, reflecting an estimated NIS 400 million value for the bank. Of that amount, NIS 250 million of shares would be issued at closing and NIS 150 million of milestone shares would be issued to a trustee and released only if milestones are met. On top of that, the parties would define events that could entitle the bank’s shareholders to aggregate cash payments of up to NIS 100 million over the five years following completion.

The technology leg is different. Isracard would buy 25% of eOS through a USD 40 million investment, subject to adjustments. But this leg also carries a revealing option: subject to approvals, the company may assign the binding agreement for the eOS purchase to its controlling shareholders. In other words, the leg that is explicitly cash, rather than shares, is also the one that can in principle be moved outside Isracard.

Next to those two elements sits the clause the market should not ignore: Isracard may also carry out a private placement of up to about 5% of its shares to Delek, at the same allocation price, to allow the controlling shareholder to comply with the conditions of the Bank of Israel control permit.

The economic reading of that structure is sharp:

• The bank acquisition is built to preserve cash and capital through share consideration.
• The eOS leg is the one that can require direct cash, which is exactly why it also carries an assignment option.
• The possible Delek placement is not only a control-permit clause. It is also a potential capital valve.

Using the issued and paid share count at the end of 2025, the bank consideration alone equates to about 25.1 million shares, roughly 7.7% of the current base. If the Delek placement is also used in full, the visible share layers around the transaction move toward roughly 12% to 13% of the current share count, before anything else changes.

This is exactly where the common reading error begins. Dilution is not an incidental side effect of the structure. It is probably one of the structure’s main funding tools. If Isracard were sitting on a genuinely comfortable capital cushion, it would be much easier to build such a deal with more cash and fewer share-based layers. The fact that the main leg is paid in shares, that milestone shares are used, that the eOS leg can be pushed outside the company, and that a Delek placement is also available tells you that the company is preserving capital through the deal structure itself.

Where the room can disappear, and where it can be rebuilt

This is where the discussion has to move from headlines to order of magnitude. The roughly NIS 257 million of static room above the internal CET1 target is not trivial, but it is also not large enough to justify saying there is “plenty of room.” Several already-known items are moving on roughly the same scale.

The clearest example is the possible Delek placement. On a simple year-end 2025 share-count basis, using the same NIS 15.96 allocation price, a 5% placement implies roughly NIS 260 million. That is almost the same size as the entire current room above the internal CET1 target. So this is not just a formal compliance mechanism for the control permit. Economically, it is almost a full extra layer of current capital headroom.

By contrast, the BuyMe transaction disclosed in Note 30 can help, but on a different scale. If the transaction closes, Isracard estimates a net profit of NIS 60 million to NIS 90 million, most of it expected close to completion. That is useful, but even the top end is materially smaller than the current static CET1 room, and much smaller than the overall size of the ESH package.

The possible cash payment of up to NIS 100 million to the bank’s sellers also cannot be brushed aside, even though it is contingent and spread over five years. In order-of-magnitude terms, that is already large enough to consume a meaningful part of today’s room if it materializes. And in the documents published so far, the memorandum and Note 30, there is no pro forma capital bridge for the transaction. So there is no clean basis for assuming that all of these pieces can be absorbed without either additional dilution or more pressure on the balance sheet.

That brings the central insight into focus. The deal does not read like a pure “do we have enough capital or not” question. It reads like a question of who absorbs the price, the balance sheet or the shareholders. If Isracard wants to defend its capital cushion, the natural structure is more shares, more deferred share consideration, more reliance on interim offsets such as BuyMe, and possibly moving the eOS leg outside the listed entity. If it wants to minimize dilution, the pressure comes back to the balance sheet and the capital cushion.

Bottom line

This continuation’s conclusion is sharper than the narrow question of whether Isracard “can” do the deal. It can build a structure that gives it a chance to do it. But the year-end 2025 numbers show that the company is entering that discussion with measured capital room, not with obvious surplus.

The cushion above the internal CET1 target is about NIS 257 million. Even total capital is only about NIS 186 million above the internal target. Midroog has already moved the outlook to negative, and its base case keeps CET1 around 10.6% to 10.7%, while the documents published so far still do not include a pro forma capital bridge for ESH. At the same time, the memorandum itself is set up so that the bank is bought mainly with shares, the eOS leg can sit outside Isracard, and Delek may inject more equity through a private placement.

This is not a memorandum for an easy transaction. It is a memorandum for a structure trying to turn capital pressure into controlled dilution. Put simply, what remains before the ESH deal is not room that allows many mistakes. It is room that forces Isracard to decide very carefully which part of the price sits on capital, which part sits on dilution, and which part is pushed into the future.