Bank Jerusalem's Mortgage Sale Ceiling Limits Capital-Release Profit
The February mortgage portfolio sale proved that Bank Jerusalem's capital-release model works and moved the bank close to its approved limits. With sold mortgage portfolios already at about 24% of the managed mortgage book versus a 25% ceiling, first-quarter profit looks less repeatable without portfolio growth or broader approval.
Export already received first-quarter evidence that Bank Jerusalem's profit and cash can move up to the holding-company layer, but this continuation isolates the quantitative constraint inside that mechanism. Bank Jerusalem sold about 90% of a roughly NIS 680 million mortgage portfolio in February 2026, recognized NIS 37.5 million of pre-tax profit, and retained 10% of the portfolio alongside servicing and management fees. That is a real capital-release transaction, not only a convenient earnings line. The issue is that immediately after the transaction, sold mortgage portfolios already represented about 24% of the mortgage book against an approved 25% ceiling, and about 33% including syndication against a 35% ceiling. In business terms, the tool that lifted first-quarter profit remains active, but its approved operating room is nearly full. The quality mix of the sold portfolio also matters: 87.5% of the sold mortgage portfolio is at an LTV of up to 60%, compared with 62.6% in the bank's retained balance-sheet mortgage portfolio. The next proof point is therefore not whether the bank can sell another portfolio, but whether it can grow the denominator, secure wider approval, or build recurring servicing income without leaving a less conservative mortgage mix on balance sheet.
The Sale Ceiling Makes the Quarter Harder to Repeat
The prior credit portfolio sale analysis already framed this as a strategic model rather than a one-off event. The first quarter of 2026 confirms that part: the bank did not only sell a portfolio. It remained the sponsor and servicer, retained 10% of the rights and obligations, and continues to service the buyer's portion for management fees. This is how a relatively small bank can turn a constrained balance sheet into a platform that originates credit, transfers part of the risk and continues earning service income.
The less comfortable part for first-quarter earnings is the ceiling. The regulatory framework prevents selective cherry-picking or lemon-picking of mortgages, requires the bank to retain at least 10% of each loan in a sale transaction, and limits mortgage sales or joint origination with institutional investors. The bank has wider Bank of Israel approval: up to 35% of the managed mortgage portfolio when syndication is included, and up to 25% for outright sales excluding syndication. As of March 31, 2026, the bank was close to both limits.
| Mortgage Capital-Release Route | Use at End-March 2026 | Approved Ceiling | Meaning |
|---|---|---|---|
| Mortgage portfolio sales excluding syndication | about 24% | 25% | Little room for another large transaction through the same route without denominator growth or additional approval |
| Sales and syndication combined | about 33% | 35% | The model remains active, but room for a series of similar deals is already narrow |
The disclosure does not provide a single shekel amount of remaining sale capacity, so the right read is not to invent a residual cap in NIS terms. The disclosed percentages are enough to change the quality of the quarter's profit.
The bank can grow the denominator through new mortgage originations, and the 2024 syndication agreement is still producing volume. In the first quarter, joint mortgage loans of about NIS 235.3 million were originated under the agreement, and the outstanding balance originated under it reached about NIS 1.16 billion at quarter-end. Even so, the large February sale accelerated use of the existing approval. NIS 37.5 million of pre-tax profit is not a quarterly run rate that can simply be extended without asking how many approved portfolios remain available for sale.
The Sold Portfolio Is More Conservative Than the Retained Book
The quantitative approval is not the only constraint. The mortgage portfolio risk table shows that the exiting portfolio is more conservative by loan-to-value ratio. In the sold portfolio, 50.0% of loans are at LTV of up to 45%, and another 37.5% are in the 45% to 60% band. In the retained balance-sheet mortgage portfolio, those two bands together account for only 62.6%. By contrast, mortgages above 60% LTV are 12.6% of the sold portfolio and 37.4% of the retained portfolio.
That gap does not mean the retained book is weak. The bank explains that sales increase the average LTV left on balance sheet because buyers restrict purchases of loans with LTV above 60%, while the average LTV remains relatively low in light of the bank's underwriting policy. The bank also states that the sales had a positive impact on profitability and no material negative impact on net interest income.
That is the strongest counterpoint, and it is reasonable. Capital release is not necessarily a deterioration in credit quality, especially when the bank retains part of the portfolio on the same seniority and continues to service the loans. The point for parent-company investors is to track the next mix, not only the next gain. If future sales continue to remove mostly lower-LTV mortgages, the bank will need to prove that the credit remaining on balance sheet does not create more credit pressure or margin erosion over time.
For Export, the Constraint Matters Through Dividends
For shareholders of the parent company, the portfolio sale matters because it can improve profit, capital and dividends at the bank. It is less powerful as evidence of recurring profit. In the first quarter, the bank recognized a large gain from a portfolio sale, and the parent benefited from a dividend that moved cash up. In the coming quarters, that chain will need different proof: updated utilization of the approved ceilings, a clearer split between sale gains and recurring servicing income, and stability in the quality of the retained book.
The current read is that the bank's capital-release model works, but the first quarter moved it close to the quantity limit approved for it. That does not erase the model's value, because syndication, servicing fees and natural mortgage-book growth can still extend the runway. It does change how to weigh the profit: the February transaction is proof of capability, not proof that a similar gain can be released every quarter. The next evidence will be how much new headroom appears in the next disclosure, and what happens to LTV and credit quality in the portfolio that remains.
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.