Mizrahi Tefahot 2025: Profit Is Still Strong, but the CET1 Cushion Remains Tight

Getting to Know the Company

Mizrahi Tefahot is not just "Israel's mortgage bank." That is still the core engine, with a NIS 244.2 billion mortgage book and a 36.7% market share in mortgages, but in 2025 the faster growth actually came from business credit, which rose 20.4% to NIS 115.0 billion. So the right way to read this report is not to ask whether the bank can generate profit. It clearly can. The real question is whether it can broaden itself from a dominant mortgage franchise into a wider growth bank without burning through its capital cushion.

What is working right now is easy to see. Net credit to the public rose 11.9% to NIS 400.5 billion, deposits from the public rose 14.0% to NIS 448.4 billion, credit-loss expense fell to just NIS 228 million, and the bank closed the year with NIS 5.63 billion in net profit and 17.0% ROE. Credit quality also looks strong: the NPL ratio fell to 0.97%, total problematic debt fell to NIS 3.9 billion, and the credit-loss charge fell to 0.06% of loans.

But this is not a clean story. Total revenue slipped 0.9% to NIS 14.59 billion, net interest income fell 0.7%, and in the fourth quarter total revenue declined to NIS 3.421 billion from NIS 3.491 billion a year earlier, even as net profit rose to NIS 1.404 billion. In other words, the bottom line still looks excellent, but the engine underneath it is already benefiting less from the 2024 rate tailwind.

That matters at the market level too. At a market cap of roughly NIS 60.3 billion in early April 2026, the bank is trading at roughly 1.7 times year-end equity. The market is not paying for mere stability here. It is already paying for sustained high profitability. That makes even modest slippage in revenue or capital more important than a superficial reading might suggest.

And that leads straight to the active bottleneck. The CET1 ratio stands at 10.24%, against a regulatory requirement of 9.60% and an internal floor of 9.80% set by the board in the third quarter. That means only 64 basis points of headroom above the regulatory minimum, and just 44 basis points above the internal floor. So even after the January 2026 financing layer, the key question for 2026 is not liquidity. It is common equity. That is what separates a comfortable growth story from one that has to prove itself again every quarter.

Events and Triggers

The strategic plan changed the scorecard

In June 2025 the board approved a new 2025-2027 strategic plan with unusually clear targets: ROE of about 17% to 18%, a credit market share of about 23% to 24%, a domestic business-credit market share of about 15% to 16%, a deposit market share of about 20% to 21%, and an average efficiency ratio not above about 35%. On top of that, the plan opened the door to payout of up to 50% of profit, subject to CET1 compliance and adequate buffers.

That means Mizrahi Tefahot can no longer be read through the income statement alone. It also has to be read against the framework it set for itself. The picture is mixed: reported ROE landed exactly at the low end of the target range, but the efficiency ratio rose to 35.9%, so the first year of the plan started slightly above the level the bank wants to see on average.

The fourth quarter already previews 2026

The fourth quarter matters because this is where the market sees the new rate environment in practice. In that quarter, net interest income fell to NIS 2.689 billion from NIS 2.753 billion a year earlier, total revenue fell to NIS 3.421 billion from NIS 3.491 billion, yet net profit rose to NIS 1.404 billion from NIS 1.306 billion, mainly because credit-loss expense dropped to just NIS 25 million from NIS 105 million and operating expenses fell to NIS 1.267 billion.

That is the key point. A reader focused only on headline profit could conclude that the engine is still accelerating. In practice, fourth-quarter profit improved even though revenue weakened. That is a materially different quality of earnings, because sustained improvement from low provisions is much harder to build on than renewed revenue acceleration.

January 2026 added a new funding layer

After the balance-sheet date, the bank moved through an aggressive financing sequence. On January 9, ratings were assigned to a USD 750 million Tier 2 issue due April 15, 2036, rated BBB by Fitch and BBB- by S&P. On January 22, an expected A-(EXP) rating was published for a new senior unsecured issue. On January 23, the bank completed pricing on a USD 700 million senior unsecured issue at a fixed 5.049% coupon, due January 28, 2031. By January 26, Fitch had assigned the notes a final A- rating, and the terms sheet also showed a BBB+ rating from S&P.

This sequence matters in two ways. On one hand, it improves maturity structure, adds long-dated funding, and lowers the risk that the bank enters 2026 with no fresh financing flexibility. On the other hand, it does not create CET1. Senior debt solves funding. Tier 2 supports total capital. The common-equity bottleneck remains.

Rates are a trigger, not just background

The Bank of Israel cut rates twice by 25 basis points, in November 2025 and January 2026, from 4.50% to 4.00%, and its January 2026 research forecast points to an average policy rate of 3.50% in the fourth quarter of 2026. That is good news for borrowers and for housing-related activity, but it is also a clear sign that the peak rate support for financing revenue is now behind the bank.

Efficiency, Profitability and Competition

What really drove profit

At first glance, 2025 looks like another clean record year. That is too shallow a read. Total revenue slipped to NIS 14.588 billion from NIS 14.721 billion, net interest income fell to NIS 11.727 billion from NIS 11.814 billion, and non-interest financing income fell to NIS 387 million from NIS 574 million. Higher fees, at NIS 2.251 billion, helped, but they did not reverse the overall direction.

By contrast, credit-loss expense fell by NIS 291 million to just NIS 228 million, while total operating and other expenses rose by only NIS 17 million to NIS 5.239 billion. Salaries and related expenses actually fell 2.4% to NIS 3.348 billion. So the improvement in profit depended far more on credit quality and cost discipline than on a broader revenue engine.

There is also a finer point inside the provision line. The mortgage book itself moved from a NIS 64 million expense in 2024 to a NIS 119 million gain in 2025, while business activity still generated NIS 85 million of expense and overseas activity generated NIS 69 million. So 2025 benefited not only from a quiet credit year, but also from a direct positive swing in housing-credit provisioning. That does not mean the quality is fake. It does mean the comparison base for 2026 is unusually easy.

One more important layer is management framing. The bank asks investors to look at a "cleaner" number: excluding the special levy and voluntary customer relief, net profit would have been about NIS 6 billion and ROE about 18%. That is a legitimate framing device, but it should not be smoothed into the base case. Capital, payout and buffer room are judged on reported numbers. The adjustment is useful context, not the core lens.

Growth is broader now, but mortgages still dominate

The clearest positive in the report is that growth is no longer riding on the same old engine alone. Housing credit rose 8.9% to NIS 244.2 billion, but business activity rose 20.4% to NIS 115.0 billion. Within that, mid-sized businesses rose 22.3%, large businesses rose 24.5%, and institutional activity more than doubled to NIS 7.9 billion from NIS 3.8 billion. In other words, the bank is genuinely moving toward its strategic goal of broadening the business franchise.

Even so, mortgages remain the center of gravity. About 61% of total loans still sit in housing credit. In the market itself, the bank remains highly dominant, with a 36.7% share in mortgages, 33.6% in retail banking, 21.9% in total loans and 18.8% in deposits. That means the story is improving, but not changing architecture overnight. The bank is broader, not fundamentally different.

Efficiency is good, but not finished

Mizrahi Tefahot still shows impressive expense discipline relative to the system. The efficiency ratio stood at 35.9%, and operating expenses rose just 0.3%. But maintenance and depreciation rose 9.0%, partly because of tariff increases and one-off items, and the bank continues to spend heavily on technology: IT expense reached NIS 1.241 billion, with another NIS 398 million capitalized as additions to assets.

That means the efficiency story is not finished. The Lod campus, automation and branch-footprint optimization are still supposed to deliver future benefits. As long as the efficiency ratio stays around 36%, it is hard to argue that the bank is already in true operating cruise mode.

Cash Flow, Debt and Capital Structure

For a bank, this is not an FCF story

In this case, the right framework is not industrial-style free cash flow. The relevant question is how much flexibility exists on the funding side, how much capital cushion sits against growth, and whether liquidity can absorb a less friendly scenario. So this bank has to be read first through deposits, liquidity, debt structure and capital ratios, and only then through the income line.

What clearly improved

Deposits from the public rose 14.0% to NIS 448.4 billion, faster than loan growth. The deposits-to-loans ratio stood at 112.0% at year-end, up from 109.9% in the fourth quarter of 2024. Core deposits reached NIS 255 billion. At the same time, the securities portfolio jumped 68.1% to NIS 47.9 billion, while cash and deposits with banks stood at NIS 82.8 billion. On liquidity, the LCR was 129% and the NSFR 112%.

That is why flexibility is the right word here. Funding sources look solid, liquidity buffers are high, and the January 2026 issues added more depth. On funding and liquidity, the bank does not look squeezed.

What remains unresolved

The gap sits in CET1. At the end of 2025, the CET1 ratio was 10.24%, against a 9.60% requirement and a 9.80% internal target. Total capital was 13.05% against a 12.5% minimum, and the leverage ratio was 5.88% against 4.5%. So leverage and liquidity are not close to the wall. Common equity is.

That matters especially because the board has already lifted potential payout to as much as 50%, and the 2025 dividend came to NIS 2.686 billion. Once the bank wants to grow, pay out and still maintain a comfortable buffer above a 9.80% internal floor, every quarter of fast business-loan growth automatically becomes a capital question.

That is the single most important distinction in this report. The January 2026 financing layer improves funding diversification, duration and total capital structure. It does not solve the common-equity constraint. So the story is not "the bank issued debt and the issue is gone." It is "the bank bought time and flexibility, and now has to show that profit, pricing and growth can rebuild CET1 room."

Guidance and Forward View

Four non-obvious points need to stay on the screen before looking at 2026:

• Profit grew, but the revenue base softened. The top line is no longer moving forward cleanly, and the gap was filled by lower provisions and tight costs.
• The bank is meeting the ROE target, but not from a wide capital cushion. This is meeting the floor, not operating with comfort.
• The healthier growth is now in business lending. That improves franchise quality and diversification, but usually costs more capital.
• The January 2026 issues improve flexibility, not certainty. They buy time. They do not remove the need to prove the next stage of the story.

That makes 2026 look like a transition year. It is not a reset year, because the bank enters it with very strong operations, high-quality credit, strong deposits and proven access to debt markets. But it is not a clean breakout year either, because the rate tailwind is fading and growth now has to prove itself through fees, business lending and capital discipline, not just through an unusually supportive rate environment.

What has to happen for the read to improve? First, financing revenue from current operations has to stay broadly stable as rates fall. Here the 2025 figure was NIS 11.304 billion, only slightly above NIS 11.236 billion in 2024, so the line is still being held, but only just. Second, business-loan growth has to keep broadening without unpleasant surprises in provisions. Third, the efficiency ratio needs to move toward 35% in practice, otherwise the strategic-plan narrative will keep outrunning the reported numbers. Fourth, CET1 has to rebuild away from the internal floor even if payout remains high.

The point the market could miss is that a bank can look excellent precisely when the next challenge is just beginning. Mizrahi Tefahot currently benefits from high-quality credit, growing deposits, leadership in mortgages and open access to capital markets. But the shift from high rates to falling rates changes the rules. If rates decline, borrowers may look healthier and the book may even remain benign, yet the revenue engine will need to come from a deeper place: volume, pricing, fees and business mix.

That is also where 2025 sends a mixed signal. On one side, business lending expanded nicely, and the bank is moving beyond mortgages. On the other, construction and real estate still account for about 18.5% of public-credit risk, and the bank itself highlights ongoing monitoring of non-linear home-purchase contracts and cancellations. So even if no material stress has shown up there yet, 2026 will be a year in which the market needs to see not just returning demand, but returning demand of acceptable quality.

In the near term, the first headline will likely remain positive: high profit, high payout and smooth access to debt markets. The second headline will be more demanding. The market will want proof that the bank can move through rate cuts without slipping materially in net interest income, without leaning again on unusually low provisioning, and without tightening the common-equity cushion even further. So the next reports will be less of a "is there growth?" test and more of a "what kind of growth, and at what capital cost?" test.

Risks

CET1 remains the key structural risk

The first risk is not liquidity and not refinancing. It is a common-equity cushion that is not wide. An internal floor of 9.80% versus an actual 10.24% leaves limited room for mistakes, especially if the bank wants to keep growing and paying out as much as 50% of profit. This is not a distress setup, but it is clearly a setup in which every acceleration in risk-weighted assets carries a cost.

Mortgages look good, but that does not make the market risk-free

The risk profile of the mortgage book was left unchanged at low-to-medium. Average LTV fell to 54.7% from 55.1%, and deferred payments in housing credit fell to just NIS 11 million out of roughly NIS 238 million of modified loans, versus NIS 115 million out of roughly NIS 5.2 billion at the end of 2024. Those are very good numbers, and they support the argument that the sharper housing stress did not materialize.

But two thoughts have to coexist. On one side, underwriting looks strong. On the other, this is still a NIS 244.2 billion book, and the bank itself continues to flag uncertainty around the long-term effects of the war while monitoring arrears, non-linear contracts and cancellations in construction and real estate.

Business credit quality is also good, but it is being judged after an unusually easy year

Construction and real estate in Israel account for about 18.5% of public-credit risk. Within that sector, about 59.2% of on-balance-sheet credit risk and 71.4% of off-balance-sheet exposure are tied to closed-end project finance, overwhelmingly for residential construction in demand areas. The bank says it has not identified risk realization so far in non-linear contracts or cancellations, and that much of the risk is backed by real-estate collateral, guarantees or insurance.

That clearly supports the positive case. Even so, 2025 was a year in which total credit-loss expense fell to an almost unusually low 0.06%. When the starting point is that benign, the risk is not only a sharp deterioration. It is also normalization. Provisions do not need to blow up to pressure profit. They only need to move back toward more ordinary levels.

Strong liquidity does not eliminate rate sensitivity

The bank describes liquidity risk as low-to-medium, with no breaches, and with LCR at 129% and NSFR at 112%. That is strong. But the risk report also makes clear that a mortgage-heavy balance sheet creates medium- and long-duration assets against shorter-dated funding, and that behavioral models around current accounts, deposits and prepayments are a meaningful part of managing rate sensitivity. So here too, simplicity would be a mistake: very strong liquidity does not mean financing revenue is insulated from falling rates.

Short Interest View

Short interest is very low. At the end of March 2026, short float stood at just 0.11%, with SIR at 0.34, versus a sector average of 0.27% and average SIR of 0.821. Even at the latest peak in early March, short float reached only 0.21%. This is not what a market worried about an imminent credit or funding event looks like.

That does not mean the market loves the stock at any price. It means something narrower and more useful: the debate around Mizrahi Tefahot is currently about earnings quality and capital, not about survival or tail-risk funding stress. That reinforces the view that the key variable in the next reports will be the ability to keep returns high without consuming capital, not a question of liquidity pressure.

---

Conclusions

Mizrahi Tefahot exits 2025 from a position of strength. Loans grew, deposits grew faster, credit quality looks excellent, and debt markets remain open to it. But this is not an end point. It is a transition point. The big test for 2026 is whether the bank can move from an unusually favorable rate tailwind into a cleaner, more durable growth phase without eroding a CET1 cushion that is already not especially wide.

Current thesis: this remains one of the strongest banks in the system, but the earnings engine is shifting from dependence on a favorable environment toward dependence on execution, capital discipline and funding structure.

What changed versus the earlier way of reading the bank? In 2025 the story became less "another record year" and more "how do you hold a record when the tailwind is fading." The strongest counter-thesis says the concern is overstated: credit quality is exceptional, capital is adequate, and the January 2026 issues show the bank can fund itself comfortably even while growing and paying out. That is a serious objection. The problem is that it still does not solve the CET1 question or the profitability question in a lower-rate world.

What could change the market reading in the short to medium term? First, any sign that financing revenue is stabilizing even after the rate cuts. Then, any quarter in which business lending keeps growing without leakage in credit quality or capital. On the other hand, if revenue keeps softening while provisions normalize, the read of 2025 as a "clean peak year" will erode quickly.

The hurdle over the next 2 to 4 quarters is clear. The bank has to show that business loans, fees and expense discipline can fill the gap left by a fading rate tailwind while CET1 rebuilds toward a more comfortable margin. If that happens, 2025 will look in hindsight like a successful transition year. If it does not, it will become clear that too much of the current peak relied on a trough in provisions and on a market backdrop that is no longer here.