Follow-up on Mizrahi Tefahot: how much growth fits inside a 50% payout

Mizrahi Tefahot does not currently need to prove that it earns enough to pay a dividend. It needs to prove that common equity can fund both the payout and the growth in risk-weighted assets. The first quarter gives a sharp test: CET1 stood at 10.17%, RWA reached NIS 351.0 billion, and the surplus above the regulatory requirement was only 0.57 percentage points. The dividend declared on May 18, 2026, NIS 619 million, is worth roughly 0.18 percentage points of CET1 and consumes about one-third of the regulatory surplus that existed at the end of March. The half of quarterly profit retained by the bank supports about NIS 6.1 to 6.5 billion of RWA, almost the same order of magnitude as actual RWA growth in the first quarter. A 50% payout can therefore still fit inside growth, but only under one condition: RWA growth has to stay close to the pace of internal capital generation. If credit keeps growing faster, or if profit weakens, the dividend may remain accounting-feasible while becoming an economic constraint on rebuilding the cushion.

The dividend consumes the cushion before it looks expensive

The main first-quarter article already framed the broader issue: bank profitability remains strong, but the CET1 surplus is narrowing. The test here is narrower. At the end of March, CET1 capital stood at NIS 35.692 billion against RWA of NIS 351.030 billion. The regulatory CET1 requirement is 9.60%, so the bank had about NIS 2.0 billion of common-equity surplus above that requirement.

The dividend declared after the balance-sheet date is not minor relative to that surplus. The board approved a NIS 619 million dividend, 50% of first-quarter net profit, and the amount will be deducted from retained earnings in the second quarter. Mechanically, before second-quarter profit and before any further movement in RWA, the dividend lowers the CET1 ratio from about 10.17% to about 9.99%. The surplus above the requirement falls from about 0.57 percentage points to about 0.39 percentage points.

The bank's sensitivity disclosure makes this concrete: a NIS 100 million change in CET1 moves the ratio by about 0.03 percentage points, and a NIS 1 billion change in RWA also moves it by about 0.03 percentage points. A NIS 619 million dividend is therefore not just "half of profit". It is roughly six capital-sensitivity units, or about NIS 6 billion of RWA pressure in ratio terms.

Half the profit funds roughly one quarter of RWA growth

The cleaner way to test the payout policy is not to ask whether profit covers the dividend. It is to ask how much RWA the retained half of profit can support. First-quarter net profit was NIS 1.238 billion. A 50% payout leaves about NIS 619 million of retained earnings in the bank.

At the 9.60% CET1 requirement, that amount can support about NIS 6.45 billion of new RWA. At a 10.17% CET1 ratio, meaning if the bank wants not only to meet the requirement but also to preserve the end-March ratio, the same amount supports about NIS 6.1 billion of RWA. That is very close to what actually happened: RWA increased by NIS 6.875 billion in the first quarter versus the end of 2025.

The implication is straightforward: a quarter with profit similar to Q1 and a 50% payout almost funds that quarter's RWA growth, but leaves little for widening the cushion. If the bank keeps adding roughly NIS 7 billion of RWA per quarter, the half of profit retained mainly prevents sharper erosion. It does not build a new capital surplus. For the CET1 ratio to rise while the payout remains at 50%, at least one of three things has to happen: higher profit, slower RWA growth, or a credit mix that consumes less capital.

That is the useful extension of the year-end 2025 analysis. Back then, the issue was a tight CET1 surplus relative to growth targets and dividends. The first quarter now provides the number that tests the issue: at the current profit run-rate, a 50% payout is not an unreasonable policy, but it leaves very little room for error if business credit and mortgages keep pushing RWA higher.

Tier 2 adds protection, not distributable common equity

The USD 750 million Tier 2 issuance in January 2026 remains a positive signal for the bank's access to debt markets, as discussed in the analysis of the dollar funding stack. It is visible in the numbers as well: Tier 2 capital rose to NIS 9.999 billion, and total capital stood at NIS 45.691 billion. But the ratio that matters for dividends and growth is not total capital. The total capital ratio stood at 13.02%, above the 12.50% requirement, while CET1 is the component that determines how fast the bank can grow without consuming common equity.

That distinction matters because the bank has no Additional Tier 1 instruments. Tier 1 capital is effectively the same as CET1, so there is no intermediate capital layer softening the tension between retained earnings, dividends, and RWA growth. The full early redemption of USD 600 million of CoCo instruments in April 2026 also shows active management of the lower capital layers, but it does not change the basic dividend-versus-CET1 arithmetic.

The positive read is that the bank still shows good funding access, high liquidity, and profitability that creates raw material for internal capital generation. The cautious read is that it cannot pay a high dividend, grow RWA quickly, and widen the CET1 cushion unless profitability stays strong or growth slows. The first quarter showed that the payout fits the model. It did not show that the model is building a wider margin of safety.

Conclusion

The 50% dividend is not the problem on its own. It becomes the problem if high profit is treated as enough to solve the capital constraint. At the current pace, the half of profit retained by the bank funds roughly one quarter of RWA growth, so the next proof point is simple: upcoming reports need to show whether the CET1 ratio starts moving away from 10%, or remains in a zone where every additional NIS 1 billion of RWA and every NIS 100 million of dividend are already visible in the capital ratio.