Bank Leumi reported its second quarter today, 12 August 2026.
Errors or inaccuracies are possible. Spotted something that looks wrong? Write to me and I will correct it.
Why This Report Matters Today in Particular
Yesterday Bank Hapoalim reported a decline in profit. Its credit grew 14.3%, its balance sheet passed NIS 809 billion - and quarterly profit fell 2.1%, with the half down 7.1%. The explanation was a single line: the financing margin eroded from 2.90% to 2.70%.
Today Leumi reports profit up 8.5%.
And both banks took exactly the same hit. Which makes the comparison between them more instructive than either one alone.
The Quarter
| NIS millions | The quarter | A year ago | Change |
|---|---|---|---|
| Net interest income | 4,572 | 4,540 | +0.7% |
| Credit loss expense | 291 | 223 | +30.5% |
| Non-interest income | 2,016 | 1,446 | +39.4% |
| Operating and other expenses | 1,627 | 1,610 | +1.1% |
| Pre-tax profit | 4,670 | 4,153 | +12.5% |
| Tax provision | 1,970 | 1,623 | +21.4% |
| After-tax profit | 2,700 | 2,530 | |
| Share in associates' profits | 132 | 80 | |
| Net profit | 2,832 | 2,610 | +8.5% |
And for the first half: NIS 5,178 million, against NIS 5,013 million a year ago - up 3.3%.
And the quarterly profit progression makes the picture clear:
| NIS millions | Q1 | Q2 | Q3 | Q4 |
|---|---|---|---|---|
| 2025 | 2,403 | 2,610 | 2,700 | 2,549 |
| 2026 | 2,346 | 2,832 |
The second quarter of 2026 is the highest in the table.
What Actually Happened Here
Net interest income rose 0.7%. Non-interest income rose 39.4%.
That is the entire explanation for the gap against Hapoalim.
Leumi did not escape the margin erosion - its interest income is close to frozen, exactly as at Hapoalim. What it did instead was compensate from another source: fees, non-interest financing income, and investment gains.
And in absolute terms: interest income contributed an extra NIS 32 million against last year. Non-interest income contributed NIS 570 million.
So this quarter's engine was not the classic banking core at all.
And alongside it, operating expenses rose only 1.1% - meaning here too, as at Hapoalim, management held cost discipline.
And Two Lines Pulling the Other Way
The first: the credit loss expense rose 30.5%, to NIS 291 million.
And as a rate: the credit loss expense against the average balance stood at 0.20% in the quarter, against 0.19% a year ago - and 0.17% for the half against 0.12%. So the direction is up.
And problem credit too: the rate of credit not accruing or 90 days or more past due rose to 0.48% from 0.46%, and the net accounting write-off rate rose to 0.10% from 0.08%.
All of these are still very low in absolute terms, and the allowance for credit losses actually fell to 1.20% from 1.37%. But on a book that grew 15.8% in a year, this is the line checked first.
And the second: tax. The tax provision rose 21.4% while pre-tax profit rose 12.5%. Tax grew at close to twice the pace of profit, which is the effect of the levy on the banking system.
The Balance Sheet
| NIS billions | 30.6.2026 | 30.6.2025 | Change |
|---|---|---|---|
| Total assets | 947.3 | 844.3 | +12.2% |
| Cash and deposits with banks | 129.5 | 147.1 | -12.0% |
| Securities | 186.5 | 142.6 | +30.8% |
| Net credit to the public | 566.5 | 489.2 | +15.8% |
| Total liabilities | 876.7 | 778.8 | +12.6% |
| Public deposits | 718.9 | 642.3 | +11.9% |
| Bonds and subordinated notes | 58.1 | 40.5 | +43.3% |
| Equity attributable to shareholders | 70.6 | 65.5 | +7.7% |
Credit to the public grew 15.8% - faster than Hapoalim, which grew 14.3%.
And three movements worth reading together: cash and deposits with banks fell 12%, securities rose 30.8%, and credit rose 15.8%. The bank moved money out of cash and into yielding assets - which is precisely the right response to a falling rate environment.
And on funding: bonds and subordinated notes rose 43.3%, from NIS 40.5 billion to NIS 58.1 billion. The same phenomenon we saw at Hapoalim - when credit grows faster than deposits, the difference is raised in the capital markets.
Additional data: the share price at period end was NIS 66.5, against NIS 62.6 a year ago and NIS 70.2 at the end of 2025. Cumulative dividend for the period: 142.55 agorot per share.
Return on Equity
Return on equity excluding the effect of the tax levy:
| The period | A year ago | |
|---|---|---|
| Quarter | 17.9% | 17.5% |
| Half | 16.5% | 16.9% |
In the quarter the return rose; for the half it fell. The explanation sits in the first quarter of 2026, which was relatively weak - NIS 2,346 million, the lowest in the table.
The Target Revision, and This Is the Part I Consider Most Important
In its 2025 annual report Leumi published strategic targets, which rested among other things on the assumption that the special tax on the banking system would affect 2026 and 2027 similarly to its effect on 2024 and 2025.
Following the enactment of the budget law, that assumption changed - and the targets were revised:
| Previously | Revised | |
|---|---|---|
| ROE 2026 | 14.5% to 16% | 13.75% to 15.25% |
| ROE 2027 | 14.5% to 16% | 15% to 16.5% |
| Net profit 2026 | NIS 10-12 billion for both years | NIS 9-11 billion |
| Net profit 2027 | NIS 10.5-12.5 billion |
So: 2026 was cut, 2027 was raised.
And why that is interesting: the bank is not saying the business has weakened. It is saying tax moves profit from one year to another. The combined two-year target barely changes - it is simply divided differently.
הזווית שלי
דעה אישית של אילן אברמוב - לא ייעוץ ולא המלצה
The comparison between the two large banks, reporting one day apart, is the best lesson I have seen this week.
Both took exactly the same blow: financing margin erosion. At Hapoalim it was measured explicitly - from 2.90% to 2.70%. At Leumi it shows up in the interest income line: plus 0.7%, which is frozen.
And each of them offset it differently.
Hapoalim offset on costs - its efficiency ratio improved from 32.8% to 30.6% in a single quarter, which is a very impressive move. And still its profit fell 2.1%.
Leumi offset on income - plus 39.4% in non-interest income. And its profit rose 8.5%.
And the conclusion I take: in a falling rate environment, cutting costs is a good defence - but a bounded one. You can only save so much. Alternative income, by contrast, can grow without a ceiling. Which is why Leumi finished in a better place, even though Hapoalim did the better job on the cost side.
What I do flag with an asterisk is the quality of that income. A 39.4% jump in a single line does not happen by itself, and the release does not break it down. Non-interest income includes fees, which are recurring and stable, and investment gains, which are not. The difference between the two is the difference between an engine and an event - and that breakdown has to be looked for in the full financial statements, not in the headline.
And what concerns me in measured degree is the credit loss line. Plus 30.5%, a problem credit rate rising from 0.46% to 0.48%, and write-offs rising from 0.08% to 0.10% - all on a book that grew 15.8% in a year. It is still very low, but rapid credit growth always looks excellent in the year it happens.
And what I regard as the most important item in the whole report is actually the target revision. A bank that cuts its 2026 target and raises its 2027 target in the same breath, explaining it by tax legislation, is effectively saying underlying profitability has not changed, only its distribution in time. If that is right, then anyone looking at 2026 alone sees a weakening bank, and anyone looking at the two years together sees a bank standing still. And the difference between those two readings is tax, not business.
