Melran Projects 2025: Credit quality improved, but the funding test is only starting

Getting To Know The Company

Melran is not just another non-bank lender. This is a business whose economics are driven by three variables that have to work together: the yield on the loan book, the cost of funding, and the ability to keep credit losses under control. In 2025, two of those three engines improved. The book expanded, provisions fell sharply, and net profit rose to NIS 57.1 million. But the third engine, the price of money, moved the other way. Finance expense climbed to NIS 57.0 million, while year-end cash was almost negligible.

That is exactly the point a superficial reading can miss. Anyone who looks only at the headline profit will see a company that left 2024 behind. Anyone who reads the details will see a more nuanced picture: Stage 3 really did fall sharply, but Stage 2 increased; collateral quality improved, but financial flexibility still relies far more on credit lines and bond-market access than on cash in the bank; and the book kept growing into an environment where the average yield already slipped.

What is working now? The net credit book reached NIS 1.099 billion, the allowance fell to NIS 10.6 million, and the company materially expanded its funding base. What is the live bottleneck? The spread between asset yields and funding costs is no longer especially comfortable, which means each credit mistake and each deterioration in funding terms now hits reported profitability faster.

As of April 3, 2026, the share traded at NIS 14.5, implying a market value of roughly NIS 436 million on about 30.0 million shares. That is not a tiny company, but trading liquidity is extremely weak. The latest turnover was just NIS 2,494, which is a real actionability constraint. Short interest is also not sending any strong warning signal here: on March 27, 2026, Short Float stood at just 0.02%, well below the sector average.

The right way to read Melran in 2025 is as a lender that has entered a new phase. The main question is no longer whether it can grow the book. It already proved that. The real question is whether it can carry a larger, cleaner, better-collateralized book while funding it properly and without leaning every quarter on fresh debt expansion.

The Economic Map

Events And Triggers

First trigger: 2025 was a year of funding expansion, not just loan-book expansion. During the year, the company issued Series Vav bonds with nominal value of NIS 222.981 million, total proceeds of roughly NIS 228.3 million, and a 7.95% coupon. At the same time, Melran opened a new institutional real-estate line of up to NIS 100 million, while Bank C facilities were expanded so that together they now reach up to NIS 200 million. This was not cosmetic. It is the move that allowed the company to keep growing without immediately hitting a liquidity wall.

Second trigger: the market also received an external confirmation signal. In October 2025, S&P Maalot upgraded the issuer and debt back to ilA- with a stable outlook, after the outlook had already been revised to positive in May 2025. The rating improvement did not erase the funding-cost problem, but it did improve the starting point. On floating Series D, the better rating reduced the spread by 25 basis points, and the half-year interest payment made on December 31, 2025 reflected a weighted rate of 4.31589%.

Third trigger: after the balance-sheet date, one of the most important test cases moved into the infrastructure portfolio. An exposure of roughly NIS 49.2 million to an infrastructure client became part of a restructuring process, but the company moved quickly, and on February 5, 2026, a structured framework with NTA was approved. By the time the annual report was approved, around NIS 5.5 million had already been collected and another NIS 1.5 million payment had been approved. That does not close the case, but it changes the nature of the risk. The issue is less whether a repayment source exists and more whether timing, framework execution, and actual collections hold up.

Fourth trigger: the April 2026 shareholders' meeting does not give formal guidance, but it does reveal an internal hurdle. For both the chairman and the active director, the 2026 bonus target is tied to annual net profit of at least NIS 65 million after officer bonuses. That is not a management forecast. It is still a useful governance signal: the 2025 result of NIS 57.1 million is not treated inside the compensation framework as a sufficiently high endpoint.

Fifth trigger: from the second half of 2023 into 2025, the company executed three structural moves whose impact is still running through the numbers: the exit from the US activity, a stronger management bench, and a sharp expansion of longer-duration funding. That matters because 2025 does not stand alone. It looks better partly because Melran entered the year with a cleaner and more focused setup.

Efficiency, Profitability, And Competition

The core 2025 story is the gap between asset growth and spread growth. Finance income rose to NIS 172.2 million, up about 25%, and pre-tax profit jumped to NIS 74.7 million, up about 52%. But finance expense rose 57%, to NIS 57.0 million. That is why net finance income increased only to NIS 115.2 million, or about 13% growth. In other words, business volume remained impressive, but the price of money absorbed a meaningful share of the improvement.

What Actually Drove Profit

The sharpest contributor to the earnings improvement came from credit-loss expense. In 2024, the company recorded NIS 17.0 million of expected credit-loss expense. In 2025, that number fell to just NIS 4.6 million. That NIS 12.5 million swing is the heart of the profit jump. Put differently, Melran did not just write more credit. It also carried much less bad-credit weight.

But it is important to separate a real improvement from one that still needs proof. The lower provision burden did not come only from ordinary collections. It also reflected write-offs, recoveries, estimate changes, and changes in collateral mix. That improves the 2025 picture, but it does not automatically mean 2026 will look the same.

The Spread Is Under Pressure Even As The Book Expands

The yield on the average net credit book fell to 17.5% from 18.2% in 2024. Management attributes that to continued book expansion, with emphasis on real-estate-backed credit, and to rising competition in the sector. Economically, that makes sense. Real-estate-backed credit is usually safer, but it also usually comes with lower yield. So growing deeper into real estate can improve security while compressing profitability at the same time.

That is why the 2025 read matters. Melran can no longer rely only on balance-sheet growth. To keep improving profitability, it needs to hold both ends together: preserving credit quality while preventing funding costs from rising faster than the yield it earns on new lending.

The income breakdown by channel sharpens the point. Business lending income rose to NIS 66.5 million, real estate development and construction finance rose to NIS 52.0 million, financial service providers stayed almost flat at about NIS 40.9 million, and infrastructure finance rose to NIS 12.8 million. In other words, a large part of the growth came exactly from the channels the company wants to deepen, but not necessarily from the channels that preserve the highest spread.

Credit Quality Improved, But The Risk Did Not Disappear

Stage 3 fell meaningfully. Gross Stage 3 credit dropped from NIS 63.0 million at the end of 2024 to NIS 34.4 million at the end of 2025. That is a substantial improvement. At the same time, Stage 2 rose from NIS 24.9 million to NIS 39.6 million. So the risk did not leave the system. It moved one step earlier.

The balanced reading is this: 2025 was a good cleaning year, but not a year that ended the problem-credit cycle. When Stage 3 declines and Stage 2 rises, the market has to ask whether that is a staging area before full collection, or simply a staging area before Stage 3 rebuilds.

The key outlier here is the infrastructure client. Melran did not classify that case as credit-impaired because the repayment source is tied to government-linked project cash flows and a protected collection mechanism. As of year-end, about NIS 19.7 million of that exposure was kept in Stage 2 and about NIS 29.5 million in Stage 1. That is an analytically assertive position, but not an unsupported one. There are already real receipts. The 2026 test, then, is not whether the exposure exists but whether the company can keep turning that story into cash.

The Collateral Story Is Stronger Than It First Looks

One of the more interesting facts in the report sits inside the definitions. In 2025, the company tightened its classification of collateral-backed credit: any exposure above 100% of hard collateral value is now classified as not backed by hard collateral. In earlier periods, the threshold had been 125%. That is a tightening, not a relaxation.

And despite that stricter rule, credit backed by hard collateral still rose to NIS 663.9 million from NIS 424.1 million in 2024, while total hard collateral rose to NIS 1.189 billion from NIS 810.9 million. On top of that, all registered liens are now first-ranking, whereas at the end of 2024 the company still had about NIS 19.5 million of collateral in second-ranking liens.

That matters. Anyone looking at lower provisions could immediately suspect pure accounting improvement. Here there is also real structural reinforcement of collateral quality. It does not remove credit risk, but it does widen the margin of safety.

The Fourth Quarter Was Good, But Not As Clean As The Full Year

The fourth quarter helps cool the enthusiasm. Net finance income came in at NIS 28.9 million, down from NIS 30.7 million in the third quarter. Net profit declined to NIS 14.0 million from NIS 16.8 million in the third quarter, while credit-loss expense turned positive again at NIS 2.3 million after two quarters of credit-loss income. So 2025 did not end with clean acceleration. It ended at a respectable pace, but with a reminder that earnings are still sensitive to funding costs and provision behavior.

Cash Flow, Debt, And Capital Structure

The right cash lens here is all-in cash flexibility, not normalized cash generation. The reason is simple. The Melran thesis in 2025 is not mainly about how much cash the existing business can produce before growth spending. It is about how much genuine flexibility remains after the company expands the book, pays dividends, rolls debt, and keeps the whole funding machine moving.

The All-In Cash Picture

Viewed that way, the picture is much tighter than the headline earnings line suggests. Operating cash flow was negative NIS 67.7 million. Investing cash flow was negative NIS 2.4 million. Financing activity provided net NIS 68.3 million, but that net figure hides large underlying moves: NIS 225.4 million of net bond issuance, NIS 70.8 million of long-term institutional borrowing, NIS 170.0 million of bond repayments, NIS 22.6 million of long-term credit repayment, and NIS 32.7 million of dividends.

After all that, the company ended the year with just NIS 1.37 million of cash and another NIS 1.26 million of restricted cash. That is not the cash balance of a company sitting on a comfortable liquidity cushion. It is the cash balance of a lender operating through debt recycling, committed lines, and efficient use of equity. So when people describe Melran as balance-sheet strong, the language needs to be precise: the strength is not in idle cash. It is in continued access to fresh funding, the ability to refinance existing funding, and adequate equity against the growth pace.

The Real Cushion Sits In The Credit Lines

There is also good news here. Around the date of the report, the company had total credit lines of about NIS 617 million. Roughly 70% were committed, and about NIS 390 million could be used on a longer-term basis. The presentation shows around NIS 304 million of unused capacity, including NIS 86 million of short-term availability and NIS 219 million of long-term availability.

That means Melran is not in an immediate funding squeeze. Quite the opposite. It materially expanded its funding base. Total sources made up of bank lines, bond balances, and equity rose from NIS 1.094 billion at the end of 2024 to NIS 1.406 billion at the end of 2025. Inside that, bank facilities rose from NIS 390 million to NIS 515 million, bond balances from NIS 412 million to NIS 439 million, and equity from NIS 292 million to NIS 318 million.

But this still needs the right framing. It is flexibility built on continued confidence from the bond market and the banking system. It is not self-funded flexibility. If market conditions tighten, or if credit quality disappoints, room to maneuver can shrink faster than a year-end balance sheet alone would suggest.

Covenants Are Comfortable, And That Matters

As of year-end 2025, the company does not look remotely close to covenant stress. Equity-to-assets stood at 28.3%, versus minimum thresholds of 15% to 17% across funding documents and trust deeds. Equity stood at NIS 318 million, comfortably above minimums ranging from NIS 140 million to NIS 200 million. Credit losses relative to the book stood at 0.84%, against a 5% ceiling at Institutional Lender B. The top ten clients represented 36.5% of the book against a 50% cap.

That matters because it moves the discussion to the right place. Melran is not currently a covenant-edge story. It is a funding-cost story, a credit-quality story, and a capital-allocation story. The risk is real, but it is not yet sitting on the legal red line.

Dividends Turn Into A Debate About Flexibility Quality

Across 2023 through 2025, the company distributed cumulative dividends of NIS 83.5 million, including NIS 28.5 million for 2025 once the NIS 7.0 million dividend approved in March 2026 is included. The distribution policy approved in December 2024 allows dividends of up to 50% of annual net profit, subject to corporate-law tests and financial covenants.

There is no formal discipline breach here, but there is a real economic question. When a lender ends the year with NIS 1.37 million of cash, negative operating cash flow, and structural reliance on debt markets, every dividend is not just an expression of confidence. It is also a choice to favor current shareholder yield over a wider liquidity cushion. That is not necessarily a mistake. It is simply a choice that has to be read correctly.

Outlook

The 2026 read should begin with four non-obvious findings:

• The biggest earnings gain in 2025 came from provision normalization, not from a clean improvement in funding spread.
• The collateral picture improved even under stricter classification rules, so the quality improvement is not purely accounting.
• Stage 3 fell, but Stage 2 rose. The risk moved earlier in the chain rather than disappearing.
• The company has financial flexibility, but it sits in credit lines and market access, not in cash on hand.

That leads directly to the right label for the next year. 2026 looks like a funding proof year. Not a reset year, because Melran enters it with profit, a restored A- rating, a broader equity base, and a larger book. But not a clean breakout year either, because earnings still have to prove they can hold after the book has already grown and after the era of older cheap funding is mostly behind it.

What Has To Happen For The Read To Improve

First, the funding spread has to stop compressing. In 2025, about NIS 34.3 million of additional finance income translated into only NIS 13.6 million of additional net finance income. If Melran cannot improve the funding mix or stabilize its cost of funds in 2026, it will again need lower provisions to drive net profit. That is a less durable formula.

Second, the Stage 2 book has to come down through collections, not through quiet migration back into Stage 3. The first indicator will be continuing collections from the infrastructure exposure and ongoing execution under the NTA framework. The second indicator will be whether the unsecured business book, still standing at NIS 262.4 million, avoids a new wave of specific provisions.

Third, the funding structure needs to keep improving, not just expanding. The base is already better, but the company still has to extend duration, improve pricing, and maintain adequate room versus the markets. The A- rating helps, but it does not solve everything.

Fourth, capital allocation has to remain disciplined. If the company keeps distributing dividends at a healthy pace while ending each period with almost no cash, the market may start treating shareholder yield as a drain on flexibility rather than as a sign of confidence.

What The Market Is Likely To Watch Next

The near-term market read will probably focus on three axes. First, whether the first half of 2026 shows stability in net finance income rather than only stability in accounting profit. Second, whether Stage 2 starts shrinking through collections. Third, on what terms the next layer of funding arrives. For Melran, funding terms are almost as important as origination pace.

There is also a fourth axis, smaller but still relevant: governance and capital allocation. The NIS 65 million bonus hurdle and continued dividends could sharpen the debate over whether the company is balancing growth, compensation, and flexibility in the right way.

Risks

The first risk is funding risk. Not because a covenant breach looks imminent, but because the business model structurally depends on markets and funding providers. Melran operates in a sector where growth consumes capital and funding capacity. If the fundraising window narrows, or if the price of money rises too much, profitability will get hit before the problem shows up in provisions.

The second risk is that credit quality may currently look cleaner than it will in hindsight. Stage 3 fell in a convincing way, but Stage 2 rose to NIS 39.6 million. Until that amount turns into cash, the cleanup cycle cannot be declared finished.

The third risk sits in portfolio mix. The company keeps increasing its exposure to real estate development, which already represents 29.7% of clients and 33.6% of drawers. That is not unusual for a non-bank lender, but it does mean that the quality story increasingly depends on underwriting quality and the real value of the underlying collateral in the real-estate channel.

The fourth risk is capital-allocation risk. The company ends the year with very little cash, yet continues to distribute dividends and set a high compensation bar for the chairman and active director. If 2026 is good, that will look like justified confidence. If spreads keep tightening, it may later look like unnecessary pressure on the safety cushion.

The fifth risk is market risk. The share is highly illiquid, which means even a good report may not be efficiently priced and a weak one may create exaggerated moves. That reduces the signaling value of short-term price action.

Conclusions

Melran finished 2025 with a strong report, but not with a simple one. What supports the thesis right now is a real improvement in book quality, a broader funding base, and a rating that returned to A-. What prevents the thesis from becoming cleaner is that earnings are still very dependent on the price of funding and on whether Stage 2 actually turns into collections. In the short to medium term, the market will mainly react to whether spreads stabilize and whether the infrastructure case resolves in cash rather than only in narrative.

Current thesis in one line: Melran now looks like a stronger lender operationally and a cleaner underwriter, but not yet like a company that has proven durable profitability without the support of expanding funding and very low provision burden.

What changed versus the older read of the business: Stage 3 fell sharply, collateral quality improved, and the company built a much broader funding base. That shifts the center of gravity from pure credit-cleanup risk to funding cost and Stage 2 execution.

The strongest counter-thesis: the market may still be anchoring on the 2023 to 2024 credit scare even though the company has already cleaned up a meaningful part of the book, returned to an A- rating, holds broad funding lines, and appears to be handling complex collection events better than expected.

What could change the market read in the short to medium term: a real collection trend in Stage 2, especially inside the infrastructure file, stable or slightly improving net finance income, and continued proof that real-estate growth is not coming at the cost of severe yield erosion.

Why this matters: for a non-bank lender, value is not created just because the book grows. It is created when the book grows, remains collectible, and is funded at a price that still leaves real returns for shareholders after repeated debt cycles.

What has to happen over the next 2 to 4 quarters for the thesis to strengthen: Stage 2 should decline through collections, net finance income should stabilize, and financial flexibility should remain intact without a renewed build in provisions. What would weaken the read: more spread compression, slippage in the infrastructure recovery story, or overly aggressive capital distribution.