Melran: Credit lines, maturities, and what is really left of funding flexibility

Where the flexibility actually sits

The main article argued that Melran's 2026 test is no longer just loan-book growth. It is whether the company can keep funding that growth without making profitability dependent on one more round of debt expansion. This follow-up isolates the financing question itself: how much of Melran's flexibility is genuinely hard, and how much exists only as long as banks, institutional lenders, and the bond market continue to cooperate.

That requires separating accounting comfort from real funding freedom. At the end of 2025, Melran showed working capital of about NIS 635 million, equity of about NIS 318 million, and compliance with all financial covenants. On first read, that looks strong. Under an all-in cash-flexibility frame, which is the right one here, the picture is different: operating cash flow was negative NIS 67.7 million, investing cash flow was negative NIS 2.4 million, and year-end cash was only about NIS 1.37 million, plus another NIS 1.26 million of restricted cash.

In other words, Melran's room to maneuver does not sit in free cash on the balance sheet. It sits in access to outside funding. That is not the same thing. Positive working capital says current assets exceed current liabilities. It does not mean the company holds a cash buffer large enough to absorb a temporarily closed market, tighter bank behavior, or another step-up in the price of money.

That is exactly what keeps funding as Melran's active bottleneck. This is not a story of imminent credit stress, but it is a story of a lender that needs three doors to stay open all the time: banks, institutional lenders, and the bond market.

What is committed and what stays open only as long as banks want it

Near the approval date of the annual report, Melran had total credit lines of NIS 617 million. That is a large figure, but it needs to be unpacked. Of that amount, NIS 440 million was committed and NIS 177 million was uncommitted. The difference between those two layers is the heart of the issue, because the company explicitly states that banks may reduce or cancel the uncommitted lines and may also demand repayment under certain conditions, including material deterioration in the business or breach of material obligations.

At year-end 2025, Melran had drawn NIS 287.6 million of the committed lines and NIS 25.6 million of the uncommitted lines. Near the report-approval date, committed-line usage actually fell to NIS 257.4 million, but uncommitted-line usage rose to NIS 81.3 million. That leads to a double read. On one hand, committed headroom improved from roughly NIS 152.4 million to NIS 182.6 million. On the other hand, the softer layer of flexibility shrank, with uncommitted headroom falling from about NIS 151.4 million to NIS 95.7 million.

This chart shows why the NIS 617 million headline can mislead. On paper, Melran still had total room of about NIS 278.3 million near the report-approval date. In practice, about NIS 95.7 million of that room sat in the uncommitted layer. So the flexibility exists, but part of it is conditional by definition.

That matters even more because this is exactly the layer the company itself describes as cancellable or reducible. Even in the additional lines that are not individually material, with total scope of about NIS 227 million, Melran highlights concentration limits, collateral requirements, personal guarantees from the controlling shareholder, and cross-default provisions. That does not make the lines weak. It does mean Melran's funding flexibility is not unconditional flexibility.

The duration mismatch improved, but the bank layer is still very short

Melran did try to address the duration problem during 2025. That is visible both in management's language and in the funding stack itself. The credit book is not long-duration in the sense of classic infrastructure or real-estate finance: the average duration of the total credit portfolio is about 249 days, and 83% of the portfolio is due within one year. At the same time, the company added longer-duration funding, mainly the up-to-NIS 100 million institutional B real-estate line with final maturity on June 15, 2027, the up-to-NIS 100 million real-estate bank line with maximum duration of 2.5 years, and Series Heh and Vav bonds that amortize between 2026 and 2029.

But the problem did not disappear. It was managed better. In the liquidity-risk note, the company shows bank debt with carrying value of about NIS 216.0 million, all falling into the up-to-six-month bucket, and the note explicitly adds that this bank debt is repaid every 7 to 30 days. So even if the asset book is relatively short, the bank layer still turns much faster than the underlying portfolio.

The chart captures two opposite truths. The positive one is that Melran now has a more meaningful layer of institutional funding and bonds that pushes part of the maturity stack into 2027 to 2029. The less comfortable truth is that more than NIS 318 million of funding obligations still falls within the next 12 months even before payables, and that the bank layer itself effectively has to be rolled every few weeks.

This is also where the quality gap between funding sources becomes clear. Series Heh and Vav buy time, but they are relatively expensive, with stated coupons of 8.95% and 7.95%. Institutional B buys time until mid-2027, but it comes with LTV criteria, equity requirements, a rating condition, and a requirement that at least 20% of each approved loan be funded from Melran's own sources. The bank layer is built differently: it is much shorter, more collateral-dependent, and still relies on the continued willingness of banks to roll.

So Melran's end-2025 duration picture is neither fully matched nor plainly broken. It is transitional. The company bought more time, but it has not yet escaped structural dependence on rolling the bank layer.

Covenants are comfortable, but that is not the same as funding freedom

There is very good news in Melran's covenant layer. As of December 31, 2025, equity-to-balance-sheet stood at 28.3%, well above the 17% thresholds that matter for the institutional line and the bond series. Equity stood at NIS 318 million, above minimum levels of NIS 200 million for institutional B, NIS 140 million for Series Vav, and NIS 95 million for Series D. Credit-loss expense stood at 0.84%, versus a 5% ceiling. The rating was ilA-, versus a BBB- minimum in the institutional track.

That matters because it frames the risk correctly. Melran is not currently a covenant-breach story. It is a continuity-of-access story. The question is not whether headline ratios have already deteriorated far enough to trigger acceleration. The question is whether book quality, collateral quality, and rating stability remain strong enough for the system to keep letting the company roll its funding.

That distinction shows up in the detail of the agreements. In Bank Gimel, the company committed to keep deposited checks of at least 125% of loan balances in the pledged account. In the real-estate track with the same bank, each loan is limited to NIS 10 million, to up to 75% financing, and to a minimum 1.1 coverage ratio. Institutional B adds single-loan and weighted LTV tests, acceleration triggers if debt above NIS 15 million is accelerated elsewhere, and a rating-linked repricing risk if the rating falls below BBB+.

So the big numbers are comfortable, but real funding room still passes through a long list of operational, collateral, and rating gates. That is an important distinction. A distant covenant line means there is time. It does not remove the fact that Melran still has to pass through the lenders' gate every day.

Even after the upgrade, the cost of money does not fully work in Melran's favor

Melran's funding position did improve during 2025. In October, Maalot upgraded the issuer and debt back to ilA- with a stable outlook, and Series D moved back to its original spread of Bank of Israel rate plus 3.95%, after the spread had risen to 4.20% in November 2023. But here too, it is important to distinguish between a local improvement and a full solution.

The December 21, 2025 interest report makes that clear. The December 31, 2025 interest payment on Series D was set at 4.31589%, because the interest period still included both the pre-upgrade and post-upgrade parts. So the rating relief is real, but it did not instantly reset the entire funding stack.

And the sensitivity structure is still net-negative. At the end of 2025, variable-rate assets stood at about NIS 192.1 million, while variable-rate liabilities stood at about NIS 283.4 million. That leaves Melran with a net variable-rate liability position of about NIS 91.3 million. The sensitivity test shows that a 1% increase in the Bank of Israel rate would add about NIS 1.9 million to the variable-rate part of the loan book, but would reduce profit by about NIS 2.16 million on bank debt and another NIS 0.67 million on Series D. Net, the effect is still negative.

That is a subtle but critical point. The rating upgrade improved the starting point, and most importantly it prevented the funding story from getting worse. It still has not made the liability side cheap enough to eliminate the spread problem.

What this means for 2026

The practical conclusion is that Melran's funding looks better today than it did a year ago, but it has still not become a self-standing source of strength. The flexibility exists, but it is access-based flexibility rather than cash-based flexibility. There are more lines, more longer-duration debt, covenant cushions that remain wide, and a better rating. On the other side, part of the line structure is still uncommitted, the bank layer is still very short, and funding cost remains high and rate-sensitive.

That is why 2026 is still a funding proof year. For the read on Melran to improve materially, it will not be enough for the book to keep expanding. The company will need to show that committed lines keep displacing reliance on the uncommitted layer, that banks and institutional lenders remain open, that the rating holds, and that the spread between book yield and funding cost stops eroding. Until that happens, Melran's funding strength is real, but still conditional.