Zur Shamir: Direct Finance After the Revenue Growth, Higher Credit Losses, and More Structured Funding

Growth, But Not All of the Same Quality

The main article already established that Direct Finance is the group’s second major engine, but also the part of the story where it is easy to confuse balance-sheet growth with cleaner economics. This continuation isolates that distinction. In 2025 consumer-credit revenue rose 7% to about NIS 1.507 billion, yet total profit before tax fell to about NIS 211.8 million from about NIS 234.5 million a year earlier. There was more business, but less profit per layer of risk and cost.

That is the core point. The increase in revenue did not come only from more recurring spread on a larger book. Roughly NIS 65 million of the uplift came from higher fair-value income on loan portfolios, including about NIS 38 million from the first-time fair-value measurement of mortgage portfolios. Another roughly NIS 37 million came from higher fee income, mainly from higher average commissions on vehicle loans and from a pricing change with vehicle agencies that shifted economics from interest to fixed commissions. Interest and CPI-linked income increased by only about NIS 5 million, even though the average vehicle and mortgage books that Direct Finance kept on balance sheet expanded by about NIS 0.9 billion combined.

That does not make the revenue fake. It does mean the mix changed. A larger part of 2025 growth moved away from plain recurring carry on retained loans and toward valuation gains, portfolio assignments, and distributor economics. That type of income can be highly efficient when institutional markets are open and funding is available, but it is also more exposed to discount rates, transaction pricing, and distributor bargaining power.

The balance sheet already showed that shift in accounting form. Starting in the third quarter of 2025 Direct Finance classified mortgage loan portfolios as held for sale and began measuring them at fair value. That helps explain why part of the revenue improvement looked like operating growth while also reflecting a change in how mortgage economics were measured and monetized.

The bridge makes the point more clearly than the headline. Revenue added more than NIS 100 million, but the benefit was more than absorbed by three frictions: higher credit losses, higher funding costs, and much heavier selling and operating costs needed to keep volume moving.

The Cost of Growth Went Up

The more problematic part of 2025 is that the friction rose exactly where the book was expanding. Credit-loss expense increased to about NIS 260 million from about NIS 224 million in 2024. In vehicle loans, the credit-loss rate rose to about 4.08% from about 3.32%, mainly because recovery rates on defaulted loans deteriorated while the average retained vehicle book still grew by about 5%. In mortgages, the credit-loss rate rose to about 0.19% from about 0.12%, alongside about 30% growth in the average retained mortgage portfolio.

The more concerning signal sits in the fourth quarter. Mortgage credit-loss expense jumped about 120% year over year, and the mortgage loss rate rose to about 0.39% from about 0.21%, alongside a 19% increase in the average mortgage portfolio and higher customer defaults. That is not just a full-year accounting blur. It is a fresher sign that the fastest-growing part of the book is also beginning to show more stress.

The mix issue in vehicle loans matters as well. Direct Finance says that, as part of its model, it assigns by way of sale about half of the vehicle loans it originates, and that the credit-loss rate on assigned loans is materially lower. The implication is that the on-balance-sheet book is not a clean mirror of all origination activity. It is more sensitive to which loans stay on the balance sheet and which are sold away. That makes the loss-rate signal more important, not less.

Funding costs also moved higher, to about NIS 455.9 million from about NIS 435.7 million. The increase came from about 31% higher credit utilization in mortgages and about 7% higher utilization in vehicle lending. At the same time, Direct Finance and the mortgage subsidiary ended the year with variable-rate liabilities exceeding variable-rate assets by about NIS 2.755 billion. That means a falling-rate environment can help, but 2025 still closed with a larger growth engine leaning on funding that had not yet been repriced low enough.

The commercial side became more expensive too. Selling, marketing, G&A, and other expenses rose by about NIS 76 million. Around NIS 25 million of that came from higher commissions, partly because vehicle-loan volumes were higher and average commissions moved up as competition intensified. Add to that more headcount in mortgages, about NIS 8 million of extra advertising tied to rebranding, and another about NIS 7 million linked to mortgage growth in a crowded market. Put simply, Direct Finance bought part of its growth at a higher cost.

That is exactly why the July 2025 efficiency plan matters. Direct Finance’s board approved a focused program in the vehicle-loan segment that cut headcount by about 13% and reduced other expenses. The expected annual saving is about NIS 40 million before tax, but management said the benefit was negligible by the end of 2025 and should be felt in full only from 2026. In other words, 2025 contains the transition cost, not the benefit.

More Funding, But Also More Structured Funding

What happened after year-end matters just as much, because it shows where management is trying to take the model. At the start of January 2026 the mortgage subsidiary amended its credit facility with the Menora group. The utilization period was extended through December 31, 2026, the spread on fixed CPI-linked loans was cut to 1.75% to 2.25% from 2.5% to 3.5%, and the spread on prime-based loans was cut to 0.25% to 0.75% from 0.5% to 0.9%. In addition, NIS 180 million of unused capacity remained available through the end of the facility period, in a linked floating structure that is repriced annually.

That is not cosmetic. If 2025 showed that faster loan growth pushed funding expense up, the first move of 2026 was already aimed at compressing those margins. Together with the late 2025 and early 2026 Bank of Israel rate cuts, that creates a more supportive setup for volume growth to stop feeding almost automatically into higher funding cost.

At the same time, the mortgage unit moved from warehouse-style funding to a more institutional securitization structure. On January 25, 2026 it said it was examining a sale of a roughly NIS 400 million residential mortgage portfolio to a wholly owned issuing vehicle through a true-sale transfer. On February 18 it completed the institutional book-building for Series A with NIS 303.9 million par value and Series B with NIS 64.5 million par value, for total consideration of about NIS 401.1 million. In that same filing management said Direct Finance expected a net profit of about NIS 13.5 million to NIS 17.5 million from completion, with Zur Shamir’s share at about NIS 5.2 million to NIS 6.8 million.

On February 23 the assignment was executed and the bonds were issued, and on February 24 the proceeds were received. The annual report then adds a broader picture: after the 2025 balance-sheet date, Direct Finance assigned loan portfolios totaling about NIS 1.814 billion, including about NIS 399 million of mortgages and about NIS 1.415 billion of vehicle loans.

That is the number that sharpens the transition. In a short period, Direct Finance did not just grow the book. It also moved out a volume equivalent to almost one-fifth of the loan portfolio that still sat on the balance sheet at the end of 2025. Its growth engine is becoming more dependent on assignment pricing, institutional market access, and the capital structure of each transaction.

The other side matters too. The mortgage securitization was not just a clean sale for cash. The annual report describes a rated senior layer of about NIS 368 million par value, secondary layers of about NIS 22 million par value for the mortgage company, an additional junior layer, and more financing from borrower groups. Direct Finance gained a more efficient funding path and better capacity, but at the cost of greater dependence on institutional-market pricing and a more complex funding stack.

What Has to Happen Now

By 2026 the question is no longer whether Direct Finance can grow. It already proved that it can. The real question is whether it can turn that growth into cleaner and less noisy earnings. Four checkpoints matter most:

• Lower funding costs need to show up in the income statement, not only in financing headlines.
• Mortgage credit losses need to move back to a milder range after the fourth-quarter spike, or else growth is outrunning underwriting quality.
• The roughly NIS 40 million efficiency plan has to flow through vehicle-loan operating costs during 2026.
• Revenue mix needs to lean more on recurring spread and less on fair-value marks, newly priced commissions, and transaction gains.

Bottom line: 2025 was a year of real growth, but not a year that cleaned up the economics of growth. Direct Finance expanded volume, pushed mortgages harder, opened a securitization channel, and lowered funding spreads after year-end, but it paid for that with higher credit losses, higher commissions, and heavier operating costs. That is why the right framing for 2026 is not “more growth.” It is whether the new funding structure and the efficiency program are enough to improve earnings quality.