Peninsula 2025: Profit Is Up, but Risk Has Moved Deeper into the Book

Getting To Know The Company

Peninsula is no longer just a check-discounting lender. By the end of 2025 it had become a broader non-bank credit platform with three engines: business credit, urban-renewal project finance, and a guarantees activity that expanded very quickly during the year. A surface-level read looks fairly clean: revenue rose to NIS 224.9 million, net profit climbed to NIS 71.9 million, equity reached NIS 559.7 million, and the company still maintained an aggressive payout policy.

But that is only part of the story. What is working now is the revenue engine, the ability to expand the book, and some improvement in the funding mix after the Series D bond issuance. The active bottleneck sits somewhere else: credit quality inside the book, especially around real-estate-linked borrowers and specific files, while the company is also moving into a longer-duration credit mix that needs more funding and leaves less room for mistakes. That matters now because the fourth quarter showed that the issue is no longer sitting only at the edge of the book. Credit-loss expenses jumped to NIS 11.6 million in one quarter alone, 44% of the full-year total.

There is another point a casual reader can miss. The credit book that looked strong at December year-end had already fallen to NIS 1.704 billion near the report approval date, versus NIS 1.793 billion at year-end. That does not automatically signal weakness, but it does mean the headline closing number looks stronger than the run-rate entering 2026. The discussion therefore shifts from growth to the terms of growth.

At the market layer there is an interesting disconnect. Market cap is around NIS 631 million, slightly above book value, and short interest is almost non-existent at 0.06% of float at the end of March 2026. On the other hand, trading is extremely thin, and turnover on April 3, 2026 was only about NIS 3.7 thousand. That means the lack of short interest is not real confirmation of the thesis. It mostly reflects the fact that the stock is not an easy venue for expressing a bearish view.

To understand 2025, four non-obvious findings matter first:

1. Revenue and profit improved, but risk moved inward. Exposure to financing companies fell to 9.6% of credit extended, versus 15.6% in 2023 and 47.6% in 2021, yet 58.1% of the business-credit book is still tied to developers, real estate, and contractors.
2. Project finance has not yet proven economics comparable to the legacy core. The segment accounted for 12.7% of segment assets, but only 5.6% of segment operating profit.
3. The cash story is much weaker than the accounting story. Cash flow from operations was negative NIS 173.7 million, and year-end cash fell to NIS 1.3 million.
4. The issue is not immediate covenant pressure. The issue is recurring dependence on funding markets if Peninsula wants to keep growing, keep paying dividends, and still absorb credit volatility.

Peninsula's economic map at the end of 2025 looked like this:

Peninsula is entering 2026 as a lender with real earnings power, but also with a book that is more complex, longer-duration, and more sensitive to underwriting quality than in the past. This is no longer just a spread story. It is an underwriting-discipline story.

Events And Triggers

The first trigger: the Series D bond issuance in June 2025 changed the liability structure. The company raised NIS 200 million at a fixed 5.4% rate, with first principal repayment only from September 2027, and later swapped that exposure into floating rate. This is important because it lengthens liabilities against a book that is itself becoming longer-duration, and it reduces relative dependence on commercial paper.

The second trigger: project finance and guarantees moved during 2025 from option-like activities into meaningful business lines. The project-finance book reached NIS 227.3 million net, while guarantees exposure jumped to NIS 294 million from just NIS 45 million in 2024. That creates a route beyond traditional discounting, but it also pulls the company deeper into the risk cycle of developers and urban renewal.

The third trigger: the fourth quarter showed a sharp spike in credit-loss expenses. The company links that increase mainly to specific clients in the fourth quarter, real-estate risk, and a higher group allowance. That is probably the single most important trigger for how the market will read 2026: was this a one-off clean-up, or the start of a structurally higher cost of risk.

The fourth trigger: controlling shareholder Meitav is examining the feasibility of entering banking through a small-bank license and is also examining whether Peninsula could be part of that move. This is not an operating thesis yet and should not be the base case, but it is an important strategic signal. The group is clearly looking for a better long-term funding architecture.

The fifth trigger: in December 2025 Maalot reaffirmed the ilA / ilA-1 rating and revised the outlook from negative to stable. That does not erase the rise in credit risk, but it does suggest that, from the funding side, there is no immediate story of structural deterioration.

This chart is the core of the year. Until the fourth quarter, the story still looked controlled. Once provisions jumped, profitability remained positive, but it started to look different. That is why 2026 will be judged first through the direction of provisions, not through book-growth alone.

Efficiency, Profitability, And Competition

The central story of 2025 is healthy revenue growth alongside some deterioration in earnings quality. Finance income rose to NIS 224.9 million, up 23.3%, and net finance income rose to NIS 157.9 million, up 28.0%. That is strong. Profit before tax rose to NIS 93.6 million, and net profit was up about 40.3%.

But the quality of that profit is less clean than it looks at first glance.

What Actually Drove Profit

Most of the profit still comes from the legacy business-credit segment. That segment generated NIS 88.4 million of operating profit on NIS 203.7 million of revenue. Project finance generated only NIS 5.2 million of operating profit on NIS 21.2 million of revenue. That means segment operating margin was about 43.4% in business credit versus only about 24.7% in project finance.

This is a key point. If Peninsula keeps steering capital and funding toward project finance, it needs either better returns from that business over time or clearly better risk quality that justifies a lower accounting yield. As of year-end it had not fully proven either.

Funding Spread Improved, But That Is Not The Whole Story

Average bank funding cost fell to 5.5% from 6.13% in 2024. That helps, and the bond issuance adds some flexibility. The interest-rate sensitivity table also shows that a 10% change in the Bank of Israel rate affects profit by only about NIS 1.6 million, so the company is less exposed to a simple rate shock than one might assume.

But that must be set against book quality. In business credit, hard-collateral loans increased to NIS 746 million, yet their share of the segment fell to 40% from 45% in 2024. At the same time, long-term loans jumped to NIS 324.3 million from just NIS 80.0 million a year earlier. So Peninsula is presenting a book that sounds safer, yet in practice it is also becoming longer-duration and more dependent on collateral realization and real-estate stability.

Competition Is Not Only About Price, But About Underwriting Standards

The company describes very intense competition, especially in project finance, including a risk of a race to the bottom. That matters because in credit markets competitive pressure does not always show up immediately in lower revenue. Sometimes it shows up through looser underwriting, tighter collateral ratios, or borderline projects being accepted in order to preserve growth.

That sign is already visible. Four new projects in the project-finance portfolio moved into default during 2025, totaling about NIS 31 million. In addition, NIS 48.5 million of the project-finance book is already classified as default, another NIS 23.9 million is classified as significant increase in credit risk, and only NIS 78.9 million sits in the over-35% absorption-capacity bucket.

This is especially important because it shows the new book has not yet reached a stage where it can be described as clean and seasoned. It is simply still young, concentrated, and being tested in a difficult competitive environment.

Earnings Quality Was Hit Through Provisions

Credit-loss expense rose to NIS 26.2 million from NIS 21.8 million in 2024, an increase of 20.1%. That is slower than revenue growth and therefore manageable at first glance. But again, that is incomplete. The fourth quarter alone accounted for NIS 11.6 million of credit-loss expense, up 70.6% year over year. On a full-year basis, the credit-loss ratio was 1.56%, while the fourth quarter alone ran at 2.63%.

That is the key test for the coming year. If the fourth quarter was a one-off clean-up, Peninsula can still benefit from better funding and larger scale. If it signaled that the move into real estate, urban renewal, and longer-duration books is structurally raising underwriting pressure, the market will give less credit to net profit and more weight to the cost of risk.

Cash Flow, Debt, And Capital Structure

Cash Flow

This is where the biggest yellow flag of the year sits. Cash flow from operations was negative NIS 173.7 million, versus negative NIS 330.4 million in 2024. The company is right that negative operating cash flow does not automatically mean a liquidity problem for a lender, because loan growth uses cash upfront. Still, the practical point should not be softened: Peninsula grew on external funding, not on cash accumulation.

It is important to separate two cash lenses here. In a normalized view of the core earning engine, the NIS 71.9 million of net profit still suggests there is a solid underlying business. But in an all-in cash flexibility view, meaning how much cash remains after actual cash uses, the picture is far tighter: negative NIS 173.7 million of operating cash flow, less NIS 1.7 million of investment, less NIS 2.5 million of lease-principal repayment, and less NIS 43.0 million of dividends. That is roughly negative NIS 220.9 million.

So the question is not whether the business makes money. It does. The question is who funds the gap between profit, book growth, and distributions.

Debt And Covenants

On the positive side, the funding structure improved. Commercial paper fell to NIS 200 million from NIS 300 million in 2024, bonds appeared for the first time at almost NIS 199 million on the balance sheet, and the company still has broad bank lines. In the regular business it has NIS 1.345 billion of bank lines, of which NIS 778 million was utilized at year-end and NIS 627 million near report approval. In project finance it has three NIS 100 million lines, with NIS 57 million utilized in each line at year-end.

On the other hand, most of the bank debt remains short-term. Fifty-nine percent of the bank loans are on-call and 41% are credit lines. That is manageable as long as banks remain comfortable, the rating is stable, and covenant room stays wide. As of year-end, covenant headroom looked comfortable: adjusted equity-to-balance-sheet ratio of 30.47% versus a 15% minimum, equity of NIS 559.7 million versus a NIS 275 million minimum, and single-borrower exposure of 3.57% versus a 10% cap. In the project-finance subsidiary, tangible equity ratio stood at 26.3% versus a 20% minimum.

The conclusion is that there is no immediate covenant story here. The issue is not tomorrow-morning survivability. The issue is next-year flexibility. If credit costs remain elevated, and if Peninsula wants both to keep growing and to keep distributing around 60% of earnings, it will rely on banks and capital markets more than on cash already in hand.

Dividend Versus Flexibility

Peninsula paid NIS 43.0 million of dividends in 2025 and already approved another NIS 9.8 million distribution in March 2026. Its policy is to distribute at least 60% of net profit, subject to limitations. That matters because it shows a degree of confidence from management and the board.

But this point cuts both ways. A payout ratio of about 59.7% on net profit can make sense for a stable lender, yet it is less comfortable when year-end cash is only NIS 1.3 million, the book is becoming longer-duration, and credit provisions spiked precisely at year-end. In other words, the dividend strengthens the confidence narrative, but weakens the flexibility buffer.

Outlook

2026 looks like a proof year. Not a breakout year, and not a calm stabilization year. Peninsula has several engines that are working, but each of them still needs another layer of confirmation before the market is likely to assign a cleaner earnings multiple.

1. Was The Fourth Quarter An Exception Or A New Level

This is the number-one test. The company explains that higher provisions came from specific clients, risk in the real-estate sector, and a higher general allowance. It also acknowledges that real-estate risk has already materialized, mainly in business credit and to a lesser extent in project finance. That wording matters. This is no longer a theoretical risk discussion. It is already passing through the numbers.

What needs to happen next? Over the coming quarters the market will first look for a lower cost of risk than in the fourth quarter, and no further deterioration in real-estate-linked and project-finance exposures. If that happens, 2025 may look like a year of absorption and clean-up. If not, 2025 may later look like the year in which the credit cycle changed.

2. Can Project Finance Become A Value Engine Rather Than Just A Volume Engine

The Value portfolio acquisition and the build-out of the activity drove fast growth. That matters because it expands Peninsula's operating space and gives it exposure to a relatively active niche. But by the end of 2025 the book was still highly concentrated: the ten largest exposures represented 87.4% of the project-finance book, and the largest exposure stood at NIS 66.8 million. On top of that, a meaningful part of the book already sits in higher-risk buckets.

So the question is not whether the activity can grow. It already has. The question is whether it can generate profitability, diversification, and underwriting quality that justify that growth. For the market, that is a very clear checkpoint for the next 2 to 4 quarters.

3. Will Guarantees Start Contributing Profit Rather Than Only Exposure

This may be the most interesting underappreciated story in the file set. Guarantees exposure jumped to NIS 294 million, while the equity-to-exposure ratio stands at 26.8% versus a regulatory-style minimum of 4%. In other words, the company is operating here with a relatively thick capital cushion. If this activity keeps growing and can convert exposure into revenue efficiently, it could become a strong return engine with moderate capital intensity.

For now it is still an emerging engine. Revenue from guarantees was only NIS 5.7 million in 2025. That means 2026 will measure not only exposure growth, but also conversion into revenue and profit.

4. What The Controlling Shareholder Does With The Banking Option

The examination of a possible move into small-bank licensing is not an operating fact yet, but it does signal that the group sees funding cost and funding architecture as strategic levers. If 2026 brings a real update on that front, the market may read it as structural upside. If not, it will remain background noise. For now it should be treated as optionality only.

5. What Happens To The Large Specific File

One of the most sensitive issues sits in a specific business-credit file that went through a settlement. Outstanding balance at the end of 2025 was NIS 38.6 million after a partial NIS 35 million repayment funded by an additional financer. At first glance that looks like de-risking, but the structure is more complicated. The new financer received a first-rank pari passu mortgage on the collateral, realization proceeds prioritize repayment of the new financing until September 2026, and Peninsula undertook to repay that new financing if it is not repaid within 18 months from the start of enforcement, unless extended.

That does not mean there must be a loss here. The company itself expects full recovery based on the collateral. But it does mean the risk did not disappear. It changed shape. If this file closes well, it becomes a relief point. If not, it can remain a significant source of noise in 2026.

Risks

Real-Estate Concentration Runs Deeper Than It First Appears

Lower exposure to financing companies is good news, but it is not the same as true diversification. In business credit, 35.1% of the book is directed to developers and real estate, and another 23.0% to contractors and subcontractors. On top of that, the entire project-finance segment sits in urban renewal. Peninsula is clearly more diversified than it used to be, but a large part of its economic system is still tied to the health of the real-estate market.

A Longer Book Is Also Less Forgiving

The move into longer-term loans and project finance is supposed to deepen client relationships and create a more substantial lending platform. That is the positive side. The other side is that exposure duration has increased, rapid turnover is lower, and risk assessment depends more on market scenarios, collateral realization, and the developer's ability to complete projects. For a lender, this is not just a growth issue. It is a resilience issue.

Reliance On Funding Markets Remains Part Of The Model

Peninsula shows broad credit lines, comfortable covenants, and a stable rating. That matters. Still, year-end cash is only NIS 1.3 million against more than NIS 1 billion of bank debt, commercial paper, and bonds. The model therefore depends on continuous line availability and on funders remaining willing to provide liquidity. That is a structural risk even if it does not look urgent today.

Project-Finance Concentration Risk

The ten largest exposures account for 87.4% of the book. The largest project exposure stands at NIS 66.8 million, and average actual credit per borrower is NIS 15.1 million. For a company of Peninsula's size, that means any slippage in a small number of projects can affect quarterly results in a meaningful way.

Low Liquidity Distorts The Market Read

The low short-interest figure may look like a positive signal, but in a thinly traded stock it means less. That makes liquidity not just a technical detail, but a real limit on how efficiently price can absorb new information and how easily the market can express a bearish or bullish read.

Conclusions

Peninsula ends 2025 as a profitable lender, more diversified and better funded than it was a year earlier. That is the strong side of the story. The other side is that the company has deliberately moved deeper into areas where the cost of mistakes is higher: real estate, project finance, longer-duration credit, and activities that need more capital and more funding support. So 2025 looks good on profit, but less clean on earnings quality and financial flexibility.

In the short to medium term, market interpretation will likely hinge on three points: whether credit-loss expense comes back down toward a more normal level, whether the project-finance book proves stability rather than only growth, and whether the company can keep paying dividends without tightening its own funding flexibility.

Current thesis: Peninsula has successfully built a broader credit platform, but the core question has shifted from whether it can grow to whether it can grow without eroding book quality.

What changed relative to the older Peninsula story? The narrative used to be mainly about moving from classic discounting into a broader credit model. In 2025 it became a test of underwriting quality inside that new model, especially around real estate and project finance.

Counter-thesis: the market may be overreacting. Funding cost came down, covenant headroom is wide, exposure to financing companies has been cut sharply, and the fourth-quarter provision spike may prove to be a one-off clean-up rather than a new risk level. If so, 2025 was a successful transition year rather than the start of deterioration.

What could change the market read in the short to medium term? A sharp improvement in credit-loss expense, stability in real-estate-linked exposures, and proof that guarantees and project finance can increase revenue without materially increasing noise in the book.

Why does this matter? Because Peninsula is no longer being judged only as a small lender with an attractive spread. It is now being judged as an expanding risk platform. The difference between good profitability and good business quality will be decided over the next 2 to 4 quarters.

What must happen for the thesis to strengthen? Cost of risk has to normalize, funding sources need to keep diversifying, and project finance plus guarantees need to contribute more clearly to profit. What would weaken it? More slippage in real-estate-linked exposures, deterioration in the large settled file, or growth that continues to lean mainly on leverage rather than on book quality.