Paz in 2025: The Operating Engine Improved, But Not All the Value Is Equally Accessible

Getting to Know the Company

Paz at the end of 2025 is no longer a company you read through a refinery. After the spin-off of Baz"a, what remains on the market is a combined platform of transport energy, food retail, Pazgaz, and real estate. It operates without a controlling shareholder, with a market cap of about NIS 8.5 billion, 264 energy sites, 313 food retail points, and 3,941 employees on a full-time-equivalent basis. That matters because the Paz story no longer rests on one engine. The year itself makes that clear: attributable net profit fell to NIS 578 million, but recurring net profit rose to NIS 518 million, free cash flow reached NIS 637 million, and net financial debt fell to NIS 2.273 billion.

What is working now is fairly clear. Transport energy improved margins despite lower fuel prices, Yellow kept growing even as supermarkets became more cautious, Pazgaz improved EBITDA, and the balance sheet strengthened meaningfully with net debt to CAP at just 40% against covenant thresholds of 75% to 77%. What is still not clean? A meaningful part of value sits in self-use real estate that is not recorded at fair value on the balance sheet, part of 2025 earnings still relied on real-estate disposals and revaluations, and in its core fuel business Paz still depends on outside refineries precisely as the new supply model enters its first real year of testing.

That is the active bottleneck in the story. A superficial read may look at lower net profit and conclude that 2025 was weak, or look at NIS 3.151 billion of real-estate fair value and conclude that Paz has a free cushion. Both readings are incomplete. 2025 looks more like a year in which the operating engine got stronger, while the gap between value created and value actually accessible to shareholders remained in place. It simply moved from an internal refining issue to a new proof point: can Paz hold transport margins, keep scaling charging and retail, and fund a generous payout policy without leaning too heavily on optimistic assumptions around real estate and monetization?

It is also worth stating what this is not. This is not a stock driven by weak liquidity or aggressive short pressure. At the end of March, short interest stood at only 0.27% of float with an SIR of 1.56, below the sector average. In other words, the market is not debating immediate financial stress or a “stuck” stock. The debate is higher quality than that: how representative 2025 really is of a durable earnings base, and whether 2026 will show that Paz can operate well without a refinery inside the group while still depending on external refining infrastructure.

This chart matters because it frames 2025 correctly. Paz did not finish the year with business deterioration. It finished with lower reported profit, but with better recurring profit and better free cash flow. If you stop at the headline line, you miss the main point.

Events and Triggers

The central insight from 2025 is that Paz did not just improve operationally. It also changed the kind of questions investors need to ask. It used to be natural to ask what was happening at the refinery. Now the right question is what a retail-and-energy company looks like after the spin-off, when operations are stronger but dependence on outside supply terms and on capital-allocation quality is greater.

Plan 500 was achieved within a year, but 2025 is not a fully clean year

In November 2024, the board approved Plan 500, which was designed to bring recurring annual net profit to NIS 500 million within three years. In practice, that target was reached in the first reporting year. The company attributes the result both to continued growth in food retail, energy, and convenience, and to an efficiency plan that included HQ consolidation and fewer management layers. This was not cosmetic. The decline in G&A to NIS 249 million from NIS 268 million in 2024 is one of the clearer signs that the process was real.

But two thoughts need to be held together. On the one hand, NIS 518 million of recurring net profit is a real operating achievement. On the other hand, 2025 still includes gains from real-estate disposals, revaluations, and other noise. The 2024 comparison was also flattered, especially in food retail, by a NIS 100 million non-compete compensation item and the Pazkar sale. The right thesis, then, is not that Paz has become completely “clean,” but that the base level of the business improved and becomes visible only after you break earnings apart.

2026 is the first year in which the external supply model will really be tested

During 2025, most of the petroleum products sold by Paz were purchased from Baz"a, with the balance mainly from BAZAN. That is still a heavy supply dependence. The year also reminded investors why that dependence matters: in June 2025 BAZAN reported a shutdown after damage at its site during Operation Rising Lion, and in July and August Baz"a reported equipment faults that shut units down for certain periods. As of the reporting date, both suppliers were still filling Paz’s orders, but the events themselves show that this is not a theoretical risk.

That is why the December 11, 2025 immediate report matters. Paz signed agreements with both Baz"a and BAZAN for gasoline, diesel, and jet fuel, so that starting in 2026 the two refineries are expected to become the main suppliers of the transport-energy business. Management says the change is not expected to have a material effect on results. That is an important statement, but it is also a live checkpoint. 2026 will be the first full year in which Paz has to prove that the post-refinery retail-and-marketing model does not erode margins and does not reduce operating flexibility.

Real-estate enhancement is moving from talk to execution, but the cash is not all there yet

Paz continued in 2025 to monetize assets that are not part of its core operating base. In February it completed the sale of the Raem Junction and Be’er Sheva properties and recorded roughly NIS 41 million of net profit on those deals. The Ashdod transaction, at NIS 52 million, is still pending. In September the company launched a sale process for a privately owned 104-dunam site in Haifa Bay, while Paz Shemen is simultaneously working to shrink its plant footprint to roughly 40 dunams and to find an alternative northern dispensing site. In addition, Paz Baka completed the sale of part of its excess rights and recorded about NIS 10 million of capital gain.

The implication is two-sided. On the one hand, management is signaling that it is not just “sitting” on real-estate value, but trying to turn it into action. On the other hand, part of the process is still at the stage of data rooms, non-binding offers, and conditions precedent. So 2025 proved that there is real value here, but did not yet prove that all of that value is liquid or close enough to the common-shareholder layer to be treated as flexible cash.

EV charging and Pazgaz are no longer side stories

EV charging and Pazgaz matter here not because of their 2025 revenue line alone, but because of the direction in which they are pushing the group. In EV charging, Paz sold 28.3 million kWh in 2025, ended the year with 294 ultra-fast chargers, and reached 312 chargers across 112 sites by the reporting date. The average installation cost of a charger stands at roughly NIS 500 thousand, so this is not a free growth engine, but it is already a business with real scale that is beginning to affect the earnings mix.

Pazgaz, for its part, delivered NIS 150 million of EBITDA in 2025 versus NIS 141 million in 2024, and Paz itself says it had signed up tens of thousands of private electricity customers by the reporting date. The Sunray Texas solar project, where the company holds about 25% of equity and 50% of management, contributed another NIS 24 million of EBITDA. This matters because Paz no longer depends only on fuel at the station and coffee at Yellow. It now has more than one energy engine. At the same time, friction remains here too: Pazgaz depends on Katsa"a services for imported LPG, and in February 2026 it received a labor-dispute notice with potential organizational action.

Efficiency, Profitability and Competition

The most interesting part of 2025 is not the decline in net sales. It is the improvement in gross-profit quality and operating efficiency. Sales fell by NIS 776 million to NIS 11.13 billion, mainly because fuel prices were lower and Pazkar had been sold. But gross profit rose by NIS 138 million to NIS 2.596 billion, and operating profit excluding inventory losses reached NIS 853 million. That is not the result of one accounting trick. It is the result of mix, efficiency, and the maturation of operating engines that were not large enough to matter as much two years ago.

Transport energy: less revenue, more margin

This is the segment that is easiest to misread. Transport-energy revenue fell to NIS 7.045 billion from NIS 7.848 billion. If you stop there, you get a picture that is too weak. Gross profit rose to NIS 1.123 billion from NIS 1.042 billion, EBITDA rose to NIS 466 million from NIS 397 million, and operating profit excluding inventory losses rose to NIS 237 million from NIS 168 million. Fuel volumes also increased slightly to 2.731 billion liters from 2.696 billion liters. In other words, 2025 was not a contraction year. It was a quality-change year.

The company itself explains the improvement through three levers: higher EV-charging activity, stronger station-fuel profitability, and a recovery in aviation activity within direct marketing. That matters because the improvement did not come from an unusual volume boom. It came from better gross economics per liter and per visit. In plain terms, Paz made more money from the station and from the customer even without a particularly strong fuel market.

This is exactly where readers can miss the quality of the number. Volumes barely changed, but gross profit and EBITDA jumped. That is a sign that the station, service, and aviation mix worked better.

Still, there is no room for an overly rosy read. The Palestinian Authority was the group’s largest customer in 2025 and a material one, with sales of about NIS 1.4 billion. Paz explicitly says that profitability and volumes from this customer declined in 2025 because of the war’s effects. The agreement was extended through the end of 2027 on similar terms, which is good for continuity but not for risk elimination. On top of that, direct marketing and Pazomat are highly competitive businesses and sensitive to price and payment terms, while the company does not hedge petroleum inventory. So the margin improvement in 2025 is real, but it does not sit in a risk-free business.

Food retail: the headline is weaker than the underlying picture

At first glance, the segment looks weaker. Operating profit fell to NIS 268 million from NIS 339 million. But that is misleading because 2024 included NIS 100 million of non-compete compensation. Excluding that item, Paz itself says segment operating profit increased by roughly NIS 29 million, mainly because of better gross margin and efficiency. That is a critical point. It means the improvement did not depend on a booming food market. It depended on better economics inside a tough market.

The underlying data sharpens the point. Food-retail turnover including franchisees came to NIS 3.276 billion, almost unchanged from NIS 3.284 billion in 2024. Same-store sales fell 0.49%, and monthly sales per square meter edged down to NIS 4.7 thousand. So Paz did not benefit here from volume growth. It earned more because the segment’s gross margin rose to 39% from 36.7%, and because the synergy and efficiency work under Plan 500 started to show through.

This chart tells the entire segment story. Yellow generated NIS 1.171 billion with 3.42% same-store growth and a 34% gross margin. Supermarkets generated more revenue, NIS 2.008 billion, but suffered a 2.83% same-store decline. So the key 2026 question is not whether Paz knows how to run food retail. It does. The question is whether it can keep lifting profit quality while the supermarket format remains under consumer and competitive pressure.

The app and loyalty-data layer is no longer a side note. Identified sales through the app exceeded NIS 3.5 billion, loyalty-club members reached 1.8 million, and identified purchase rate rose to 31%. That means Paz is building a commercial asset that ties together fuel, food, coffee, gas, and electricity. That does not change 2025 on its own, but it does strengthen pricing, targeting, and cross-sell quality.

What supports Paz beyond fuel: Pazgaz and real estate

Real estate and Pazgaz are both important support layers, but they are not the same quality. In real estate, operating profit rose to NIS 242 million from NIS 166 million, and EBITDA rose to NIS 253 million from NIS 190 million. But 2025 also included around NIS 51 million of gains from asset sales, more revaluation gains, and higher rental income. So the real-estate layer is very supportive, but part of the support is less repeatable than the operating business.

In gas and renewables, the picture is cleaner. Operating profit rose to NIS 106 million from NIS 102 million, and EBITDA rose to NIS 174 million from NIS 151 million. In the company presentation, that was split into NIS 150 million from Pazgaz and NIS 24 million from the Sunray Texas project. This matters because the layer is now delivering a real contribution, not just optionality.

The chart makes the main point clear: Paz is now built on four legs. It also shows why investors should not blur together different kinds of profit. Transport and food explain the core operating story. Real estate adds value and support, but with lower repeatability. Pazgaz and renewables add diversification, but still do not replace the core.

Cash Flow, Debt and Capital Structure

This is the section where quick rules of thumb are most dangerous. Paz does not look like a company backed into a financing corner. Far from it. But it is also not right to read 2025 as if every shekel of free cash flow were a free cushion. Cash was generated, but it was also distributed, invested, and partly absorbed by recurring operating commitments.

normalized / maintenance cash generation: the cash engine itself is strong

Within a recurring-cash-generation frame, 2025 was a good year. Cash flow from operations reached NIS 1.279 billion. After neutralizing working-capital changes and the effect of IFRS 16, adjusted operating cash flow reached NIS 862 million, up from NIS 772 million in 2024. After NIS 225 million of capex, free cash flow reached NIS 637 million versus NIS 544 million a year earlier. That is precisely why reported net profit is not the full story. The business itself generated more free cash than in 2024.

The strong side of Paz shows up clearly here: less net debt, more FCF, and an operating business that continues to generate cash at a high pace. This is not the picture of a business living on paper gains alone.

all-in cash flexibility: after leases, capex, and dividends, the cushion is narrower

This is where the frame has to change. If you look at all-in cash flexibility, 2025 started with NIS 1.279 billion of operating cash flow, but it also included NIS 327 million of lease cash and NIS 225 million of capex. That leaves roughly NIS 727 million before debt-principal movements and other financing flows.

That is still a healthy cushion, but not an endless one. During 2025, the board approved dividends of NIS 75 million in March, NIS 130 million in May, NIS 120 million in August, and NIS 120 million in November, or NIS 445 million approved inside the year itself. After year-end, on March 11, 2026, it approved another NIS 200 million dividend for 2025. So the right read is this: Paz knows how to generate cash, but it also chooses to pass a large share of that cash through to shareholders while still investing in charging, stations, food, and asset enhancement.

The distinction matters because it prevents a common analytical mistake: treating FCF as if it were the same as the cash left over after all real cash uses. At Paz, it is not the same thing.

debt and covenants are not the problem in the story right now

The good news is that the balance sheet truly improved. Net financial debt fell to NIS 2.273 billion from NIS 2.588 billion, cash reached NIS 745 million, and gross debt fell to NIS 3.352 billion from NIS 4.246 billion. The liquidity layer also looks better than the working-capital deficit alone suggests: the company has NIS 300 million of committed bank lines through the end of 2026, and management explicitly notes that NIS 303 million of current lease liabilities under IFRS 16 are a major factor behind the working-capital deficit.

More important still is the covenant headroom. Net debt to CAP stood at only 40% at the end of 2025, against thresholds of 75% to 77% in the bond indentures. Equity stood at NIS 3.392 billion against a required floor of NIS 1.9 billion, and NIS 2.3 billion for dividend-related tests. The rating agency also provided external confirmation, with an ilA+ issuer rating and positive outlook, and ilAA- ratings for the bonds. So the Paz discussion is not a “can it refinance?” story. It is a capital-allocation and earnings-quality story built on top of balance-sheet room that already exists.

created value versus accessible value: this is one of the most important points in the filing

The real-estate layer carries real value. Fair value reached NIS 3.151 billion against book value of NIS 2.005 billion, implying a surplus of NIS 1.146 billion. Within that, self-use properties generated NIS 151 million of NOI and third-party yielding assets generated NIS 60 million. But Paz states clearly that it does not record the revaluation of fixed assets in the books, and that the value of self-use stations is based on business value or market rent, under an assumption that station profitability does not decline in the visible future.

That is exactly the difference between value created and value accessible. You can look at NIS 3.151 billion and say there is a strong support layer here, and that is true. But you cannot treat that number as if it were cash in hand. To turn it into something truly available to shareholders, Paz still has to sell, rezone, enhance, obtain permits, and sometimes relocate existing activity. In other words, real estate supports the thesis, but it does not solve it on its own.

Forward View

Before getting into the details, four non-obvious findings need to frame 2026:

• First finding: Paz has already achieved Plan 500 at the recurring-profit level, but 2026 will be the first year in which investors can test that level without hiding behind the distorted 2024 base and while the new supply structure is in force.
• Second finding: Food retail is better than the NIS 268 million operating-profit headline suggests, but the real test has now moved from HQ efficiency to supermarket stabilization.
• Third finding: Paz’s problem right now is not a covenant or rating issue. It is how much of its strong cash generation remains flexible after distributions, leases, and investment needs.
• Fourth finding: The real-estate layer has proved hidden value exists, but has not yet proved that all of that value is accessible on a timeline that matters to shareholders.

That leads to a simple conclusion: 2026 looks more like a proof year than a breakout year. It is not a rescue year, because the balance sheet is far from stressed. It is not a clean harvest year either, because part of the support in 2025 came from revaluations and asset sales. It is a proof year because the three engines that matter most now need to deliver evidence at the same time: transport must hold margin under the new supply model, food must show that efficiency holds up in a cautious consumer market, and real estate must continue creating value without becoming a temporary substitute for the operating business.

The fourth quarter of 2025 already gave a constructive early signal. Gross profit rose to NIS 655 million from NIS 602 million, operating profit excluding inventory losses rose to NIS 201 million from NIS 155 million, and continuing-operations net profit rose to NIS 141 million from NIS 104 million. EBITDA also improved across all four major segments.

The chart does not prove that everything is solved. It does show that Paz exited 2025 with operating tailwinds, not with deterioration.

What has to happen over the next 2 to 4 quarters

Checkpoint one: transport-energy margins need to hold under the new Baz"a and BAZAN agreements. Continuous supply alone will not be enough. Investors need to see that the new commercial terms do not give back the profitability gained in 2025.

Checkpoint two: supermarkets need to stabilize. Yellow is proving that Paz has a strong convenience engine, but the market will not give full credit if supermarkets continue to post negative same-store sales. 2025 showed that profitability can improve without sales growth. 2026 now needs to show that the improvement is durable.

Checkpoint three: the real-estate moves need to progress from relatively simple transactions toward proof that more complex monetization is also possible. Ashdod, Haifa Bay, and continued rights sales will be judged not by theoretical value, but by how quickly they mature into cash.

Checkpoint four: the newer growth engines, EV charging and Pazgaz, need to keep scaling without creating an outsized new friction point. For Pazgaz, that means getting through 2026 without an escalation in labor relations and without operating disruption. For EV charging, it means continuing to justify a meaningful capital base.

What the market may miss in the short to medium term

The market may try to force Paz into a false choice between “old fuel company” and “modern retailer.” In practice, 2025 shows the company is sitting in the middle. Transport still carries the biggest operating weight, but the incremental value is increasingly built through food retail, Pazgaz, charging, and real estate. That means the positive surprise path in 2026 is not just a profit number. It is a combination of three things: holding transport margin, keeping food stable, and continuing to reduce debt despite a generous distribution policy.

The negative surprise path is the mirror image. If the new supplier structure trims margins, if supermarkets remain soft, and if real-estate value continues to look good on paper but slow in cash, then the market can quickly re-read 2025 as a year that was better on paper than in the shareholder’s pocket.

Risks

• Refinery dependence remains the key operating choke point. Even after the new agreements, Baz"a and BAZAN remain the main suppliers of the fuel products Paz buys. 2025 already showed what happens when outages and shutdowns occur.
• The Palestinian Authority is a very large single customer. Roughly NIS 1.4 billion of sales to one customer is both an anchor and a risk, especially when Paz itself says profitability and volumes from that customer declined in 2025.
• In food retail, the issue is growth quality. The market did not grow materially, overall same-store sales were down 0.49%, and consumers became more price-sensitive. Better internal economics matter, but they do not eliminate market toughness.
• Real estate is support, not a free pass. NIS 3.151 billion of fair value looks strong, but a meaningful part sits in self-use assets that are not recorded on the balance sheet and whose valuation depends on stable station economics.
• Older stations carry an environmental cost. 63% of Paz’s public stations were built before 1997, and the estimated cost to remediate one contaminated site ranges between NIS 200 thousand and NIS 650 thousand.
• Paz does not hedge petroleum inventory. In 2025 it recorded NIS 8 million of inventory losses before tax. In a faster price-down year, that noise could be larger.
• Pazgaz adds diversification, but also its own friction layer. Dependence on Katsa"a services, dynamic regulation, and the February 2026 labor-dispute notice all make clear that this engine is not frictionless either.

Taken together, Paz’s risk list now looks less like an existential-risk map and more like a set of frictions that can change the quality of a year. That is an important distinction. The company is not on the edge. But if several of those frictions materialize at once, it is easy to see how 2026 could look less clean than 2025.

---

Conclusions

Paz exits 2025 in better shape than the reported net-profit line alone suggests. The operating engine improved, debt came down, the rating outlook turned positive, and the company is no longer built only around fuel at the station. The main blocker did not disappear. It changed shape. Instead of a question about a refinery inside the group, the question now is how much of the margin and value created will actually reach shareholders through an external supply model, a generous payout policy, and a real-estate layer that is still only partly monetized.

Current thesis: Paz has become a stronger retail-and-energy platform, but 2026 will be judged less by reported profit and more by its ability to convert a stronger operating engine into flexible cash and accessible value.

What changed in the way Paz needs to be read after the spin-off is not the quality of the assets themselves, but the location of the friction. The center of gravity used to be refining. Today it is supply, food-retail quality, and the gap between real estate that looks excellent and cash that actually moves up the stack.

The strongest counter-thesis: the caution here may be overstated. Paz has already delivered NIS 518 million of recurring net profit, NIS 637 million of FCF, a material debt reduction, and very wide covenant headroom. If the new supplier structure works as planned and real estate continues to be monetized, the market may actually be underestimating the strength of the new platform.

What could change the market’s reading in the short to medium term? First, the 2026 transport-energy quarters. Then the quality of the numbers in food retail. And finally, whether real-estate monetization keeps turning value into cash rather than remaining a comfortable promise in the filings.

Why this matters: Paz is an excellent example of a company where the business itself is already better, but investors still need to distinguish carefully between earnings created, value created, and cash that is truly left free.

For the thesis to strengthen over the next 2 to 4 quarters, Paz needs to show four things together: stable transport margins under the new supply structure, continued improvement or at least stabilization in supermarkets, real-estate monetization that produces cash rather than just headlines, and payout discipline that does not erode flexibility. What would weaken the thesis is not one dramatic event, but a combination of slightly weaker transport margin, continued food softness, and slower conversion of real-estate value into accessible cash.