Elbit Systems 2025: Peak demand is showing up in profit, not yet in cash

Getting to Know the Company

Elbit Systems no longer needs to prove that demand exists. 2025 settled that question with $7.94 billion of revenue, 16.3% growth and a $28.1 billion backlog. But anyone reading the year only through that headline is missing the core point: Elbit is looking more and more like a heavy industrial defense platform, not a light-capital technology story. About 92% of revenue came from product sales and only about 8% from services. That is why the real question is no longer whether there is a market, but how much capital, inventory, capacity and execution discipline are required to turn this demand wave into profit and clean cash.

What is working right now is clear. Operating margin rose to 8.5% from 7.2%, net income attributable to shareholders jumped to $534.3 million from $321.1 million, and backlog climbed by 24.4%. What is still not clean is that this expansion continues to consume working capital, and the strong operating cash flow still sits on top of customer advances, equity funding and broader use of operating-finance tools. That is the active bottleneck: converting backlog into deliveries and cash without layering on more balance-sheet support.

This matters now because Elbit already trades at roughly NIS 129.3 billion of market value, with daily turnover above NIS 107 million. This is a large, liquid local stock, so the main actionability constraint is not trading liquidity. Short interest is also not signaling deep structural skepticism. The latest short float was only 0.05% versus a sector average of 0.84%, with SIR at just 0.17. In other words, the market is not arguing about the existence of demand. It is testing the quality of conversion.

There is one more point to keep in mind. In May 2025 the company issued 1.57 million shares at $375 per share and raised about $573 million net. At the same time it kept expanding capacity in Israel, the U.S. and Europe, with the new UAV assembly site in Israel already operational, the new munitions site in Ramat Beka partially operational, and new facilities being established in Sweden and Germany. This is not a footnote. It is evidence that the very large backlog requires both a wider industrial base and a thicker capital layer.

The four non-obvious findings from 2025 are these:

• Finding one: The profit step-up was not broad-based across the whole group. Land and ESA together generated about 86% of the growth in segment operating income, so the real improvement is concentrated in specific engines rather than evenly spread across the portfolio.
• Finding two: The huge backlog improves visibility, but it also lengthens risk duration. About 46% of backlog is scheduled for 2028 and beyond, while cumulative catch-up estimate revisions cut net income by about $87.5 million, or $1.86 per diluted share, in 2025.
• Finding three: Strong operating cash flow does not mean the cash issue is solved. The $778 million of operating cash flow was supported in part by a $651 million increase in customer advances, while inventories rose by $358 million and receivables plus contract assets rose by $660 million.
• Finding four: The equity raise, hedge book and very large guarantees are part of the production machine, not cosmetic finance. At year-end Elbit carried $5.245 billion of guarantees, $4.601 billion of purchase commitments and $5.629 billion of notional FX forwards.

The economic map of Elbit in 2025 looks like this:

Two more layers matter. The first is customer concentration. IMOD accounted for 28% of revenue, and the U.S. government for 15%. On a broader basis, sales to government customers in Israel, the U.S. and other countries made up roughly 93.4% of revenue, so Elbit is fundamentally a story about government budgets, military programs, foreign military financing and state procurement priorities. The second layer is labor footprint. Elbit ended 2025 with 20,537 employees, including 14,805 in Israel and 3,267 in the U.S. Revenue per employee rose to about $386.6 thousand from about $346.4 thousand, meaning growth outpaced headcount. That is a positive efficiency signal, but also evidence that system load rose quickly.

Events and Triggers

The first trigger is demand environment. The wars in the Middle East and Ukraine changed defense procurement priorities toward ammunition, land systems, EW, C4ISR and broader multi-domain capabilities. That helps explain why Israel rose to 32.2% of 2025 revenue, but just as importantly why Europe still accounted for 27% and North America 20.9%. Elbit is not a purely domestic story. It is benefiting from a broader budget cycle across Israel, Europe and the U.S.

The second trigger is the quality of wins, not just the quantity. During 2025 the company received, among other awards, a roughly $1.6 billion contract for a European country, a $2.3 billion international strategic solution contract, aggregate contracts of about $260 million for advanced airborne munitions for IMOD, and about $210 million of tank upgrade contracts. The point is not to recite press releases. The point is that backlog is built on both very large, multi-year programs and shorter-cycle domestic demand. That improves visibility, but it also raises dependence on timing, cost estimates, subcontractors and revenue recognition.

The third trigger is localization. Customers increasingly want more than a product. They want local manufacturing, local partners and sometimes technology transfer or local production lines. The EU in 2025 adopted SAFE, a EUR 150 billion loan framework aimed specifically at investing in European defense companies. That can help Elbit because the company already has a U.S. and European footprint and is still adding facilities in Germany and Sweden. But the same trend also raises capex, operating complexity, offset costs and the risk of delays. That is the duality of 2026 and beyond: what expands the market also raises the execution burden.

The fourth trigger is capacity proof. The new UAV assembly site in Israel is already operational, the new Ramat Beka munitions site is partially operational, and a new U.S. facility for structural aircraft operations was built during the year. If these additions move smoothly from capacity build to actual delivery, they support growth and protect margin. If they introduce bottlenecks or cost slippage, a larger backlog simply pushes the issue forward in time.

The key point is that none of these triggers is one-directional. Higher defense budgets are good news only if the company can expand production, recruit people, hold inventory and absorb localization demands without giving up margin or cash conversion. That is why the near-term trigger is no longer the next headline contract award by itself, but whether those awards keep flowing through margins and cash in the next reporting cycles.

Efficiency, Profitability and Competition

Profitability improved, but not evenly. Revenue rose to $7.94 billion, gross profit rose to $1.94 billion, gross margin improved to 24.4% from 24.0%, and operating margin rose to 8.5% from 7.2%. Net income attributable to shareholders also jumped to $534.3 million from $321.1 million. This is no longer just volume growth. It is a real improvement in business economics.

Who actually drove the step-up

Land is the core of the story. Segment revenue rose 38% to $2.25 billion, while operating income jumped 75% to $263.7 million. Segment operating margin reached 11.4% versus 9.0% a year earlier. This is not just growth. It is profitable growth, driven mainly by stronger ammunition and munitions sales in Israel and Europe.

ESA is the second engine. Revenue rose only 7%, but operating income almost doubled to $122.8 million from $56.2 million, and margin rose to 7.2% from 3.5%. In other words, Elbit did not just sell more in the U.S. It sold better, helped by new night-vision and maritime contracts and a better program mix.

ISTAR and EW also improved, with 16% revenue growth and 34% operating-income growth, mainly in maritime systems, electro-optics and electronic warfare. By contrast, C4I and Cyber delivered 16% revenue growth but a $6.1 million decline in operating income, with margin falling to 6.0% from 7.8%. Aerospace was broadly stable, with small revenue growth and essentially flat margin.

The implication is straightforward: not every segment is moving at the same speed. Anyone reading only the consolidated line could conclude that the whole portfolio is entering the demand wave in the same way. That is not accurate. The real economic improvement in 2025 came from the parts of the portfolio where urgency is higher, manufacturing is heavier and backlog can be turned into profit more effectively.

What actually changed inside the margin

There is an important under-the-surface shift. Wages and related benefits rose in dollar terms to about $1.968 billion, but their share of cost of revenue fell to about 32% from about 36%. At the same time, subcontractors and material consumed rose to about $3.417 billion, and their share of cost of revenue rose to about 55% from about 52%. This is the signature of a production year: more output, more munitions, more subcontractors, more material, and a lower relative labor share per dollar of sales.

The good news is that scale is working. Net R&D fell to 6.5% of revenue from 6.8%, marketing and selling fell to 5.0% from 5.5%, and G&A fell to 4.4% from 4.6%. The more complicated point is that this is still an industrial model that depends on material, subcontractors, inventory and delivery schedules. It is less a software story and more a capacity story.

The number in the notes that cools the headline

This is one of the most important data points in the year. Cumulative catch-up adjustments on significant contracts hurt 2025 results. The effect on cost of revenue was a negative $98.8 million, and the effect on net income was a negative $87.5 million, or $1.86 per diluted share. That does not fit a clean, frictionless record-year narrative.

The point is not that backlog is weak. Quite the opposite. The point is that as the company moves deeper into long-duration, more complex programs, estimate sensitivity remains meaningful. The real question for 2026 and beyond is not whether Elbit can win. It is whether it can execute without giving back part of the improvement through cost revisions, delays or tighter competitive terms.

Cash Flow, Debt and Capital Structure

This is where the main friction sits. The $778 million of operating cash flow looks strong in isolation, but its composition matters as much as its size. The company itself breaks it down into a roughly $651 million increase in customer advances and a $172 million increase in non-cash operating items, offset by about $358 million of inventory growth and about $660 million of growth in receivables and contract assets. Demand is generating cash, but it is also pulling hard on working capital.

Working capital, backlog and cash conversion

The balance-sheet numbers sharpen the point. Trade receivables plus contract assets rose to $3.332 billion from $2.943 billion. Inventory rose to $3.130 billion from $2.774 billion. At the same time, gross contract liabilities rose to $3.617 billion from $2.966 billion, leaving net contract liabilities at $868.7 million versus $560.3 million a year earlier. This is a mixed picture. Customers are still funding part of the load through advances, but the company is simultaneously carrying much more inventory and much more revenue recognized ahead of billing.

Another hidden point is that IMOD-related trade receivables and contract assets rose to $1.03 billion from $761.5 million. That is another sign that strong domestic demand in Israel is helping the business, but it is also enlarging the working-capital layer tied to it.

The notes add two more building blocks. Factoring expenses on sold receivables were $27.9 million, and supplier-finance obligations rose to $269.0 million from $241.8 million. This does not point to immediate financial distress. It does show that the machine is actively using operating-finance tools to support the expansion.

Cash framing: all-in cash flexibility

When the thesis is about balance-sheet flexibility, operating cash flow alone is not enough. The right frame here is all-in cash flexibility based on the uses that management explicitly broke out. Operating cash flow was $778 million. Against that sat reported capex of $226 million, dividends of $112 million, and combined repayments of commercial paper, loans and notes of about $479 million. On that disclosed-use basis, 2025 sits around a slightly negative balance before the equity raise.

That means the cash generated by operations still did not fully fund the company’s chosen expansion, distributions and debt reduction on its own. The May 2025 equity raise, which brought in roughly $573 million net, is not a side detail. It is part of how Elbit chose to fortify the balance sheet while stepping into a larger industrial footprint.

What the balance sheet says after the equity raise

At year-end the company held $635 million of cash and another $181 million of short-term bank deposits. Against that, it had $81 million of bank borrowings, $48 million of commercial paper and $308 million of listed notes. Working capital stood at $1.758 billion, the current ratio at 1.29, and the company was in material compliance with all covenants.

That is a real improvement, but it is not a light balance sheet. Elbit carries $5.245 billion of guarantees, $4.601 billion of purchase commitments, $575.2 million of operating lease liabilities and $116.2 million of actual lease cash payments in 2025. On top of that, the FX hedge book is large, with forwards on euro, sterling and other currencies totaling $5.629 billion notional, and a roughly $239 million unrealized loss sitting in accumulated other comprehensive income at year-end.

The conclusion from this layer is that the balance sheet is stronger today, but it is stronger precisely because the company combined demand, customer advances, equity capital, hedging and operating-finance tools. This is a sophisticated system, not a self-running cash box.

Guidance and Outlook

Four points need to frame 2026.

• Point one: Backlog reached $28.1 billion, about 3.5 times 2025 revenue, but only 54% is scheduled for 2026 and 2027. The remaining 46% sits in 2028 and beyond. There is visibility here, but also a long time window in which estimates, capacity and pricing can move.
• Point two: Land and ESA are the current positive-surprise engines. If they hold margin, the story remains strong. If either weakens, the group will feel it quickly.
• Point three: Negative contract estimate revisions show that execution is still not smooth enough to assume a straight-line conversion of backlog into profit.
• Point four: The balance sheet is less pressured than before, but it is no longer “free”. After the equity raise, the market will want to see stronger internal funding of the next leg of growth.

That leads to the central conclusion: 2026 looks like a conversion-proof year, not a demand-proof year. Demand is already here. What now has to be proven is the ability to translate backlog into deliveries, a stable margin framework and cleaner cash generation without expanding inventory, contract assets and balance-sheet support at the same pace.

What has to happen for the thesis to strengthen

First, Land needs to remain a double-digit-margin engine even while capacity keeps expanding. That matters because the segment already generates about 39% of group operating income. Second, ESA needs to show that the 2025 margin rebound was not just a favorable one-off mix effect, but a more durable base in U.S. programs, night vision and maritime systems. Third, C4I and Cyber needs to stop margin erosion, otherwise the consolidated improvement remains too dependent on only two engines.

Another central test is working capital. If revenue keeps rising in 2026 and 2027 while inventory and receivables plus contract assets begin to grow more slowly, the read on the company improves meaningfully. If instead the next growth leg again shows up through more inventory, more receivables, more supplier finance or more factoring, the concern that Elbit is partly balance-sheet-funding its growth becomes stronger.

What can break the story

There are three near- to medium-term pressure points the market is likely to measure in upcoming quarters. The first is currency. The shekel appreciated by 12.5% against the U.S. dollar in 2025, and the company itself says a significant portion of costs, especially labor and operations, is NIS-denominated. The hedge book is large, but management also says it cannot always fully and cost-effectively hedge all exposures.

The second is localization. SAFE in Europe, domestic-procurement preferences and industrial-participation requirements can expand the opportunity set, but they can also bring investment needs, delays and offset costs. The same force that supports demand can erode part of its economics.

The third is the regulatory overlay in the U.S. The company states that under a January 7, 2026 U.S. Executive Order, the U.S. Secretary of War could seek to limit the company’s ability to pay cash dividends or conduct buybacks if the company is deemed to have underperformed or insufficiently prioritized investment or production speed under U.S. government contracts. This is not generic macro risk. It is a concrete outside signal that operating execution could also become a capital-allocation issue.

Put together, the market will likely continue to value the backlog, but it will increasingly focus on the quality of conversion. A positive surprise will come from strong deliveries with a slower build in working capital. A negative surprise will come if more and more of the backlog keeps appearing as inventory, receivables or estimate revisions that drag on earnings.

Risks

The first risk is customer and sovereign concentration. IMOD accounted for 28% of revenue and the U.S. government for 15%, while government customers overall drove almost the entire business. That gives Elbit demand depth, but it also leaves the company exposed to procurement priorities, FMF, budget timing, policy changes and contract modifications or terminations.

The second risk is execution on long-duration contracts. The company itself describes a world of competitively awarded contracts, pricing proposals submitted before design completion, and dependence on cost estimates, schedules and supply-chain availability. The negative estimate revisions in 2025 show that this is not just risk-factor language. It is already affecting results.

The third risk is FX with only partial protection. A meaningful share of costs is NIS-based, many contracts are in euro and other currencies, and the company explicitly says it cannot always hedge every exposure adequately and economically. A $5.629 billion hedge book is an important shield, but it is also evidence of how large the built-in exposure is.

The fourth risk is localization and offset. Local manufacturing requirements can be both fuel and friction. If Elbit meets them well, it deepens its place in key markets. If it does not, it can face higher cost, delays, contractual penalties or reduced access to future programs.

The fifth risk is that the balance sheet can look stronger than it feels. There is no covenant-breach story here today, and the company remained in material compliance. But the guarantees, purchase commitments, leases, supplier finance and the fact that 2025 still relied on equity capital are reminders that the step-up in scale is not costless.

There is also one risk the market may underappreciate in the other direction. As of year-end the company was not involved in significant legal proceedings, and short interest is negligible. That means the center of gravity is not a legal overhang or a hostile short case. It is a much more basic operating and financial question: how cleanly backlog turns into delivery.

Conclusions

Elbit exits 2025 operationally stronger than it looked a year earlier. Demand is broad, backlog is large, Land and ESA are already translating it into profit, and the balance sheet is better fortified after the equity raise. But this is still not a clean cash machine. Working capital, contract estimates and the need to build local capacity across several geographies at once are the tests that will determine whether 2026 looks like a natural continuation of the record year, or more like a bridge year between demand and cleaner economics.

Current thesis: Elbit is benefiting from real structural defense demand, but the heart of the thesis has shifted from backlog existence to the ability to convert backlog into profit and cash without absorbing more capital.

What changed: The debate used to be about demand depth. Now demand is already proven, and the balance sheet has already been reinforced. The focus has moved to working capital, execution and margin quality.

Counter thesis: A reasonable opposing view is that the cash concern is overstated, because the company has already strengthened the balance sheet, still benefits from customer advances, remains in covenant compliance and carries a backlog large enough to absorb temporary working-capital noise.

What could change the market read in the short to medium term: Strong deliveries paired with slower growth in inventory and receivables would strengthen the thesis. More negative estimate revisions, weaker C4I margin or heavier reliance on operating-finance tools would weaken it.

Why this matters: Elbit has already proved it has demand. What defines the quality of the business now is whether it can carry its new scale cleanly.

Over the next 2-4 quarters the thesis strengthens if Land and ESA sustain margin, if inventory and contract-asset growth slows relative to revenue, and if capacity expansion turns into smoother delivery. It weakens if more and more demand keeps showing up as working capital, if the NIS keeps pressuring the cost base beyond what hedging can absorb, or if estimate revisions continue to eat into net income.