Arit Industries in 2025: The Cash Is In, Now It Has to Prove This Wasn't a One-Off Peak Year

Getting to Know the Company

Arit is no longer a story about a small defense supplier. The listed company sits above Reshef, the real operating engine of the group, and through Reshef it develops, manufactures, and markets electronic fuzes for the defense market in Israel and abroad. In 2025, almost every positive driver lined up at once: a major North American order was fully delivered, the Sderot plant expanded capacity, and the Reshef capital transaction turned the balance sheet into something that looks unusually strong relative to the company's own history.

What is working today, and this matters up front, is execution. Arit proved it can ride a global rearmament wave, add labor, expand production lines, and deliver a 271.3 million shekel project to a North American customer within the same year. That is no longer a promise. It is demonstrated capability.

But the active bottleneck for 2026 is no longer production. It is durability. 2025 looks like a peak year built on a rare combination of a very large one-off order, strong local demand, and an aggressive capital event at Reshef. For this to become a multi-year story rather than an exceptional year, Arit now has to show that India, Europe, and the U.S. move beyond legal entities and signed frameworks into repeat orders with acceptable economics.

The easiest thing to miss is that the listed company itself changed this year. On one hand, Reshef raised capital and Arit sold part of its stake, so meaningful cash came into the group. On the other hand, Arit's ownership in Reshef fell to 77.6% by the report date. In other words, value was created and liquidity improved, but Arit shareholders now own a smaller slice of the main operating asset.

It would be a mistake to look only at the consolidated cash pile and conclude that the risk has disappeared. That is only part of the picture. Some of the improvement sits in mark-to-market gains on securities, some of the future value still depends on converting geographic expansion into real revenue, and part of the India backlog is still contingent on local production requirements. The opposite quick conclusion, that everything was "just war demand," also misses something important: capacity has already been expanded, the customer base is identifiable, and the company now has a balance sheet that can absorb a proof year without financing pressure.

Events and Triggers

The North America order delivered the peak year, but not yet a new base

First trigger: the North America project. In 2025 the group completed deliveries of fuzes totaling NIS 271.3 million to that customer, following an order received in November 2024 worth about NIS 310 million. This was the first time Arit executed a project of that scale for a North American customer, and it was delivered within the year.

The implication cuts both ways. On one hand, it proves the company can work with a large Western customer, on short timelines, and at a scale that moves the entire report. On the other hand, by the report date no additional orders had been received from that customer, and the company explicitly says it is not dependent on it. Anyone trying to treat 2025 as a steady run-rate is ignoring the fact that the largest contributor to the year has not yet repeated.

The Reshef capital cycle dramatically strengthened the balance sheet, but diluted the parent

Second trigger: the partial registration of Reshef shares on the institutional trading platform and the December 2025 package transaction. The deal included a private placement of Reshef shares, a sale of shares held by Arit, and the grant of warrants and put options. Initial proceeds totaled about NIS 600 million, rising to about NIS 900 million once all options were exercised.

By the report date all of those options had already been exercised, so in practice about NIS 450 million went to Arit and about NIS 450 million went to Reshef. That changed the liquidity picture completely, but it also reduced Arit's stake in Reshef to 77.6%. This is the heart of the story. The market now has a better-funded operating asset, but a listed parent that owns less of it.

India, Europe, and the U.S. are now a monetization test, not just a geography story

Third trigger: the international footprint. In the U.S., a wholly owned subsidiary, Flash Fuzes USA, was established in September 2025 to build a fuze manufacturing plant, and the company estimates that the production license should be received in the second quarter of 2026. In India, a subsidiary was established after the balance-sheet date to meet Made in India requirements. In Europe, a commercial agreement was signed in January 2026 with a Western European company to market fuzes to NATO countries and transfer know-how for local production.

All three moves serve the same strategic goal, shifting from Israeli manufacturing plus export to more local market access. But each one also shifts part of the economics to a local partner, to licensing, or to manufacturing outside Israel. In the European case, management even estimates that the model would leave Reshef with about 70% of the selling price on average, which means market access comes with some economic give-up.

The domestic market is still strong, but not free of noise

Fourth trigger: local demand. During 2025 the company received Ministry of Defense orders totaling about NIS 40 million in May and NIS 50 million in August, in addition to further orders from Israeli defense companies. At the same time, the report references the Nagel Committee conclusions and the emphasis on munitions independence and long production continuity.

The counterpoint is that in February 2026 investigators from the Israel Securities Authority searched the offices of the company and Reshef over suspicions of insider-trading offenses. As of the report date, the company says there are no suspicions against the company or Reshef, and it is unaware of further developments. This does not break the operating thesis, but it does add an overhang the market will watch if anything else emerges.

Efficiency, Profitability, and Competition

2025 was a scale year, and that explains almost everything

Group revenue rose to NIS 523.8 million from NIS 126.5 million in 2024. Gross profit jumped to NIS 359.9 million and operating profit to NIS 322.7 million. In the second half alone, revenue reached NIS 368.7 million and operating profit NIS 238.0 million. The company also says that from the second half of 2025 onward, plant capacity was already sufficient to meet the existing order book.

That matters because it shows the step-up did not come from deteriorating quality, inventory pushing, or rushed procurement bypassing the plant. On the contrary, the operating system expanded in time, and the company even says potential capacity now exceeds what is currently required to fulfill known and expected orders.

Margins jumped, but management itself warns against drawing a straight line

Gross margin reached 69%, versus 56% in 2024 and 39% in 2023. According to the report, the improvement came from higher scale, better efficiency, and a more profitable mix. That makes sense, but it also contains a built-in warning: the company itself says its overall gross margin in recent years had generally been around 40% to 50%, and there is no certainty those levels, or 2025's level, will hold in future years.

That means 2025 was not just a growth year. It was also a very unusual mix year. Large production series lift profitability because a meaningful portion of fixed cost is spread over many more units. In plain terms, what looks like a very wide economic moat in 2025 may partly be the result of an exceptional volume year.

Even in a peak year, you can see who paid to get the sales

Selling and marketing expenses rose to NIS 21.1 million from NIS 1.8 million in 2024. The jump came mainly from NIS 18.8 million of sales commissions, primarily tied to the North America order. So the right reading is that the mega-order was highly profitable, but it was not won without a meaningful market-entry cost. That does not weaken the commercial achievement. It just improves the quality of the interpretation.

Competition did not disappear, it just moved to another layer

In Israel, the company presents itself as the only manufacturer of electronic fuzes for the Ministry of Defense, with only potential competitors rather than active ones in its current product range. In India, by the report date, BEL and just one other company are the approved suppliers of electronic artillery fuzes in closed tenders. In the U.S. and Europe, Arit describes only a limited number of manufacturers.

But the real entry barrier is not only technological. It is local. Local manufacturing requirements, local partners, export licenses, and sometimes know-how transfer. So Arit's edge is not just the fuze itself. It is the ability to design a market-entry model that does not destroy the economics along the way.

Labor and the plant explain why 2025 was even possible

Near the report date, the group employed 208 people directly and another 133 through manpower-service companies. Of those, 172 direct employees were in production and quality control. In addition, the company says the plant labor base had risen to about 250 workers, work areas were expanded, production lines were added, and work moved into shifts.

These numbers matter because they show the step-up was not just a paper backlog. It also rested on fast operational organization. But once capacity is already higher than current needs, investors have to move to the next question: will the plant stay full, or will fixed costs start to weigh again if order flow slows down.

Cash Flow, Debt, and Capital Structure

The right cash frame here is all-in cash flexibility

The right way to read 2025 is through all-in cash flexibility, meaning how much cash remained after actual cash uses, rather than through a narrower normalized cash view. On that basis, Arit ended the year in an unusually strong position: NIS 697.7 million of cash and cash equivalents plus NIS 340.8 million of securities at fair value, against only NIS 170.4 million of current liabilities. Working capital stood at roughly NIS 1.04 billion, and the company states explicitly that it did not need, and does not currently need, external financing from financial institutions.

This is not the balance sheet of a company scrambling to finance growth. It is the balance sheet of a company that now has to decide how to use surplus flexibility.

Profit was stronger than cash flow, but not because collections broke down

Cash flow from operations was NIS 204.4 million, below net profit of NIS 346.8 million. At first glance that gap looks sharp, but it is mainly explained by a NIS 95.1 million increase in receivables and a NIS 38.8 million decline in customer advances. In other words, part of the year's profit sat in working capital, largely because sales were very high in the fourth quarter and customer advances were converted into revenue as orders were delivered.

The key point is this: 2025 earnings quality is still strong, but not magically insulated from cash conversion. The company grew very quickly, and that growth pulled working capital along with it. Even after that, more than NIS 200 million of operating cash flow is still a very strong outcome.

Not all of 2025 earnings were operating, and that actually sharpens the thesis

Net finance income reached NIS 55.6 million, versus NIS 5.4 million in 2024. Of that, NIS 49.0 million came from fair-value gains on financial assets, NIS 4.7 million from FX gains, NIS 2.5 million from interest income, and NIS 3.5 million from dividends. Against that, finance expenses were NIS 4.0 million, including NIS 2.7 million of interest on customer advances.

That means the 2025 net-profit line includes a securities-market contribution that may not repeat at the same scale. At the same time, even without it, operating profit alone was already NIS 322.7 million. So 2025 was not a year built on financial engineering, but it was a year in which excess liquidity began to create more volatility in the bottom line.

More value became accessible at the parent level, but it still does not map one-for-one to the consolidated balance sheet

The parent-only financial statements add an important layer. At December 31, 2025, Arit itself had NIS 204.2 million of cash and NIS 340.8 million of marketable securities, against NIS 53.8 million of current taxes and NIS 15.6 million of put-option liability tied to subsidiary shares. So a meaningful part of the liquidity is not fully trapped inside Reshef.

That is a major improvement over a surface reading of the consolidated balance sheet. Still, investors should not confuse higher liquidity with full ownership of Reshef. The parent is more liquid, but it also owns less of the main operating engine. Flexibility improved, but ownership did not stay intact.

Outlook and What Comes Next

First finding: 2025 outperformed the company's own revenue expectation. Actual revenue reached NIS 523.8 million, versus company guidance of about NIS 485.1 million, and management says the gap came from accelerated deliveries after production capacity was expanded. That matters because it suggests part of 2025's performance was also timing, not just underlying demand.

Second finding: the big backlog is not necessarily the firm backlog. At December 31, 2025, backlog stood at NIS 784.8 million including a contingent order. Excluding that contingent component, regular backlog was NIS 617.4 million, sharply below NIS 1.16 billion at the end of 2024. Near the report date, backlog was already down to about NIS 729 million including about NIS 167 million contingent. That is a clear reminder that the book can refill and empty quickly.

Third finding: India is still a long way from being a firm USD 200 million story. Under the multi-year order agreement, Reshef's share can reach up to about USD 200 million, but only USD 141 million represents components included in the regular backlog, and only about USD 39 million is already supported by an operational order and letters of credit for the first two years. Another roughly USD 58 million depends on meeting local-production requirements in India. That is exactly why the company books that portion as contingent.

Fourth finding: Europe is an opportunity with friction. The January 2026 agreement opens a route into NATO markets, but it also requires know-how transfer, export licensing, partial production in Europe, and a shared economics model with a local partner built around cost plus and a base commission. That can become a large market, but not necessarily one with the same peak economics seen in 2025.

Fifth finding: 2026 looks like a proof year, not a clean breakout year. The company expects the U.S. production license in the second quarter of 2026, completion of the electronic loitering-munition fuze in the first half of 2026, and separate battery sales only starting in 2027. In other words, a meaningful part of the future growth case is still not sitting inside the 2025 revenue line.

That leaves the market with four simple questions. Can India move from promise to execution beyond the first two years. Can Europe actually turn into approvals and orders rather than a good-looking agreement. Can the U.S. get licensed and then get ordered. And after the North America year, can the plant land another wave of projects of comparable scale, or does the business drop back to a much lower base.

The right label for 2026 is therefore a proof year. There is balance-sheet strength, there is capacity, there are open doors, and there is even an additional India award worth about USD 4.4 million beyond the ten-year framework. What is still missing is proof that the external growth platforms can produce a new recurring order pattern without another one-off event on the scale of 2025.

Risks

Extreme customer concentration

In 2025, three customers accounted for 100% of reported revenue among customers above the disclosure threshold: the Israel Ministry of Defense at 39%, BEL at 9%, and the North America customer at 52%. That does not mean other customers do not exist. It means they did not reach the reporting threshold. Economically, the implication is straightforward: visibility into 2026 depends first on what those three demand pockets do next.

Project backlog with high timing sensitivity

The company explicitly says that even a delay of a few days in delivering a production series can shift revenue recognition between periods. It lists the drivers itself: failed incoming inspections, shipping delays, dependence on external testing schedules, delays at other points in the chain, and even export embargoes on components.

Dependence on BEL and local-content requirements

In the multi-year India framework, a meaningful portion of the long-term economics depends on Reshef meeting local-production requirements. The company has three possible routes and is currently pursuing the first, through an Indian subsidiary. Any delay in licensing, production setup, or the shift from the first two years into years 3 through 10 effectively pushes out the contingent part of the value.

Margin normalization

Management itself notes that gross margin in recent years had generally been around 40% to 50%, while 2025 reached 69%. That does not mean 2025 margins will collapse, but it does mean the forward comparison base is very demanding.

Regulatory and governance noise

As of the report date there were no material legal proceedings pending against the company, but the February 2026 ISA search will remain in the background until there is full clarity. As long as nothing else develops, this is a governance overhang rather than an economic event. If anything does develop, the market will not ignore it.

FX and securities exposure

At December 31, 2025, the company had a net U.S. dollar asset exposure of NIS 80.2 million, driven mainly by the North America receivable. The company says that receivable was collected and converted into shekels after the reporting date, so part of the FX exposure has already come down. At the same time, the securities portfolio is now large enough to matter to the income statement, and the company itself presents a sensitivity of about NIS 34.1 million for a 10% move in the market value of that portfolio.

Conclusions

Arit entered 2026 from a position of strength. 2025 execution was exceptional, the balance sheet is very strong, and the company has already shown it can expand capacity and deliver large projects. The central constraint is no longer financing or production. It is whether this peak year was the start of a new multi-year earnings path, or a peak that will be hard to repeat without another unusual order.

In one line, the current thesis is this: Arit has already passed the capability test, and now it has to pass the durability test. What changed versus earlier years is not only the size of profit, but also the fact that the listed company now has real balance-sheet flexibility and a better-funded Reshef, at the cost of owning less of that operating asset.

The strongest counter-thesis is that 2025 already represents the start of a new normal, because the world is in a long rearmament cycle, India is opening, Europe is opening, and the U.S. is next. That is a reasonable counter-thesis, but for now it still rests more on future backlog conversion than on a visible, recurring replacement cycle.

What will change the market's near- to medium-term interpretation is less another strong quarter by itself, and more the quality of that quarter: whether follow-on orders arrive from North America or India, whether the U.S. license is received on time, whether Europe turns into approvals and revenue, and whether margins remain good enough to justify treating 2025 as more than a one-off peak.

Over the next 2 to 4 quarters, the company has to show three things: replacement orders that offset the North America comparison gap, practical progress on local-production models in India and Europe, and good profitability even after the 2025 peak year is behind it. What would weaken the thesis is some combination of backlog not refilling, delays in licensing and local production, or margins snapping back too quickly toward the historical range without a new external engine replacing the one-off project.