Spuntech 2025: Cash Flow Jumped, But Margins Are Still Stuck Between FX, Tariffs, and Large Customers

Getting To Know The Company

Spuntech is a small public industrial company with a market value of roughly NIS 184 million, but its economics are global. It produces spunlace nonwoven fabric, mostly as an input for wipes, and sells almost all of its output outside Israel. In 2025 about 96% of sales came from abroad, about 81% of revenue came from North America, and 66.4% of sales were attributed to the U.S. production site. This is not a domestic-demand story. It is an export manufacturer whose results are highly sensitive to FX, supply chains, fiber prices, and a small number of large customers.

What is working now is demand. Annual revenue fell only 1.9% to NIS 651.8 million, volumes sold increased by about 1%, and the fourth quarter looked much better than the comparable quarter, with revenue up 8.1% and operating profit swinging to NIS 5.9 million from an operating loss of NIS 9.5 million. The U.S. segment also recovered, generating NIS 15.1 million of segment profit versus NIS 2.9 million in 2024.

The bottleneck is that better demand has not yet translated into a clean earnings story. Spuntech operates in a setup where the final margin is filtered through three hard variables: exchange rates, the ability to pass through raw-material and tariff inflation, and customer concentration that limits pricing power. In 2025 the strong shekel alone reduced revenue by about NIS 45.8 million and gross profit by about NIS 12.6 million. The U.S. tariff program cut another NIS 5 million-plus from gross profit in the second half. That is the core read: demand held up, but the conversion of volume into economic value deteriorated.

Another easy mistake is to look only at the NIS 11.2 million net profit and miss what happened at the equity layer. Translating the dollar business back into shekels produced NIS 39.7 million of other comprehensive loss, so the year actually ended with a total comprehensive loss of NIS 28.5 million. For Spuntech, FX does not only hit margins. It also compresses reported equity and changes how resilient the business looks from the shareholder layer.

The economic map in 2025 looked like this:

Events And Triggers

FX was an operating event, not just an accounting line

The shekel strengthened in 2025 to an average of NIS 3.4529 per dollar versus NIS 3.69 in 2024. Management quantified the damage: about NIS 45.8 million was lost from revenue and about NIS 12.6 million from gross profit. In the fourth quarter alone, with the average dollar rate at NIS 3.2491 versus NIS 3.6128 a year earlier, the revenue hit was NIS 21.3 million and the gross-profit hit about NIS 4.5 million. The sensitivity table shows how material this is: a 10% move in the dollar changes pre-tax profit by about NIS 7.1 million and equity by about NIS 36.6 million.

Tariffs hurt on both sides of the cost structure

Starting in April 2025, the U.S. imposed a 10% tariff on company products exported from Israel, and in August that rose to 15%. At the same time, imported raw materials into the U.S. faced tariffs ranging from 15% to 35%. The damage was not merely mechanical. The company states that it tries to maintain uniform pricing for identical items supplied from Israel and from the U.S., which meant it absorbed a meaningful portion of roughly USD 2.4 million of Israeli-export tariffs into the U.S. This matters because a uniform-service model helps protect customer relationships, but it also reduces pricing flexibility during external shocks.

The U.S. fire hit utilization, but did not break the system

On April 23, 2025, a fire broke out in an energy room at the U.S. plant, shutting production until May 4. Net direct damage to the company was about NIS 1.5 million of deductible cost, while other income included about NIS 2.7 million for property damage and NIS 1.3 million for lost profit. That means the real operating damage was mainly utilization-related, while part of the financial impact was offset through other income.

The late-2025 and early-2026 customer contracts changed the starting point

This is the main trigger for the year ahead. On December 29, 2025, the U.S. subsidiary signed two supply agreements with a material customer: one extension for 2026 with expected revenue of about USD 24 million, and another agreement for 2026 and 2027 with an option into 2028 and expected revenue of about USD 12 million per year. On January 27, 2026, another material-customer agreement was extended through the end of 2028, with expected supply volumes of about USD 25 million in 2026, USD 23 million in 2027, and USD 27 million in 2028. These contracts stabilize the volume base, but they do not guarantee better margins if FX and tariff conditions remain unfavorable.

What the market can easily miss on first read is that the fourth quarter really was better, but the comparison base was distorted. The prior-year quarter included a NIS 15.4 million revenue reduction tied to a troubled customer, so the jump is not a purely organic apples-to-apples recovery. Even so, the move in utilization to 87.5% from 81.6% and the return to positive operating profit are important signals.

Efficiency, Profitability, And Competition

Price, volume, and mix did not move in the same direction

At first glance 2025 looks like a year of relative top-line stability, but profitability deteriorated much faster. Revenue fell 1.9%, gross profit fell 9.9% to NIS 88.4 million, operating profit fell 21.4% to NIS 25.0 million, and EBITDA fell 15.0% to NIS 53.2 million. Gross margin compressed to 13.6% from 14.8%, and operating margin to 3.8% from 4.8%.

The crucial point is that volumes sold actually increased by about 1%. The main problem was not demand, but the quality of the sale. Part of the damage came from FX, part from tariffs, and part from lower utilization. The company notes that most raw-material price changes are passed to customers quarterly, semiannually, or monthly, but 2025 shows that this mechanism is not frictionless. There are lags, there are markets where not everything can be passed on, and there are customers where uniform pricing across Israel and the U.S. has an economic cost.

Utilization fell, and Q4 does not erase that yet

Capacity utilization declined to 82.33% from 86.17% in 2024 and 89.71% in 2023. The company attributes that to two main factors: the loss of a material customer at the end of 2024, which hurt first-quarter 2025 throughput, and the fire event in the U.S. that shut production for roughly 11 days. That explains why 2025 is a year where the top line held up reasonably well, but the economics underneath became less efficient.

The positive news is that the fourth quarter already looked better. Gross profit jumped to NIS 22.7 million from NIS 7.1 million, and EBITDA turned to NIS 12.6 million from minus NIS 1.9 million. The less positive news is that the company remains far from 2023, when gross margin was 18.5% and operating margin 8.8%. So the story is not a full recovery to peak economics. It is a stabilization after a bad year.

The pressure shifted inside the group

In 2024 the Israel segment generated NIS 28.9 million of segment profit, versus only NIS 2.9 million in the U.S. In 2025 that picture partly flipped: the U.S. rose to NIS 15.1 million and Israel fell to NIS 9.9 million. That is important because it shows the group is leaning more heavily on the U.S. plant to stabilize performance just as trade between Israel and the U.S. became more expensive. It also helps explain the warehouse investment in the U.S. This is not just a real-estate decision. It is an effort to reinforce the side of the business where most of the economics now sit.

Real competitive advantages, but not full pricing freedom

Spuntech does have a real industrial moat: five production lines, a two-continent footprint, meaningful entry barriers in spunlace manufacturing, and customer relationships that run 15 to 22 years with major accounts. Supplier relationships also stretch beyond 20 years. But 2025 shows that this moat works mainly at the market-access and volume-retention level, not necessarily at the margin-protection level. When the top five customers account for 65.1% of sales and raw materials represent 65.9% of production cost, even a well-run industrial business can run out of room when FX and tariffs move against it at the same time.

Cash Flow, Debt, And Capital Structure

Cash flow improved, but the framing matters

Operating cash flow rose to NIS 64.7 million from NIS 27.9 million in 2024. That is a sharp improvement, but it does not, by itself, prove a step change in recurring cash generation. The main driver was a NIS 42.4 million decline in inventory, while receivables actually increased by NIS 8.4 million and payables to suppliers fell by NIS 19.0 million. In other words, cash came first and foremost from working-capital contraction rather than from a margin recovery.

And even that needs one more layer of care. Inventory fell from NIS 169.3 million to NIS 114.1 million, but the company explains that the move came from a mix of lower finished-goods quantities, raw-material price changes, mix changes, and a stronger shekel that reduced the translated value of inventory. At the same time, average raw-material inventory days actually rose to 67 from 65. So the cash improvement was real, but it does not necessarily mean the business has already moved to a structurally lighter working-capital model. Part of the release was translation-driven and part was operational.

The all-in cash picture is still conservative

To understand financing flexibility, the right frame here is all-in cash flexibility, meaning cash after actual uses rather than just before them. In 2025 the company generated NIS 64.7 million from operations, but used NIS 20.6 million for investing activity, NIS 12.3 million for interest payments, NIS 5.8 million for lease principal, and NIS 40.5 million for financing activity. Year-end cash only increased from NIS 15.7 million to NIS 17.0 million. Put differently, the business generated more cash, but almost all of that improvement was absorbed by capex, debt service, and the broader financing system. This is not a liquidity stress case. It is also not a case of wide-open excess capital.

The U.S. warehouse makes sense, but it is not free

During 2025 the company bought adjacent land in the U.S. and started building a main warehouse of roughly 9,000 square meters, expected to eliminate the need for external warehouses beginning in the second quarter of 2026. Total cost is estimated at about USD 8 million, and by the report date about USD 5.6 million had already been invested. To fund part of that, the subsidiary took a USD 6 million long-term loan in December 2025. Strategically the move is logical, because the company currently rents four temporary buildings in North Carolina and Virginia totaling about 26,000 square meters. But it also means 2026 will not start from a lighter base. It starts with a project that still needs to prove real operating savings.

Debt is lower, and covenant headroom looks comfortable

On the positive side, the debt structure looks cleaner than a year earlier. Bank credit and bank loans together fell from roughly NIS 190.9 million to NIS 162.2 million, and trade payables fell from NIS 85.6 million to NIS 61.3 million. The company also had NIS 51 million of unused credit lines. The covenant picture looks comfortably away from pressure: consolidated financial credit to EBITDA stood at 2.73 versus a ceiling of 5, the equity ratio stood at 47.6% versus a minimum of 20%, and at the subsidiary level debt-service coverage improved to 2.51 versus a 1.2 requirement while net debt to EBITDA fell to 0.49 versus a 4.5 ceiling.

The yellow flag is not covenant pressure, but sensitivity. About 96% of financial liabilities are floating-rate, and a 1% change in the variable rate affects cash financing expense by about NIS 1.2 million a year. That is manageable, but not trivial when total operating profit for 2025 was only NIS 25 million.

Outlook

First finding: 2026 does not start from zero. The volume base improved thanks to material-customer agreements signed in late 2025 and early 2026, and the fourth quarter already showed better utilization and better results.

Second finding: better volume does not automatically mean better margins. Even in 2025 the company sold more volume, yet still finished the year with a lower gross margin because FX, tariffs, and pricing conditions outweighed the volume benefit.

Third finding: 2025 cash flow bought time, but did not settle the question. As long as the working-capital release does not also show up in lower inventory days and better pricing quality, it is hard to treat 2025 as a normalized cash base.

Fourth finding: tariff relief helps, but does not remove the issue. According to the report, after the U.S. Supreme Court ruling in February 2026, a 10% tariff still remained on imports of company products into the U.S. That is better than 15%, but far from a return to the old regime.

That leads to the main read on the year ahead: 2026 looks like a bridge year with a margin test. Not a breakout year, because too many variables remain open. But not a reset year either, because the customer base, contracts, and utilization trend already look better than they did at the start of 2025.

What must happen for the read to improve

First, the new contracts need to lift utilization for several quarters rather than just plug a temporary hole left by the customer loss at the end of 2024. Second, the company needs to show better pass-through of tariffs and raw-material inflation without sacrificing realized pricing. Third, the new U.S. warehouse needs to reduce real logistics and storage costs rather than simply enlarge the asset base. Fourth, working capital needs to stay controlled even as volume recovers.

What the market may focus on next

A positive surprise would be one or two quarters showing that even under a 10% tariff regime, the company is moving back toward mid-2024 gross-margin territory on the back of utilization and warehouse-related efficiency. A negative surprise would be volume growth that still fails to translate into margin because uniform pricing and FX keep absorbing the benefit.

The right way to read 2026 is to ask less “how much will the company sell?” and more “how much of each sale will it keep?” That is the difference between operational stability and a real earnings proof year.

Risks

The first risk is customer concentration. The top five customers account for 65.1% of sales, and the top ten for 83.3%. The company explicitly says that losing a material customer could require 4 to 6 quarters to find replacements. This is not a one-quarter repair job.

The second risk is geographic concentration and external exposure. About 96% of sales are exports, and all raw materials for the Israeli plants are imported. Port disruptions, war, freight issues, or changes in U.S. import tariffs all hit sales, inventory, and margin directly.

The third risk is FX. The company executed only six forward transactions in 2025, generating total profit of NIS 564 thousand. That means hedging exists, but it is limited. When a 10% move in the dollar changes pre-tax profit by NIS 7.1 million, FX is not a side note. It sits inside the core thesis.

The fourth risk is receivable quality. Even though the doubtful-debt allowance fell to zero, about NIS 16.9 million of receivables were still outside credit insurance at the end of 2025, and in 2024 the company already had to reduce revenue by NIS 15.4 million because of a customer in distress. Any reading of renewed growth has to keep that history in mind.

The fifth risk is interest-rate sensitivity. The company is not under covenant pressure, but 96% of its financial liabilities are floating-rate. If the rate backdrop turns again, part of the financing relief seen in 2025 will disappear quickly.

Conclusions

Spuntech entered 2026 from a better place than where it stood at the beginning of 2025. The fourth quarter was stronger, the contract base with major customers improved, and operating cash flow recovered. But the picture is still not clean: margins compressed, equity was hit by FX translation, and the company still has to prove that better volume can survive alongside tariffs, rates, and large-customer bargaining power.

That means the market is not really waiting for another contract headline. It is waiting for proof that the new contracts, the new warehouse, and the better utilization level finally translate into a steadier margin structure.

Current thesis in one line: Spuntech no longer looks like a demand problem. It looks like a demand base that still struggles to convert into acceptable margins.

What changed versus the previous read: The 2026 customer base looks stronger, the U.S. operation is contributing again, and cash flow improved. On the other hand, 2025 proved that commercial strength alone is not enough to neutralize FX and tariffs.

Strongest counter-thesis: If the effective U.S. tariff burden remains manageable, the new warehouse lowers costs, and the new contracts keep utilization elevated, then 2025 may prove to have been mostly an externally distorted bridge year rather than a sign of structural erosion.

What could change the market reading in the short to medium term: Two reports with better gross margin, sustained utilization, and no renewed inventory build could materially improve the interpretation. Renewed FX pressure or fresh tariff absorption would bring the skepticism back quickly.

Why this matters: This is a real industrial business with real customers, but its earnings quality is determined less by shipment volume and more by the trading terms wrapped around that volume. Missing that distinction leads to the wrong read of 2025.

What must happen over the next 2 to 4 quarters: The new contracts need to support utilization without a price war, the new U.S. warehouse needs to show real operating benefit, working capital needs to stay controlled even in days terms, and FX cannot keep taking this much out of profits. What would weaken the thesis is a situation where volume rises but margins remain stuck around 2025 levels.