Had-Asaf follow-up: is Romania a real recovery engine or just a temporary trading-spread story
Romania moved from a NIS 9.3 million segment loss in 2024 to a NIS 0.7 million segment profit in 2025, but 88% of revenue there already came from trading and import quotas are set to be cut by about 50% from July 1, 2026. The real durability test still lies ahead.
Romania's Durability Test
The main article argued that Had-Asaf came out of 2025 weaker than the annual profit line suggests. This follow-up isolates Romania because that is where the only clear improvement showed up: revenue reached NIS 129.7 million, up 48%, and the segment moved from a NIS 9.3 million loss to a NIS 0.7 million profit.
The question here is not whether Romania improved. It did. The question is what kind of improvement this is. To become a real recovery engine, Romania has to produce earnings the group can build on. Right now the picture is narrower: an operation that became much more trading-based, much less manufacturing-based, and still earned only a near-breakeven segment margin.
That matters because the next bottleneck is already clear. Steel inventory in Romania usually represents two to three months of sales and remains exposed to price moves. Once imports rise above the European Union's quarterly quota limits, tariffs can reach about 25% of the value of the steel purchased. And from July 1, 2026, those quotas are due to be cut by about 50%. Romania's durability test is still ahead of it, not behind it.
| Key metric | 2024 | 2025 | What it really says |
|---|---|---|---|
| Romania revenue | NIS 88.2m | NIS 129.7m | Sharp growth, but still only about 7% of consolidated revenue |
| Segment result | loss of NIS 9.3m | profit of NIS 0.7m | A return to profit, but on only about a 0.6% segment margin |
| Trading activity as a share of Romania revenue | about 76% | about 88% | The improvement leaned more on trading and less on manufacturing |
| Employees in Romania | about 35 | about 25 | The cost base became lighter, but also less industrial |
| Depreciation and amortization in Romania | NIS 7.8m | NIS 0.8m | The segment returned to profit alongside a much lower depreciation charge |
That chart sharpens the issue. Revenue did build a multi-year trend, but the profit line still does not describe a new earnings engine. After a NIS 4.9 million segment loss in 2023 and a deeper NIS 9.3 million loss in 2024, 2025 only brought the business slightly above zero.
What Actually Improved
The improvement deserves real credit. Romania's gross profit moved from a gross loss of NIS 0.9 million in 2024 to a gross profit of NIS 5.7 million in 2025. The other-expense line also fell to NIS 1.3 million from NIS 4.9 million. This is not just a classification story. There was genuine operating improvement versus the prior year.
But the key number is not the fact that Romania turned profitable. It is the size of the earnings cushion it created. On NIS 129.7 million of revenue, the segment kept only NIS 0.7 million of segment profit. That is a very thin margin pool, one that could disappear on a relatively small move in raw-material cost, delivery timing, or import economics.
The source of the improvement is also clear. Starting in 2020, the company gradually cut back steel-cable and steel-wire manufacturing, fully stopped that production in 2022, and shifted toward trading raw steel and producing wire fences. By 2025, about 88% of Romania revenue already came from trading activity. In other words, Romania did not recover through a comeback of the old factory model. It recovered through a lighter trading model.
That is the core of the continuation. In the same year that the trading share of revenue rose, headcount fell by about 10 employees. The decline came from fewer manufacturing and logistics workers alongside expanded trading activity. Depreciation and amortization in Romania also fell sharply. Put differently, Romania in 2025 was a lighter, more commercial operation. That can help it turn profitable faster, but it also means the improvement depends less on manufacturing advantage and more on buying and selling conditions.
Why This Still Looks Like a Trading-Spread Story
The first yellow flag is the customer and product structure. In Romania's trading activity, the company sells raw steel bars and coils to local traders and steel processors, while its fence-production business sells into the local Romanian market as well as Europe and Israel. This is not a long-contract business or an anchor-customer model that stabilizes volume and price. No customer accounts for 10% or more of revenue, and the company explicitly says it has no dependence on any one customer.
The second yellow flag is the lack of time cushion. Order backlog in the trading business runs from a few weeks to about two months, and there are effectively no binding orders that have not yet been recognized as revenue. Romania is therefore entering 2026 without an order book that protects margin if import conditions tighten.
The third yellow flag is that the remaining manufacturing layer is not currently opening a new earnings tier. The company says the wire-fence plant is already operating at its production potential. At the same time, there is still potential production capacity in steel cables and steel wires, but that activity was stopped because it was not economically viable. There is no evidence here that Romania is on the verge of a new industrial step-up.
The July 1, 2026 Quota Test
The risk mechanism itself is fairly clear. If imports exceed the European Union's quarterly quotas, tariffs can reach about 25% of the value of the steel purchased. The company also says it has no certainty, when buying raw material, about quarterly delivery timing or quota compliance because other importers operate inside the same quota framework.
There is, in theory, an alternative: source raw material from within the European Union and avoid the tariff. But that side is just as clear. Steel bought inside the Union is priced about 15% to 20% above steel from Turkey. So the roughly 50% quota cut from July 1, 2026 does not look like a technical footnote. It goes directly to the heart of the model that brought Romania back to profit.
This is where the distinction between recovery and a favorable trading window becomes sharp. If the company can hold a positive margin after the quota reduction, then Romania can start to look like a genuine secondary recovery engine. If margin disappears once the choice becomes a high tariff or higher raw-material cost, then 2025 will look, in hindsight, more like a year in which the trading spread simply worked in Romania's favor.
The test should be strict. Romania does not only need to keep growing. It needs to prove that its profitability survives a change in import conditions. Without that, the 2025 growth tells more of a volume story than a quality story.
Bottom Line
Romania at Had-Asaf is a real improvement, but not yet a recovery engine the group can rely on. In 2025 it benefited from stronger demand and a better gap between selling prices and raw-material costs, but it did so through a model that became more trading-heavy, leaner, and more exposed to import conditions.
That is exactly why the NIS 0.7 million segment profit matters less than what happens after July 1, 2026. If Romania stays profitable under narrower quotas, that will be the first sign of a secondary engine that really works. If not, 2025 will be remembered more as a temporary trading-spread story than as a structural recovery.
Disclosure: Deep TASE analyses are general informational, research, and commentary content only. They do not constitute investment advice, investment marketing, a recommendation, or an offer to buy, sell, or hold any security, and are not tailored to any reader's personal circumstances.
The author, site owner, or related parties may hold, buy, sell, or otherwise trade securities or financial instruments related to the companies discussed, before or after publication, without prior notice and without any obligation to update the analysis. Publication of an analysis should not be read as a statement that any position does or does not exist.
The analysis may contain errors, omissions, or information that changes after publication. Readers should review official filings and primary sources before making decisions.