ICL in 2025: Why EBITDA Did Not Turn Into Cash
ICL presented $664 million of free cash flow in 2025, yet the same presentation also admitted that its usual cash metric was only $236 million. That gap is not technical. It shows how 2025 was a year of stable EBITDA on paper, but much weaker cash conversion in practice.
What This Follow-Up Is Isolating
The main article argued that ICL’s test has moved from potash and phosphate prices to cash. This follow-up isolates only that thread: how a company that reports $1.488 billion of adjusted EBITDA and $664 million of “free cash flow” ends 2025 with only $236 million on its usual cash metric, and with a net $54 million decline in cash and cash equivalents.
The short answer is that in 2025 ICL told three different cash stories. The first is the presentation story: adjusted EBITDA minus CAPEX equals $664 million. The second is the footnote in that same presentation: the company’s usual metric is cash flow from operations minus CAPEX, and in 2025 that was $236 million. The third is the cash flow statement itself: using the rounded reported figures, $1.056 billion of operating cash flow minus $824 million of purchases of property, plant and equipment and intangible assets leaves roughly $232 million. The message is the same under any reasonable reading: more than $400 million disappears the moment the reader moves from a presentation definition to a cash definition.
That is exactly where 2024 and 2025 diverge. In 2024, adjusted EBITDA was $1.469 billion and operating cash flow was $1.468 billion. The shortcut used in the presentation still tracked economic reality reasonably well. In 2025, that shortcut stopped being a harmless simplification and became a much more generous framing.
45% on the Slide, About 16% on the Usual Metric
The March 2026 presentation does not only show $664 million of free cash flow. It builds a capital-discipline story around it. In the peer comparison slide, ICL is shown with 45% free cash flow conversion, almost in line with the 46% peer average. Yet the same presentation states that the company’s usual metric for 2025 was only $236 million. If that $236 million is placed over the same $1.488 billion adjusted EBITDA denominator, conversion comes out at roughly 16%, not 45%.
That is the heart of the issue. This is not a debate over a small technical adjustment. It is a shift between two different languages. One language measures adjusted EBITDA minus CAPEX and produces a story of almost peer-average cash conversion. The other runs through the cash flow statement and shows that 2025 was, in practice, a much weaker year in the transition from reported profitability to cash.
| Measurement layer | 2025 | What it actually captures |
|---|---|---|
| Adjusted EBITDA | $1.488 billion | A non-GAAP operating metric after add-backs above reported earnings |
| “Free cash flow” in the presentation | $664 million | Adjusted EBITDA minus CAPEX |
| Company’s usual metric | $236 million | Cash flow from operations minus CAPEX |
| Net change in cash and cash equivalents | $(54) million | What remained after actual investing and financing flows |
In 2024 this shortcut was still reasonable, because adjusted EBITDA and operating cash flow were almost identical. In 2025 it created a materially different story. Anyone reading only the $664 million headline sees a company with decent cash conversion. Anyone reading the cash flow statement sees a company where the gap between EBITDA and cash has become too large to ignore.
Why the Gap Opened
The starting point was already too adjusted
The first issue is that the starting metric is much more generous than reported earnings. Operating income in 2025 was $580 million, while adjusted operating income was $873 million. The $293 million gap consists of $131 million of impairment and write-offs, $80 million of legal proceedings, disputes and other settlements, $54 million of charges related to the security situation in Israel, and $28 million of early-retirement provisions.
So even before working capital, taxes and investment, the presentation is already starting from a softer story than the one running through the income statement. That does not make adjusted EBITDA meaningless. It does mean the bridge from that EBITDA to actual cash is much longer than the $664 million headline suggests.
Working capital and taxes squeezed cash flow
The second explanation sits directly in the cash flow statement. The company itself says that the $412 million decline in operating cash flow was driven mainly by changes in operating assets and liabilities and by higher taxes paid. The line items make the friction easy to see:
| Item | 2025 | 2024 | What it means |
|---|---|---|---|
| Change in inventories | $(210) million | $(7) million | Inventory absorbed cash instead of being almost neutral |
| Net change in operating assets and liabilities | $(63) million | $205 million | Working capital moved from supporting cash flow to consuming it |
| Income taxes paid, net of refund | $(151) million | $(98) million | Another $53 million of cash left the system |
| Operating cash flow | $1.056 billion | $1.468 billion | A $412 million decline |
The real point is that the gap did not open because of a single one-off line. It opened from several directions at once: inventory, working capital, taxes, and a starting point that was already highly adjusted. That is why 2025 does not look like a one-line accident. It looks like a year in which cash conversion weakened across the whole bridge.
There is also a smaller but important contradiction inside management’s framing. The presentation describes a financial profile with predictable, low-volatility CAPEX. Yet in the chart itself, 2025 CAPEX is the highest level in the five-year series at $824 million, and it rises to 12% of sales from 10% in 2024. So the investment leg also did not help the cash story this year.
Even After the Usual Metric, the Cash Was Not Truly Free
This is the part that is easiest to miss. Even $236 million is not clean free cash flow to equity. In 2025 the company paid $117 million of interest, $224 million of dividends to shareholders, $64 million of dividends to non-controlling interests, and $12 million related to business combinations. Once those items are deducted from the company’s own usual metric, the residual turns negative by roughly $180 million, even before deposit movements and before debt refinancing.
That explains why the bottom line in the cash flow statement does not look like a year in which $664 million was left sitting in the cash balance. Net investing activities consumed $915 million, net financing activities consumed $195 million, and the net change in cash and cash equivalents was negative $54 million. At the same time, the company increased short-term investments and deposits to $205 million from $115 million, so total liquid resources reached $496 million. Liquidity did not break, but that is already a balance-sheet and funding story, not proof that EBITDA truly turned into free cash.
The debt movements make the same point. In 2025 the company received $1.666 billion of long-term debt, repaid $1.599 billion of long-term debt, and received another $146 million net from short-term debt. That means there is no immediate liquidity stress here. But it also means the cash balance was not replenished by $664 million of truly spare cash. It also depended on refinancing and active liquidity management.
What Needs to Be Measured Next
First check: whether working capital, and especially inventory, starts supporting cash flow again instead of consuming it. If EBITDA stays firm while CFO remains hundreds of millions below it, the market will focus on the gap rather than the headline.
Second check: whether CAPEX returns to a level consistent with the stable-investment story, or whether the company will have to admit that its mineral platform requires a structurally higher investment base than the presentation suggests.
Third check: whether the $293 million gap between reported and adjusted operating income starts to narrow. As long as EBITDA relies on add-backs of that scale, the cash story will remain exposed to skepticism.
Fourth check: whether the company can restore a better fit between the cash number sitting in the footnote and the cash number sitting in the headline. In 2024 that still worked. In 2025 it no longer did.
The bottom line is fairly sharp. In 2025 ICL’s EBITDA did not fail because the business collapsed. It did not turn into cash because the starting point was already too adjusted, because working capital and taxes compressed CFO, and because whatever remained then moved on to CAPEX, interest, dividends and acquisitions. So the $664 million figure explains how the year looked through the presentation. The cash flow statement explains how it looked through cash.
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.