Ratio Energies (Finance): Does The IFRS 9 Model Really Capture The Bond Risk
The main article already showed that Ratio Energies (Finance)'s bond rests on Leviathan, while the pledged royalty is a stress-case backstop. This follow-up isolates the IFRS 9 engine that created 2025 profit, and shows that it leans more on market spreads and broad recovery assumptions than on the bond's actual shutdown and collection chain.
The main article already made the core point: the bond of Ratio Energies (Finance) rests on Leviathan cash flow, while the pledged royalty is a backstop rather than a current debt-service engine. This continuation isolates the layer that almost single-handedly decided the issuer's 2025 numbers, the IFRS 9 model.
Why does that matter? Because at the issuer level, very little changed economically. Interest income from the partnership loan and interest expense on the bond were both $5.678 million, while annual net profit was only $113 thousand. That exact amount also appeared as expected credit loss income. In other words, 2025 profit did not come from a new operating engine. It came from a softer allowance model, with the reserve falling from $581 thousand at year-end 2024 to $468 thousand at year-end 2025.
The short answer to the headline question is: only partly. The model explains quite well how the company translated the year-end 2025 market and macro picture into a lower accounting reserve. It is far less convincing as a full credit map for a single-asset bond that depends on one field, one operator, sensitive export infrastructure, and a collection right that becomes truly relevant only in a stress scenario.
The Model That Created 2025 Profit
The company itself defines the valuation approach as a market-comparison approach. That is not a technical side note. It is the center of the methodology. First, the company asks whether credit risk increased materially since initial recognition. Only then does it decide whether the allowance should be measured over the life of the instrument or only over the next 12 months.
In Ratio Energies (Finance), the company concluded that no such significant increase occurred. The practical result was a 12-month ECL path rather than a lifetime ECL path. From there, the road to the $113 thousand annual profit was short: the allowance rate eased slightly, the reserve balance fell, and the full profit came out of that accounting line.
This chart shows why the allowance model is not a footnote here. At the issuer level, it is almost the whole story. Without the lower reserve, annual profit would have been effectively zero.
Why The Spread Test Calms The Accounting, Not Necessarily The Bond
The company tested the risk spread of Ratio Energies (Finance) Series D against a tradable AAA-rated dollar deposit and compared three reference points: initial recognition in July 2021, the series expansion in December 2024, and the balance-sheet date of December 31, 2025. The result looked comfortable from the company's perspective. The spread fell from 5.1% in July 2021 to 1.8% at year-end 2025, while relative to the December 2024 expansion it rose from 0.5% to 1.8%. The report also notes that the average of the four quarter-end readings in 2025 was 1%, so the December move is framed as a point event rather than a structural deterioration.
From an accounting perspective, that is a coherent read. If the market is not pricing a major deterioration relative to the recognition point, it is hard to justify an automatic move into lifetime ECL. The problem is that this spread test answers a narrow question only: how the market priced the series at the cut-off point of December 31, 2025.
It does not answer the broader question that matters to the bondholder: what happens to a single-asset credit structure when Leviathan can be shut by a security instruction, when force majeure is declared toward customers, and when even at the approval date management still cannot estimate the cumulative net impact of the February 2026 shutdown on the discounted cash flow of the asset base.
That is where the gap opens between accounting comfort and credit comfort. The bond market can stay relatively calm until an operating event is already happening in the real asset. And the model itself effectively admits that this is what it measures: it translates market pricing and macro conditions into a balance-sheet date estimate, not into a full shutdown simulation of the bond's only cash source.
A 2.34% PD Is A Fine-Tuning Exercise, Not A Repricing Of The Credit
To understand the center of the model, the PD needs to be unpacked. The base general-risk layer is made of only two components: a 0.95% oil-and-gas sector risk and a 0.85% Israel-country-risk component. Together they produce a 1.8% general PD.
From there, the model moves into specific risks, but even here it does not rebuild cash flow from the bottom up. It works through fine-tuning. Each specific factor can raise or lower the base rate by 10%, and all factors carry equal weight. The valuation report says this explicitly and also states that about 90% of the allowance comes from "the market" while only about 10% comes from direct effects.
That may be the single most important line in this continuation. If 90% of the allowance comes from market and macro factors, then the model is not really trying to reprice the bond around its own collection structure in depth. It is mainly asking whether the world, Israel, and the sector look a bit better or a bit worse, and only then adding issuer-specific adjustments.
Those issuer-specific factors tell a mixed story:
| Component | Model conclusion | What it means |
|---|---|---|
| Geopolitical risk | Raises the allowance | Shutdowns, insurance pressure, and physical-risk concerns are treated as real credit factors |
| Sales-price risk | Raises the allowance | Lower Brent in 2025 weakens price visibility |
| Operator and partner risk | Lowers the allowance | Stable ownership and Chevron as operator are treated as a comfort factor |
| Operating risk | Raises the allowance | Delays around the third pipeline, export infrastructure, and shutdowns did not disappear |
| Competition | Raises the allowance | Tamar, Karish, and growing regional supply matter for forward risk |
| Major-customer dependence | Raises the allowance | Egypt and Jordan remain essential but also vulnerable |
| Royalty-eligibility risk | Lowers the allowance | Registration of the royalty in the petroleum register is treated as risk mitigation |
The arithmetic behind the 2.34% PD becomes fairly clear from that table. Five factors push upward, two push downward, and each carries the same 10% weight. That is not a broken model, but it is a choice. A platform shutdown and regional competition are not weighted through their actual cash-flow severity. They are fed into the same standardized adjustment step.
That is why 2.34% does not describe a fresh market discovery of much higher risk. Quite the opposite. It describes a model that stayed very close to the 1.8% macro base and then added a net 30% adjustment on top.
The 25% LGD Comes From Broad Recovery Studies, Not From This Bond's Own Waterfall
If the PD shows how macro-heavy the model is, the LGD shows how generic it is. The company kept the recovery rate at 75%, meaning a 25% LGD, based on broad international recovery studies. The report refers to Global Credit Data with 67% to 80% recovery ranges, a European banking study with 73% recovery in infrastructure, and another study with 82% recovery in the wider energy sector.
In IFRS 9 terms, that is understandable. In terms of this bond, it deserves more caution. The filings do not present a recovery simulation that starts from this bond's own payment waterfall: state royalties, overriding royalties, current payments to Leviathan's financial lenders, and only then the royalty right of Ratio Energies (Finance). In addition, the loan agreement makes clear that the royalty is the company's sole remedy in a default scenario, with no broader claim against the partnership beyond that right.
That is a material distinction. The 25% LGD is a broad market-study assumption, not a bespoke liquidation model for this series. That does not automatically mean the figure is too low. The filings do not provide enough evidence to assert a different number. But it does mean that the reassuring part of the model is not derived from a security-by-security recovery map. It is derived from broad recovery literature applied to a narrower structure.
So here too, the model answers a narrower question than the bondholder probably wants answered. It asks what average recovery looks reasonable for energy and infrastructure credits. It does not fully answer how much value would really reach the bondholders if this specific structure ever had to rely on the pledged royalty in practice.
Where The Model Helps, And Where It Needs To Be Read Carefully
The model is useful in two clear ways. First, it explains why the company still did not see a significant increase in credit risk at December 31, 2025 relative to initial recognition. Second, it forces the reader to recognize that geopolitical, operating, competitive, and customer-concentration risks have already entered the allowance calculation, at least to some degree.
But that is exactly where the reader should stop and become more demanding. Once the model itself says that most of the influence comes from market and macro factors, once it uses a market-comparison approach, and once the issuer-specific layer is framed as fine-tuning, it becomes a mistake to equate "the allowance fell" with "the bond risk eased."
In 2025 the allowance fell because year-end conditions looked easier than the original recognition points, because Israel's average risk reading eased relative to 2024, and because the recovery assumption stayed high. But in 2026 the system already reminded the market what spreads do not always capture in real time: production shutdowns, uncertainty about duration, and delay risk in regional transport infrastructure.
So the right way to read IFRS 9 here is not "the filings prove the series is safe." It is something narrower and more useful: as of year-end 2025, the company did not see market and macro evidence strong enough to justify a harsher accounting stance. That matters. It is just not a full bond-risk map.
Bottom Line
What the IFRS 9 model captured well was the market mood and macro backdrop at the balance-sheet date. What it captured less well was the true structure of the series itself, an unrated bond issued by a financing shell, with one exposure, one cash engine, one operator, and a recovery right that becomes active only at the end of the chain.
That is why 2025 profit needs to be read carefully too. It does not say that credit quality improved dramatically. It says that the accounting model became somewhat more comfortable, mainly because the year-end market spread did not signal a structural jump in risk and because the assumed recovery rate remained high.
If the question is whether the model really captures the bond risk, the most accurate answer from the filings is this: it captures the market and macro risk of December 31, 2025 quite well, and captures the bond's own shutdown and recovery risk only partially. That is why, even after the lower reserve, the most important reading point for future reports is not only the ECL number itself. It is whether Leviathan proves operating continuity, export continuity, and capacity expansion in a way that reality, not only the model, can validate.
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.