The EU ETS Market Stability Reserve keeps its ratchet: what the 700 million threshold means for carbon-exposed industry

What was actually agreed

The Council of the EU and the European Parliament reached a provisional agreement on 20 October 2026 to amend the Market Stability Reserve (MSR) of the EU Emissions Trading System. The headline change is narrow but consequential: the threshold above which allowances held in the reserve are permanently cancelled rises from 400 million to 700 million allowances.

That single number deserves more attention than it usually receives. It is the last remaining quantitative ratchet in the EU's carbon market plumbing, and it has just been loosened rather than removed. The European Commission had proposed suspending the invalidation mechanism indefinitely; the two co-legislators refused, retaining cancellation but granting the reserve roughly 300 million extra allowances of headroom before anything is destroyed.

The deal still requires endorsement by both institutions, legal-linguistic revision and formal adoption, so the 700 million figure is not yet law. Treat the substance as provisional but the direction as settled.

Why invalidation matters more than intake or release

The MSR has three moving parts, and they are frequently conflated. Only one of them shifts the permanent supply of allowances.

Intake and release manage price volatility. Invalidation changes the total number of allowances that will ever exist under the cap. That is why the Commission's proposal to suspend it — flagged on the Commission's MSR page as a 1 April 2026 proposal — was the most market-moving element of the file, and why the decision to retain it, even at a higher threshold, is the story.

On the Commission's published figures, invalidations have totalled 2,500 million on 1 January 2023, 381 million on 1 January 2024 and 271 million on 1 January 2025 — a cumulative 3,152 million allowances, consistent with the "more than three billion" cited in reporting of the deal. That is a reduction of the same order as several years of the cap's linear reduction factor, delivered in three annual steps. The new threshold does not reverse those cancellations; it slows the pace of future ones.

Reading the number: softness now, optionality later

The practical effect of moving the threshold to 700 million depends on how large MSR holdings actually are. The Commission does not publish a single "holdings" number in the same place as the TNAC, but the arithmetic of the three documented invalidations is informative:

• In each of 2023, 2024 and 2025, the cancelled volume equals holdings minus the applicable threshold.
• The 2024 and 2025 cancellations (381m and 271m) are substantially smaller than the 2023 event (2,500m), which occurred at the peak of the historical oversupply.
• A 381m overshoot above 400m implies holdings of roughly 781m at the start of 2024; a 271m overshoot implies roughly 671m at the start of 2025.

If holdings were continuing on that trajectory — and nothing in the record suggests a sharp rebuild — the reserve by 2026–27 would sit comfortably below 700 million. In that case the higher threshold would, for the next few years, amount to a de facto pause in invalidation without formally abandoning the instrument.

That is the analytical crux. The Commission wanted to abandon the instrument; the co-legislators chose to make it dormant instead, with an explicit commitment to consider gradual threshold reduction in the broader ETS review. The ratchet remains in the drawer. Whether it is ever turned is a political choice deferred to the 2027–2030 review cycle, not a mechanical outcome of market conditions.

The mid-2030s argument is the one to take seriously

The Council's stated rationale — strengthening the MSR's ability to respond to potential supply constraints after the mid-2030s as the cap declines — is not a short-term talking point. It reflects a genuine structural problem at the intersection of three converging policies:

1. The linear reduction factor continues to shrink annual issuance.
2. Free allocation is being phased down in parallel with the CBAM phase-in.
3. The 2030–2034 period brings the end of free allocation for most sectors and the full CBAM regime.

Under those conditions, a reserve holding a few hundred million allowances is a thin cushion against a demand-supply mismatch during a period of unprecedented industrial investment in electrification, hydrogen-based ironmaking and carbon capture. A 700 million threshold effectively keeps that cushion available for longer. The Commission's suspension proposal would have achieved something similar, but by removing the cancellation clause entirely — leaving an open-ended stock that markets might read as a permanent overhang. The Council–Parliament solution keeps the stock for the medium term but signals it will be drained, one year at a time, if conditions change.

What this means for steel and CBAM

The ETS review track referenced in the agreement also carries the free-allocation redesign, the CBAM extension and the Industrial Decarbonisation Bank proposals tabled in July 2026. Steel sits at the intersection of all three.

Three practical channels matter:

CBAM certificate pricing. CBAM certificate prices are benchmarked to EUA auction clearing prices. Any softening of the EUA scarcity signal feeds through, with a lag, into CBAM costs for importers of steel, aluminium, cement and fertilisers. A higher invalidation threshold therefore modestly reduces the expected escalation path of CBAM costs over the 2026–2030 phase-in. That is a benefit for importers, but it also narrows the cost gap that European low-emission steel needs in order to compete against higher-carbon imports.

Banked allowances and balance sheet optionality. Steelmakers with surplus free allocation banked on their balance sheets are, in effect, long EUA. A softer scarcity signal reduces the mark-to-market value of those positions. Conversely, producers facing a future shortfall gain more time to close it at a non-disruptive cost — which matters disproportionately for capital-intensive DRI-EAF conversions now under construction.

Investment signals for decarbonisation. The economics of green steel depend on the spread between the EUA price and the cost of abatement. A reserve that absorbs rather than destroys surplus allowances will, at the margin, compress that spread in the 2027–2032 window. It is a cost-of-capital headwind for first movers, not a fatal one, but one that reinforces the sector's reliance on the Innovation Fund, the Modernisation Fund and the proposed Industrial Decarbonisation Bank rather than on the carbon price alone.

What is not yet known, and why it matters

The provisional agreement has not been formally published in retrievable form. The following remain unclear and could materially change the interpretation:

• The precise instrument (amending Decision (EU) 2015/1814) and any transitional provision governing the timing of the first 700-million-threshold invalidation.
• Whether the review clause commits to a specific threshold-reduction schedule, or merely to "consider" one — the wording implies the latter, which is the weaker commitment.
• The relationship between this MSR amendment file and the broader ETS review tabled in July 2026. The two are distinct legislative tracks, and the deal explicitly defers the threshold path into the review, so the sequencing is deliberate.
• How the EUA curve reacted in the days after 20 October. No market data was available for this analysis.

The absence of a published legal text is not unusual at this stage, but it does mean the gap between "retained" and "effectively suspended" is currently a matter of arithmetic rather than of settled law.

Bottom line

The EU has chosen to slow the carbon market's supply-ratchet rather than remove it. For industrial consumers of allowances, that is less bullish for 2027–2030 EUA prices than the Commission's suspension would have been, and more bullish for 2035 onwards than an outright abolition would have been. The real signal is procedural: the co-legislators have kept a policy lever that the executive wanted to discard, and parked it in the ETS review where it can be a live question in the next political cycle. Steel decarbonisation economics — already dependent on subsidies, off-take contracts and CBAM protection — should be underwritten on the assumption that the carbon price ratchet will survive, but be turned more slowly than previously legislated.