The Numbers: One Reference, Two Ladders, Two Prices

Every regulation eventually comes down to a handful of numbers someone in your office will have to defend in a budget meeting. The framework has five that matter: one reference, two target ladders, and two prices.

All of them are quoted here from the draft Chapter 5 text that MEPC 83 approved in April 2025 and circulated as Circular Letter No.5005. That text has not been adopted, and since MEPC 84 it is open to amendment again, so these are the current draft's numbers rather than the law's. Lesson 6 gives the record. Regulation references below are to that draft.

The reference: 93.3

The anchor for everything is 93.3 gCO2eq/MJ, fixed by regulation 35.2: the well-to-wake GHG fuel intensity of the average energy used by international shipping in 2008. Every target in the framework is written as a percentage reduction from this single value. For calibration: conventional fuel oil sits in the low-to-mid nineties well-to-wake, which is to say the world fleet of 2008 burned almost nothing else, and a ship burning only fuel oil operates at roughly the reference to this day.

Two ladders, not one

Each compliance year carries two targets, both carved from 93.3 (DNV's GFI explainer). Each is written as a reduction factor Z applied to the reference:

GFI_target = (1 − Z/100) × 93.3

where Z is the year's percentage reduction and the result is in gCO2eq/MJ. Table 4 of the draft text carries a Z for each ladder for every year from 2028 to 2035, and for no year after that:

Two features of that table are worth stopping on. The first is the step from 2030 to 2031: the base ladder jumps from 8.0% to 12.4%, and the direct ladder from 21.0% to 25.4%. The schedule is deliberately back-loaded past the first review point, which is a political fact as much as a technical one. The second is the one almost nobody says out loud: the two ladders are always exactly 13 percentage points apart, which on a 93.3 reference is a constant 12.13 gCO2eq/MJ, every single year. That constant has a large consequence, and lesson 5 spends the money it implies.

Beyond 2035 Table 4 stops, and what follows it is mostly a deadline rather than a number. Regulation 35.3 reads: "By 1 January 2032, the Committee shall determine the Z-factor for the Base target and Direct compliance target for the years 2036 to 2040. The 2040 ZT for the Base target shall be set at 65%." Two things follow that are easy to get wrong, and both turn up wrong in slide decks every week.

The 65% figure, about 32.7 gCO2eq/MJ, is real, and it is not a row of Table 4. It sits in the prose of regulation 35.3 as the single value the Committee is told in advance where to land, and it belongs to the base ladder only. There is no 2040 direct compliance target anywhere in the text, and no value for either ladder in 2036, 2037, 2038 or 2039. A chart showing a 2040 direct compliance figure has filled in a blank the regulation deliberately leaves open, and the Committee has until 1 January 2032 to fill it in properly. That is the twin of the price deadline further down this lesson, and neither deadline binds anyone until the amendments are adopted.

Two ladders create three zones, and the zone your attained GFI lands in decides what the year costs:

• At or below the direct compliance target: fully compliant, and the margin below the target becomes surplus units (SUs), tonne for tonne of CO2-equivalent avoided.
• Between the two ladders: a Tier 1 deficit: the shortfall against the direct target, priced at USD 100 per tCO2eq.
• Above the base target: the portion above the base ladder is a Tier 2 deficit at USD 380 per tCO2eq, stacked on top of the full Tier 1 band beneath it.

Both figures are initial prices. Regulations 36.8 and 36.9 fix them for the reporting periods 2028 to 2030 and no further, and the text gives itself a deadline for the rest: regulation 36.10 requires the mechanism by which the Tier 1 and Tier 2 prices from 2031 onward are reviewed and set to be settled by 1 January 2028. Everything past 2030 is therefore unpriced, which is a real limit on how far ahead anyone can model. The money paid for remedial units flows to the Net-Zero Fund (lesson 4).

Which tier can be settled how

Here is the rule most summaries of the framework leave out, and it decides who can help you and who cannot. The two tiers are not settled the same way.

• A Tier 1 deficit can only be settled by buying Tier 1 remedial units from the IMO Net-Zero Fund. Surplus units are no use against it, no matter how many you hold or how cheaply you could buy them. For any ship above the direct compliance target, Tier 1 is in practice an unavoidable levy on the gap between the two ladders.
• A Tier 2 deficit has three routes: surplus units transferred from another ship, surplus units you banked yourself in an earlier reporting period, or Tier 2 remedial units from the Fund. Only here does the market get to compete with the Fund's price.

Two consequences follow. Trading in surplus units is a Tier 2 business only, so the pool of demand for them is smaller than a casual reading suggests. And because the ladders sit a constant 12.13 gCO2eq/MJ apart, a ship that stays above the base target owes the same Tier 1 tonnage every year, at the same USD 100, no matter how much the targets tighten. Cleaning up the fuel shrinks Tier 2 and eventually removes it; nothing but getting under the direct compliance target removes Tier 1.

Run the fuel-oil ship through 2028 to see the design's intent. At an attained GFI of roughly the reference, she is above the base target from day one: a thin Tier 2 band (93.3 down to 89.6) at the painful price, and the full Tier 1 band (89.6 down to 77.4) at the moderate one. The expensive tier bites first and shallowly: a deliberate early-years signal that says start moving rather than shut down. By the mid-2030s the same ship's Tier 2 band has grown wide, and the signal says something much louder.

Surplus units: the market half

Surplus units are what turn the standard into a market. A ship that beats the direct compliance ladder, say a methanol dual-fuel newbuild running a large green-methanol fraction, has three things it can do with the units it earns (MMMCZCS framework hub). It can transfer or sell them to another ship, including one in someone else's fleet, for whatever price the two parties agree. It can bank them, but only for a limited life: a surplus unit is valid for two calendar years after the year the registry issues it, which is the year after the reporting period that earned it, and then it expires: a 2028 unit, issued in 2029, is good to the end of 2031. Or it can voluntarily cancel them, retiring the units instead of selling them, which is what a company wanting a clean sustainability disclosure would do.

The two-year clock is the part that catches people out. Surplus units are not a savings account. A ship that runs a heavy green-fuel fraction in one strong year and holds the units back waiting for a better price can watch them expire unsold, and the framework has no provision for reviving them.

Because only Tier 2 deficits can be settled with surplus units, the tier prices act as a guardrail on one side of the market only. Nobody rationally pays more than USD 380 for a surplus unit, because a Tier 2 remedial unit from the Fund covers the same tonne at that price. Below that ceiling, price finds its level on supply: the more over-compliant tonnage exists, the cheaper Tier 2 compliance becomes for everyone else. Tier 1 has no such market, because the Fund is the only seller. If the Tier 2 half of that logic feels familiar, it should: FuelEU Maritime's pooling works on the same principle, and lesson 5 leans on the comparison.

Two habits to take from this lesson. First, always convert percentages to absolute intensities before reasoning about fuels. "30% below reference" is 65.3 gCO2eq/MJ, a number you can hold up against a bunker delivery note. Second, remember which ladder a quoted target belongs to. The gap between base and direct compliance is not pedantry; it is the difference between paying USD 100 and USD 380 per tonne, and between a deficit you can buy your way out of on the open market and one you cannot.