Waste to Biofuels Runs on a Credit With a Ceiling Almost Nobody Models

Every waste to biofuels model that has crossed my desk in the last four years carries the same line, and the line is wrong. It's the D3 RIN price: pinned somewhere between $2.50 and $3.00, held flat for twenty years or escalated at CPI, sized to close whatever gap the gate fee couldn't. That line isn't a forecast. It's a bet against a formula Congress wrote into the Clean Air Act, and the formula usually wins.

The cellulosic RIN has a ceiling. It's arithmetic, it's published, and it moves in the opposite direction from the thing most sponsors assume protects them.

The waiver credit sets the lid

When EPA projects that cellulosic biofuel supply will fall short of the statutory volume, two things follow. The agency cuts the applicable standard, and it has to offer obligated parties cellulosic waiver credits. That credit price isn't negotiated. Per EPA, it's the higher of $0.25 a gallon or the amount by which a $3.00 per gallon reference price exceeds the average wholesale price of gasoline, adjusted for inflation. For 2025 the agency set it at $1.91.

So a refiner carrying a cellulosic obligation has two ways to discharge it: buy a D3 RIN, or buy a waiver credit and pair it with an advanced RIN. Those are substitutes, and traders price them as substitutes. A D3 print can't wander far above the sum of the other two. That's the lid.

Now look at what drives the waiver credit. Gasoline. Expensive gasoline narrows the gap to the reference price, which makes the waiver credit cheaper, which pulls the D3 ceiling down behind it. Your cellulosic revenue line is structurally short crude. In the year when every other input on the plant's P&L is shouting that energy is dear, the credit meant to carry your debt service gets capped lower. What else in a project's revenue stack gets thinner when energy gets more expensive?

I've never seen that correlation in a sponsor's sensitivity table. Most waste to biofuels models flex the RIN up and down as though both directions were equally open, downside at $1.50, upside at $3.50, symmetric and tidy. They aren't symmetric. The upside is fenced by statute. The downside is fenced by nothing at all.

Ninety-nine percent of the cellulosic pool is a pipe

Awkward fact for anyone marketing cellulosic ethanol from MSW: renewable natural gas accounts for 99% of every cellulosic RIN generated, according to EIA's read of the program data. In 2022, RNG qualifying under that category came to 55 billion cubic feet, roughly 84% of all the natural gas burned as transportation fuel in the country. The D3 market is a landfill and digester gas market wearing a liquid fuels name over the door.

That's a metering verdict more than a technology verdict. EPA counts 77,000 Btu of biogas as one gallon equivalent, which is one RIN, and the generating event is a meter reading plus a contract showing the gas moved a vehicle. Behind it sits a wellfield, a compressor, an upgrading skid, an interconnect. Those projects fail too, usually at the pipeline rather than in the process (we've written before about why RNG projects stall at the injection meter), but they fail small and they fail early, which is the cheapest way to fail.

There's a sequencing problem sitting underneath all of this. A producer can't generate a RIN until EPA has approved the specific feedstock, process and fuel combination being run, so a new RFS waste feedstock pathway means a petition, a lifecycle analysis, and a wait measured in quarters rather than weeks. Landfill gas has settled pathways and a decade of precedent behind it. A novel route from household waste to jet fuel is asking a regulator to bless something nobody has yet run at scale, while the construction loan accrues interest in the meantime.

A thermal MSW-to-liquids plant has to do all of that and then build a first-of-kind reactor, hold a syngas spec through feedstock swings a landfill never sees, and run a Fischer-Tropsch train downstream. Same credit per gallon. The RFS pays for delivered volume, not for degree of difficulty, so capital walks toward the easy molecule. Gasification and pyrolysis systems aren't the binding constraint here. The payment mechanism is.

What Fulcrum actually proved

Fulcrum BioEnergy's Sierra plant outside Reno was built to take 175,000 tons of household waste a year and turn it into roughly 11 million gallons of jet fuel. It shipped a small volume of synthetic crude in late 2022, then shut down for repairs, restarted, and hit further trouble on the gasification island. Most of the staff went in May 2024.

Chapter 11 followed that September, and according to the filing coverage at the time it listed liabilities between $100 million and $500 million against assets under $10 million. A WM subsidiary took the feedstock prep building out of the estate in November.

Most people read that as a gasifier problem. I read it as a capital-structure problem. A first-of-kind plant's worst operating years are its first two, which is exactly when debt service starts and exactly when a credit-weighted revenue line pays least. RINs settle on gallons that leave the gate. They pay nothing at 30% availability, and 30% availability in year one is the base case, not the stress case.

Notice which assets cleared the auction. The building that sorts and prepares the waste went to a waste company. The site where the fuel was supposed to be made went to a data center operator.

One revenue line on that project had a counterparty's signature under it, and it wasn't the credit. It was the tip fee, which is the only thing on the page that survives a bad reactor.

And the waiver credit itself can't be traded or banked for a future year, so no secondary market forms around it and no arbitrage grinds the spread. A credit you can't bank or trade is a footnote in somebody else's compliance filing, not a revenue line in yours.

The objection I get most

Renewable diesel from waste clears fine. The D4 market is deep, the plants run, the fuel sells. Why should municipal waste be any different?

Because "waste" is carrying enormous weight in that sentence. Used cooking oil and rendered fat arrive by tanker with an assay, a spec, and a published price curve. They're byproducts with a commodity market already wrapped around them, and the conversion step is hydrotreating inside a refinery that already exists. Municipal solid waste arrives in a packer truck with whatever the route happened to collect (and the route mix moves week to week), where the first unit operation is a Vecoplan grinder and an optical sort line trying to manufacture a consistency that was never there.

EPA's final standards for 2026 and 2027, issued in March 2026, set biomass-based diesel at 9.07 billion RINs for 2026 against 1.36 billion for cellulosic. Read that split as the program's own forecast of where the gallons come from. It isn't from municipal solid waste.

The same rule struck renewable electricity out of the program altogether. Anyone who spent 2023 writing a business case around eRINs learned the lesson from the other side: an instrument created by rulemaking can be deleted by rulemaking. That rule also reallocates 70% of the small refinery exemptions granted for 2023 through 2025, which shuffles compliance demand in ways no sponsor influences and few pro formas bother to model.

Where this argument breaks

There are limits to how far I'd push this. The lid only binds while EPA keeps waiving. If cellulosic supply ever caught the statutory volume, waiver credits would stop being offered and D3 would price off genuine scarcity. I don't expect that this decade. But a view I hold strongly and can't prove is still a view, and anyone underwriting a twenty-year tenor should carry the scenario where the ceiling lifts.

It's a federal-only argument, too. In California, Oregon and Washington the low-carbon fuel credit is often the marginal dollar rather than the RIN, and carbon intensity scoring does reward the difficult feedstocks in a way the RFS never has. A deeply negative CI pathway changes the answer outright. Definitions decide eligibility more often than chemistry does, which is the same fight we picked through on the power side over when renewable energy from waste counts as clean power.

Scale cuts the other way as well. Below a few million gallons a year the fixed cost of participating starts to swamp the value: registration, an approved quality assurance plan, third-party engineering review, annual attest engagements. On a small plant that's a line you can see from the boardroom.

I've been wrong on escalation before, in a way that maps straight onto this. In 2022 I worked a gate-fee indexation dispute where a CPI-linked contract lagged actual tipping inflation by 14 months and ate most of the equity return over the period. Nothing about it was hidden. I'd modeled the escalator and skipped the lag mechanism sitting underneath it. Modeling a RIN price without modeling the formula that caps it is the same mistake wearing different clothes.

What I'd actually underwrite

Size the debt on a waste to biofuels project off the tipping fee and the metered molecule. Underwrite the D3 RIN at the waiver-credit ceiling rather than last quarter's print, and treat anything above that as equity upside nobody promised you. If the credit clears higher for a decade you'll have been conservative, and conservative is a survivable way to be wrong. The reverse is not.

Lenders already work this way, which is the part sponsors keep missing. Every credit committee I've presented to haircuts uncontracted credit revenue toward zero when it sizes debt, because a compliance credit has no counterparty and no term. So the RIN was never going to carry the leverage. It lands in the equity case instead, where a statutory ceiling hurts more rather than less, because equity is the tranche that owns the upside and the upside is the part Congress already fenced.

For sponsors whose actual mandate is diversion rather than fuel, it's worth asking whether a fuels pathway is even the shortest route. Plenty of zero-waste-to-landfill solutions reach the same tonnage target without asking a first-of-kind reactor to carry a leveraged balance sheet, and they get there on contracted revenue instead of a compliance credit set by a gasoline formula.

Waste to biofuels isn't a fuels business with a waste input. It's a waste business that sells a fuel, and the credit market has been saying so for a decade in the only language it has.

Sources & Notes

• Volumes for 2026 and 2027, the eRIN removal and the small refinery reallocation all come from EPA's final rule on the 2026 and 2027 standards, finalized 27 March 2026.
• Waiver credit mechanics, the $1.91 figure for 2025, and the fact the credits can't be traded or banked: EPA's page on cellulosic waiver credits.
• For the 99% share and the 55 Bcf volume I used EIA's Today in Energy note on RFS targets and RNG production, published December 2023.
• The 77,000 Btu per RIN conversion is EPA's own, set out in its landfill methane program guidance on the RFS.
• Fulcrum's timeline, plant design basis and bankruptcy figures come from Waste Dive's reporting on the Chapter 11 filing and its follow-up on the asset auction. The indexation figures are mine, from a 2022 engagement.