EPA’s ‘Set 3’ Rule Nears Release, Carrying the Biofuel Sector’s Biggest Questions Into 2028 and Beyond

Forthcoming proposal will set post-2027 RVOs under a transformed policy framework — with SRE treatment, reallocation math and 45Z tax-credit mechanics all converging on the same rulemaking

The Environmental Protection Agency (EPA) is expected to release, perhaps within weeks, its proposed “Set 3” rule establishing Renewable Fuel Standard (RFS) volume requirements (RVOS) for 2028 — and potentially for multiple years beyond. The clock is unforgiving: under the Clean Air Act’s set authority, EPA must finalize applicable volumes no later than 14 months before the compliance year to which they apply, putting the statutory deadline for the 2028 standards at Oct. 31, 2026. A summer proposal is therefore not a courtesy to markets but a procedural necessity — leaving room for a comment period, a hearing and interagency review before an on-time October final. For agriculture, the stakes arguably exceed those of the record-setting 2026-27 volumes finalized in March, because Set 3 will establish the first Renewable Volume Obligations (RVOs) under a fundamentally restructured program — one in which import penalties, small refinery exemption (SRE) policy, reallocation mechanics and the Section 45Z Clean Fuel Production Credit all interact in ways the market has never priced before.

Note: EPA has not yet sent its proposed rule to the Office of Management & Budget (OMB).

— Why Set 3 is different

When EPA finalized the Set 2 rule on March 27, it set 2026 and 2027 volumes at the highest levels in the program’s 20-year history, maintained the 15-billion-gallon conventional ethanol level, and pushed biomass-based diesel (BBD) requirements to levels the agency estimates will require production and use to climb more than 60% above 2025. But the agency deliberately deferred the most structurally consequential piece of its June 2025 proposal: the import RIN reduction (IRR). Under that policy, renewable fuels produced from imported feedstocks — or imported as finished fuels — would generate only half the Renewable Identification Number (RIN) compliance value of fuels made from qualifying domestic feedstocks. EPA said in the final Set 2 rule that it intends to establish IRR provisions taking effect “beginning in the 2028 compliance year or shortly thereafter,” but that implementing them requires a new rulemaking. Set 3 is that vehicle. 

That means the coming proposal must do two hard things simultaneously: set headline volumes for 2028 (and possibly 2029-2030), and stand up the operational machinery of the half-RIN import penalty — feedstock tracing, country-of-origin verification, and possibly a country-by-country application that EPA floated for comment in the original Set 2 proposal.

— The RVO calculus

The central volume question is whether EPA maintains the aggressive growth trajectory it established for 2026-27 or moderates it once the IRR takes hold. The domestic-feedstock pivot cuts both ways. On one hand, penalizing imported used cooking oil, foreign tallow and imported canola oil should channel demand toward U.S. soybean oil, distillers corn oil and domestic animal fats — supporting the billions of dollars in Midwest crush capacity built on expectations of sustained renewable diesel growth. On the other, EPA must now judge how much renewable fuel the market can realistically deliver when a meaningful share of the current feedstock slate is devalued by half for compliance purposes. 

The key: Set aggressive BBD and advanced targets for 2028-2030, and the agency validates the crush-expansion wave and supports soybean oil demand, crush margins and farm income. Take a conservative tack, and the market will begin asking whether recent processing investments mark the start of another growth cycle or the high-water mark of the renewable diesel boom. 

A second structural question is scope — and here the history is widely misread. The big difference between Set 3 and prior rulemakings that finalized multiple years at once is that EPA is not working from behind. The multi-year approach was born of delinquency: the agency bundled 2023-25 into Set 1 under a court-approved consent decree after missing the Oct. 31, 2021, statutory deadline for the 2023 volumes, and swept 2026-27 together in Set 2 after blowing past the October 2024 deadline for 2026 and drawing another litigation threat. Bundling years was how EPA dug out of a hole, not an effort to provide extended certainty — certainty was the byproduct of being late, not the motive. Having finalized 2026 and 2027 in March, the agency enters this cycle current for the first time under the set authority. Only 2028 is due, and an on-time final by Oct. 31 delivers obligated parties the full 14-month statutory lead time to plan, no bundling required. That reality argues for a single-year 2028 rule as the cleaner path; a 2028-2030 package remains possible if EPA wants to lock in the import RIN framework across multiple years at once, but the historical precedent for multi-year rules does not predict one here, because the driver behind those precedents is gone.

— SREs: the recurring wildcard

No RFS rulemaking in recent memory has escaped the gravitational pull of small refinery exemptions, and Set 3 will be no exception. EPA’s August 2025 action on the backlog of 175 SRE petitions — 63 full grants, 77 partial grants under the restored 50% partial-hardship framework, 28 denials and seven ineligibility findings — re-established a policy of granting relief generously relative to the Biden-era posture, and the agency has since continued granting additional exemptions, including reconsideration of previously denied petitions. 

For Set 3, the operative questions are threefold. 

First, how many exemptions will EPA project for the 2028 compliance year when it converts volumes into percentage standards? The agency’s SRE projection methodology directly inflates or deflates the effective blending burden on non-exempt obligated parties. 

Second, will EPA continue its practice of returning previously retired RINs to refineries granted retroactive exemptions — a mechanism that injects supply into the RIN market and softens effective mandates? 

Third, and most consequential for agriculture: what happens to volumes exempted in 2026 and 2027, the years now underway, when EPA grants the petitions that will inevitably be filed for them? On that last question, the agency’s own words in the Set 2 final rule supply an answer — one that reshapes the reallocation debate entirely.

— Reallocation: how much of the leakage gets restored

The Set 2 final rule answered the reallocation question for 2023-25 exemptions with a compromise: a 70% partial reallocation of exempted RVOs into the 2026 and 2027 standards, roughly 0.99 billion gallons added in 2026 and 1.04 billion in 2027. EPA framed the 70% figure as balancing biofuel demand protection against the need to preserve a liquid carryover RIN bank and a smoothly functioning credit market. Biofuel and farm groups — which had pressed for 100% reallocation, with Clean Fuels Alliance America warning that unreallocated exemptions could cost soybean farmers as much as 40 cents per bushel and strip billions in crop value — accepted the outcome as directionally positive but incomplete. Refiners, led by American Fuel and Petrochemical Manufacturers, denounced reallocation as a mandate for higher costs. 

But on the question of whether 70% becomes the template, EPA has already shown its hand. The final Set 2 rule states that the equations incorporating the SRE reallocation volume will be used only for the 2026 and 2027 percentage standards, and that the agency intends to continue prospectively accounting for exempted volumes of gasoline and diesel going forward — such that reallocation volumes of this kind will never need to be included again. In plain terms: the 70% adder was a one-time cleanup of the 2023-25 backlog, not a recurring mechanism. Under prospective accounting, projected exempt volumes are built into the percentage-standard formula up front, automatically shifting the burden to non-exempt obligated parties before the compliance year begins. That retires the reallocation fight in its old form but replaces it with a new one: everything now rides on the accuracy of EPA’s exemption projection. If the agency under-projects 2028 exemptions and then grants relief beyond what the formula assumed, the resulting demand leakage is never recovered — there is no retroactive true-up in this framework. That is the biofuel and farm groups’ exposure, and it is why the projection methodology in the Set 3 proposal, not a reallocation percentage, will be the number they scrutinize hardest. Refiners carry the mirror-image concern that over-projection inflates the effective mandate on everyone else. The battleground has moved from restoring past leakage to preventing future leakage before it happens.

— The 45Z overlay

Set 3 will not be written in a vacuum. The Treasury Department’s proposed Section 45Z regulations, published Feb. 4 and now moving toward finalization following a May 28 public hearing, govern the tax-credit side of the same production economics the RFS governs on the mandate side — and the two programs are converging on the same domestic-feedstock logic. 

The One Big Beautiful Bill Act extended 45Z through 2029, removed indirect land use change from lifecycle emissions calculations (a change that materially improves corn ethanol and soybean-based fuel scores), and restricted eligible feedstocks to those produced or grown in the United States, Canada and Mexico. 

The RFS import RIN reduction, by contrast, draws the line at the U.S. border. That asymmetry matters: Canadian canola oil, for example, retains 45Z eligibility but would take the half-RIN haircut under the IRR as proposed. 

Producers optimizing across both programs face a matrix in which feedstock origin determines both tax-credit value and compliance-credit generation, and where the highest-value play — domestic soybean oil, corn oil and animal fats — is the same in both. That alignment is precisely what soybean groups and the National Oilseed Processors Association have lobbied for, with NOPA explicitly tying its 45Z support to a “strong RFS” that includes the import RIN mechanism.

— Where the programs mesh — and where they grind

The interaction runs deeper than feedstock geography. RIN prices and 45Z credit values are economically fungible components of the same producer margin: a robust 2028 RVO raises RIN values and reduces the burden on 45Z to carry project economics, while generous 45Z monetization (credits currently trade at roughly 88-93 cents on the dollar in the transfer market) allows producers to tolerate softer RIN prices. 

EPA’s volume-setting analysis for 2028-2030 will implicitly embed assumptions about 45Z’s stimulative effect on supply — yet key 45Z variables remain unsettled, including annual updates to the 45ZCF-GREET model and the credit’s scheduled expiration after 2029. One major piece did just fall into place: on June 25, alongside a presidential executive order on regenerative agriculture, USDA announced its final Regenerative Feedstock Rule — published in the Federal Register June 29 — and released the operative version of its Feedstock Carbon Intensity Calculator (FD-CIC), covering corn, soybeans, sorghum and spring canola and letting farmers monetize no-till and reduced tillage, cover crops and nutrient management through lower feedstock CI scores. 

But the plumbing into the tax credit is not yet connected — DOE and Argonne must still incorporate the calculator as a module in 45ZCF-GREET, and Treasury must issue guidance allowing taxpayers to claim the enhanced credit values. Treasury’s February proposed regulations anticipated the calculator may be used for fuel produced and sold in 2025, and by industry accounts the agency intends to permit its use for 2026 as well. 

That last point creates a cliff problem for a multi-year Set 3: if EPA sets 2030 volumes assuming 45Z-supported production economics, and Congress lets the credit lapse, the mandate could land on a market suddenly stripped of its per-gallon tax support. Conversely, the certainty of a multi-year RVO could strengthen the case on Capitol Hill for extending 45Z. 

Producers, blenders and feedstock suppliers should read the two rulemakings as a single policy architecture — one that Washington is assembling piece by piece, on parallel tracks, with the seams still visible.

Bottom line

Set 3 is shaping up as the most consequential biofuel rulemaking since the original statutory tables ran out. It will determine whether the domestic-feedstock realignment announced in Set 2 gets operational teeth, whether SRE-driven demand leakage is structurally repaired or perpetually relitigated, and whether the RFS and 45Z function as reinforcing pillars or misaligned overlays. With the 14-month statutory clock requiring a final rule by the end of October, the proposal-to-final window will be compressed — and the pressure to resolve the IRR mechanics, SRE projections and reallocation policy in a single pass is correspondingly intense. What EPA puts in it will tell corn growers, soybean farmers, crushers and renewable diesel investors whether Washington intends the biofuel boom to keep compounding — or to plateau.