The Defense Production Act Just Entered the Refining Business — Refiners Want the Money for Retrofits, Not New Plants

On September 11, Reuters reported the White House is weighing how to use the Defense Production Act — a Korean War-era statute normally reserved for military procurement — to expand US oil refining. The statute has never been used to add refining capacity at scale, but the legal scaffolding is already in place: on April 20, 2026, the President signed five presidential determinations under Section 303 covering petroleum production, refining, and logistics, coal supply chains, natural gas and LNG, grid infrastructure, and large-scale energy systems, all pursuant to Executive Order 14156's declared national energy emergency.

The DPA hands the executive branch three concrete tools. Title I lets the government prioritize and allocate contracts — forcing refiners to put defense orders ahead of commercial customers and adjust product slates toward diesel, jet fuel, and military-grade products. Title III authorizes loans, loan guarantees, purchase commitments, and direct federal purchases for capacity expansion. Title VII, via §708 voluntary agreements, gives antitrust immunity to refiners who coordinate production and distribution with each other under federal supervision. Combined, these tools reframe refiner capex from a return-on-capital decision into a defense procurement contract.

The political pressure is concrete. National average diesel crossed $6 a gallon for the first time on record in the week of September 8, 2026, up 61% from $3.71 a year earlier, as a six-month conflict with Iran disrupts seaborne crude flows. US refinery utilization is 98.0% as of the EIA weekly release for August 28, 2026 — there is no spare capacity to absorb a supply shock. With the November midterms approaching, the administration is searching for an instrument that delivers visible action without waiting for a permitting cycle or a greenfield project.

The DPA already has a refining-adjacent precedent. On March 14, 2026, the Energy Secretary invoked the DPA to compel Sable Offshore to restart the Santa Ynez Pipeline System in California — the first DPA order aimed at a specific oil and gas operator. California sued; on August 21, 2026 a federal judge dismissed the state's challenge and authorized Sable to continue operating. That ruling established that DPA orders override state environmental and pipeline-safety law when the executive branch certifies a defense-related shortfall.

According to the Reuters report, in a recent White House meeting with nearly a dozen refiners, executives told officials that federal money would be better deployed retrofitting existing refineries — debottlenecking units, adding hydroprocessing capacity, or expanding adjacent plants — than financing an entirely new refinery. A greenfield build, refiners argued, would be considerably more costly and take years to come online. That advice is consistent with the data: at 98% utilization the bottleneck is not how much crude the system can process but how efficiently each barrel is converted into high-value products.

Marathon Petroleum is the standout spender, with TTM capex of $4.23B against a Q2 2026 65/35 split between value-enhancing and sustaining capital, and 25% of refining value-enhancing capital directed at its Garyville site. Phillips 66 is mid-build — net crude capacity rose 45 MBD to 1,993 MBD as of January 1, 2026 — and Q2 2026 reported earnings of $3.8B, roughly twenty times the prior quarter on higher crack spreads. Valero Energy runs a structurally leaner model at under $1B annual capex. Title III funding would lower the WACC for MPC's Garyville-style expansions and PSX's midstream-tied debottlenecking — those are the projects the refiners themselves want financed.

Whether the federal instrument finances a greenfield or a retrofit, the engineering and equipment supply chain captures a disproportionate share of the spend. Honeywell's UOP business is the dominant licensor of refining process technology — hydrocracking, fluid catalytic cracking, alkylation, and the catalyst packages that determine yield — and any Title III-funded expansion would translate into licensing fees, key equipment, and long-tail catalyst supply contracts. TechnipFMC is the largest pure-play EPC contractor in the refining and petrochemical space. Dover supplies pumps, compressors, and process equipment that go into every unit operation.

America First Refining's proposed 168,000 bpd facility at the Port of Brownsville, Texas — announced by President Trump in March 2026 as the first new US refinery in nearly 50 years — has emerged as the immediate test case for whether the DPA flows to greenfield or retrofit. The project is anchored by Reliance Industries's 20-year offtake commitment for the refinery's output. Investor disclosures name Donald Trump Jr. as a passive minority stakeholder and Cantor Fitzgerald — the firm founded by Commerce Secretary Howard Lutnick — as financial adviser, a political-alignment footnote that will shape whether DPA funds ultimately flow to this site.

If the administration treats Brownsville as a national-security priority, Title III financing could fund a meaningful share of the build. If refiners' retrofit-first advice prevails, Brownsville is left to commercial capital while existing refiners absorb the federal support. The split matters less for absolute refining capacity — Brownsville's 168,000 bpd is roughly 0.9% of US operable capacity — and more for what it signals about the administration's willingness to override market structure. The April 20, 2026 PD waives the normal Section 303(a)(1)–(a)(6) requirements, clearing the legal path for either outcome.

The DPA is also being used to secure crude feedstock rather than refining capacity. The Trump administration secured a 35% US government equity stake in North American Blue Energy Partners, a private Venezuelan operator tied to ~65 billion barrels of proven reserves across 17 oil fields. The deal gives the US government the right to purchase Venezuelan crude, including 20% of the company's output at production cost. The White House has said millions of barrels of new Venezuelan production will eventually be processed through US refineries — meaning ExxonMobil, Chevron, and Occidental Petroleum, all with prior Venezuelan asset experience, become the natural offtake counterparties as soon as those barrels land.

Refiners with Gulf Coast and mid-Continent exposure — Marathon Petroleum, Phillips 66, and Valero Energy — are best positioned to process discounted Venezuelan crude. The Venezuela angle effectively decouples crude supply from the Persian Gulf exposure that drove the Iran-driven diesel spike, reducing the strategic justification for an emergency refining buildout in the first place. That contradiction — using the DPA to add Venezuelan barrels while also using it to add refining capacity — is the most underappreciated tension in the current policy mix.

The short-term investment play is not refining volume — it is the capex visibility the DPA confers on the refining complex. A federal backstop for retrofit projects changes the discount rate on projects like MPC's Garyville expansion, PSX's 1,993 MBD footprint extensions, and idled-plant restarts such as the St. Croix facility that Port Hamilton directors confirmed in May 2026 they intend to restart. The 1–3 year story is whether the administration actually channels Title III dollars into new builds or uses the statute's authority as leverage to extract voluntary expansion commitments without spending. The Sable Offshore precedent — invoked March 2026, litigated, upheld August 21, 2026 — shows the statute is usable as a threat even when no money changes hands.

The risks cut both ways. Refiners' retrofit-first advice may simply mean no Title III money flows at all — the statute authorizes funding but does not appropriate it, and Congress has not yet weighed in. California's litigation against the Sable order shows the statute is contestable. The Brownsville project's political entanglements raise a separate conflict-of-investment risk. And the Venezuelan crude side door could deflate the refining-capacity argument before any new build breaks ground. Refiners and engineering vendors should price in policy optionality, not policy commitment.