U.S. Oil, Gasoline and Diesel History

U.S. Oil, Gasoline and Diesel History Dossier

Stoney Hughes — October 1, 2026

1970–October 1, 2026
Evidence, annual statistics, inflation, presidential policy and project status

Prepared for Stoney Hughes. Source-based research with primary government and company records. Completed years are separated from partial 2026 observations and forecasts.

Reading guide

1–2 Findings, methods and evidence rules

3–5 History, Reagan and current conditions

6–9 Projects, refining, taxes, profits and SPR

10–12 Presidential authority, policy and claim ratings

A–D Annual numeric tables and decadal summaries

E Reproduction instructions and linked source register

1. Findings and substantive answer

U.S. pump prices are the outcome of a global crude market and a regional fuel-manufacturing and delivery system. Presidents influence both, but no president sets the world clearing price, compels profitable private investment, or turns an undeveloped lease into next week’s diesel. The annual data show large price swings under both parties, alongside production changes that often began before an administration took office. Price movements alone are not a causal evaluation of presidential policy. [1–4,20,43–44]

Reagan’s January 28, 1981 order genuinely ended the principal federal petroleum price and allocation controls. It accelerated a transition Carter had already begun; it was not simply an order to drill more. The largest 1980s price break occurred in 1986, five years later, when weak demand and expanding non-OPEC supply met Saudi Arabia’s shift toward defending market share. U.S. composite crude acquisition cost fell from $26.75/barrel in 1985 to $14.55 in 1986; city-average unleaded gasoline fell from $1.202 to $0.927/gallon. Those are declines of 45.6% and 22.9%. Domestic crude output also fell, rather than surging to cause that collapse. [1–3,14–17]

Biden’s day-one pipeline action was revocation of the Keystone XL cross-border permit in section 6 of EO 13990 on January 20, 2021. It was a Canada-to-U.S. crude project, not an Alaska pipeline, and it had never become an operating 830,000-barrel/day system. TC Energy terminated the project in June 2021. TAPS kept operating; Willow was approved in revised form in March 2023; Alaska LNG retained a Biden-era export authorization. The 2026 Bridger border permit concerns a separate new project. [24–31,36–42]

As of October 1, 2026, the latest EIA national weekly observations are $4.465/gallon regular gasoline and $6.382 diesel, for September 28. High domestic crude production has not insulated buyers from the international disruption. EIA’s September outlook identifies Middle East supply/shipping constraints and tight global distillate availability. Restoring disrupted supply, shipping and reliable refinery output offers a faster route to relief than a multiyear new Arctic field or pipeline. A project can improve security and transportation economics without producing a large, immediate or measurable national pump-price reduction. [11–12,28,35,38–42]

The strongest policy package combines emergency supply coordination when warranted; reliable refining, terminals and product transportation; predictable permitting for economically viable supply; and sustained reductions in fuel required per mile or unit of freight. Tax relief can cut an identifiable part of the bill if passed through, but it costs revenue and does not remove crude or refinery scarcity. Environmental and safety protections carry costs and benefits; relaxing a rule should be judged against a specific bottleneck and public-health tradeoff. Neither “drill everywhere” nor “one canceled pipeline explains everything” is an adequate account. [43–51,54–55]

2. Scope, definitions and evidence rules

Cutoff: October 1, 2026, America/Chicago. This is a source-based historical and economic dossier, not a claim of professional licensure or an investment forecast. “Verified” means directly supported by the cited record, not immune to later statistical revision. Administrative announcements establish what an agency announced; operative orders establish legal scope; developer statements establish company plans and reported progress, not guaranteed commissioning dates.

Evidence vocabulary

A barrel is 42 U.S. gallons. “Crude production” here includes lease condensate, not natural-gas plant liquids, refinery processing gains, biofuels or all liquids. Imports can be crude alone, total petroleum gross imports, or imports minus exports. These measures answer different questions. Refinery capacity is operable atmospheric crude distillation in barrels per calendar day, measured January 1. It is not a gasoline/diesel output guarantee or a count of finished-fuel facilities. [3–7,56]

The dossier uses composite refiner acquisition cost (RAC) for a long crude-price history. RAC is the volume-weighted cost of domestic and imported crude acquired by U.S. refiners; it is neither WTI nor Brent spot price. Domestic first-purchase prices are shown separately because controls and domestic grade differences mattered. An annual average masks intra-year peaks; comparing an annual average with a weekly high exaggerates the historical change. [1]

The early gasoline record is not one seamless modern survey. Table A preserves leaded regular city-average, unleaded regular city-average and modern EIA regular all-area prices in separate columns. For real-price comparisons only, a marked historical regular series uses leaded city average in 1970–1975, unleaded city average in 1976–1991, and EIA regular from 1992. This splice changes grade and survey population. It is useful for broad purchasing-power comparisons, not precise cross-break causal estimates. EIA’s diesel survey begins in 1994, but the first complete annual average in these downloads is 1995. Earlier diesel cells are unavailable, not zero; heating oil and wholesale distillate are not substituted. [2]

Inflation method: real price in August 2026 dollars = nominal annual price × 334.980 ÷ the year’s published annual CPI-U. CPI is all items, U.S. city average, not seasonally adjusted, 1982–84=100. This expresses August 2026 purchasing power; it is not a full-year 2026 CPI base. For 2026 monthly comparisons, the corresponding monthly CPI is used. BLS did not collect an October 2025 observation; its published 2025 annual average is retained with that caveat. The September 2026 CPI and full-year 2026 averages were unavailable at cutoff. [8–10]

Annual EIA records are downloaded CSV observations with YYYYMM=YYYY13, rather than averages recomputed from rounded months. Flow values are divided by 1,000 to express million barrels/day; refinery capacity is divided by 1,000,000. Tables round prices to three decimals at retail and two at crude; computations use source precision. Decadal averages are simple means of available annual values, with explicit sample counts, not population-weighted prices or averages of rounded displayed cells. The 2020s comparison covers 2020–2025 only. Historical revisions may change later downloads. [1–7]

3. Price history and the world oil market

1970s: domestic controls meet external shocks

The United States entered the 1970s with domestic crude output near 9.64 million barrels/day, then faced declining conventional output and rising imports. Nixon’s 1971 wage-price program and subsequent oil-specific controls treated categories of domestic oil differently. The Emergency Petroleum Allocation Act of 1973 provided the allocation/control framework later used in EO 12287. Controls could redistribute costs and restrain particular posted prices, but could not manufacture additional imported barrels at controlled prices. Queueing, shortages and allocation rules were distinct from the world-price shock. [3–4,14–15]

The 1973–1974 Arab oil embargo was undertaken by Arab exporting countries, often discussed as OAPEC; it should not be equated with every OPEC member. Production cuts, embargo destinations and OPEC price decisions reinforced market stress. RAC rose from $4.15 in 1973 to $9.07 in 1974. Even when supplies were rerouted, a smaller pool of globally available oil and fear of further cuts raised costs beyond directly embargoed buyers. [1,20]

The second shock began with strikes and the Iranian Revolution in late 1978–1979. EIA documents a dramatic initial loss of Iranian production; other producers offset part of it. The Iran–Iraq War began in September 1980 and damaged both countries’ supply capability. The initial disruption, subsequent annual losses, precautionary inventories and panic buying are different measures. It is wrong to add every quoted disruption figure together as a simultaneous global loss. RAC increased from $12.46 in 1978 to $17.72 in 1979 and $28.07 in 1980. [1,18–19]

1980s: high prices create the conditions for a collapse

High 1970s prices encouraged conservation, substitution away from petroleum, more efficient vehicles and industrial equipment, and non-OPEC exploration. Alaska, the North Sea and other new supply took years to develop. Recessions and monetary tightening reduced demand. OPEC attempted to sustain prices through output restraint; Saudi Arabia absorbed a large share of the reductions until that strategy became economically and politically difficult. This is a supply-demand sequence spanning multiple governments and countries. [16–17,19–20,30]

EIA’s first-quarter 1986 outlook reported Saudi output rising from about 2.4 million barrels/day in June 1985 to roughly 4.6–4.8 million in January 1986, alongside a shift away from acting as swing producer. Its third-quarter report described the large decline in OECD oil demand since 1979. Demand reductions, competitors’ supply and Saudi pricing/output strategy are directly documented contemporary explanations. Exact attribution of each factor would require a counterfactual model, not a chart alone. [16–17]

1990s and 2000s: war risk, weak demand and renewed demand growth

Iraq’s 1990 invasion of Kuwait created a sharp supply-risk episode; annual averages conceal the rapid rise and later reversal. By 1998, a weak international demand environment coincided with an exceptionally low annual RAC of $12.52/barrel. The following decade brought strong global demand growth, supply constraints, geopolitical risk and refinery interruptions. Hurricanes can remove finished-product capacity even if world crude remains available. The 2008 boom ended in a worldwide financial crisis and demand contraction: annual RAC fell from $94.74 in 2008 to $59.29 in 2009. [1,20,43–44]

2010s: shale changes trade, but global prices still matter

Horizontal drilling, hydraulic fracturing, private investment and infrastructure expansion raised U.S. output from 5.48 million barrels/day in 2010 to 12.34 million in 2019. The 2014 price decline reflected rising supply, changing expectations and weak demand growth; EIA documented Brent futures falling from $108 in early January to $57 at year-end. OPEC’s restraint choices mattered, but no single U.S. president created or controlled this multinational adjustment. Added pipelines also narrowed particular inland crude discounts without necessarily lowering every destination refinery’s crude cost. [3,21,26,56]

The OPEC+ framework adds cooperating non-OPEC producers, notably Russia, to coordination among OPEC members. An announced quota is not actual output: compliance, sanctions, field capacity, shipping and voluntary cuts can all differ. OPEC’s influence is greatest when spare capacity is concentrated and other producers and consumers cannot quickly adjust. It is not an unrestricted ability to pick any price without losing demand or market share. [20,12]

2020–2025: pandemic, recovery and war

COVID-19 restrictions sharply reduced travel and oil demand in 2020. The expiring WTI futures contract went negative on April 20 amid delivery/storage stress; this did not make annual U.S. crude acquisition cost or retail gasoline negative. As demand recovered in 2021, supply and inventories adjusted more slowly. Russia’s 2022 invasion and disrupted trade added to a recovery already straining fuel markets. RAC reached $95.29 in 2022; EIA regular gasoline averaged $3.951 and diesel $4.989. Diesel’s separate refining and international-trade balance helps explain its disproportionate increase. [1–2,22–23,44]

By 2025 U.S. crude production averaged 13.662 million barrels/day, while EIA regular gasoline averaged $3.097 and diesel $3.660. These outcomes undercut the claim that Biden’s Keystone action stopped U.S. oil production. They do not prove every Biden policy was costless or beneficial. Production reflects prior leases, private and state land, drilling productivity, capital discipline, prices and project lead times; a fair policy evaluation must account for those channels. [2–4,24–26]

4. Reagan: what changed, what did not, and what can transfer to today

EO 12287 was signed January 28, 1981, and published January 30 at 46 FR 9909. Section 1 exempted crude oil and refined petroleum products from price and allocation controls under the Emergency Petroleum Allocation Act of 1973. It retained a limited set of provisions temporarily, including transition-related requirements and certain state middle-distillate set-asides, until March 31. “Reagan removed all government controls on energy” is therefore too broad. He did not abolish environmental law, pipeline safety, fuel taxes or every petroleum rule. [14]

Carter announced phased decontrol on April 5, 1979, with completion planned for October 1981. Reagan accelerated the endpoint by about eight months. This removed distortions between controlled categories of domestic oil, allowed price signals to encourage supply and conservation, and changed allocation incentives. These mechanisms are economically plausible; their magnitude cannot be recovered from the subsequent decline in prices alone. World crude was expensive around decontrol, and the major global collapse came in 1986. [14–17]

Reagan-era annual record; gasoline is the city-average unleaded series [1–3,57]

Oil-and-gas rig counts are not oil-only drilling or completed wells. Even on that broad measure, activity peaked early in the 1980s and fell sharply after the price collapse. U.S. annual crude production reached 8.971 million barrels/day in 1985 and fell to 8.140 million in 1988, below the 1980 level. It is false to explain cheaper late-1980s fuel simply as Reagan continually increasing domestic drilling and oil output. Alaska’s TAPS entered service in 1977, before Reagan; its later peak reflected a long investment cycle. [3,30,57]

What transfers to 2026 is the principle of clear price signals, credible permitting and incentives for economically productive investment. What does not transfer is a second removal of the same 1970s federal petroleum price-control system: those principal controls are already gone. Today’s marginal constraints can be global shipping disruptions, distillate refinery output, infrastructure, labor, reservoir decline, finance or market risk. Removing an unrelated rule does not recreate Saudi spare capacity or commissioning-ready refineries. Lower prices themselves discourage drilling, so a promise of more supply and permanently lower producer revenues contains an investment tension. [12,14,43–46]

5. Current conditions as of October 1, 2026

The latest national weekly retail survey at cutoff is September 28, published September 29. It is an observed point in time, not the 2026 annual average. The September STEO is a forecast issued earlier in the month, based on information through September 3; it does not incorporate every later event. EIA’s outlook reported August Brent averaging $91/barrel versus $84 in July and estimated August Middle East crude shut-ins at 6.720 million barrels/day versus 4.980 million in July. Its estimated March–June average shut-ins were 9.435 million barrels/day. These are EIA estimates of disrupted production, not a permanent reduction in installed capacity. [11–12]

EIA described constraints around the Strait of Hormuz, attacks affecting shipping, and disruptions to alternative export routes. It also identified tight distillate availability and high U.S. diesel exports. Its expectation that U.S. distillate stocks would fall below 100 million barrels in September was a forecast, not a completed weekly inventory measurement. Crude production at home can remain high while globally traded diesel becomes scarce. [12]

Observed regional retail prices, September 28, 2026 [11]

The national diesel premium over regular gasoline was $1.917/gallon in this observation. California’s gasoline price exceeded the Gulf Coast average by $2.265. These comparisons illustrate distinct regional and product markets; they cannot be decomposed into tax, climate-program and refinery-cost shares from the pump survey alone. State-specific blend rules, isolated transportation, refinery availability and tax differences all matter. [11,43–48]

Full-year outlook versus completed annual data [2–3,12]

Forecasts are not targets or guarantees. The September 28 retail readings can exceed the STEO annual averages because the latter include lower-price months and assumptions about the rest of the year. The annual averages of 2026 and any October 1 “daily national EIA price” are not substituted for the weekly observation. [11–12]

6. Keystone XL, TAPS and Alaska: exact project and legal distinctions

The canceled northern Keystone XL extension would have linked Alberta supply to Steele City, Nebraska, for onward movement through other pipelines to the Gulf Coast. DOE defines its U.S. border-to-Steele City extension as 875 miles and planned capacity as 830,000 barrels/day, including 730,000 for western Canadian and 100,000 for Bakken crude. Larger route-length descriptions include the Canadian portion or original southern proposal. The Cushing-to-Gulf segment became a separate operating project in January 2014. Treating every “Keystone” segment as canceled is wrong. [26–27]

Project comparison at the cutoff

Why the old XL project has not simply restarted

The reviewed record establishes termination by its developer, not a mothballed, fully permitted operating pipeline waiting for a valve to be opened. Restoring a presidential policy preference is not equivalent to restoring the company’s investment decision, financing, construction contracts, easements, route approvals or Canadian connection. A new venture must address the laws and permits applicable to its design. The April 30, 2026 Bridger permit explicitly retains other legal requirements and applies to a 36-inch border segment extending less than 2,000 feet to the first valve or pumping station. It should not be described as reinstatement or completion of the former XL route. [25–29]

Consumer-price impact is uncertain. A functioning pipeline can reduce transport expense, displace rail, improve refinery feed reliability and change regional crude discounts. But transport capacity is not identical to additional world oil production. Some transported oil would otherwise move by other routes; a higher source-region crude price may transfer part of the gain to producers. DOE’s 2022 review found consumer-price estimates inconclusive in changed market conditions and cited earlier modeling with no significant change in overall U.S. refining or trade costs. That is neither proof of exactly zero benefit nor support for a guaranteed dollar-per-gallon saving. [26,56]

ANWR: leasing and barrels are different stages

Congress directed Coastal Plain leasing in the 2017 tax law. The first sale occurred January 6, 2021; Biden’s day-one EO ordered review and a temporary moratorium, followed by a June 2021 suspension. Interior canceled seven remaining AIDEA leases on September 6, 2023. The January 2025 sale attracted no bids. Trump’s EO 14153 on January 20, 2025 directed renewed Alaska development policies, with subsequent agency implementation. Those actions concern leasing and resource development, not removal of existing fuel from a functioning line. [24,32–34]

BLM’s June 5, 2026 recap records 58 tracts covering 688,829 acres offered, nine bids from two participants, and five tracts covering 72,049 acres receiving bids. High bids totaled $3,741,528. These figures establish bidding interest; they do not establish discovery size, commercially recoverable reserves, permits to drill, financed development or a first-oil date. ANWR’s prospective resource estimates cannot responsibly be quoted as current daily production capacity. [35]

Willow and Alaska LNG: approvals did not eliminate lead times

Willow is in the National Petroleum Reserve–Alaska, not the Arctic Refuge. The March 2023 revised decision approved three drill pads and denied two. ConocoPhillips made its investment decision in December 2023; its 2026 first-quarter filing reports 50% completion after the winter construction season. First oil is anticipated in early 2029. Its target peak of 180,000 barrels/day is about 1.3% of 2025 U.S. crude production, but a peak target cannot be added to 2026 production as if already delivered. [3,36–39,61]

On June 13, 2025, the Ninth Circuit largely upheld the Willow decision but required further explanation of the agency’s alternatives analysis. It remanded without vacating the 2023 approval. That distinction allowed the operative approval to remain while the agency addressed the procedural deficiency; neither “all litigation disappeared” nor “the court canceled Willow” describes the ruling. This dossier does not claim every later docket filing has been exhausted. [60]

Alaska LNG is a natural-gas transportation and export undertaking. FERC authorized it in May 2020; DOE’s April 2023 supplemental decision retained export authorization with conditions. Glenfarne’s September 30, 2026 announcement welcomed more than $50 billion of Korean investment support and described progress toward FID. Announced support is not proof that all funds have been disbursed or the entire system commissioned. LNG could change Alaska gas availability and international gas trade; it does not directly feed gasoline or diesel refineries as crude. [40–42]

7. Refining, diesel and delivery constraints

Crude oil is an input, not a ready-to-use fuel. Distillation separates fractions; cracking and other conversion units change yields; hydrotreating removes sulfur; blending produces compliant finished products. A refinery’s configuration, crude assay, energy supply, hydrogen, maintenance and downstream logistics determine what can be made. Light tight oil and heavy sour Canadian oil are not interchangeable on a one-for-one basis for every refinery. Importing one grade while exporting another can be economically sensible even when the country is a net petroleum exporter. [43–44,56]

The annual count of refineries is a poor standalone supply measure: fewer, larger or upgraded plants can have greater capacity than many older plants. January 1 operable distillation capacity rose from 12.021 million barrels/day in 1970 to 18.160 million in 2026 while the count fell from 276 to 130. Capacity includes idle but operable equipment, and stream-day capacity exceeds calendar-day capacity because it excludes normal downtime. Utilization is a ratio with its own denominator, not a promise that a plant can operate indefinitely above its sustainable rate. [5–7]

The 2026 annual survey shows a decline from 18.423 million barrels/day on January 1, 2025 to 18.160 million on January 1, 2026. Houston’s LyondellBasell refinery and Phillips 66’s Los Angeles refinery ended petroleum refining during 2025; the Benicia closure affects the monthly 2026 fleet after the January snapshot. The July monthly series reported about 18.044 million barrels/calendar day and utilization of 96.5%. This is strong operation of the available fleet, not evidence that every region has surplus diesel capacity. [6–7,45–46]

Diesel competes with heating oil and other distillate uses, including international markets. Freight, agriculture, construction, industrial demand and winter heating can pull in different directions from passenger driving. Global refinery outages or trade restrictions can increase U.S. export demand and lift domestic opportunity costs. U.S. federal diesel tax is also six cents/gallon higher than gasoline’s, but that fixed difference cannot explain a suddenly much larger premium. Ultra-low sulfur requirements affect production cost and logistics; inventory scarcity and distillate margins can change far faster. [44,47–50]

A “crack spread” is the difference between product values and crude cost for a chosen yield basket. It is a gross indicator, not net profit after energy, labor, depreciation, financing, compliance and distribution. The simple calculation $10/barrel ÷ 42 = 23.8 cents/gallon is useful to illustrate an input scale, but it is not a prediction of a 23.8-cent retail change: actual yields, other costs, inventories, margins and pass-through matter. [43–44]

January 1, 2026 operable refinery fleet [7]

Moving crude to a Gulf refinery does not automatically move diesel to California or New England. Product pipelines, terminals, ships, rail, trucks, blend compatibility and geographic isolation can bind. Gulf Coast hurricanes are consequential because a large share of U.S. capacity is concentrated there. Adequate spare manufacturing and storage resilience can be more valuable during a disruption than another barrel of crude upstream of an unavailable refinery. [7,43–46,56]

8. Taxes, environmental rules and corporate earnings

Fuel taxes and fees: January 2026 snapshot, cents/gallon [47–49]

The federal figures include 18.3 cents gasoline/24.3 cents diesel excise plus a 0.1-cent leaking underground storage tank fee. Federal rates have been unchanged since October 1993. EIA state averages include generally applicable state excise, environmental and inspection fees, but exclude gross/net-receipts taxes and county/local taxes. They are not a complete state-by-state tax burden or the exact average tax embedded in every national gallon sold. January 2026 is the verified snapshot used here, not a claim that every state rate remained unchanged to October 1. [47–49]

At September 28 national prices, the federal component alone was about 4.1% of regular gasoline and 3.8% of diesel. Even removing the full federal tax could only directly remove 18.4 or 24.4 cents/gallon before market responses. Tax holidays may be partially absorbed by sellers or offset by stronger demand in a tight market; long-term infrastructure funding needs a replacement. Tax differences explain part of regional price levels, but cannot be invoked as the sole explanation of a national crude shock. [11,47–49; arithmetic and economic inference]

EPA’s highway diesel sulfur limit is 15 parts per million, with the transition beginning in 2006. Hydrotreating and cleaner fuel formulations require equipment, hydrogen and processing; regional gasoline blends reduce interchangeability. These costs must be weighed against pollution reduction and the ability to operate modern emission controls. A short emergency waiver can address a demonstrated compatibility shortage; blanket repeal is a different long-term policy and does not guarantee a commensurate reduction in crude prices. [43–44,50–51]

Other policy channels include the Renewable Fuel Standard and renewable identification number compliance, seasonal volatility specifications, state clean-fuel programs and emissions requirements. Their legal authority, implementation and cost vary. The AEO 2026 assumptions document is a useful record of rules included in a model, not a substitute for each operative rule or court decision. This report does not assign a universal cents-per-gallon “regulation cost,” because that would require region-, time- and refinery-specific evidence. [51]

Illustrative company results, global GAAP earnings—not U.S. pump margins [52–53]

Large earnings establish that owners benefited in the reported period; they do not by themselves establish illegal collusion or the fraction of each U.S. gallon attributable to profit. Scarcity can raise crude producer earnings while squeezing particular refiners or retailers, and product scarcity can raise refining margins even when crude is relatively cheaper. Market concentration deserves investigation when specific conduct or local evidence warrants it. “All high prices are greed” and “profits never matter” both bypass the needed market analysis. [43–44,52–53; economic inference]

9. Strategic Petroleum Reserve and national security

The SPR was established following the 1970s shocks under the Energy Policy and Conservation Act of 1975. It stores crude, primarily in Gulf Coast salt caverns, for emergency use; storage capacity and withdrawal capability are different from inventory actually on hand. The latest EIA weekly stock at cutoff was 283.767 million barrels on September 25, 2026. One must not substitute roughly 714 million barrels of authorized storage capacity for current available stocks. [13,55,58]

The Biden administration’s major 2022 emergency sale reduced stocks alongside other authorized sales and subsequent repurchases. Trump’s DOE on March 11, 2026 announced authorization to release 172 million barrels as part of an IEA-coordinated 400-million-barrel response. An authorization, solicitation, contract award, physically delivered barrel and future promised return are different stages. DOE’s announced future return of additional barrels is not counted as current inventory. [13,54–55]

SPR releases can add near-term crude supply, calm expectations and buy time during a physical disruption. They cannot indefinitely replace lost world production, and crude still needs transportation and refining. Their retail effect depends on whether crude or finished-product capacity is the binding constraint. Replenishment has fiscal and logistical costs; maintenance can limit withdrawal capability. The stock figure supports a smaller emergency cushion than a full reserve, but it is not a simple national “days until gasoline runs out” clock because normal production, imports, commercial stocks and changing demand continue. [13,54–55]

Security policy therefore includes diversified import and export routes, dependable allies, navigable chokepoints, refinery and grid resilience, cyber/physical protection, emergency inventories and efficient fuel use. A net-export position provides economic and geopolitical advantages, but globally connected prices mean an overseas disruption still reaches U.S. consumers. Gasoline affordability and national energy security overlap without being identical objectives. [12,20,56; synthesis]

10. Presidential energy-policy timeline and authority

Selected actions; policy dates do not identify price causation

Presidential authority can direct executive agencies within statute, propose legislation, decide certain border permits, administer authorized leases and reserves, and pursue diplomacy. It is bounded by Congress, appropriations, administrative procedure, courts, state/local authority, private property and commercial decisions. Congress sets federal tax law; state governments set state taxes. A presidential announcement does not override all those constraints. EO 12287 relied on an existing statutory control framework; the 2026 Bridger permit explicitly preserves other laws. [14,24,28,34,49,54]

Presidents can affect expectations immediately, emergency flows over weeks or months, and infrastructure over years. They cannot set OPEC’s willingness to supply, eliminate reservoir decline, guarantee a refinery’s profitability, or force global buyers to value U.S. fuel below export opportunities without an intervention that has further consequences. A meaningful judgment asks which instrument changed which constraint, how much, when, and relative to what counterfactual. [12,20,43–44,56; analysis]

11. Policy matrix: mechanisms, timing and tradeoffs

Evidence strength refers to the mechanism, not an estimated guaranteed price reduction. “High” indicates strong physical/accounting support; “moderate” means the mechanism is credible but size depends on conditions; “low” means a broad promise is not established. Directions below are conditional economic judgments, not an econometric forecast.

Price-relief and resilience options

A useful distinction is the household or freight bill versus the price per gallon. Efficiency can reduce annual expense even if the world oil price barely changes. Conversely, a policy that makes gasoline cheaper by suppressing economic activity is not automatically a desirable affordability strategy. Recommendations should be judged by reliability, total cost, health/environmental effects, fiscal expense and distribution of benefits, alongside pump prices. [16–17,43–44,50–51; synthesis]

12. Claim-by-claim evidence ratings

Claims tested against the record

Claims tested against the record (continued)

Appendix A. Annual nominal prices, 1970–2025

Crude: dollars/barrel. Retail: dollars/gallon, taxes included. “City” refers to the historical city-average price series, distinct from EIA regular all areas. Preserve parallel columns; do not blend them. — means no annual observation in the selected source series. Tables are divided into decade panels for readability. [1–2]

A1970. Annual nominal observations

A1980. Annual nominal observations

A1990. Annual nominal observations

A2000. Annual nominal observations

A2010. Annual nominal observations

A2020. Annual nominal observations

Appendix B. Annual purchasing-power comparisons

All real prices below are expressed in August 2026 dollars, using CPI-U 334.980 as the dollar base. Crude is RAC; historical regular gasoline uses the explicitly marked splice. L=leaded city regular (1970–1975), U=unleaded city regular (1976–1991), E=EIA regular (1992–2025). Diesel remains unavailable before 1995. 2025 uses the published BLS annual average despite missing October collection. [1–2,8–10]

B1970. Inflation-adjusted annual prices

B1980. Inflation-adjusted annual prices

B1990. Inflation-adjusted annual prices

B2000. Inflation-adjusted annual prices

B2010. Inflation-adjusted annual prices

B2020. Inflation-adjusted annual prices

Inflation changes the interpretation: the 1981 city-average unleaded price of $1.378 equals approximately $5.078 in August 2026 dollars. The 1986 $0.927 equals $2.833; the 1998 EIA regular price of $1.030 equals $2.117; the 2022 EIA regular annual average of $3.951 equals $4.522. Comparing these with a single September 2026 week still mixes frequencies and the older city/EIA surveys. The real table supports broad context, not a claim of exact historical equivalence in fuel quality or household affordability. [2,8–11]

Appendix C. Production, imports and refinery capacity by year

All flows in million barrels/day. Crude production includes lease condensate. Petroleum imports are gross crude plus products/other petroleum categories; net petroleum imports = gross imports − exports, so negative means net exports. Crude imports are gross and should not be subtracted directly from total-petroleum exports. Refinery capacity is operable atmospheric distillation, million barrels/calendar day, at January 1. Counts include idle operable facilities. — in 1996 and 1998 means no refinery survey. [3–7]

C1970. Supply, trade and refining

C1980. Supply, trade and refining

C1990. Supply, trade and refining

C2000. Supply, trade and refining

C2010. Supply, trade and refining

C2020. Supply, trade and refining

In 2025, petroleum imports were 7.937 million barrels/day and exports 10.785, leaving net imports of −2.848. Yet gross crude imports were 6.168 million barrels/day, alongside domestic crude output of 13.662. Crude quality, refinery configuration, geography and product trade explain why all four statements can be true. Natural-gas liquids and other non-crude streams help satisfy total liquids use; refinery volume gain also means a barrel of crude is not simply a barrel of identical final products. [3–4,56]

Appendix D. Decadal summary and partial 2026 record

D1. Simple means of annual price observations [1–2,8–10]

Gasoline means use the historical regular splice, so the 1970s and 1990s contain series breaks. Diesel’s 1990s mean covers only 1995–1999 (five years), not the whole decade. Prices are in $/barrel for crude and $/gallon for retail. Real columns use August 2026 dollars; 2020–2025 is six completed years.

D2. Simple annual means of physical measures [3–7]

Refinery decade means average January 1 snapshots; the 1990s capacity mean has eight survey years, excluding 1996 and 1998. It is not a time-weighted operating-capacity calculation.

D3. Available 2026 monthly prices and inflation conversion [1–2,10]

Monthly price availability differs by series. A blank RAC cell reflects publication lag in the downloaded table, not an assumed unchanged price. Monthly retail averages and weekly September readings differ in frequency; no full-year 2026 observation is manufactured.

D4. 2026 monthly flow record in source downloads, mb/d [3–4]

Recent Monthly Energy Review quantities can be preliminary or estimated, especially the newest months; they are displayed as the available source record, not declared final measured annual data. The January 1, 2026 refinery survey reports 130 operable refineries and 18.160493 million barrels/calendar day. These two survey values should not be averaged with monthly flow observations to invent a 2026 annual statistic. [3–7]

Appendix E. Reproducibility and source register

Sources were accessed or checked for this dossier on October 1, 2026. Government datasets and operative legal documents take precedence for numbers and legal scope; company releases are used for termination, investment and developer targets. Forecasts retain their issue date. CRS R42465 is used as a historical explanation of trade and refinery quality; its 2012 legal environment is not treated as current law.

Reproduction: download EIA MER CSV tables T09.01, T09.04, T03.01 and T03.03B using https://www.eia.gov/totalenergy/data/browser/csv.php?tbl=TABLE. Filter YYYYMM ending in 13 for annual data and select the MSN codes in sources 1–4. Divide quantity units as specified above. Use EIA annual refinery table 5.9 and the historical capacity/count series, preserving the two unsurveyed years; use Table 1 of the 2026 report for the latest January snapshot. Apply published annual CPI-U using the stated August dollar-base formula. EIA Table T05.01 supplies OGNRPUS annual oil-and-gas rotary rig counts used in section 4.

Limits: no causally identified cents-per-gallon impact is assigned to a president, a canceled pipeline, tax holiday or regulatory package. No complete annual 2026 average is available at cutoff. The diesel retail series does not support annual 1970s/1980s values. No full-system commissioning is inferred from a border permit, lease bid or investment announcement. A legal-status summary cannot substitute for a complete litigation docket audit; unverified restarts are reported as unverified, not proven impossible.

[1] EIA Monthly Energy Review, Table 9.1: crude oil prices. Annual records YYYY13; CODPUUS domestic first purchase and RACPUUS composite refiner acquisition cost. Retrieved October 1, 2026.

[2] EIA Monthly Energy Review, Table 9.4: retail petroleum prices. RLUCUUS leaded city regular; RUUCUUS unleaded city regular; RMRTUUS EIA regular all areas; DFONUUS on-highway diesel. Taxes included; series and survey coverage differ.

[3] EIA Monthly Energy Review, Table 3.1: petroleum overview. PAPRPUS crude production, PAIMPUS imports, PAEXPUS exports, PANIPUS net imports. Thousand barrels/day in downloads.

[4] EIA Monthly Energy Review, Table 3.3b: petroleum imports by type. COIMPUS crude imports; gross, not net imports.

[5] EIA Annual Energy Review, Table 5.9: refinery capacity. Earlier January 1 operable capacity and refinery counts; no survey in 1996 or 1998.

[6] EIA historical refinery operable atmospheric distillation capacity. January 1 barrels per calendar day; paired counts: s=8_NA_8O0_NUS_C. Calendar-day capacity differs from stream-day capacity.

[7] EIA Refinery Capacity Report 2026, Table 1. Released June 26, 2026; January 1, 2026 snapshot, not the September operating fleet.

[8] BLS CPI-U historical table, 1913–2017. All items, U.S. city average, not seasonally adjusted, 1982–84=100; annual averages used through 2016.

[9] BLS CPI annual and semiannual averages. 2017–2025 published CPI-U annual averages. October 2025 collection gap: use BLS published annual average, not an invented observation.

[10] BLS CPI news release: August 2026. CPI-U all items NSA 334.980. September index scheduled for October 14; no full-year 2026 CPI available at cutoff.

[11] EIA Gasoline and Diesel Fuel Update. September 28, 2026 weekly observations, published September 29. Prices include taxes; regular gasoline and on-highway diesel.

[12] EIA Short-Term Energy Outlook, September 2026. Released September 9; forecast assumptions through September 3. Forecasts are explicitly distinguished from observed statistics in this dossier.

[13] EIA Strategic Petroleum Reserve stocks, weekly. September 25, 2026: 283,767 thousand barrels.

[14] Executive Order 12287, January 28, 1981. 46 Federal Register 9909, January 30, 1981; exact scope and transition provisions.

[15] DOE historical timeline, 1971–1980. Carter announced phased oil-price decontrol April 5, 1979, to finish in October 1981.

[16] EIA Short-Term Energy Outlook, first quarter 1986. Contemporaneous account of Saudi output increases and abandonment of swing-producer role.

[17] EIA Short-Term Energy Outlook, third quarter 1986. Contemporaneous evidence of demand reductions and 1986 world-price collapse.

[18] EIA: Iranian Revolution and petroleum supply. Published March 8, 2011; initial Iranian disruption and subsequent production changes.

[19] Federal Reserve History: Oil Shock of 1978–79. Historical monetary-policy and oil-shock context; figures depend on disruption window.

[20] EIA oil market analysis: supply and spot prices. Global crude pricing, OPEC supply and geopolitical risk framework.

[21] EIA review of 2014 crude oil prices. Historical 2014 supply, demand, and infrastructure analysis; used only for 2014.

[22] EIA: negative WTI futures in spring 2020. January 5, 2021 retrospective. Negative expiring futures contract is not negative annual retail fuel prices.

[23] EIA: crude demand returned faster than supply in 2021. Demand, production, refining and inventory recovery after COVID-19.

[24] Executive Order 13990, January 20, 2021. 86 FR 7037, January 25, 2021; section 6 Keystone XL permit revocation; section 4 Arctic Refuge review/moratorium.

[25] TC Energy: termination of Keystone XL, June 9, 2021. Original company announcement, distributed through its news-release service.

[26] DOE: Keystone XL Extension Permit Revocation—Energy Costs and Job Impacts. December 2022 report to Congress; pp. iii–iv and 1–11; distinguishes route segments and inconclusive consumer-price effects.

[27] Canada Energy Regulator: existing Keystone pipeline profile. Existing operating system is distinct from uncompleted XL extension; South Bow ownership after 2024 spin-off.

[28] Presidential permit for Bridger Pipeline Expansion, April 30, 2026. 91 FR 24333, May 5, 2026. Border facilities only; applicable laws, easements and other permits remain required.

[29] Bridger Expansion: developer project overview. Approximately 650-mile, 36-inch proposed border-to-Guernsey route; connection to proposed South Bow Prairie Connector. Developer description, not independent completion certification.

[30] Alyeska: TAPS system overview. 800-mile North Slope–Valdez oil pipeline, operating since 1977; peak throughput approximately 2.1 million barrels/day.

[31] Alyeska: TAPS throughput. 2025 daily average 462,821 barrels; August 2026 daily average 457,788; 2026 year-to-date 458,570 through August.

[32] BLM: Coastal Plain Arctic National Wildlife Refuge program. Leasing history, environmental review and current program; a lease is not a producing field.

[33] Interior: Arctic lease cancellations and protection, September 6, 2023. Cancellation of seven remaining AIDEA leases; administration announcement describes its stated rationale.

[34] Executive Order 14153, Unleashing Alaska’s Extraordinary Resource Potential. January 20, 2025; 90 FR 8347, January 29. Policy directives require subsequent implementation; not a production order.

[35] BLM June 5, 2026 Coastal Plain sale, bid recap. 58 tracts/688,829 acres offered; 5 tracts/72,049 acres receiving bids; 9 bids, 2 bidders, high bids $3,741,528.

[36] Interior: revised Willow decision, March 13, 2023. Three approved drill pads, two denied; separate NPR-A project, not ANWR or Keystone.

[37] BLM Willow Master Development Plan Record of Decision, March 2023. Operative administrative approval; environmental conditions apply.

[38] ConocoPhillips Alaska: 2025 year-end review, February 18, 2026. Developer expected nearly 50% completion during the winter season and anticipated early-2029 first oil. Subsequent Q1 filing reports achieved 50% completion; source 61.

[39] ConocoPhillips: Willow final investment decision, December 2023. Developer peak target 180,000 barrels/day; approximately 600 million barrels over project life.

[40] FERC authorizes Alaska LNG, May 21, 2020. CP17-178; approximately 807 miles of pipeline, 3.9 billion cubic feet/day, 20 million tonnes/year LNG.

[41] DOE Alaska LNG export orders, docket 14-96-LNG. Order 3643-A August 20, 2020; Order 3643-C April 13, 2023. Biden-era supplemental decision retained export authorization with conditions.

[42] Glenfarne: Alaska LNG investment announcement, September 30/October 1, 2026. Developer announcement of $50 billion-plus Korean investment support; project still advancing toward final investment decision. Also indexed at https://glenfarnegroup.com/news-insight/.

[43] EIA: factors affecting gasoline prices. Crude input, refining, distribution/marketing and taxes; blend requirements and seasonality.

[44] EIA: factors affecting diesel fuel prices. Distillate demand, international markets, refining, transportation and taxes.

[45] EIA: U.S. refinery capacity decreased in 2025. 2026 refinery report discussion of Houston and Los Angeles closures and early-2026 Benicia closure.

[46] EIA refinery utilization and monthly capacity. July 2026 operating capacity approximately 18.044 million barrels/calendar day; utilization 96.5%. Monthly series differs from January 1 survey.

[47] EIA federal and state fuel tax FAQ. January 2026 snapshot: federal 18.40/24.40 cents, average state 33.27/35.50 cents for gasoline/diesel; exclusions stated in text.

[48] EIA: state fuel tax changes, corrected March 4, 2026. Original February 12 release corrected national averages and Michigan change. California/Alaska extremes; federal rates unchanged since October 1993.

[49] IRS Publication 510: excise taxes. Statutory excise-tax mechanics and special exemptions; do not treat a general retail average as a tax bill for every transaction.

[50] EPA diesel fuel standards and rulemakings. 15 ppm sulfur highway diesel transition began 2006; protects air quality and enables emission-control systems.

[51] EIA AEO 2026 legislation and regulations assumptions. Documents RFS, regional fuel formulations, taxes and model assumptions; assumptions are not promises of future policy.

[52] ExxonMobil full-year 2025 results, January 30, 2026. GAAP global net income $28.844 billion vs $33.680 billion in 2024; $37.2 billion shareholder distributions.

[53] ConocoPhillips full-year 2025 results. Global earnings approximately $8.0 billion vs $9.2 billion in 2024; upstream company, not a measure of U.S. retail margins.

[54] DOE: 172-million-barrel SPR release authorization, March 11, 2026. Part of coordinated IEA 400-million-barrel response; announced schedules and future returns are not completed deliveries.

[55] GAO: Strategic Petroleum Reserve, 2026 review. Reserve management, drawdowns and maintenance context; EPCA 1975 origins and capacity distinct from current stocks.

[56] CRS R42465: U.S. Oil Imports and Exports, April 4, 2012. Historical trade/refining and crude-quality analysis. Its then-current export restrictions are not represented as 2026 law.

[57] EIA Monthly Energy Review, Table 5.1: drilling activity. OGNRPUS total oil-and-gas rotary rigs, annual average; does not measure completed wells or oil-only rigs.

[58] DOE Strategic Petroleum Reserve quick facts. Authorized storage capacity 714 million barrels; this is not the inventory or a guarantee of immediate usable withdrawal capability.

[59] BLM and Montana DEQ: proposed Bridger pipeline scoping. 2026 environmental scoping: 647 miles, 36 inches, about 550,000 barrels/day proposed; NEPA DOI-BLM-MT-C020-2026-0054-EIS.

[60] Ninth Circuit Willow decision, June 13, 2025, Nos. 23-3624/23-3627. Affirmed in part, reversed in part, remanded without vacating 2023 approval; procedural alternatives-analysis deficiency, pp. 61–64.

[61] ConocoPhillips 2026 first-quarter Form 10-Q. Management discussion, p. 28: Willow achieved 50% completion during winter construction season. Company-filed report, not independent engineering certification.