Oil Chartpack
The world oil balance and the US gas market on a few pages: who produces, who consumes, what is left over, where it is refined and stored, and what the futures curves say about it. The balance tables are EIA’s Short-Term Energy Outlook, read straight from EIA’s workbook, so they update the day each outlook is published; prices, curves and weekly data refresh daily.
STEO released October 2026; history runs through September 2026 and shaded columns are EIA forecasts. Quarters and years are averages of the months, except inventories, which are end of period.
World balance
Table 1. The three monthly outlooks: EIA, IEA and OPEC, with each agency’s revision on its previous report
| EIA Oct 2026 | IEA Sep 2026 | OPEC Sep 2026 | ||||
|---|---|---|---|---|---|---|
| Forecast | vs Sep 2026 | Forecast | vs Aug 2026 | Forecast | vs Aug 2026 | |
| World demand | ||||||
| 2026 | 102.4 | -0.2 | 102.4 | – | 105.8 | +0.1 |
| 2027 | 104.6 | -0.3 | 105.0 | – | 108.2 | +0.3 |
| Demand growth | ||||||
| 2026 | -2.0 | -0.3 | -2.5 | -0.9 | +0.4 | -0.2 |
| 2027 | +2.2 | -0.2 | +2.6 | +0.2 | +2.4 | +0.2 |
| Non-OPEC+ supply growth | ||||||
| 2026 | -0.6 | -0.1 | -0.7 | – | +0.8 | +0.2 |
| 2027 | +3.8 | -0.3 | +3.4 | – | +0.7 | +0.1 |
| Call on OPEC+ crude | ||||||
| 2026 | 30.9 | 0.0 | – | – | – | – |
| 2027 | 28.5 | +0.2 | – | – | – | – |
The market reads each report against the same agency’s previous edition, so read the revision columns first. EIA is computed from the Short-Term Energy Outlook and EIA’s archive of the previous one; IEA and OPEC are the headline figures from each report’s free summary, entered by hand, and a dash means the figure is not in the free material (the IEA’s balance tables are subscription-only). The levels are not strictly comparable across agencies: each has its own demand baseline and its own membership list, and since the UAE left OPEC+ in May 2026, EIA counts it outside the group. Call on OPEC+ crude is EIA demand less all supply other than OPEC+ crude, and OPEC’s own “demand for DoC crude”.
Growth rather than levels, because each agency counts demand on its own baseline and revises history, so a level moves when nothing about the outlook has. EIA is computed from every archived Short-Term Energy Outlook from July 2013 (world consumption, Table 3a); IEA and OPEC are read from each Oil Market Report and Monthly Oil Market Report from January 2024, with a handful of OPEC months taken from press coverage where the report could not be downloaded. The IEA publishes the next year only from about the second quarter, so its line has gaps early in each year. None of the three forecasts more than one calendar year ahead, so there is no two-year-forward series to draw: every horizon ends in December of the following year.
Table 2 · EIA STEO 3a. World Petroleum and Other Liquid Fuels Production, Consumption, and Inventories
| 2024 | 2025 | 1Q26 | 2Q26 | 3Q26 | 4Q26 | 2026 | 2027 | |
|---|---|---|---|---|---|---|---|---|
| Production (million barrels per day) | ||||||||
| World total | 103.15 | 106.27 | 103.82 | 96.51 | 101.25 | 102.97 | 101.14 | 109.53 |
| Crude oil | 76.61 | 78.99 | 77.45 | 71.59 | 75.69 | 77.02 | 75.44 | 81.49 |
| Other liquids | 26.55 | 27.28 | 26.37 | 24.92 | 25.56 | 25.95 | 25.70 | 28.04 |
| World total | 103.15 | 106.27 | 103.82 | 96.51 | 101.25 | 102.97 | 101.14 | 109.53 |
| OPEC total | 28.43 | 29.33 | 27.56 | 20.37 | 23.59 | 24.38 | 23.97 | 29.17 |
| Crude oil | 23.81 | 24.63 | 23.18 | 17.31 | 20.41 | 20.98 | 20.47 | 24.41 |
| Other liquids | 4.62 | 4.70 | 4.39 | 3.06 | 3.17 | 3.40 | 3.50 | 4.76 |
| Non-OPEC total | 74.72 | 76.94 | 76.26 | 76.15 | 77.67 | 78.59 | 77.16 | 80.37 |
| Crude oil | 52.79 | 54.36 | 54.28 | 54.29 | 55.28 | 56.04 | 54.97 | 57.08 |
| Other liquids | 21.93 | 22.58 | 21.98 | 21.86 | 22.39 | 22.55 | 22.19 | 23.29 |
| Consumption (million barrels per day) | ||||||||
| World total | 103.10 | 104.41 | 102.92 | 100.05 | 103.13 | 103.66 | 102.44 | 104.65 |
| OECD total | 46.22 | 46.33 | 45.94 | 44.58 | 46.18 | 46.05 | 45.69 | 46.17 |
| Canada | 2.52 | 2.55 | 2.57 | 2.44 | 2.61 | 2.57 | 2.55 | 2.52 |
| Europe | 13.56 | 13.52 | 13.01 | 13.18 | 13.87 | 13.42 | 13.37 | 13.48 |
| Japan | 3.16 | 3.10 | 3.29 | 2.59 | 2.80 | 3.14 | 2.95 | 2.95 |
| United States | 20.46 | 20.73 | 20.72 | 20.54 | 20.61 | 20.54 | 20.60 | 20.81 |
| U.S. Territories | 0.14 | 0.14 | 0.11 | 0.11 | 0.12 | 0.12 | 0.12 | 0.14 |
| Other OECD | 6.37 | 6.29 | 6.23 | 5.72 | 6.17 | 6.27 | 6.10 | 6.26 |
| Non-OECD total | 56.89 | 58.08 | 56.98 | 55.47 | 56.95 | 57.61 | 56.75 | 58.48 |
| China | 16.37 | 16.58 | 16.57 | 15.81 | 15.68 | 16.06 | 16.03 | 16.30 |
| Eurasia | 5.07 | 5.11 | 4.87 | 4.91 | 5.30 | 5.23 | 5.08 | 5.03 |
| Europe | 0.80 | 0.80 | 0.76 | 0.80 | 0.81 | 0.81 | 0.80 | 0.80 |
| Other Asia | 14.46 | 14.95 | 15.10 | 14.27 | 14.54 | 15.24 | 14.79 | 15.56 |
| Other non-OECD | 20.18 | 20.64 | 19.69 | 19.69 | 20.62 | 20.26 | 20.07 | 20.79 |
| Total crude oil and other liquids inventory net withdrawals (million barrels per day) | ||||||||
| World total | -0.05 | -1.87 | -0.90 | 3.54 | 1.88 | 0.69 | 1.30 | -4.89 |
| United States | -0.07 | -0.20 | 0.07 | 1.74 | 0.05 | 0.47 | 0.58 | -0.54 |
| Other OECD | 0.02 | -0.09 | 0.25 | 0.04 | 0.89 | 0.39 | 0.39 | -0.98 |
| Other inventory draws and balance | -0.01 | -1.58 | -1.22 | 1.77 | 0.95 | -0.17 | 0.33 | -3.36 |
| End-of-period commercial crude oil and other liquids inventories (million barrels) | ||||||||
| OECD total | 2,744 | 2,832 | 2,798 | 2,730 | 2,683 | 2,645 | 2,645 | 3,102 |
| United States | 1,236 | 1,289 | 1,281 | 1,215 | 1,249 | 1,247 | 1,247 | 1,344 |
| Other OECD | 1,508 | 1,543 | 1,518 | 1,516 | 1,434 | 1,398 | 1,398 | 1,758 |
Supply
Table 3 · EIA STEO 3b. Non-OPEC Petroleum and Other Liquid Fuels Production
| 2024 | 2025 | 1Q26 | 2Q26 | 3Q26 | 4Q26 | 2026 | 2027 | |
|---|---|---|---|---|---|---|---|---|
| Petroleum and other liquid fuels production | ||||||||
| Non-OPEC total | 74.72 | 76.94 | 76.26 | 76.15 | 77.67 | 78.59 | 77.16 | 80.37 |
| North America total | 30.88 | 31.85 | 32.02 | 32.57 | 32.83 | 33.14 | 32.64 | 33.50 |
| Canada | 6.00 | 6.26 | 6.43 | 6.27 | 6.34 | 6.49 | 6.38 | 6.47 |
| Mexico | 2.01 | 1.87 | 1.88 | 1.89 | 1.87 | 1.82 | 1.86 | 1.78 |
| United States | 22.88 | 23.72 | 23.72 | 24.42 | 24.62 | 24.83 | 24.39 | 25.25 |
| Central and South America total | 7.39 | 7.92 | 8.14 | 9.07 | 9.24 | 8.69 | 8.79 | 9.13 |
| Argentina | 0.89 | 0.98 | 1.06 | 1.08 | 1.09 | 1.10 | 1.08 | 1.13 |
| Brazil | 4.28 | 4.65 | 4.64 | 5.55 | 5.71 | 5.13 | 5.26 | 5.33 |
| Colombia | 0.80 | 0.78 | 0.77 | 0.75 | 0.76 | 0.75 | 0.76 | 0.77 |
| Guyana | 0.62 | 0.75 | 0.90 | 0.91 | 0.91 | 0.94 | 0.92 | 1.15 |
| Europe total | 3.86 | 3.99 | 4.10 | 3.87 | 3.73 | 3.85 | 3.89 | 3.81 |
| Norway | 2.01 | 2.05 | 2.18 | 2.04 | 2.02 | 2.08 | 2.08 | 2.08 |
| United Kingdom | 0.74 | 0.77 | 0.77 | 0.70 | 0.61 | 0.66 | 0.68 | 0.62 |
| Eurasia total | 13.45 | 13.68 | 13.38 | 13.57 | 13.11 | 13.65 | 13.43 | 13.62 |
| Azerbaijan | 0.60 | 0.58 | 0.56 | 0.55 | 0.55 | 0.55 | 0.55 | 0.53 |
| Kazakhstan | 1.90 | 2.15 | 1.75 | 2.24 | 2.07 | 2.26 | 2.08 | 2.19 |
| Russia | 10.53 | 10.53 | 10.65 | 10.36 | 10.11 | 10.47 | 10.40 | 10.52 |
| Middle East total | 7.38 | 7.54 | 6.58 | 5.06 | 6.63 | 7.00 | 6.32 | 8.07 |
| Oman | 1.00 | 1.01 | 1.04 | 1.09 | 1.09 | 1.09 | 1.08 | 1.09 |
| Qatar | 1.85 | 1.88 | 1.36 | 0.39 | 0.84 | 0.92 | 0.88 | 1.43 |
| United Arab Emirates | 4.24 | 4.34 | 3.93 | 3.40 | 4.47 | 4.72 | 4.13 | 5.20 |
| Africa total | 2.55 | 2.58 | 2.58 | 2.64 | 2.60 | 2.70 | 2.63 | 2.68 |
| Angola | 1.16 | 1.06 | 1.03 | 1.06 | 1.06 | 1.14 | 1.07 | 1.15 |
| Egypt | 0.64 | 0.61 | 0.61 | 0.62 | 0.60 | 0.60 | 0.61 | 0.57 |
| Asia and Oceania total | 9.21 | 9.39 | 9.45 | 9.37 | 9.53 | 9.56 | 9.48 | 9.56 |
| China | 5.33 | 5.45 | 5.58 | 5.52 | 5.52 | 5.56 | 5.54 | 5.55 |
| India | 0.94 | 0.99 | 0.97 | 0.96 | 1.00 | 1.00 | 0.98 | 1.02 |
| Indonesia | 0.84 | 0.85 | 0.82 | 0.82 | 0.83 | 0.83 | 0.83 | 0.83 |
| Malaysia | 0.57 | 0.60 | 0.57 | 0.57 | 0.57 | 0.57 | 0.57 | 0.55 |
| Unplanned production outages | ||||||||
| Non-OPEC total | 1.22 | 1.07 | 2.54 | 4.07 | 2.78 | - | - | - |
Table 4 · EIA STEO 3c. World Petroleum and Other Liquid Fuels Production
| 2024 | 2025 | 1Q26 | 2Q26 | 3Q26 | 4Q26 | 2026 | 2027 | |
|---|---|---|---|---|---|---|---|---|
| Petroleum and other liquid fuels production | ||||||||
| World total | 103.15 | 106.27 | 103.82 | 96.51 | 101.25 | 102.97 | 101.14 | 109.53 |
| OPEC+ total | 38.74 | 39.36 | 37.41 | 31.15 | 34.57 | 36.34 | 34.87 | 39.49 |
| United States | 22.88 | 23.72 | 23.72 | 24.42 | 24.62 | 24.83 | 24.39 | 25.25 |
| Non-OPEC+ excluding United States | 41.53 | 43.19 | 42.69 | 40.94 | 42.07 | 41.80 | 41.87 | 44.79 |
| OPEC total | 28.43 | 29.33 | 27.56 | 20.37 | 23.59 | 24.38 | 23.97 | 29.17 |
| Algeria | 1.38 | 1.40 | 1.44 | 1.45 | 1.45 | - | - | - |
| Congo (Brazzaville) | 0.25 | 0.25 | 0.26 | 0.28 | 0.28 | - | - | - |
| Equatorial Guinea | 0.10 | 0.08 | 0.08 | 0.07 | 0.07 | - | - | - |
| Gabon | 0.21 | 0.24 | 0.23 | 0.23 | 0.22 | - | - | - |
| Iran | 4.63 | 4.77 | 4.60 | 3.66 | 3.01 | - | - | - |
| Iraq | 4.57 | 4.50 | 3.58 | 1.60 | 3.49 | - | - | - |
| Kuwait | 2.77 | 2.80 | 2.36 | 0.93 | 1.91 | - | - | - |
| Libya | 1.18 | 1.38 | 1.39 | 1.39 | 1.40 | - | - | - |
| Nigeria | 1.56 | 1.67 | 1.60 | 1.72 | 1.71 | - | - | - |
| Saudi Arabia | 10.87 | 11.21 | 11.02 | 7.84 | 8.78 | - | - | - |
| Venezuela | 0.90 | 1.01 | 1.00 | 1.18 | 1.25 | - | - | - |
| OPEC+ total | 38.74 | 39.36 | 37.41 | 31.15 | 34.57 | 36.34 | 34.87 | 39.49 |
| OPEC members subject to OPEC+ agreements | 21.72 | 22.17 | 20.57 | 14.13 | 17.92 | 19.16 | 17.95 | 22.35 |
| OPEC+ other participants total | 17.03 | 17.19 | 16.84 | 17.02 | 16.65 | 17.17 | 16.92 | 17.14 |
| Azerbaijan | 0.60 | 0.58 | 0.56 | 0.55 | 0.55 | 0.55 | 0.55 | 0.53 |
| Bahrain | 0.19 | 0.19 | 0.12 | 0.04 | 0.11 | 0.13 | 0.10 | 0.19 |
| Brunei | 0.10 | 0.11 | 0.12 | 0.10 | 0.11 | 0.11 | 0.11 | 0.11 |
| Kazakhstan | 1.90 | 2.15 | 1.75 | 2.24 | 2.07 | 2.26 | 2.08 | 2.19 |
| Malaysia | 0.57 | 0.60 | 0.57 | 0.57 | 0.57 | 0.57 | 0.57 | 0.55 |
| Mexico | 2.01 | 1.87 | 1.88 | 1.89 | 1.87 | 1.82 | 1.86 | 1.78 |
| Oman | 1.00 | 1.01 | 1.04 | 1.09 | 1.09 | 1.09 | 1.08 | 1.09 |
| Russia | 10.53 | 10.53 | 10.65 | 10.36 | 10.11 | 10.47 | 10.40 | 10.52 |
| South Sudan | 0.08 | 0.11 | 0.13 | 0.15 | 0.14 | 0.15 | 0.14 | 0.15 |
| Sudan | 0.04 | 0.03 | 0.03 | 0.03 | 0.03 | 0.03 | 0.03 | 0.03 |
Table 5 · EIA STEO 3d. World Crude Oil Production
| 2024 | 2025 | 1Q26 | 2Q26 | 3Q26 | 4Q26 | 2026 | 2027 | |
|---|---|---|---|---|---|---|---|---|
| Crude oil production | ||||||||
| World total | 76.61 | 78.99 | 77.45 | 71.59 | 75.69 | 77.02 | 75.44 | 81.49 |
| OPEC+ total | 32.75 | 33.32 | 31.55 | 26.24 | 29.58 | 30.93 | 29.57 | 33.38 |
| United States | 13.27 | 13.66 | 13.58 | 13.88 | 13.95 | 14.05 | 13.87 | 14.30 |
| Non-OPEC+ excluding United States | 30.59 | 32.01 | 32.32 | 31.47 | 32.16 | 32.04 | 32.00 | 33.81 |
| OPEC total | 23.81 | 24.63 | 23.18 | 17.31 | 20.41 | 20.98 | 20.47 | 24.41 |
| Algeria | 0.91 | 0.93 | 0.97 | 0.98 | 0.98 | - | - | - |
| Congo (Brazzaville) | 0.24 | 0.24 | 0.25 | 0.27 | 0.27 | - | - | - |
| Equatorial Guinea | 0.06 | 0.05 | 0.05 | 0.04 | 0.04 | - | - | - |
| Gabon | 0.22 | 0.24 | 0.23 | 0.23 | 0.22 | - | - | - |
| Iran | 3.31 | 3.38 | 3.35 | 2.82 | 2.30 | - | - | - |
| Iraq | 4.40 | 4.33 | 3.44 | 1.54 | 3.38 | - | - | - |
| Kuwait | 2.46 | 2.49 | 2.09 | 0.83 | 1.77 | - | - | - |
| Libya | 1.09 | 1.29 | 1.29 | 1.29 | 1.31 | - | - | - |
| Nigeria | 1.28 | 1.41 | 1.36 | 1.48 | 1.48 | - | - | - |
| Saudi Arabia | 9.02 | 9.33 | 9.22 | 6.71 | 7.49 | - | - | - |
| Venezuela | 0.83 | 0.94 | 0.93 | 1.11 | 1.18 | - | - | - |
| OPEC+ total | 32.75 | 33.32 | 31.55 | 26.24 | 29.58 | 30.93 | 29.57 | 33.38 |
| OPEC members subject to OPEC+ agreements | 18.58 | 19.02 | 17.61 | 12.09 | 15.63 | 16.64 | 15.49 | 19.04 |
| OPEC+ other participants total | 14.17 | 14.30 | 13.94 | 14.16 | 13.95 | 14.29 | 14.09 | 14.34 |
| Azerbaijan | 0.48 | 0.46 | 0.45 | 0.44 | 0.44 | - | - | - |
| Bahrain | 0.17 | 0.18 | 0.11 | 0.03 | 0.09 | - | - | - |
| Brunei | 0.08 | 0.09 | 0.09 | 0.07 | 0.09 | - | - | - |
| Kazakhstan | 1.50 | 1.75 | 1.39 | 1.85 | 1.75 | - | - | - |
| Malaysia | 0.34 | 0.37 | 0.34 | 0.35 | 0.35 | - | - | - |
| Mexico | 1.55 | 1.43 | 1.43 | 1.43 | 1.43 | - | - | - |
| Oman | 0.76 | 0.77 | 0.81 | 0.86 | 0.85 | - | - | - |
| Russia | 9.16 | 9.12 | 9.16 | 8.94 | 8.78 | - | - | - |
| South Sudan | 0.08 | 0.11 | 0.13 | 0.15 | 0.14 | - | - | - |
| Sudan | 0.04 | 0.03 | 0.03 | 0.03 | 0.03 | - | - | - |
| Crude oil production capacity | ||||||||
| OPEC total | 27.37 | 28.06 | 24.91 | 17.33 | 20.43 | 21.00 | 20.92 | 26.06 |
| Middle East | 22.65 | 22.89 | 19.80 | 11.90 | 14.93 | 15.57 | 15.55 | 20.40 |
| Other | 4.73 | 5.17 | 5.11 | 5.43 | 5.50 | 5.43 | 5.37 | 5.65 |
| Surplus crude oil production capacity | ||||||||
| OPEC total | 3.56 | 3.43 | 1.73 | 0.02 | 0.02 | 0.02 | 0.45 | 1.64 |
| Middle East | 3.46 | 3.36 | 1.70 | 0.00 | 0.00 | 0.00 | 0.43 | 1.61 |
| Other | 0.10 | 0.07 | 0.03 | 0.02 | 0.02 | 0.02 | 0.02 | 0.03 |
| Unplanned production outages | ||||||||
| OPEC total | 1.35 | 0.98 | 3.68 | 10.72 | 7.58 | - | - | - |
Table 6. Average API gravity of crude produced, by country
The world’s crude averaged about 33.3 degrees API in 2025, against 32.5 in 2015. OPEC’s barrel, at about 32.0, is heavier than the rest of the world’s at 33.9. Most of the world’s rise is the US: its crude averaged 40.1 degrees in 2023, the lightest of any large producer bar Kazakhstan and Algeria, and its share of supply grew. OPEC’s own line rises for a different reason: Venezuela’s extra-heavy crude, at about 18 degrees, fell to a smaller share of OPEC output.
| Country | Output 2025, kb/d | Share of world | API gravity | Confidence | Main grades (API, share of output) |
|---|---|---|---|---|---|
| United States | 13,586 | 16.1% | 40.1 | measured, 2023 | EIA production by API gravity band, Lower 48 and Gulf of Mexico, 2015 to Feb 2024; Alaska North Slope 32 |
| Russia | 9,885 | 11.7% | 32.3 | medium-low | Urals 31-32 (about 70%), ESPO 34.5-35.6 (17%), Sokol, Novy Port, Varandey 35-37 (8%) |
| Saudi Arabia (OPEC) | 9,556 | 11.3% | 33.0 | medium | Arab Light 33.3 (about 55%), Arab Extra Light 38 (15%), Arab Medium 30.5 (10%), Arab Heavy 27.7 (20%) |
| Canada | 4,964 | 5.9% | 29.0 | medium | Dilbit and WCS 20.5-21.8 (45%), synthetic crude 31-32 (24%), conventional light about 40 (10%), conventional heavy 20.5 (9%), condensate and other about 55 (12%) |
| Iraq (OPEC) | 4,388 | 5.2% | 28.0 | low-medium | Basrah Medium about 28 (57%), Basrah Heavy 23.7 (31%), Kirkuk about 35 (7%) |
| China | 4,325 | 5.1% | 27.0 | low | Daqing 32-36 (about 14%), other CNPC about 29 (46%), Shengli and Sinopec about 25 (16%), CNOOC Bohai about 21 (21%) |
| Iran (OPEC) | 4,051 | 4.8% | 30.8 | low-medium | Iranian Heavy 29.5 (about 52%), Iranian Light 33.6 (30%), Forozan 28.5 (8%) |
| UAE | 3,771 | 4.5% | 38.0 | low-medium | Murban 39.6-40.5 (about 50%), Upper Zakum 33.1 (20%), Das 39.1 (12%), Umm Lulu 38.9 (8%) |
| Brazil | 3,769 | 4.5% | 27.4 | medium | Pre-salt (Tupi, Buzios, Mero) about 29 (77%), post-salt Campos about 22 (23%) |
| Kuwait (OPEC) | 2,584 | 3.1% | 30.8 | high | Kuwait Export Crude, a single national blend, 30.5-31.4 |
| Kazakhstan | 2,047 | 2.4% | 42.0 | medium | Tengiz 46.4 (32%), Kashagan about 44 (22%), Karachaganak condensate about 46 (14%), other about 35 (32%); CPC Blend 45.3 |
| Norway | 1,858 | 2.2% | 33.0 | low-medium | Johan Sverdrup 28.7 (40%), Ekofisk, Oseberg, Gullfaks, Statfjord, Troll about 36 (60%) |
| Mexico | 1,726 | 2.0% | 30.0 | medium | Maya 21.5-22 (53%), Isthmus 32.9-34 (23%), Olmeca 37.3-39 (9%), condensate about 50 (15%) |
| Nigeria (OPEC) | 1,605 | 1.9% | 36.4 | medium | Bonny Light 35, Forcados 32, Qua Iboe 38, Escravos 32, Bonga 29.1, plus lighter deepwater and condensate grades |
| Libya (OPEC) | 1,363 | 1.6% | 38.5 | low-medium | Es Sider about 37, Sharara 43.1, Brega 40, Zueitina 41, Amna 36, Bouri 26 |
| Qatar | 1,297 | 1.5% | 32.5 | low-medium | Al-Shaheen 29.5 (about 50%), Qatar Land about 40 (25%), Qatar Marine 31.2 (25%) |
| Algeria (OPEC) | 1,143 | 1.4% | 45.0 | high | Saharan Blend, over 90% of output |
| Angola | 1,031 | 1.2% | 29.0 | low-medium | Cabinda 32, Dalia 23.7, Girassol 30.8, Nemba 38.7, Pazflor 25.6 |
| Oman | 1,002 | 1.2% | 33.0 | high | Oman Blend, a single national stream, 33.2 |
| Venezuela (OPEC) | 973 | 1.2% | 18.0 | low | Merey 16 (60% or more), Hamaca upgraded crude about 26, Boscan |
| Argentina | 794 | 0.9% | 36.0 | medium | Medanito and Vaca Muerta 40.8 (74%), Escalante 24.1 (23%) |
| Guyana | 747 | 0.9% | 32.0 | medium | Liza 31.9, Unity Gold 33.9, Payara Gold 28, Yellowtail 36.5 |
| Colombia | 746 | 0.9% | 22.0 | medium | Castilla Blend 18.8 (about 50%), Vasconia 21-22 (40%), Cusiana 43.2 (10%) |
| United Kingdom | 616 | 0.7% | 29.5 | medium | Clair Ridge 23.7, Buzzard 32.6, Schiehallion 25.2, Mariner and Kraken about 14, lighter fields 38-48 |
| Indonesia | 582 | 0.7% | 33.0 | low-medium | Banyu Urip 32.8 (about 39%), Duri 20.8, Minas 34-35 |
| India | 581 | 0.7% | 32.0 | low | Bombay High about 38 (22%), Mangala 25-30 (25%), other about 32 |
| Azerbaijan | 561 | 0.7% | 35.5 | medium | Azeri Light 35.6, BTC Blend 36.6-39; ACG is over 90% of output |
| Malaysia | 513 | 0.6% | 39.0 | low | Tapis 43-46, Kikeh 36.7-37.6, Kimanis 38.6, Labuan 29.9 |
| Egypt | 505 | 0.6% | 37.0 | medium | Western Desert Blend 41.1 (56%), Suez Blend 30.4 (23%), other about 28 (21%) |
No agency publishes the average gravity of the world’s crude, so this is built rather than read. The United States is measured: EIA published Lower 48 and Gulf of Mexico production by API gravity band from January 2015 to February 2024, weighted here at each band’s midpoint, with Alaska added at Alaska North Slope gravity; after EIA stopped the series the US is held at its 2023 value. Every other country carries a gravity for its crude mix built from its main grades, held fixed and weighted by EIA production of crude and condensate. So the world, OPEC and non-OPEC lines move only with who produces, and a country whose own barrel changes (Guyana ramping up, Venezuela running down) is not tracked within itself. The grade gravities are mostly from producer assays; the shares of each grade in national output are estimates for most countries, which is what the confidence column grades. The countries listed produced 95% of world crude in 2025; the rest are left out of the averages. OPEC is its eleven members after the UAE’s exit in May 2026, applied to every year; Congo, Gabon and Equatorial Guinea have no gravity here and are outside the OPEC line.
Demand
Table 7 · EIA STEO 3e. World Petroleum and Other Liquid Fuels Consumption
| 2024 | 2025 | 1Q26 | 2Q26 | 3Q26 | 4Q26 | 2026 | 2027 | |
|---|---|---|---|---|---|---|---|---|
| Petroleum and other liquid fuels consumption | ||||||||
| World total | 103.10 | 104.41 | 102.92 | 100.05 | 103.13 | 103.66 | 102.44 | 104.65 |
| OECD total | 46.22 | 46.33 | 45.94 | 44.58 | 46.18 | 46.05 | 45.69 | 46.17 |
| Non-OECD total | 56.89 | 58.08 | 56.98 | 55.47 | 56.95 | 57.61 | 56.75 | 58.48 |
| World total | 103.10 | 104.41 | 102.92 | 100.05 | 103.13 | 103.66 | 102.44 | 104.65 |
| North America total | 24.90 | 25.20 | 25.20 | 24.94 | 25.14 | 24.97 | 25.06 | 25.19 |
| Canada | 2.52 | 2.55 | 2.57 | 2.44 | 2.61 | 2.57 | 2.55 | 2.52 |
| Mexico | 1.92 | 1.91 | 1.90 | 1.95 | 1.92 | 1.86 | 1.91 | 1.85 |
| United States | 20.46 | 20.73 | 20.72 | 20.54 | 20.61 | 20.54 | 20.60 | 20.81 |
| Central and South America total | 6.88 | 6.98 | 7.00 | 7.11 | 7.28 | 7.23 | 7.16 | 7.23 |
| Brazil | 3.27 | 3.36 | 3.32 | 3.34 | 3.46 | 3.47 | 3.40 | 3.43 |
| Europe total | 14.36 | 14.31 | 13.77 | 13.98 | 14.68 | 14.23 | 14.16 | 14.28 |
| Eurasia total | 5.07 | 5.11 | 4.87 | 4.91 | 5.30 | 5.23 | 5.08 | 5.03 |
| Russia | 3.80 | 3.82 | 3.61 | 3.62 | 4.00 | 3.88 | 3.78 | 3.77 |
| Middle East total | 9.35 | 9.55 | 8.28 | 8.45 | 9.17 | 8.61 | 8.63 | 9.11 |
| Africa total | 4.67 | 4.82 | 5.01 | 4.75 | 4.80 | 5.03 | 4.89 | 5.11 |
| Asia and Oceania total | 37.87 | 38.43 | 38.79 | 35.93 | 36.77 | 38.36 | 37.46 | 38.69 |
| China | 16.37 | 16.58 | 16.57 | 15.81 | 15.68 | 16.06 | 16.03 | 16.30 |
| India | 5.60 | 5.67 | 5.74 | 5.58 | 5.40 | 5.84 | 5.64 | 5.98 |
| Japan | 3.16 | 3.10 | 3.29 | 2.59 | 2.80 | 3.14 | 2.95 | 2.95 |
Diesel
Diesel and gasoil are the largest single product in the world’s oil barrel, and the one the current crisis has squeezed hardest. How much a country pays depends mostly on tax; how exposed it is depends on how much of its fleet runs on diesel and how much of its oil use is diesel.United States
Table 8. US crude oil production by basin, million barrels per day
| Basin (main states) | Sep 2026 | A year earlier | Change | Dec 2027 (EIA forecast) |
|---|---|---|---|---|
| Permian (TX, NM) | 6.87 | 6.71 | +0.15 | 7.31 |
| Bakken (ND, MT) | 1.21 | 1.23 | −0.02 | 1.20 |
| Eagle Ford (TX) | 1.11 | 1.14 | −0.04 | 1.25 |
| Federal Gulf of America (offshore LA, TX) | 2.05 | 2.05 | −0.01 | 1.88 |
| Rest of Lower 48 | 2.04 | 2.11 | −0.07 | 2.17 |
| Alaska (AK) | 0.46 | 0.42 | +0.04 | 0.53 |
| Appalachia (PA, WV, OH) | 0.21 | 0.20 | 0.00 | 0.20 |
| Haynesville (LA, TX) | 0.03 | 0.03 | 0.00 | 0.03 |
Rest of Lower 48 covers the remaining onshore fields, the largest of them in the Anadarko and Denver-Julesburg basins (OK, CO) and California. The Federal Gulf of America figure is crude plus lease condensate, as EIA reports it.
Footage per well is measured depth, the vertical section plus the horizontal lateral, so it rises with the lateral: an oil well averaged 4,700 feet in 2005 and 17,500 feet in 2025; a gas well 7,000 feet and 20,300 feet. There is no free national series of lateral length alone; the best published one, for the Permian, is licensed from Enverus and shows the average new lateral going from 6,149 feet in 2015 to 10,867 feet in 2025 (EIA, Today in Energy). EIA’s trajectory split counts horizontal and directional wells together and does not separate oil from gas; the rig chart does, from 2013, and covers the rigs working now rather than wells finished.
Hydraulic fracturing has no equivalent series: no agency publishes the share of new wells fractured each year. In practice almost every horizontal shale well is fractured, and EIA’s one count, for 2016, put fractured horizontal wells at 69% of US wells drilled and 83% of footage. Nor is there a world figure for either measure: Baker Hughes publishes trajectory only for the US and Canada, and outside North America horizontal fracturing at scale is limited to a handful of plays, chief among them Vaca Muerta in Argentina and the Sichuan shale gas fields in China.
Table 9. US rigs by basin, week to Sep 18, 2026
| Basin (main states) | Rigs | A year earlier | Change | Share of US |
|---|---|---|---|---|
| Permian (TX, NM) | 269 | 254 | +15 | 45% |
| Haynesville (LA, TX) | 56 | 39 | +17 | 9% |
| Eagle Ford (TX) | 51 | 42 | +9 | 9% |
| Williston, Bakken (ND, MT) | 35 | 29 | +6 | 6% |
| Marcellus (PA, WV) | 23 | 24 | −1 | 4% |
| Cana Woodford (OK) | 21 | 18 | +3 | 4% |
| DJ-Niobrara (CO, WY) | 10 | 9 | +1 | 2% |
| Utica (OH, PA) | 10 | 13 | −3 | 2% |
| Other US | 120 | 111 | +9 | 20% |
| United States | 595 | 539 | +56 | 100% |
Baker Hughes North America rotary rig count, weekly; US rigs 595 (452 oil, 134 gas, 9 other). The file is downloaded by hand from rigcount.bakerhughes.com, which blocks automated access, so it is refreshed periodically rather than every Friday; this edition runs from Jan 4, 2013 to Sep 18, 2026.
+0.6 vs 5-yr avg, +0.2 y/y
−0.0 vs 5-yr avg, −0.0 y/y
−0.18 vs 5-yr avg, −0.06 y/y
+0.13 vs 5-yr avg, +0.10 y/y
−0.03 vs 5-yr avg, −0.02 y/y
+0.00 vs 5-yr avg, −0.01 y/y
Table 10 · EIA STEO 4a. U.S. Petroleum and Other Liquid Fuels Supply, Consumption, and Inventories
| 2024 | 2025 | 1Q26 | 2Q26 | 3Q26 | 4Q26 | 2026 | 2027 | |
|---|---|---|---|---|---|---|---|---|
| Supply (million barrels per day) | ||||||||
| U.S. total crude oil production | 13.27 | 13.66 | 13.58 | 13.88 | 13.95 | 14.05 | 13.87 | 14.30 |
| Alaska | 0.42 | 0.42 | 0.42 | 0.42 | 0.44 | 0.52 | 0.45 | 0.51 |
| Federal Gulf of America | 1.79 | 1.94 | 2.01 | 2.04 | 2.05 | 2.00 | 2.03 | 1.90 |
| Lower 48 States (excl GOA) | 11.05 | 11.30 | 11.15 | 11.42 | 11.47 | 11.53 | 11.39 | 11.89 |
| Appalachia region | 0.15 | 0.19 | 0.19 | 0.20 | 0.20 | 0.20 | 0.20 | 0.20 |
| Bakken region | 1.25 | 1.23 | 1.18 | 1.19 | 1.21 | 1.20 | 1.20 | 1.19 |
| Eagle Ford region | 1.15 | 1.15 | 1.11 | 1.11 | 1.10 | 1.13 | 1.11 | 1.21 |
| Haynesville region | 0.03 | 0.03 | 0.03 | 0.03 | 0.03 | 0.03 | 0.03 | 0.03 |
| Permian region | 6.32 | 6.58 | 6.63 | 6.83 | 6.86 | 6.92 | 6.81 | 7.14 |
| Rest of Lower 48 States | 2.15 | 2.11 | 2.00 | 2.05 | 2.06 | 2.04 | 2.04 | 2.12 |
| Total Supply | 20.46 | 20.73 | 20.75 | 20.54 | 20.61 | 20.54 | 20.61 | 20.81 |
| Crude oil input to refineries | 16.22 | 16.36 | 16.21 | 16.71 | 17.15 | 16.56 | 16.66 | 16.29 |
| U.S. total crude oil production | 13.27 | 13.66 | 13.58 | 13.88 | 13.95 | 14.05 | 13.87 | 14.30 |
| Transfers to crude oil supply | 0.61 | 0.50 | 0.63 | 0.66 | 0.58 | 0.61 | 0.62 | 0.60 |
| Crude oil net imports | 2.49 | 2.20 | 2.38 | 0.73 | 2.63 | 1.52 | 1.82 | 1.78 |
| SPR net withdrawals | -0.11 | -0.05 | -0.01 | 1.02 | 0.42 | 0.45 | 0.47 | -0.28 |
| Commercial inventory net withdrawals | 0.03 | 0.00 | -0.49 | 0.54 | -0.25 | -0.06 | -0.06 | -0.12 |
| Crude oil adjustment | -0.08 | 0.05 | 0.13 | -0.12 | -0.18 | -0.00 | -0.04 | 0.01 |
| Refinery processing gain | 0.97 | 0.98 | 0.96 | 0.94 | 0.99 | 1.02 | 0.98 | 0.98 |
| Natural Gas Plant Liquids Production | 7.04 | 7.51 | 7.58 | 7.92 | 7.97 | 8.02 | 7.87 | 8.20 |
| Renewables and oxygenate production | 1.38 | 1.37 | 1.37 | 1.45 | 1.48 | 1.51 | 1.45 | 1.54 |
| Fuel ethanol production | 1.06 | 1.08 | 1.10 | 1.08 | 1.10 | 1.11 | 1.10 | 1.12 |
| Petroleum products adjustment | 0.22 | 0.21 | 0.22 | 0.23 | 0.22 | 0.22 | 0.23 | 0.22 |
| Petroleum products transfers to crude oil supply | -0.61 | -0.50 | -0.63 | -0.66 | -0.58 | -0.61 | -0.62 | -0.60 |
| Petroleum product net imports | -4.77 | -5.05 | -5.54 | -6.24 | -6.50 | -6.27 | -6.14 | -5.68 |
| Hydrocarbon gas liquids | -2.72 | -2.88 | -3.09 | -3.52 | -3.36 | -3.42 | -3.35 | -3.48 |
| Unfinished oils | 0.12 | -0.03 | -0.05 | -0.03 | 0.09 | 0.06 | 0.02 | 0.17 |
| Other hydrocarbons and oxygenates | -0.08 | -0.18 | -0.20 | -0.20 | -0.19 | -0.19 | -0.19 | -0.17 |
| Total motor gasoline | -0.21 | -0.22 | -0.51 | -0.26 | -0.53 | -0.63 | -0.48 | -0.37 |
| Jet fuel | -0.11 | -0.09 | -0.19 | -0.27 | -0.27 | -0.21 | -0.23 | -0.13 |
| Distillate fuel oil | -1.14 | -1.09 | -1.01 | -1.42 | -1.61 | -1.27 | -1.33 | -1.09 |
| Residual fuel oil | -0.03 | 0.00 | 0.04 | 0.02 | -0.02 | -0.03 | -0.00 | -0.00 |
| Other oils | -0.61 | -0.57 | -0.53 | -0.55 | -0.60 | -0.59 | -0.57 | -0.60 |
| Petroleum product inventory net withdrawals | 0.01 | -0.15 | 0.57 | 0.18 | -0.12 | 0.08 | 0.18 | -0.14 |
| Consumption (million barrels per day) | ||||||||
| U.S. total petroleum products consumption | 20.46 | 20.73 | 20.72 | 20.54 | 20.61 | 20.54 | 20.60 | 20.81 |
| Hydrocarbon gas liquids | 3.74 | 3.93 | 4.26 | 3.68 | 3.77 | 4.08 | 3.95 | 4.09 |
| Other hydrocarbons and oxygenates | 0.33 | 0.22 | 0.21 | 0.29 | 0.31 | 0.33 | 0.29 | 0.38 |
| Motor gasoline | 8.96 | 8.93 | 8.57 | 8.91 | 8.85 | 8.66 | 8.75 | 8.75 |
| Jet fuel | 1.69 | 1.74 | 1.65 | 1.78 | 1.80 | 1.69 | 1.73 | 1.74 |
| Distillate fuel oil | 3.79 | 3.91 | 4.05 | 3.75 | 3.72 | 3.87 | 3.85 | 3.89 |
| Residual fuel oil | 0.29 | 0.32 | 0.32 | 0.36 | 0.28 | 0.30 | 0.32 | 0.33 |
| Other oils | 1.65 | 1.68 | 1.66 | 1.76 | 1.88 | 1.60 | 1.73 | 1.64 |
| Total petroleum and other liquid fuels net imports | -2.27 | -2.85 | -3.16 | -5.50 | -3.88 | -4.76 | -4.32 | -3.91 |
| End-of-period inventories (million barrels) | ||||||||
| Total commercial inventory | 1,236 | 1,289 | 1,281 | 1,215 | 1,249 | 1,247 | 1,247 | 1,344 |
| Crude oil (excluding SPR) | 413 | 411 | 453 | 404 | 427 | 432 | 432 | 476 |
| Hydrocarbon gas liquids | 226 | 274 | 222 | 250 | 291 | 252 | 252 | 278 |
| Unfinished oils | 77 | 81 | 82 | 77 | 80 | 78 | 78 | 80 |
| Other hydrocarbons and oxygenates | 35 | 35 | 38 | 35 | 34 | 36 | 36 | 37 |
| Total motor gasoline | 238 | 244 | 243 | 219 | 204 | 223 | 223 | 237 |
| Jet fuel | 44 | 44 | 45 | 49 | 43 | 43 | 43 | 43 |
| Distillate fuel oil | 130 | 128 | 119 | 109 | 104 | 113 | 113 | 122 |
| Residual fuel oil | 23 | 22 | 25 | 21 | 23 | 23 | 23 | 22 |
| Other oils | 50 | 49 | 53 | 51 | 43 | 46 | 46 | 49 |
| Crude oil in SPR | 394 | 413 | 415 | 322 | 283 | 243 | 243 | 346 |
Refining
Table 11. Refining capacity and run rates by region
| Refineries | Nameplate, kb/d | 2025 runs, kb/d | % of nameplate | Runs Apr to Jun 2026, kb/d | vs year ago | % of nameplate | JODI coverage | |
|---|---|---|---|---|---|---|---|---|
| World total | 614 | 103,756 | 83,081 | 80% | 59,729 | −2,576 | 82% | 70% |
| OECD total | 267 | 43,114 | 36,268 | 84% | 38,920 | +57 | 91% | 99% |
| Canada | 18 | 1,943 | 1,737 | 89% | 2,289 | +185 | 118% | 100% |
| Europe | 79 | 13,669 | 11,284 | 83% | 12,476 | +264 | 91% | 100% |
| Japan | 23 | 3,110 | 2,367 | 76% | 2,354 | −25 | 76% | 100% |
| United States | 127 | 18,172 | 16,371 | 90% | 17,382 | +132 | 96% | 100% |
| Other OECD | 20 | 6,220 | 4,509 | 73% | 4,419 | −500 | 75% | 95% |
| Non-OECD total | 347 | 60,642 | 46,813 | 77% | 20,809 | −2,633 | 70% | 49% |
| China | 47 | 18,829 | 14,747 | 78% | 12,806 | −1,589 | 68% | 100% |
| Eurasia | 51 | 8,387 | 5,929 | 71% | - | - | - | 1% |
| Europe | 15 | 710 | 508 | 72% | 498 | +16 | 110% | 64% |
| Other Asia | 71 | 12,249 | 10,832 | 88% | 3,004 | −150 | 79% | 31% |
| Other non-OECD | 163 | 20,467 | 14,797 | 72% | 4,384 | −903 | 66% | 32% |
Energy Institute Statistical Review. Capacity is atmospheric distillation capacity at year end, from 1965 (Russia from 1980, before which it sits inside the USSR); throughput is the annual average, from 1980 (Russia from 1990). Utilisation divides throughput by the average of the year-end capacity either side of the year, so a unit commissioned in December does not count as a full year of capacity. The European Union is its current 27 members throughout. In 2025 the world ran 80% of capacity, the United States 89% and Russia 73%. Throughput is counted differently from EIA’s gross inputs, so the US figure sits a few points below EIA’s weekly utilisation.
−3.1pp vs median, −1.6pp y/y
+0.2pp vs median, +0.2pp y/y
+0.3pp vs median, +0.0pp y/y
+0.3pp vs median, +0.3pp y/y
EIA Weekly Petroleum Status Report: refiner and blender net production of each product divided by refiner net input of crude oil, both summed over four weeks. Gasoline is finished motor gasoline, so it includes blended ethanol and blending components bought in rather than made from crude, which is why the four shares can add up to slightly more than 100% of crude runs. The reference line is the median of the eight years before 2026leaving out 2020, whose pandemic runs distort every product’s share; the band is the range over the same years.
LNG: the IEA Gas Market Report’s year-on-year fall in Qatar and UAE loadings (March), and in Middle East LNG supply (April to June, net of higher Omani exports and cargoes delivered inside the Gulf; the two measures agree within rounding over March to June, 35 bcm). A bcm a month is about 1.16 Bcf/d; the right axis is the share of world LNG trade, 48.2 bcm a month in 2025.The IEA publishes no monthly figure after June, but Qatar has all but halted shipments since one of its tankers was attacked in late July, and the IEA’s own rule of thumb is about 10 bcm lost for every month without transits, so the missing months are not smaller. Two Ras Laffan trains, 12.8 Mt a year, were damaged in March and will take three to five years to repair. Iran’s pipeline exports to Iraq recovered quickly and to Turkey were unaffected, and no source puts a number on Russian gas export losses from Ukrainian strikes, so neither is charted. LPG: Gulf exports against their 2025 average of 1.5 mb/d, from the IEA Oil Market Report (March is Hormuz transits only; April and May all routes); the right axis is the share of world seaborne LPG, 145 Mt in 2025, taken as about 4.6 mb/d at an assumed 11.6 barrels a tonne. The IEA gives no LPG-only figure after May; Kpler reported LPG crossings of Hormuz at zero in mid-July. Russian LPG from Novatek’s Ust-Luga complex was suspended after strikes in March, August and September 2026, with no published volume.
Table 12. Refining capacity knocked out by war, published estimates; world capacity 103.8 mb/d
| Region | When | Offline | Capacity, mb/d | % of capacity | % of world | What it counts | Source |
|---|---|---|---|---|---|---|---|
| Russia | Aug 2026 | 3.8 mb/d | 6.72 Russia | 57% | 3.7% | unplanned outages at 17 refineries, 44 crude units (July: 4.3). Gross capacity of units affected, including partial outages. | Industrial Info Resources |
| Russia | Jun 2026 | 2.14 mb/d | 6.72 Russia | 32% | 2.1% | nearly one-third of capacity offline. Capacity offline. | Energy Intelligence |
| Russia | May 2026 | about 1.7 mb/d | 6.72 Russia | 25% | 1.6% | 83 Mt/yr halted or cut, nearly a quarter of capacity. Capacity halted or cut. | Reuters, via Moscow Times |
| Persian/Arabian Gulf | Apr 2026 | 2.4 mb/d peak, 1.8 at month end | 10.57 Gulf states and Iran | 23% | 2.3% | Middle East unplanned refinery outages. Capacity shut. | Industrial Info Resources |
| Persian/Arabian Gulf | Mar 2026 | more than 3 mb/d | 10.57 Gulf states and Iran | over 28% | over 2.9% | shut by attacks and the loss of export outlets, with more than 4 mb/d at risk. Capacity shut, including run cuts forced by blocked exports. | IEA Oil Market Report, 12 Mar 2026 |
| Persian/Arabian Gulf | Mar 2026 | about 1.9 mb/d | 8.12 Gulf states excluding Iran | 23% | 1.8% | Saudi Arabia, Iraq, UAE, Bahrain, Kuwait and Qatar; excludes Iran. Capacity shut. | Industrial Info Resources, via Pipeline & Gas Journal |
| Ukraine | since 2022 | about 0.37 mb/d | 0.37 Kremenchuk before 2022 | 100% | 0.4% | Kremenchuk, Ukraine's largest refinery, destroyed in April 2022 and struck again in June 2025; Naftogaz said in June 2022 that all Ukrainian refining had shut. Nameplate capacity (18.6 Mt/yr). | Wikipedia, Kremenchuk Oil Refinery; Rigzone, 16 Jun 2025 |
The total is drawn only in months where both regions have a published figure. The chart measures lost processing, which is what reaches product markets: crude runs against a normal of 5.4 mb/d for Russia and 9.5 mb/d for the Middle East (the IEA region, which includes Iran), from the IEA Oil Market Report and Bloomberg. Only months with a published figure are plotted, and the IEA’s monthly detail after June 2026 is not public, so the Gulf line stops there; the IEA said in July that Middle East export refineries had yet to restart. Capacity offline is a different and larger number than runs lost, because a refinery with a damaged unit can run its others harder and spare capacity elsewhere picks up some of the loss: in autumn 2025 about a fifth of Russian capacity was offline while runs fell only about 6%. Estimates of capacity offline also disagree with each other, depending on whether they count whole units affected (Industrial Info’s 3.8 mb/d for Russia in August) or capacity actually lost, and whether they include run cuts forced by blocked exports (the IEA’s more than 3 mb/d for the Gulf in March). Ukraine lost its refining in 2022, before this window: Kremenchuk, about 0.37 mb/d of nameplate, was destroyed that April, so the war costs Ukraine imported fuel rather than a monthly swing in runs.
Table 13. Refinery yields by region, share of gross output, Jul 2025 to Jun 2026
| Refineries | Output, kb/d | Light ends | Gasoline | Distillate | Fuel oil and other | Reporting countries | Coverage | |
|---|---|---|---|---|---|---|---|---|
| Americas | 216 | 22,176 | 2.9% | 41.9% | 36.6% | 18.5% | 5 | 86% |
| Europe | 101 | 13,237 | 12.3% | 20.9% | 48.0% | 18.7% | 33 | 97% |
| Africa | 44 | 1,153 | 18.8% | 26.2% | 38.2% | 16.8% | 2 | 60% |
| Middle East | 58 | 3,928 | 8.3% | 17.9% | 50.4% | 23.4% | 3 | 44% |
| Russia | 33 | - | - | - | - | not reported | 0 | 0% |
| China | 47 | 15,287 | 23.7% | 23.1% | 33.0% | 20.2% | 1 | 100% |
| Other Asia and Oceania | 113 | 17,236 | 15.9% | 20.1% | 45.3% | 18.7% | 14 | 86% |
| World | 614 | - | 12.6% | 27.5% | 40.7% | 19.2% | - | 77% |
JODI gross refinery output by product, averaged over the twelve months. Light ends are LPG and naphtha; distillate is kerosene, jet and gasoil/diesel; fuel oil and other is residual fuel oil plus bitumen, coke, lubricants and refinery gas. Refineries counts every operating refinery in the region on Wikipedia’s list, whether or not the country reports yields, and like the counts in the table above it is a floor: the list names 47refineries in China, which has more than 200. Coverage is the share of the region’s refining capacity in countries reporting at least nine of the twelve months, and it matters. Russia reports no refinery output to JODI at all. The Middle East row is Bahrain, Kuwait, Saudi Arabia only, without the UAE, Iran, Iraq, Qatar or Oman; Africa is Algeria and Nigeria; and the Americas has no Brazil or Argentina. The Americas row is mostly the United States, where naphtha goes into gasoline rather than being sold, which is why its light ends are small and its gasoline share is double Europe’s. Europe and Asia run for distillate; China’s large light-ends share is naphtha for its petrochemical plants.
Table 14. Petrochemical cracking: ethane against naphtha, by region
Where a steam cracker sits decides what it eats. Regions with gas liquids crack ethane and LPG; regions without them crack naphtha from crude. The choice decides the product slate as well as the cost: an ethane cracker makes about 50% more high-value chemicals from each tonne of feed, almost all of it ethylene, while a naphtha cracker makes a balanced mix of ethylene, propylene and aromatics. That is why the ethane regions, the US and the Middle East, have also had to build dedicated propylene plants, and why naphtha is the swing feed of the oil barrel in Europe and Asia.
| Share of world HVC capacity | Ethane | LPG | Naphtha | What sets it apart | |
|---|---|---|---|---|---|
| North America | 16% | 55% | 34% | under 10% | Shale gas liquids; about 40% of the world’s ethane-based capacity is in the United States |
| Europe | 15% | small | small | close to 75% | Once the largest producer; about 40% of its naphtha cracking sits inside refinery complexes |
| Middle East | 12% | over half, with LPG | - | minor | Cheapest ethane crackers in the world; exports over 90% of its naphtha |
| China | 20% | very small | - | over 90% | Few gas liquids at home, so coal- and methanol-to-olefins were built as the alternative route |
| India | 4% | - | - | - | Small today and expected to grow fastest to meet domestic demand |
IEA, The Future of Petrochemicals (2018), figures for 2017: shares of world capacity for high-value chemicals (ethylene, propylene, butadiene and aromatics) and shares of each region’s cracker feedstock. These are the latest public regional figures that could be verified; the split has moved since, towards US and Middle East ethane and Chinese coal-to-olefins, so read them as the shape of the market rather than current shares. Where a cell is blank the report gives no separate number, and the Middle East’s ethane and LPG are given only together.
in these 20
- 1. China18.1%
- 2. United States17.5%
- 3. Russia6.5%
- 4. India4.7%
- 5. South Korea3.4%
- 6. Saudi Arabia3.2%
- 7. Japan3.0%
- 8. Iran2.4%
- 9. Brazil2.2%
- 10. Germany1.9%
- 11. Canada1.9%
- 12. Mexico1.8%
- 13. Italy1.6%
- 14. Spain1.5%
- 15. Kuwait1.4%
- 16. Iraq1.3%
- 17. Venezuela1.3%
- 18. Singapore1.2%
- 19. Netherlands1.2%
- 20. Indonesia1.2%
- Rest of world22.7%
Table 15. The twenty largest refining countries
| Table 3a region | Refineries | Nameplate, kb/d | 2025 % of nameplate | Runs Apr to Jun 2026, kb/d | vs year ago | Gasoline | Distillate | |
|---|---|---|---|---|---|---|---|---|
| China | Non-OECD China | 47 | 18,829 | 78% | 12,806 | −1,589 | 23.1% | 33.0% |
| United States | OECD United States | 127 | 18,172 | 90% | 17,382 | +132 | 44.7% | 38.1% |
| Russia | Non-OECD Eurasia | 33 | 6,718 | 73% | - | - | - | - |
| India | Non-OECD Other Asia | 24 | 4,924 | 111% | - | - | 18.9% | 46.4% |
| South Korea | OECD Other OECD | 5 | 3,570 | 79% | 2,783 | −596 | 14.7% | 43.8% |
| Saudi Arabia | Non-OECD Other non-OECD | 8 | 3,289 | 83% | 2,365 | −344 | 23.0% | 50.5% |
| Japan | OECD Japan | 23 | 3,110 | 76% | 2,354 | −25 | 28.4% | 44.5% |
| Iran | Non-OECD Other non-OECD | 11 | 2,460 | 82% | - | - | - | - |
| Brazil | Non-OECD Other non-OECD | 17 | 2,296 | 86% | - | - | - | - |
| Germany | OECD Europe | 12 | 1,951 | 84% | 1,830 | −51 | 24.8% | 47.7% |
| Canada | OECD Canada | 18 | 1,943 | 89% | 2,289 | +185 | 27.2% | 31.3% |
| Mexico | OECD Other OECD | 7 | 1,847 | 55% | 1,078 | +34 | 32.9% | 24.1% |
| Italy | OECD Europe | 10 | 1,673 | 75% | 1,355 | −23 | 26.3% | 47.1% |
| Spain | OECD Europe | 8 | 1,532 | 84% | 1,255 | +115 | 18.5% | 57.5% |
| Kuwait | Non-OECD Other non-OECD | 3 | 1,430 | 80% | 558 | −623 | 7.2% | 48.9% |
| Iraq | Non-OECD Other non-OECD | 14 | 1,389 | 65% | - | - | - | - |
| Venezuela | Non-OECD Other non-OECD | 5 | 1,303 | 21% | - | - | - | - |
| Singapore | Non-OECD Other Asia | 3 | 1,289 | 73% | - | - | - | - |
| Netherlands | OECD Europe | 5 | 1,260 | 85% | 1,320 | +283 | 6.7% | 44.4% |
| Indonesia | Non-OECD Other Asia | 8 | 1,259 | 69% | 911 | −24 | 26.4% | 42.9% |
China’s monthly runs are the National Bureau of Statistics measurement rather than JODI’s, which China fills in as a balancing item (see Section 6). A dash in the monthly columns means the country has not reported those months to JODI: Russia, India, Iran, Brazil, Iraq, Venezuela, Singapore. Canada’s JODI intake runs above its nameplate, so the two sources count refinery input differently there; read any monthly percentage above 100 as a trend rather than a level. Regions where fewer than 25% of capacity reports monthly show a dash rather than a total built from a handful of refineries. The refinery counts come from Wikipedia’s list of oil refineries and are a floor rather than a census: it names 47 Chinese refineries against well over 200 in operation, most of the missing ones independent “teapots” in Shandong.
Refining margins
+41.0 vs 5-yr avg, +42.8 y/y
+32.6 vs 5-yr avg, +31.7 y/y
+57.9 vs 5-yr avg, +65.0 y/y
Spot prices for conventional gasoline, ultra-low sulphur diesel and kerosene-type jet fuel at the US Gulf Coast, and WTI at Cushing, from EIA via FRED. A crack is the gross margin a refiner earns converting a barrel of crude into that product, before operating costs.
Inventories
Against the last five years
−152 vs 5-yr avg, −120 y/y
−246 vs 5-yr avg, −182 y/y
+49 vs 5-yr avg, +31 y/y
−197 vs 5-yr avg, −151 y/y
−171 vs 5-yr avg, −174 y/y
−282 vs 5-yr avg, −223 y/y
+13 vs 5-yr avg, −7 y/y
−269 vs 5-yr avg, −231 y/y
−14 vs 5-yr avg, −14 y/y
−15 vs 5-yr avg, −21 y/y
−12.1 vs 5-yr avg, −11.7 y/y
−28 vs 5-yr avg, −33 y/y
−13 vs 5-yr avg, −16 y/y
−35 vs 5-yr avg, −22 y/y
−11.7 vs 5-yr avg, −12.6 y/y
−47 vs 5-yr avg, −35 y/y
+0.8 vs 5-yr avg, −1.8 y/y
+7 vs 5-yr avg, +6 y/y
+1.9 vs 5-yr avg, −0.1 y/y
+9 vs 5-yr avg, +6 y/y
−2.2 vs 5-yr avg, +2.7 y/y
−16.2 vs 5-yr avg, −10.5 y/y
−0.1 vs 5-yr avg, −1.4 y/y
−16 vs 5-yr avg, −12 y/y
The grey band is the lowest and highest level in the same week or month across the five previous years, and the dashed line their average; a band point is drawn only where all five years have data. The United States is EIA weekly data, crude and total including the Strategic Petroleum Reserve. OECD, non-OECD and global are JODI monthly closing stocks, which include government stocks, summed over a fixed set of countries that reported every month since January 2021, so a country starting or stopping reporting does not show up as a build or a draw. That rule costs the non-OECD most of its coverage: India stopped reporting in 2026, and China, Russia, Brazil, the UAE and Singapore do not report stocks at all, so the non-OECD and global panels are the reporting countries rather than the region. Days of cover divide stocks by demand (JODI’s total demand, EIA’s product supplied for the US) and, for crude, by refinery crude runs, over the countries that report both. Distillate is gasoil and diesel; jet fuel is jet kerosene. China’s crude is the estimate described below.
Table 16. Stocks by product and region, end of July 2026, million barrels
| Crude oil | Gasoline | Distillate | Jet fuel | Light ends | Residual fuel oil | All products | Total oil | Crude vs year ago | Total vs year ago | Countries | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| World total | 1,915.0 | 423.3 | 595.5 | 146.2 | 219.9 | 107.1 | 1,976.4 | 4,030.3 | −150.9 | −230.5 | 42 |
| OECD total | 1,648.2 | 389.5 | 549.1 | 138.2 | 207.1 | 95.4 | 1,842.1 | 3,641.6 | −181.8 | −223.3 | 33 |
| Canada | 58.9 | 14.4 | 16.0 | 3.4 | 2.8 | 1.8 | 54.8 | 134.4 | +1.5 | −1.4 | 1 |
| Europe | 482.3 | 129.6 | 374.1 | 74.4 | 50.4 | 55.3 | 749.1 | 1,306.0 | −4.5 | −21.2 | 26 |
| Japan | 267.9 | 14.4 | 17.2 | 5.1 | 43.8 | 7.0 | 106.7 | 420.5 | −74.1 | −76.2 | 1 |
| United States | 711.8 | 209.7 | 107.2 | 46.9 | 79.5 | 21.4 | 813.7 | 1,525.5 | −117.6 | −135.0 | 1 |
| Other OECD | 127.4 | 21.3 | 34.6 | 8.5 | 30.5 | 10.0 | 117.7 | 255.2 | +13.0 | +10.5 | 4 |
| Non-OECD total | 266.8 | 33.9 | 46.4 | 8.0 | 12.8 | 11.6 | 134.3 | 388.7 | +30.9 | −7.1 | 9 |
| China | - | - | - | - | - | - | - | - | - | - | - |
| Eurasia | 4.3 | 0.5 | 1.1 | 0.2 | - | 0.2 | 2.5 | 2.0 | +4.1 | −0.4 | 1 |
| Europe | 4.0 | 1.7 | 5.4 | - | - | 1.7 | 10.1 | - | +0.7 | - | 2 |
| Other Asia | 50.8 | 9.2 | 18.9 | 7.0 | 9.4 | 9.7 | 57.1 | 108.1 | +2.4 | +5.2 | 3 |
| Other non-OECD | 207.6 | 22.4 | 21.1 | 0.9 | 3.4 | 0.1 | 64.5 | 278.6 | +23.6 | −12.0 | 3 |
The same data and the same countries as the seasonal charts above, so the OECD, non-OECD and world rows are the latest points of those charts. JODI closing stocks, which include government stocks, for the countries that reported that product every month since January 2021; the United States is EIA’s last weekly reading of the month, including the SPR, in place of its JODI submission. Each column counts its own reporting countries, so the products columns need not add up to all products. Distillate is gasoil and diesel; light ends are LPG and naphtha; total oil is crude and other refinery feedstocks plus all products. Countries is the number reporting crude. Large refiners not reporting crude stocks: Greece, Mexico, Israel, China, Belarus, Kazakhstan, Russia, Turkmenistan, Uzbekistan, Romania, India, Indonesia, Malaysia, Pakistan, Singapore, Vietnam, Argentina, Brazil, Colombia, Peru, Venezuela, Bahrain, Iran, Iraq, Oman, Qatar, United Arab Emirates, Algeria, Egypt, South Africa. The non-OECD totals are what is visible, not what exists.
Table 17. Strategic and commercial stocks, end of July 2026, million barrels
| Total stocks | Strategic (public) | Commercial (industry) | Strategic share | Countries | |
|---|---|---|---|---|---|
| OECD total | 3,628.9 | 1,154.0 | 2,474.9 | 32% | 28 |
| Canada | 134.4 | 0.0 | 134.4 | 0% | 1 |
| Europe | 1,303.9 | 532.2 | 771.7 | 41% | 22 |
| Japan | 420.5 | 216.2 | 204.3 | 51% | 1 |
| United States | 1,525.5 | 304.8 | 1,220.7 | 20% | 1 |
| Other OECD | 244.6 | 100.8 | 143.8 | 41% | 3 |
| China (estimate) | 1,500.0 | 360.0 | 1,140.0 | 24% | 1 |
United States: EIA weekly data for the last week of the month, the Strategic Petroleum Reserve against all other crude and product stocks (304.8 of 1,525.5). Other OECD importers: the IEA’s split of each country’s emergency stocks into public and industry holdings (June 2026, in days of net imports), applied to its JODI stock level. That is an approximation, because the IEA figure includes stocks held abroad under bilateral tickets while JODI counts what is on national territory. Canada, Mexico and Norway are net exporters with no stockholding obligation and hold no public stocks, so all of their stock counts as commercial. China: crude only, the level estimated below with the government-held share EIA gave for December 2025. No comparable split exists for the rest of the non-OECD.
China
China publishes no inventory data. The analysts who estimate it (EIA, the IEA, Kpler, Vortexa) mostly work the same way: crude supply, meaning production plus imports minus exports, less the crude refineries actually ran, with the remainder assumed to have gone into tanks. Kpler and Vortexa add satellite and ship-tracking observations of the tanks themselves, which are subscription products; the IEA publishes its China figures in the paid Oil Market Report. EIA published a level: nearly 1.4 billion barrels at the end of December 2025, about 360 million of it government-held and the rest commercial stocks the national oil companies have been directed to hold for emergencies (EIA, Today in Energy, 20 April 2026). The charts below take that level and move it month by month with the same balance, using JODI for supply and the National Bureau of Statistics’ measured refinery runs. Measured over 2025 this method gives average builds of 1,088 kb/d, against EIA’s estimate of about 1.1 million barrels a day, which is the check that it tracks.
The residual also absorbs crude burned without refining and any error in the trade data, so treat a single month with caution and the trend with more confidence. China’s runs come from the NBS rather than JODI: the JODI refinery-intake figure for China equals production plus imports minus exports to the barrel every month, which means it is filled in as a balancing item and a balance built from it is zero by construction. NBS reports runs in tonnes, converted here at 7.3 barrels a tonne.
Curves and spreads
Today and a month ago: NYMEX settlements via ComCurv. A year ago: the same delivery months as priced then (Yahoo Finance), so the dashed line shows how the market has repriced these exact contracts.
Today and a month ago: NYMEX settlements via ComCurv. A year ago: the same delivery months as priced then (Yahoo Finance), so the dashed line shows how the market has repriced these exact contracts.
Today and a month ago: NYMEX settlements via ComCurv. A year ago: the same delivery months as priced then (Yahoo Finance), so the dashed line shows how the market has repriced these exact contracts.
Today and a month ago: NYMEX settlements via ComCurv, every leg on the same trade date. A year ago: the same delivery months as priced then (Yahoo Finance). Products at 42 gallons a barrel.
Today and a month ago: NYMEX settlements via ComCurv, every leg on the same trade date. A year ago: the same delivery months as priced then (Yahoo Finance). Products at 42 gallons a barrel.
Today and a month ago: NYMEX settlements via ComCurv, every leg on the same trade date. A year ago: the same delivery months as priced then (Yahoo Finance). Products at 42 gallons a barrel.
Brent is the NYMEX Brent (BZ) contract, which settles on ICE Brent, from Yahoo Finance; WTI is the NYMEX settlement for the same named month. Same-month pairs are how the spread is quoted, although a Brent contract expires about three weeks before the WTI contract of the same name.
Timespread history is EIA’s daily first-to-fourth contract prices, which EIA stopped publishing after April 2024, then NYMEX settlements via ComCurv from mid-March 2026 (June 2026 for ULSD and RBOB). In between, both spreads are rebuilt from every individual contract’s daily settlement (TradingView), taking contract one as the nearest month that has not yet expired, so there is no roll to work around and no day is left out. Where the rebuilt series overlaps ComCurv’s sound data it matches to the cent. Products are converted from $/gal at 42 gallons a barrel.
The Brent timespreads are stitched from every individual ICE Brent contract’s daily settlement (TradingView) from July 2020, contract one being the nearest month not yet expired, and carried forward from the individual contracts on Yahoo Finance; EIA never published Brent futures. Brent was at a discount to WTI until the shale boom stranded crude inland at Cushing; the spread peaked near $30 in 2011 and narrowed once pipelines reached the Gulf Coast, before the US lifted its crude export ban in December 2015. Since then it has traded near the cost of shipping a barrel from the Gulf to Europe. See Oil 101 Chapter 17.
What it says: in Oct 2026 stocks were 152 million barrels below their five-year average. On the usual relationship (dashed line) that implies a spread of about $1.43; the actual spread was $3.65, so the curve was $2.22 tighter than stocks alone explain, which is the market pricing a supply risk the tanks do not yet show.
Grey dots are months since 2010, blue the last six, red the latest. The dashed line is fitted to every month before 2026: each 100 million barrels of stocks below normal has gone with about $1.21 more backwardation.
What it says: in Jul 2026 stocks were 294 million barrels above their five-year average. On the usual relationship (dashed line) that implies a spread of about -$1.97; the actual spread was $3.92, so the curve was $5.89 tighter than stocks alone explain, which is the market pricing a supply risk the tanks do not yet show.
Grey dots are months since 2010, blue the last six, red the latest. The dashed line is fitted to every month before 2026: each 100 million barrels of stocks below normal has gone with about $0.73 more backwardation.
Stocks are measured against the average of the same month in the five years before, so a normal seasonal build does not look like a surplus. Spread is the WTI first-to-fourth month spread; there is no free daily Brent curve history, so both panels use WTI. Months with no spread data (May 2024 to February 2026) are missing. Both panels count strategic stocks: the US SPR releases of 2022 and 2026 moved barrels from government caverns into commercial tanks, so commercial stocks alone looked comfortable while the total fell. The second panel adds the US SPR to OECD commercial stocks (other OECD government stocks have no free monthly series in barrels and are held near a mandated level) and an estimate for China, which publishes no stock data: from December 2024, EIA’s December 2025 estimate of about 1.4 billion barrels moved by China’s implied monthly balance; before that, a straight line from 850 million barrels in mid-2016, the low end of what satellite tank counts saw through 2024, because the implied balance overstates the build over those years by about a billion barrels. China enters the panel from June 2021, the first month with five years of estimates behind it, and the panel ends at the last month China can be estimated. With China counted, 2026 sits on the loose side of the line while the curve is steeply backwardated. The two readings are consistent: China added about 400 million barrels in 2025 to stocks it holds against its own emergencies, and the curve prices the barrels the market can reach, which those are not.
US natural gas
Today and a month ago: NYMEX settlements via ComCurv. A year ago: the same delivery months as priced then (Yahoo Finance), so the dashed line shows how the market has repriced these exact contracts.
Gas timespreads are seasonal by nature: the October-November and March-April spreads carry the winter premium, so a month’s reading is best compared with the same month in earlier years. Sources as for the oil timespreads: EIA to April 2024, every individual contract on TradingView stitched into exact first-to-second and first-to-fourth spreads to September 2026, then ComCurv.
Table 18. Largest natural gas producers, 2025
| Bcf/d | mb/d oil equivalent | Share of world | Change on 2024 | |
|---|---|---|---|---|
| United States | 103.9 | 17.31 | 25.6% | +4% |
| Russia | 59.0 | 9.83 | 14.5% | -3% |
| Iran | 25.6 | 4.27 | 6.3% | +1% |
| China | 25.6 | 4.26 | 6.3% | +7% |
| Canada | 20.0 | 3.33 | 4.9% | +4% |
| Qatar | 17.8 | 2.96 | 4.4% | +2% |
| Australia | 14.4 | 2.40 | 3.5% | -2% |
| Saudi Arabia | 12.9 | 2.16 | 3.2% | +9% |
| Norway | 11.7 | 1.94 | 2.9% | -3% |
| Algeria | 9.5 | 1.58 | 2.3% | -2% |
| Malaysia | 7.8 | 1.29 | 1.9% | +1% |
| Turkmenistan | 7.0 | 1.17 | 1.7% | -1% |
| World | 406.0 | 67.7 | 100% | - |
Table 19. Largest LNG exporters, 2025
| bcm | Bcf/d | Mt LNG | Share of world | Change on 2024 | |
|---|---|---|---|---|---|
| United States | 147.0 | 14.2 | 107.3 | 25.4% | +27% |
| Qatar | 111.0 | 10.7 | 81.0 | 19.2% | +3% |
| Australia | 104.8 | 10.1 | 76.5 | 18.1% | -2% |
| Russia | 42.6 | 4.1 | 31.1 | 7.4% | -4% |
| Malaysia | 36.5 | 3.5 | 26.6 | 6.3% | +1% |
| Nigeria | 20.1 | 1.9 | 14.6 | 3.5% | +9% |
| Oman | 15.3 | 1.5 | 11.2 | 2.6% | -4% |
| Indonesia | 15.3 | 1.5 | 11.1 | 2.6% | -8% |
| Algeria | 13.1 | 1.3 | 9.6 | 2.3% | -17% |
| Papua New Guinea | 11.8 | 1.1 | 8.6 | 2.0% | +6% |
| World | 578 | 56.0 | 422 | 100% | - |
Table 20. Largest LNG importers, 2025
| bcm | Bcf/d | Mt LNG | Share of world | Change on 2024 | |
|---|---|---|---|---|---|
| China | 90.8 | 8.8 | 66.3 | 15.7% | -14% |
| Japan | 88.5 | 8.6 | 64.6 | 15.3% | -0% |
| South Korea | 63.1 | 6.1 | 46.0 | 10.9% | -1% |
| India | 35.4 | 3.4 | 25.8 | 6.1% | -8% |
| Taiwan | 31.8 | 3.1 | 23.2 | 5.5% | +9% |
| France | 30.9 | 3.0 | 22.5 | 5.3% | +20% |
| Spain | 22.3 | 2.2 | 16.3 | 3.9% | +21% |
| Italy | 20.2 | 2.0 | 14.7 | 3.5% | +39% |
| Turkey | 16.3 | 1.6 | 11.9 | 2.8% | +31% |
| Belgium | 14.4 | 1.4 | 10.5 | 2.5% | +87% |
| World | 578 | 56.0 | 422 | 100% | - |
Energy Institute Statistical Review of World Energy 2025: marketed production, excluding gas flared or reinjected, and gross LNG trade. The US is 26% of world gas production and 25% of LNG exports. Oil equivalent is at 6,000 cubic feet of gas to the barrel, the SEC reporting convention; on energy content alone a barrel of oil is closer to 5,800. LNG tonnes use the EI’s approximate 0.73 million tonnes per billion cubic metres. The EI publishes several European importers, including Germany and the Netherlands, only inside an “other EU” total, so they cannot be ranked here, and the exporter list leaves out the EI’s “other” rows, which are mostly re-exports.
+38 vs 5-yr avg, −131 y/y
+22 vs 5-yr avg, +2 y/y
+13 vs 5-yr avg, +9 y/y
−36 vs 5-yr avg, −125 y/y
+15 vs 5-yr avg, −11 y/y
+25 vs 5-yr avg, −5 y/y
Table 21. US marketed natural gas production by basin, billion cubic feet per day
| Basin (main states) | Sep 2026 | A year earlier | Change | Dec 2027 (EIA forecast) |
|---|---|---|---|---|
| Appalachia (PA, WV, OH) | 37.0 | 36.6 | +0.4 | 37.9 |
| Permian (TX, NM) | 30.2 | 28.6 | +1.6 | 33.5 |
| Haynesville (LA, TX) | 17.4 | 14.9 | +2.5 | 18.0 |
| Eagle Ford (TX) | 8.1 | 7.6 | +0.4 | 8.6 |
| Rest of Lower 48 | 25.7 | 25.4 | +0.3 | 26.2 |
| Bakken (ND, MT) | 3.5 | 3.4 | 0.0 | 3.4 |
| Federal Gulf of America (offshore) | 2.1 | 2.0 | 0.0 | 1.8 |
| Alaska (AK) | 1.0 | 1.0 | 0.0 | 1.1 |
Table 22. US LNG export capacity still to come, from EIA’s Q3 2026 list
| Project | State | Status | Capacity, Bcf/d | Expected in service |
|---|
In commercial operation: 0.0 Bcf/d of baseload nameplate. Including everything commissioning or under construction: 0.0 Bcf/d. Plants run above baseload nameplate when conditions allow, which is why exports can exceed the capacity line. Approved projects that have not taken a final investment decision are left out, because they have no date. Dates given as a half-year or a year are placed mid-period.
Table 23 · EIA STEO 5a. U.S. Natural Gas Supply, Consumption, and Inventories
| 2024 | 2025 | 1Q26 | 2Q26 | 3Q26 | 4Q26 | 2026 | 2027 | |
|---|---|---|---|---|---|---|---|---|
| Supply (billion cubic feet per day) | ||||||||
| U.S. total marketed natural gas production | 113.07 | 118.42 | 120.96 | 122.68 | 125.08 | 125.40 | 123.53 | 127.95 |
| Alaska | 1.02 | 1.01 | 1.10 | 1.02 | 0.93 | 1.05 | 1.03 | 1.02 |
| Federal Gulf of America | 1.83 | 1.93 | 2.05 | 2.04 | 2.06 | 2.02 | 2.04 | 1.88 |
| Lower 48 States (excl GOA) | 110.23 | 115.47 | 117.81 | 119.61 | 122.09 | 122.33 | 120.46 | 125.05 |
| Appalachia region | 35.55 | 36.65 | 36.91 | 37.23 | 37.22 | 37.10 | 37.12 | 37.64 |
| Bakken region | 3.28 | 3.31 | 3.26 | 3.40 | 3.44 | 3.42 | 3.38 | 3.38 |
| Eagle Ford region | 6.98 | 7.44 | 7.72 | 7.87 | 8.09 | 8.12 | 7.95 | 8.34 |
| Haynesville region | 14.48 | 15.03 | 16.01 | 16.56 | 17.23 | 17.52 | 16.83 | 17.77 |
| Permian region | 24.95 | 27.55 | 28.42 | 29.02 | 30.22 | 30.46 | 29.53 | 32.03 |
| Rest of Lower 48 States | 25.12 | 25.43 | 25.41 | 25.45 | 25.82 | 25.71 | 25.60 | 25.90 |
| Total primary supply | 90.31 | 91.96 | 107.72 | 79.03 | 87.61 | 95.39 | 92.44 | 93.90 |
| Balancing item | -0.54 | -0.45 | 0.07 | -0.91 | -1.03 | -1.21 | -0.77 | -1.46 |
| Total supply | 90.86 | 92.41 | 107.65 | 79.94 | 88.64 | 96.60 | 93.21 | 95.36 |
| U.S. total dry natural gas production | 103.08 | 107.62 | 110.05 | 111.27 | 113.59 | 113.84 | 112.19 | 116.12 |
| Net inventory withdrawals | -0.02 | 0.41 | 15.65 | -12.02 | -5.78 | 3.07 | 0.23 | 0.09 |
| Supplemental gaseous fuels | 0.26 | 0.26 | 0.30 | 0.24 | 0.26 | 0.27 | 0.27 | 0.28 |
| Net imports | -12.47 | -15.89 | -18.36 | -19.55 | -19.43 | -20.58 | -19.48 | -21.12 |
| LNG gross imports | 0.05 | 0.04 | 0.15 | 0.01 | 0.03 | 0.06 | 0.06 | 0.06 |
| LNG gross exports | 11.93 | 15.09 | 17.84 | 17.17 | 17.03 | 18.23 | 17.57 | 18.62 |
| Pipeline gross imports | 8.54 | 8.64 | 9.36 | 7.33 | 7.63 | 7.95 | 8.07 | 7.98 |
| Pipeline gross exports | 9.13 | 9.47 | 10.03 | 9.72 | 10.06 | 10.36 | 10.04 | 10.54 |
| Consumption (billion cubic feet per day) | ||||||||
| Total consumption | 90.31 | 91.96 | 107.72 | 79.03 | 87.61 | 95.39 | 92.44 | 93.90 |
| Residential | 12.00 | 13.34 | 24.25 | 6.58 | 3.40 | 15.83 | 12.51 | 12.66 |
| Commercial | 9.07 | 9.92 | 15.36 | 6.44 | 4.93 | 11.39 | 9.53 | 9.85 |
| Industrial | 23.44 | 23.65 | 25.14 | 22.89 | 22.46 | 24.79 | 23.82 | 24.47 |
| Electric power | 36.78 | 35.74 | 32.89 | 34.10 | 47.34 | 33.57 | 36.98 | 37.03 |
| Lease and plant fuel | 5.33 | 5.58 | 5.70 | 5.78 | 5.90 | 5.91 | 5.83 | 6.03 |
| Pipeline and distribution | 3.49 | 3.55 | 4.16 | 3.05 | 3.38 | 3.70 | 3.58 | 3.64 |
| Vehicle | 0.19 | 0.18 | 0.20 | 0.20 | 0.20 | 0.20 | 0.20 | 0.22 |
| End-of-period working natural gas inventories (billion cubic feet) | ||||||||
| United States total | 3,438 | 3,306 | 1,906 | 2,995 | 3,523 | 3,240 | 3,240 | 3,231 |
| East region | 747 | 705 | 275 | 601 | 860 | 757 | 757 | 745 |
| Midwest region | 893 | 829 | 351 | 720 | 1,006 | 890 | 890 | 883 |
| South Central region | 1,215 | 1,184 | 795 | 1,090 | 1,067 | 1,111 | 1,111 | 1,075 |
| Mountain region | 259 | 250 | 198 | 234 | 252 | 206 | 206 | 230 |
| Pacific region | 295 | 304 | 259 | 318 | 297 | 247 | 247 | 269 |
| Alaska | 28 | 33 | 28 | 33 | 40 | 29 | 29 | 30 |
Natural gas liquids
EIA petroleum supply and disposition for US production (field production at gas plants), exports and end-month stocks; Mont Belvieu front-month futures via ComCurv. The composite NGL barrel weights each product’s price by its share of US gas plant output over the latest twelve months (ethane 42%, propane 30%, normal butane 9%, isobutane 6%, natural gasoline 12%), so it is the value of the barrel US plants actually make, and it moves with the mix: ethane, the largest and cheapest component, is priced off natural gas rather than crude, which is why the barrel trades at a fraction of WTI (39% in October 2026). Ethane that is not worth recovering is left in the gas and sold as methane, so ethane output responds to its price against gas as well as to drilling.
Table 24. Largest NGL producers, 2025
| kb/d | Share of world | Change on 2024 | |
|---|---|---|---|
| United States | 7,516 | 50.6% | +7% |
| Saudi Arabia | 1,565 | 10.5% | +0% |
| Canada | 1,181 | 7.9% | +6% |
| Iran | 710 | 4.8% | +6% |
| Russia | 622 | 4.2% | +1% |
| United Arab Emirates | 551 | 3.7% | +0% |
| Qatar | 419 | 2.8% | +6% |
| Algeria | 260 | 1.7% | +0% |
| Kuwait | 225 | 1.5% | +0% |
| Norway | 187 | 1.3% | -11% |
| World | 14,867 | 100% | - |
EIA International Energy Statistics, natural gas plant liquids. Some countries’ recent years are EIA estimates carried from earlier reports, which is why a few rows show no change.
Supply and demand history
EIA International Energy Statistics, crude and lease condensate monthly from 1980; natural gas plant liquids, other liquids (mostly biofuels) and refinery processing gain are annual before 1993 and held flat across each year’s months. US tight oil is EIA’s estimate by play, which EIA last updated for March 2024; after that it is carried forward by the STEO's Permian, Bakken and Eagle Ford output. Canadian oil sands are upgraded plus non-upgraded bitumen from the Canada Energy Regulator, from 2000. Both are carved out of crude and condensate, so the stack still adds up to EIA’s total. The fall at the right edge is real: the 2026 Strait of Hormuz closure took world crude and condensate from a record in early 2026 to the lowest since October 2020 within three months (see Oil 101 Chapter 26).
The trend is a log-linear fit to world consumption from January 1990 to December 2004, with a factor for each calendar month, extended forward: 1.7% a year. Before 2005 consumption tracked it closely; the price shock of 2005 to 2008 broke the pattern, and by September 2026 the world was using 12.6 mb/d less than the old trend implied. Observed is EIA’s monthly series; the shaded part is EIA’s forecast.
The same series as a change on a year earlier, which is where the 2026 Strait of Hormuz closure shows: world use fell about 4 mb/d below a year earlier in April and May 2026, against about 17 mb/d at the worst of Covid. On the level chart above that is a small dip. See Oil 101 Chapter 26.
Energy Institute Statistical Review world oil consumption to 2025; the shaded years are the change in EIA’s Short-Term Energy Outlook annual averages, measured on EIA’s own all-liquids basis. Demand has fallen in 10 of the 60 years since 1965; the table below says why.
Today: WTI $91.85, NYMEX front-month settlement, Fri October 9, 2026. Every year above the dashed line was dearer in real terms than oil is today.
- 11861 Oil Creek flush: the first gushers glut a market with no tanks, and barrels are worth less than the barrels they come in
- 21864 Civil War inflation and a new federal tax: in today’s money the highest annual price in the whole series
- 31870 Rockefeller incorporates Standard Oil
- 41901 Spindletop gushes near Beaumont, Texas
- 51911 The Supreme Court breaks up Standard Oil
- 61920 Postwar demand and fears of US exhaustion
- 71931 East Texas field and the Depression: oil below 10 cents a barrel at the wellhead
- 81960 OPEC founded in Baghdad
- 91974 Arab oil embargo; the posted price quadruples
- 101980 Iranian revolution and the Iran-Iraq war
- 111986 Saudi Arabia abandons swing production and the price halves
- 121990 Iraq invades Kuwait
- 131998 Asian financial crisis; OPEC raises quotas into falling demand
- 142008 Brent reaches $147 in July, then the financial crisis
- 152011 Arab Spring and the loss of Libyan exports
- 162015 US shale glut; OPEC declines to cut
- 172020 Covid; WTI settles below zero in April
- 182022 Russia invades Ukraine
- 192026 Strait of Hormuz closed (Oil 101 Chapter 26)
Annual average crude price from the Energy Institute Statistical Review (S&P Global Energy data): the US average to 1944, Arabian Light posted at Ras Tanura from 1945 to 1983, and Brent dated since 1984. The Review restates every year in 2025 dollars using the US consumer price index; this chart carries that forward to August 2026 with the same index (FRED CPIAUCSL, ×1.038). 2026 is Brent's average from January to October 2026, from FRED, and is not a full year. The dashed line is today's price, the NYMEX WTI front-month settlement, which needs no inflation adjustment. A single CPI across 165 years is a rough instrument before 1913, when the index begins and earlier years are reconstructed, so read the nineteenth-century levels as orders of magnitude.
Table 25. Every year oil use has fallen since 1859
Falls are rare, and most have a cause that made the history books. In the 10 years since 1965 when world consumption fell, the reasons are a short list: the 1973 embargo, the 1979 revolution in Iran and the recession that followed, the 2008 financial crisis and the pandemic; in 1983 and 1993 use was essentially flat. Before 1965 the series is production, so some falls are supply disruptions (Baku in 1905, the Second World War in 1942) rather than falls in demand, and one, 1876, is a quirk of the statistics. The 2020 fall was the largest since at least 1965, 8,826 thousand barrels a day, 9.0%, and use was back above its 2019 level by 2023; the 1980 to 1983 fall took until 1988 to recover.
| Year | Measure | Before, kb/d | After, kb/d | Change, kb/d | Change | Back above the previous peak | Why |
|---|---|---|---|---|---|---|---|
| 1863 | US production | 8 | 7 | −1 | −14.6% | 1866 | Output fell after the 1861 price collapse to about 10 cents a barrel closed many of the first Oil Creek operations and water flooding forced out others, even as Civil War demand and a heavy federal tax on alcohol-based illuminants pushed prices up. source |
| 1864 | US production | 7 | 6 | −1 | −19.0% | 1866 | The same contraction continued as wells abandoned after the 1861 price collapse or lost to water flooding were not replaced fast enough, so output fell again despite wartime prices that rose to the highest real levels of the century. source |
| 1867 | US production | 10 | 9 | −1 | −7.0% | 1868 | Output slipped as the Pithole boom of 1865 collapsed, its wells going dry within months, while the postwar fall in commodity demand drove a second crude price collapse in 1866 that discouraged new drilling. source |
| 1871 | US production | 14 | 14 | −0 | −1.1% | 1872 | Uncertain: the 1 percent dip sits within the boom-and-bust drilling cycles of the early 1870s in western Pennsylvania, and no specific cause is documented. source |
| 1876 | US production | 33 | 25 | −8 | −24.9% | 1877 | Largely a statistical artifact: the national 1875 figure includes a lump of 3 million barrels credited to West Virginia that covers all of that state's production before 1876, while Pennsylvania and New York output actually rose slightly in 1876 after falling about 20 percent in 1875. source |
| 1883 | US production | 83 | 64 | −19 | −22.7% | 1889 | Output fell back after the short-lived Cherry Grove flush of 1882, which lifted Pennsylvania and New York to a record 105,000 barrels a day in July 1882 before collapsing to under 4,000 a day from that pool by year end, while the wider Bradford district declined from its 1881 to 1882 peak. source |
| 1885 | US production | 66 | 60 | −6 | −9.7% | 1889 | Pennsylvania and New York production kept declining as the Bradford and New York fields aged, with output per well falling from about 3.5 to about 2.5 barrels a day even though the number of producing wells rose by 1,600 during the year. source |
| 1888 | US production | 77 | 76 | −2 | −2.4% | 1889 | Pennsylvania and New York output fell 26 percent because producers agreed to restrict drilling to lift prices, which more than offset a near doubling of production in the new Lima field of Ohio. source |
| 1892 | US production | 149 | 138 | −10 | −7.0% | 1896 | Output fell from the 1891 record set by the short-lived McDonald field flush near Pittsburgh, as the original Pennsylvania fields entered permanent decline and prices sank to 56 cents a barrel. source |
| 1893 | US production | 138 | 133 | −6 | −4.1% | 1896 | Pennsylvania production fell from 27.1 to 19.3 million barrels and Lima, Ohio, also declined, outweighing gains in West Virginia and Indiana, a supply-side fall that came with rising crude prices and record exports. source |
| 1897 | US production | 167 | 166 | −1 | −0.8% | 1900 | A sharp fall in Lima-Indiana output and far fewer new wells completed under falling prices offset small gains in the Appalachian field, leaving national production just below the 1896 record. source |
| 1898 | US production | 166 | 152 | −14 | −8.5% | 1900 | Production fell by 5 million barrels after the depressed prices of 1897 had cut the number of new wells drilled in the Appalachian and Lima-Indiana fields, before Texas and California were large enough to compensate. source |
| 1905 | World production | 552 | 543 | −9 | −1.6% | 1907 | Riots and revolutionary disturbances in the Baku oil region cut Russian output by about 23.6 million barrels, a 30 percent fall that outweighed a large rise in US production. source |
| 1906 | World production | 543 | 541 | −2 | −0.4% | 1907 | US output fell by about 8 million barrels as the Gulf Coast flush pools fell off and Appalachian production kept declining, while Russian output recovered only partly from the 1905 disturbances. source |
| 1918 | World production | 1,285 | 1,276 | −9 | −0.7% | 1919 | Russian production collapsed amid the revolution and civil war, as Baku passed through Soviet rule, the March 1918 massacres, nationalisation, the Battle of Baku and Ottoman occupation, more than offsetting rising US output. source |
| 1924 | World production | 2,661 | 2,654 | −8 | −0.3% | 1925 | US output dipped from its 1923 peak as the flush Southern California fields at Signal Hill, Santa Fe Springs and Huntington Beach began to take water and were curtailed, while Mexican production continued its decline from the 1921 peak. source |
| 1930 | World production | 3,838 | 3,660 | −178 | −4.6% | 1934 | The onset of the Great Depression cut oil demand at the same time as huge new supply, including the East Texas field from 1930, collapsed prices and forced production cuts. source |
| 1931 | World production | 3,660 | 3,542 | −119 | −3.2% | 1934 | The Depression deepened while the flood of East Texas oil drove prices down about two thirds from 1926 leading Texas and Oklahoma to impose state limits on production. source |
| 1932 | World production | 3,542 | 3,380 | −161 | −4.6% | 1934 | Continued Depression demand weakness and the new state proration regimes that restricted output to support prices kept world production falling for a third year. source |
| 1938 | World production | 5,249 | 5,124 | −125 | −2.4% | 1939 | The sharp US recession of 1937 to 1938 cut demand and state regulators restricted output, and the boycott of Mexican oil after the March 1938 expropriation reduced Mexico's sales. source |
| 1942 | World production | 5,677 | 5,343 | −333 | −5.9% | 1943 | U-boat attacks on the small lake tankers serving Lake Maracaibo forced Venezuela to shut in production, cutting it from about 600,000 to about 400,000 barrels a day, while Dutch demolition teams wrecked the East Indies fields ahead of the Japanese invasion. source |
| 1945 | World production | 6,605 | 6,597 | −8 | −0.1% | 1946 | Uncertain: the tiny fall most likely reflects wartime destruction and dislocation in Axis-held fields such as Romania and the Dutch East Indies, together with the end of war demand after August, offsetting growth in the Americas. source |
| 1949 | World production | 8,749 | 8,699 | −50 | −0.6% | 1950 | US production was cut about 8 percent through lower Texas Railroad Commission allowables as demand stagnated in the 1948 to 1949 recession and cheaper imported crude and residual fuel oil kept rising. source |
| 1974 | World consumption | 56,076 | 55,280 | −796 | −1.4% | 1976 | The Arab oil embargo from October 1973 and the doubling of posted prices on 1 January 1974 cut consumption, followed by the recession that began in late 1973. source |
| 1975 | World consumption | 55,280 | 55,028 | −252 | −0.5% | 1976 | Consumption fell again as the world recession of 1974 to 1975 and the fourfold rise in prices since 1973 continued to suppress demand. source |
| 1980 | World consumption | 64,345 | 61,435 | −2,910 | −4.5% | 1988 | The doubling of real prices after the Iranian revolution and the outbreak of the Iran-Iraq war pushed the major economies into recession and began a lasting shift away from oil. source |
| 1981 | World consumption | 61,435 | 59,607 | −1,828 | −3.0% | 1988 | Consumption kept falling as high prices drove fuel switching and efficiency gains and a second US recession began in July 1981. source |
| 1982 | World consumption | 59,607 | 58,062 | −1,544 | −2.6% | 1988 | The long-run demand response to the price shocks of the 1970s, with oil displaced from power generation and heating, continued through the 1981 to 1982 recession. source |
| 1983 | World consumption | 58,062 | 57,977 | −85 | −0.1% | 1988 | Consumption bottomed out as the lingering effects of high prices and fuel switching roughly offset the start of economic recovery. source |
| 1993 | World consumption | 67,351 | 67,349 | −2 | flat | 1994 | The collapse of oil use in the former Soviet Union and Eastern Europe after 1991 roughly cancelled out growth elsewhere, leaving world consumption essentially flat. source |
| 2008 | World consumption | 86,242 | 85,287 | −955 | −1.1% | 2010 | A price spike to a record $147 a barrel in July, driven by strong Asian demand meeting stagnant supply, cut consumption in the first half, and the financial crisis cut it further in the second. source |
| 2009 | World consumption | 85,287 | 83,970 | −1,317 | −1.5% | 2010 | The global recession that followed the financial crisis cut oil demand for a second consecutive year. source |
| 2020 | World consumption | 98,580 | 89,754 | −8,826 | −9.0% | 2023 | Covid-19 lockdowns and travel restrictions caused the largest fall in oil consumption on record, with road and air travel collapsing in the spring. source |
| 2026 | World consumption, EIA forecast | 104,406 | 102,441 | −1,965 | −1.9% | - | The closure of the Strait of Hormuz and the price spike that followed (EIA forecast; see Oil 101 Chapter 26). source |
No single annual series covers the whole span, so three are joined and each row says which it is. Before 1900, US crude oil production (EIA), which was most of the world’s output until Russia’s Baku fields grew in the 1880s and 1890s. From 1900 to 1964, world crude oil production (Our World in Data, from The Shift Project), converted at 1.7MWh a barrel; over a year production and use differ only by stock changes, but a fall in production can also be a supply disruption rather than a fall in demand. From 1965, world oil consumption (Energy Institute Statistical Review). Back above the previous peak is the first year the series exceeded its highest level before the fall. The shaded row, where present, is EIA’s forecast for the current year.
EIA International Energy Statistics consumption over World Bank population, barrels per person per year. BRICS is its ten members as of 2025 (Brazil, Russia, India, China, South Africa, Egypt, Ethiopia, Iran, the UAE and Indonesia) applied to every year, and starts in 1992 with Russia’s series. The European Union is its current 27 members throughout.
IEA figures are the central scenario of each World Energy Outlook: Reference to 2009, New Policies 2010 to 2018, Stated Policies 2019 to 2024; WEO 2025 reintroduced Current Policies and is shown under both. They are IEA “world oil” demand, which excludes biofuels, so they sit below EIA’s International Energy Outlook reference cases, which count all liquids. Actual is EIA’s world consumption, drawn twice: all liquids, and less other liquids (mostly biofuels) for the IEA basis. EIA published no world outlook in 2012, 2014, 2015, 2018, 2020, 2022, 2024 or 2025.
Table 26. Who has been right about oil demand: track record by agency and era
Both agencies have leaned high: 62% of all the projections that can now be checked expected more oil use than happened. The IEA’s Reference-scenario editions of 2004 to 2009 were the most optimistic, +6.4% on average; its New Policies editions from 2010 to 2018 cut the bias to +1.6%. EIA’s reference cases leaned +2.0% from 2006 to 2011 and +1.7% since 2013. By average error the most accurate run has been EIA, reference case IEO 2013 to 2023, off by 2.9%, although recent editions are checked only over their first few years, which are the easiest to get right. The single closest editions have been IEO 2023 (1.1%), IEO 2021 (1.2%), IEO 2017 (2.4%); the furthest off was WEO 2004 (13.3%). Looking ahead, the IEA’s two 2025 scenarios are 15.9 mb/d apart by 2050, one still growing and one past its peak, while EIA’s latest long-term case is higher than either.
| Agency and era | Editions | Years checked | Average error | Average bias | Share too high |
|---|---|---|---|---|---|
| IEA, Reference scenario (WEO 2004 to 2009) | 6 | 9 | 7.3% | +6.4% | 89% |
| IEA, New Policies scenario (WEO 2010 to 2018) | 9 | 19 | 4.5% | +1.6% | 47% |
| IEA, Stated Policies scenario (WEO 2019 and 2020) | 2 | 2 | 2.0% | +2.0% | 100% |
| EIA, reference case (IEO 2006 to 2011) | 6 | 22 | 4.7% | +2.0% | 64% |
| EIA, reference case (IEO 2013 to 2023) | 6 | 25 | 2.9% | +1.7% | 60% |
| What they say now | Scenario | Basis | Projection |
|---|---|---|---|
| EIA Short-Term Energy Outlook | October 2026 | all liquids | 104.6 mb/d in 2027 |
| EIA International Energy Outlook 2023 | Reference | all liquids | 105.5 in 2030, 121.5 in 2050 |
| IEA World Energy Outlook 2025 | Current Policies | excludes biofuels | 105.3 in 2035, 112.8 in 2050 |
| IEA World Energy Outlook 2025 | Stated Policies | excludes biofuels | 100.4 in 2035, 96.9 in 2050 |
Error and bias are averaged over every projection year that has now happened, per edition. Share too high is the share of those checked projections that exceeded the outcome. For comparison, actual world consumption in 2025 was 104.4 mb/d on EIA’s all-liquids basis. The IEA’s Stated Policies editions since 2021 have no checked years yet, because their first projection year is 2030.
Table 27. Past long-term projections of world oil demand, edition by edition
| Edition | Scenario | Years scored | Mean error | Mean bias | 2030 projection, mb/d | Furthest projection |
|---|---|---|---|---|---|---|
| WEO 2004 | Reference | 2010, 2020 | 13.3% | +13.3% | 121.3 | 121.3 in 2030 |
| WEO 2005 | Reference | 2010 | 6.8% | +6.8% | 115.0 | 115.0 in 2030 |
| WEO 2006 | Reference | 2010, 2015 | 6.7% | +6.7% | 116.3 | 116.3 in 2030 |
| WEO 2007 | Reference | 2010, 2015 | 6.2% | +6.2% | 116.3 | 116.3 in 2030 |
| WEO 2008 | Reference | 2015 | 2.2% | +2.2% | 106.4 | 106.4 in 2030 |
| WEO 2009 | Reference | 2015 | 4.3% | −4.3% | 105.2 | 105.2 in 2030 |
| WEO 2010 | New Policies | 2015 to 2025 (3) | 4.6% | −1.8% | 96.4 | 99.0 in 2035 |
| WEO 2011 | New Policies | 2015 to 2025 (3) | 4.2% | −0.5% | 96.9 | 99.4 in 2035 |
| WEO 2012 | New Policies | 2015 to 2025 (3) | 4.0% | +1.0% | 97.7 | 99.7 in 2035 |
| WEO 2013 | New Policies | 2020, 2025 | 5.4% | +3.5% | 99.5 | 101.4 in 2035 |
| WEO 2014 | New Policies | 2020, 2025 | 5.0% | +4.5% | 101.3 | 103.9 in 2040 |
| WEO 2015 | New Policies | 2020, 2025 | 5.6% | +3.8% | 99.9 | 103.5 in 2040 |
| WEO 2016 | New Policies | 2020, 2025 | 5.5% | +4.0% | 99.8 | 103.5 in 2040 |
| WEO 2017 | New Policies | 2025 | 0.6% | +0.6% | 102.2 | 104.9 in 2040 |
| WEO 2018 | New Policies | 2025 | 2.7% | +2.7% | 104.3 | 106.3 in 2040 |
| WEO 2019 | Stated Policies | 2025 | 3.8% | +3.8% | 105.4 | 106.4 in 2040 |
| WEO 2020 | Stated Policies | 2025 | 0.2% | +0.2% | 103.2 | 104.1 in 2040 |
| WEO 2021 | Stated Policies | - | - | - | 103.0 | 103.0 in 2050 |
| WEO 2022 | Stated Policies | - | - | - | 102.4 | 102.1 in 2050 |
| WEO 2023 | Stated Policies | - | - | - | 101.5 | 97.4 in 2050 |
| WEO 2024 | Stated Policies | - | - | - | 101.7 | 93.1 in 2050 |
| WEO 2025 | Current Policies | - | - | - | - | 112.8 in 2050 |
| WEO 2025 | Stated Policies | - | - | - | - | 96.9 in 2050 |
| IEO 2006 | Reference | 2010 to 2025 (4) | 6.4% | +6.4% | 118.0 | 118.0 in 2030 |
| IEO 2007 | Reference | 2010 to 2025 (4) | 5.7% | +5.7% | 117.6 | 117.6 in 2030 |
| IEO 2008 | Reference | 2010 to 2025 (4) | 3.3% | +3.3% | 112.5 | 112.5 in 2030 |
| IEO 2009 | Reference | 2010 to 2025 (4) | 4.1% | −1.7% | 106.6 | 106.6 in 2030 |
| IEO 2010 | Reference | 2015 to 2025 (3) | 4.9% | −4.4% | 103.9 | 110.6 in 2035 |
| IEO 2011 | Reference | 2015 to 2025 (3) | 3.5% | +1.0% | 108.0 | 112.2 in 2035 |
| IEO 2013 | Reference | 2015 to 2025 (3) | 4.5% | −0.8% | 104.5 | 115.0 in 2040 |
| IEO 2016 | Reference | 2020, 2025 | 4.9% | +4.9% | 109.1 | 120.9 in 2040 |
| IEO 2017 | Reference | 2018 to 2025 (8) | 2.4% | +1.1% | 104.1 | 122.3 in 2050 |
| IEO 2019 | Reference | 2020 to 2025 (6) | 3.8% | +3.7% | 105.8 | 121.5 in 2050 |
| IEO 2021 | Reference | 2022 to 2025 (4) | 1.2% | +1.2% | 109.2 | 125.9 in 2050 |
| IEO 2023 | Reference | 2024, 2025 | 1.1% | −1.1% | 105.5 | 121.5 in 2050 |
For each projection year that has now happened, error is the projection less the actual, as a share of the actual. Mean error is how far off an edition was on average; mean bias keeps the sign, so a positive figure means the edition expected more oil to be used than was. Years are scored only once full-year data exists, and only after the edition was published. IEA editions are scored against consumption less biofuels, EIA editions against all liquids. Each edition links to its source.
Table 28. The ten largest crude producers against their own record month, June 2026
| Country | Jun 2026, mb/d | A year earlier | Change | Record, mb/d | Record month | Below record |
|---|---|---|---|---|---|---|
| United States | 13.79 | 13.66 | +0.14 | 13.99 | Oct 2025 | 1% |
| Russia | 9.58 | 9.78 | −0.20 | 11.05 | Dec 2018 | 13% |
| Saudi Arabia | 7.29 | 9.91 | −2.62 | 11.79 | Apr 2020 | 38% |
| Canada | 4.95 | 4.84 | +0.11 | 5.23 | Nov 2025 | 5% |
| Brazil | 4.47 | 3.76 | +0.72 | 4.47 | Jun 2026 | At record |
| China | 4.42 | 4.43 | −0.01 | 4.50 | Mar 2025 | 2% |
| UAE | 4.17 | 3.68 | +0.49 | 4.17 | Jun 2026 | At record |
| Iran | 3.15 | 3.89 | −0.74 | 6.68 | Nov 1976 | 53% |
| Kazakhstan | 2.15 | 2.15 | 0.00 | 2.17 | Mar 2025 | 1% |
| Iraq | 1.95 | 4.38 | −2.44 | 4.83 | Dec 2018 | 60% |
EIA International Energy Statistics, crude oil including lease condensate. Records are the highest single month since 1973. Chart OPEC, OPEC+ and everyone else: OPEC is its 11 members after the UAE left on 1 May 2026 (Algeria, Congo, Equatorial Guinea, Gabon, Iran, Iraq, Kuwait, Libya, Nigeria, Saudi Arabia, Venezuela); OPEC+ adds the ten partner countries (Russia, Kazakhstan, Mexico, Oman, Azerbaijan, Malaysia, Bahrain, Brunei, Sudan, South Sudan). Both are today’s membership applied to every year, and start in 1992 because EIA reports Russia, Kazakhstan and Azerbaijan only from the break-up of the Soviet Union. EIA’s own “OPEC” series is a fixed list of twelve that still includes the UAE.
Table 29. The largest net oil exporters, 2024
| Country | 2024, mb/d | 2025 provisional | Ten years earlier | Peak, mb/d | Peak year |
|---|---|---|---|---|---|
| Saudi Arabia | 7.18 | 7.52 | 7.79 | 9.29 | 1980 |
| Russia | 6.73 | 6.73 | 7.25 | 7.93 | 2018 |
| Iraq | 3.50 | 3.43 | 2.55 | 3.84 | 2019 |
| Canada | 3.48 | 3.70 | 1.96 | 3.70 | 2025 |
| UAE | 3.36 | 3.46 | 2.75 | 3.42 | 2022 |
| Iran | 2.63 | 2.77 | 1.54 | 3.04 | 2017 |
| United States | 2.41 | 2.99 | -4.94 | 2.99 | 2025 |
| Kuwait | 2.33 | 2.37 | 2.44 | 2.72 | 2016 |
Net exports are total petroleum and other liquids production less consumption, both from EIA International Energy Statistics, so they include refined products and count a country’s own refinery gain and biofuels. The export land model argued that a producer’s rising home consumption eats its exports faster than its output falls; this is the data it was argued from. The 2025 column is provisional: EIA has published 2025 production but not yet most countries’ 2025 consumption, so it is 2025 production less the latest consumption available. There is no 2026column because EIA’s annual country data does not reach that year.
Sources
Balances and forecasts: US Energy Information Administration, Short-Term Energy Outlook, October 2026. Agency comparison: IEA Oil Market Report and OPEC Monthly Oil Market Report headline figures, with the International Energy Forum’s monthly comparison of the three reports for the levels the free summaries leave out. Weekly stocks, demand and gas storage: EIA Weekly Petroleum Status Report and Weekly Natural Gas Storage Report. Futures curves: NYMEX settlements via ComCurv; year-ago prices of the same contracts via Yahoo Finance; timespread history before April 2024 from EIA, April 2024 to 2026 and ICE Brent from TradingView. Spot prices and cracks: EIA via FRED. LNG capacity: EIA U.S. Liquefaction Capacity (Q3 2026). Refining capacity and annual throughput: Energy Institute, Statistical Review of World Energy 2026. Monthly refinery intake, output by product, stocks and demand: JODI Oil World Database, data through June 2026 for refining and July 2026 for stocks. Refinery counts: Wikipedia, List of oil refineries (revision 1376643645). Public and industry stock split: IEA, Oil Stocks of IEA Countries (CC BY 4.0). China: EIA, Today in Energy, 20 April 2026, for the December 2025 level; National Bureau of Statistics monthly energy production releases for refinery runs; JODI for production and trade. Regions follow EIA’s Table 3a definitions, including its OECD list, which leaves out Colombia and Costa Rica. Compiled by Morgan Downey for Oil 101.