The reserves-to-production ratio, explained

Published 2026-10-08; updated 2026-10-08.

The reserves-to-production ratio, or R/P, is the simplest way to turn a figure for reserves into something readable. It answers one narrow question: at the rate the world produced in the latest year, how many years would the reserves counted for that year last? Nothing in the ratio says production will stay put, or that reserves will stay put. It is a snapshot divided by a snapshot.

How is it defined?

The Statistical Review of World Energy, in the 2018 edition published by BP (the Energy Institute has since published the Review, and this site uses its 2026 edition), defines it in the notes to its oil and gas reserves tables: take the reserves remaining at the end of a year, divide by that year’s production, and the result is the length of time those reserves would last if production continued at that rate. Its reserves are proved reserves, generally those quantities that geological and engineering information indicates with reasonable certainty can be recovered from known reservoirs under existing economic and operating conditions. The same notes warn that the series does not necessarily meet company-level definitions such as those of the U.S. Securities and Exchange Commission. As a check on the calculation here, the oil ratio this site builds for the end of 2020 (53.2 years) is within 1% of the Energy Institute’s own (53.5 years); the gas and coal ratios agree to the same tolerance.

For minerals the reserves are USGS’s. USGS defines them as the part of the reserve base, itself a part of identified resources, that could be economically extracted at the time of the estimate (Mineral Commodity Summaries 2026, Appendix C). The difference is explained on reserves vs resources.

How does this site compute it?

On how much is left:

• Minerals: USGS world reserves divided by USGS world mine production in 2025.
• Oil, gas and coal: the Energy Institute last updated its reserves for the end of 2020. We subtract production in 2021 to 2025 and divide what remains by 2025 production. No discoveries or revisions after 2020 are added, because the series has not been updated.

A worked example with real numbers

Copper. USGS puts world reserves at 980 million tonnes and 2025 mine production at 23.0 million tonnes. 980 divided by 23.0 is 42.6 years. Three years earlier, reserves of 890 million tonnes over 2022 production of 21.9 million tonnes (2022 output as revised in later USGS editions; the 2023 edition printed 22 million) gave 40.6 years. The ratio rose because reserves grew about 10% while output grew about 5%.

If output kept growing, the answer would shorten. Copper’s output grew by about 2.0% a year on average between 2013 to 2015 and 2023 to 2025. Using the formula the page applies, years = ln(1 + R × g / P) / ln(1 + g), with R of 980, P of 23.0 and g of 0.02, gives 31.1 years. That is arithmetic on a past trend, not a prediction.

Oil. The Energy Institute’s proved oil reserves (including natural gas liquids) at the end of 2020 were 1,732.4 billion barrels. World oil output (crude oil, shale oil, oil sands and natural gas liquids) for 2021 to 2025 added up to 174.9 billion barrels, leaving 1,557.5 billion barrels. Dividing by 2025 production of 36.7 billion barrels gives 42.4 years. Natural gas works the same way: 188.1 trillion cubic metres at the end of 2020, less 20.5 for 2021 to 2025, over 4.2 in 2025, gives 39.9 years.

Why hasn’t the ratio simply fallen over time?

If reserves were a fixed tank and output only rose, the ratio would shrink every year. It has not. For oil, reserves at the end of 1980 were 682.6 billion barrels against 23.0 billion barrels of production that year: 29.6 years. By 2020 reserves were 1,732.4 billion barrels against 32.6 billion barrels produced: 53.2 years. Reserves rose 154%; production rose 41%. The highest value in that series was 54.6 years, in 2011. (Natural gas started at 49.6 years in 1980 and stood at 48.5 in 2020.)

The 42.4 years oil shows is lower than 53.2, but that is mostly construction: five years of output has been subtracted and nothing has been added, because the Energy Institute’s reserves have not been updated since. A figure built this way can only fall until they are.

Lithium shows the opposite problem, a falling ratio despite rising reserves. USGS reserves rose from 26 to 37 million tonnes between the 2023 and 2026 editions, yet output doubled from 0.146 million tonnes (2022 output as revised in later USGS editions; the 2023 edition printed 0.13) to 0.29 million, so lithium went from 178 years to 128.

What can’t it tell you?

• It is not a date. The shortest exact ratios in the table are zinc at 18.5 years, gold at 20.0, lead at 21.1 and silver at 23.5. Gold’s was 16.5 in 2022; it was not a countdown. A few rows are only bounds: where USGS prints a world total as “more than” a figure, such as tin (at least 20.7 years) and chromium (at least 23.5), the years are a minimum.
• It inherits the reserve definition. Reserves are estimates made from several sources and several national systems. Rare earths show at least 193 years (USGS prints world reserves as “more than 75 million” tonnes) against 433 years in 2022, because the reserve estimate fell by about 54.7 million tonnes (only about 1.15 million tonnes were mined) and output rose from 0.30 to 0.39 million tonnes.
• It assumes steady output. Demand, price and policy change production. The ratio holds only if neither reserves nor output moves.
• It mixes bases when the data do. Platinum-group reserves cover all the metals while output counts only platinum and palladium, so their ratio runs high; the site flags this.
• It says nothing about price or location. Years of supply do not say where the material sits or what it will cost. For a comparable stock measure in crops, see stocks-to-use, and for how reserve counts relate to where production peaks, see peak oil vs peak copper.

Cite as: Commodity Origins, "The reserves-to-production ratio, explained", https://commodityorigins.com/learn/reserve-to-production-ratio/ (CC BY 4.0).