Home / Patterns / Prices and cumulative output Do prices fall as the world digs up more? Mostly the record shows no clear link. Across 17 minerals, real price is lower with more cumulative output (by more than 2% per doubling) for 9 (vanadium, potash, rare earths, lead, tungsten, fluorspar, graphite, cobalt, zinc) and higher for 5 (phosphate rock, tin, silver, manganese, gold), and the relationship explains at least a fifth of the variation in price for only 3 (potash, manganese, gold).
In some industries cost falls each time cumulative output doubles, an idea known as an experience curve. This tests whether a mineral's price did that. For each mineral it takes every year with both a real price (USGS unit values in constant 1998 dollars) and a cumulative world output (summed from the start of the production record), and fits a straight line through the logarithms. The slope is reported as the change in real price associated with each doubling of cumulative output. It describes how two trending series moved together; it is a correlation, not a cause, it ignores output before the record began, and mining prices also move with demand, ore grade, energy costs and policy. Click a heading to sort.
Change in real price per doubling of cumulative output, with fit, years used and where the production record starts Mineral Price change per doubling Fit (R²) Years used Output record starts Vanadium 1970–2022 −13.0% 0.11 53 1960 Potash 1929–2022 −10.0% 0.21 94 1919 Rare earths 1922–2020 −8.8% 0.02 99 1900 Lead 1955–2021 −7.2% 0.06 67 1945 Tungsten 1915–2019 −4.8% 0.03 105 1905 Fluorspar 1923–2006 −3.9% 0.07 84 1913 Graphite 1910–2022 −3.9% 0.01 113 1900 Cobalt 1911–2021 −2.5% 0.02 111 1901 Zinc 1910–2022 −2.1% 0.02 113 1900 Copper 1910–2020 −1.9% 0.01 111 1900 Molybdenum 1912–2022 −0.8% 0.00 111 1900 Nickel 1910–2019 +0.7% 0.00 110 1900 Phosphate rock 1910–2022 +4.0% 0.10 113 1900 Tin 1915–2021 +7.3% 0.04 107 1905 Silver 1910–2021 +12.1% 0.07 112 1900 Manganese 1910–2022 +12.7% 0.36 113 1900 Gold 1910–2022 +29.1% 0.35 113 1900
Prices: USGS Data Series 140 unit values, constant 1998 US dollars (a century of real prices ). Output: world production from Data Series 140 and the Mineral Commodity Summaries, as on how much is left . ln(real price) regressed on ln(cumulative world output since the start of the production record), from ten years into the record. Minerals need at least 30 years with both series. A low R² means the line explains little of the movement in price; where it is near zero the slope should not be read as a trend.
Each mineral's scatter One point per year, both axes on a log scale, with the fitted line.
Vanadium: real price against cumulative world output, 1970 to 2022 Scatter of 53 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 13.0% in real price for each doubling of cumulative output; R squared is 0.11. 5.0k 6.0k 7.0k 8.0k 9.0k 10k 20k 30k 40k 50k 60k 1.0e+2 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 8.0e+2 9.0e+2 1.0k 2.0k real price, 1998 US dollars per tonne (log scale) cumulative world output since 1960, tonnes (log scale) 1970 2022 Vanadium: −13.0% in real price per doubling of cumulative output, R² 0.11, 53 years (1970–2022). Potash: real price against cumulative world output, 1929 to 2022 Scatter of 94 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 10.0% in real price for each doubling of cumulative output; R squared is 0.21. 90 1.0e+2 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 8.0e+2 9.0e+2 20 30 40 50 60 70 80 90 1.0e+2 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 8.0e+2 9.0e+2 1.0k real price, 1998 US dollars per tonne (log scale) cumulative world output since 1919, tonnes (log scale) 1929 2022 Potash: −10.0% in real price per doubling of cumulative output, R² 0.21, 94 years (1929–2022). Rare earths: real price against cumulative world output, 1922 to 2020 Scatter of 99 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 8.8% in real price for each doubling of cumulative output; R squared is 0.02. 30 40 50 60 70 80 90 1.0e+2 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 8.0e+2 9.0e+2 1.0k 2.0k 3.0k 4.0k 5.0k 6.0k 7.0k 8.0k 9.0k 10k 20k 30k 40k 50k 60k 70k 80k 90k 1.0e+2k 0.050 0.060 0.070 0.080 0.090 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.0 2.0 3.0 4.0 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1922 2020 Rare earths: −8.8% in real price per doubling of cumulative output, R² 0.02, 99 years (1922–2020). Lead: real price against cumulative world output, 1955 to 2021 Scatter of 67 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 7.2% in real price for each doubling of cumulative output; R squared is 0.06. 6.0e+2 7.0e+2 8.0e+2 9.0e+2 1.0k 2.0k 20 30 40 50 60 70 80 90 1.0e+2 2.0e+2 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1945, tonnes (log scale) 1955 2021 Lead: −7.2% in real price per doubling of cumulative output, R² 0.06, 67 years (1955–2021). Tungsten: real price against cumulative world output, 1915 to 2019 Scatter of 105 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 4.8% in real price for each doubling of cumulative output; R squared is 0.03. 6.0k 7.0k 8.0k 9.0k 10k 20k 30k 40k 50k 60k 70k 80k 40 50 60 70 80 90 1.0e+2 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 8.0e+2 9.0e+2 1.0k 2.0k 3.0k real price, 1998 US dollars per tonne (log scale) cumulative world output since 1905, tonnes (log scale) 1915 2019 Tungsten: −4.8% in real price per doubling of cumulative output, R² 0.03, 105 years (1915–2019). Fluorspar: real price against cumulative world output, 1923 to 2006 Scatter of 84 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 3.9% in real price for each doubling of cumulative output; R squared is 0.07. 1.0e+2 2.0e+2 3.0e+2 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10 20 30 40 50 60 70 80 90 1.0e+2 2.0e+2 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1913, tonnes (log scale) 1923 2006 Fluorspar: −3.9% in real price per doubling of cumulative output, R² 0.07, 84 years (1923–2006). Graphite: real price against cumulative world output, 1910 to 2022 Scatter of 113 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 3.9% in real price for each doubling of cumulative output; R squared is 0.01. 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 8.0e+2 9.0e+2 1.0k 2.0k 3.0k 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10 20 30 40 50 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1910 2022 Graphite: −3.9% in real price per doubling of cumulative output, R² 0.01, 113 years (1910–2022). Cobalt: real price against cumulative world output, 1911 to 2021 Scatter of 111 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 2.5% in real price for each doubling of cumulative output; R squared is 0.02. 5.0k 6.0k 7.0k 8.0k 9.0k 10k 20k 30k 40k 50k 60k 70k 80k 90k 1.0e+2k 0.0090 0.010 0.020 0.030 0.040 0.050 0.060 0.070 0.080 0.090 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.0 2.0 3.0 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1901, tonnes (log scale) 1911 2021 Cobalt: −2.5% in real price per doubling of cumulative output, R² 0.02, 111 years (1911–2021). Zinc: real price against cumulative world output, 1910 to 2022 Scatter of 113 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 2.1% in real price for each doubling of cumulative output; R squared is 0.02. 7.0e+2 8.0e+2 9.0e+2 1.0k 2.0k 3.0k 4.0k 5.0k 8.0 9.0 10 20 30 40 50 60 70 80 90 1.0e+2 2.0e+2 3.0e+2 4.0e+2 5.0e+2 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1910 2022 Zinc: −2.1% in real price per doubling of cumulative output, R² 0.02, 113 years (1910–2022). Copper: real price against cumulative world output, 1910 to 2020 Scatter of 111 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 1.9% in real price for each doubling of cumulative output; R squared is 0.01. 2.0k 3.0k 4.0k 5.0k 6.0k 7.0k 8.0k 9.0k 10k 8.0 9.0 10 20 30 40 50 60 70 80 90 1.0e+2 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1910 2020 Copper: −1.9% in real price per doubling of cumulative output, R² 0.01, 111 years (1910–2020). Molybdenum: real price against cumulative world output, 1912 to 2022 Scatter of 111 years on log scales, one point per year, with a fitted line. The fitted line implies a fall of 0.8% in real price for each doubling of cumulative output; R squared is 0.00. 5.0k 6.0k 7.0k 8.0k 9.0k 10k 20k 30k 40k 50k 60k 70k 80k 90k 1.0e+2k 0.0020 0.0030 0.0040 0.0050 0.0060 0.0070 0.0080 0.0090 0.010 0.020 0.030 0.040 0.050 0.060 0.070 0.080 0.090 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1912 2022 Molybdenum: −0.8% in real price per doubling of cumulative output, R² 0.00, 111 years (1912–2022). Nickel: real price against cumulative world output, 1910 to 2019 Scatter of 110 years on log scales, one point per year, with a fitted line. The fitted line implies a rise of 0.7% in real price for each doubling of cumulative output; R squared is 0.00. 4.0k 5.0k 6.0k 7.0k 8.0k 9.0k 10k 20k 30k 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10 20 30 40 50 60 70 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1910 2019 Nickel: +0.7% in real price per doubling of cumulative output, R² 0.00, 110 years (1910–2019). Phosphate rock: real price against cumulative world output, 1910 to 2022 Scatter of 113 years on log scales, one point per year, with a fitted line. The fitted line implies a rise of 4.0% in real price for each doubling of cumulative output; R squared is 0.10. 20 30 40 50 60 70 80 90 1.0e+2 0.050 0.060 0.070 0.080 0.090 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1910 2022 Phosphate rock: +4.0% in real price per doubling of cumulative output, R² 0.10, 113 years (1910–2022). Tin: real price against cumulative world output, 1915 to 2021 Scatter of 107 years on log scales, one point per year, with a fitted line. The fitted line implies a rise of 7.3% in real price for each doubling of cumulative output; R squared is 0.04. 5.0k 6.0k 7.0k 8.0k 9.0k 10k 20k 30k 40k 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10 20 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1905, tonnes (log scale) 1915 2021 Tin: +7.3% in real price per doubling of cumulative output, R² 0.04, 107 years (1915–2021). Silver: real price against cumulative world output, 1910 to 2021 Scatter of 112 years on log scales, one point per year, with a fitted line. The fitted line implies a rise of 12.1% in real price for each doubling of cumulative output; R squared is 0.07. 90k 1.0e+2k 2.0e+2k 3.0e+2k 4.0e+2k 5.0e+2k 6.0e+2k 7.0e+2k 8.0e+2k 9.0e+2k 1.0M 70 80 90 1.0e+2 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 8.0e+2 9.0e+2 1.0k real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1910 2021 Silver: +12.1% in real price per doubling of cumulative output, R² 0.07, 112 years (1910–2021). Manganese: real price against cumulative world output, 1910 to 2022 Scatter of 113 years on log scales, one point per year, with a fitted line. The fitted line implies a rise of 12.7% in real price for each doubling of cumulative output; R squared is 0.36. 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 8.0e+2 9.0e+2 1.0k 2.0k 8.0 9.0 10 20 30 40 50 60 70 80 90 1.0e+2 2.0e+2 3.0e+2 4.0e+2 5.0e+2 6.0e+2 7.0e+2 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1910 2022 Manganese: +12.7% in real price per doubling of cumulative output, R² 0.36, 113 years (1910–2022). Gold: real price against cumulative world output, 1910 to 2022 Scatter of 113 years on log scales, one point per year, with a fitted line. The fitted line implies a rise of 29.1% in real price for each doubling of cumulative output; R squared is 0.35. 5.0M 6.0M 7.0M 8.0M 9.0M 10M 20M 30M 40M 7.0 8.0 9.0 10 20 30 40 50 60 70 80 90 1.0e+2 real price, 1998 US dollars per tonne (log scale) cumulative world output since 1900, tonnes (log scale) 1910 2022 Gold: +29.1% in real price per doubling of cumulative output, R² 0.35, 113 years (1910–2022). More patterns Concentration : How much of each commodity comes from one country? Producer or exporter? : Is the biggest producer also the biggest seller abroad? Prices in real terms : Which prices are records once inflation is removed? Growth and decline : Whose output is rising or falling fastest? Linked prices and seasons : Which prices move together, and which have a season? A century of real prices : Are minerals getting scarcer, judging by their real prices since 1900? Supplier exposure : Which big buyers lean on one supplier for critical minerals? Trade after export controls : What happened to a country's exports after it controlled them? Export share history : Who has gained and lost share of world exports since 2000? Prices around export controls : What happened to mineral prices after export controls? Export basket prices : What would a country's commodity exports have fetched at each month's prices? Substitution coverage : How large are other exporters' sales to other buyers next to an importer's biggest supplier? Critical risk matrix : Which critical minerals combine concentrated mining, export controls and import reliance? Country reserve life : Which producing countries have the least of what they mine? US import reliance and recycling : How much of each mineral does the United States import, and how much comes from scrap? Data confidence : How firm are each commodity's production numbers? Export controls and bans : Which governments restrict exports of minerals, and what is still in force? Chokepoints : Which narrow passages do the world's commodities depend on? By-product minerals : Which minerals are by-products of other metals? Peak production : Which minerals and fuels are still setting records, and which peaked years ago? Reserves and mining : Where are the deposits, and who digs them up? Supply concentration : Which commodities depend on the fewest suppliers? Land or yield? : Did output grow because farmers planted more or farmed better? Supply shocks and prices : When world output falls, do prices rise? Category price indices : How have energy, metals and food prices moved since 1960? Drawdowns and volatility : How far do prices fall, and do they recover? Stocks and prices : Are prices lower when the world holds more stock? Which prices follow oil? : How closely does each commodity track crude oil? How reliable is the data? : How much of world output rests on estimates? Net exporters and importers : Who is a net seller and who a net buyer? Country leaders : Which countries lead across the most commodities? Export dependency : Which countries rely on one commodity? Concentration over time : Has production spread out or narrowed? Import dependency : Which countries buy most of a commodity from one supplier? Supplier shifts : Who gained and lost importers' business? Trade price against benchmark : Does traded product sell above or below the benchmark price? Prices in other currencies : How much of a price move is the dollar? Do prices lead each other? : Does one commodity price move before another? Volatility over time : Are prices more volatile than they used to be? Seasonality : Which prices have a calendar-month pattern? Price ratios : What does one commodity buy of another, and is that high or low? El Niño and prices : Do El Niño and La Niña years move commodity prices? Who consumes it : Which countries consume each commodity, and how much per person? Self-sufficiency : Which large markets depend on imports, and which produce far more than they use? Trade flows : Who sells each commodity to whom? Every figure is computed from the same tables as the commodity pages. The full set is in patterns.json , and single facts from it appear each day on the facts page .