We’re Not Running Out of Resources… So Why Is Everyone Panicking?

Every few years, a headline shows up warning us that we have "only 20 years of copper left" or that "gold will run out by 2035." It happened with oil in the 1970s. It happened with rare earths a decade ago. It's happening right now with lithium and cobalt. And every single time, two things happen: people panic, and then, weirdly, we don't run out.

So what's going on? Are these warnings wrong? Not exactly. They're just measuring the wrong thing, or rather, they're measuring something real (reserves) and describing it as something much bigger and vaguer (resources), and the mix-up changes the whole story.

This post walks through why "running out" is the wrong question, what actually determines whether we can keep using a material, and because this is a research-led site, what that means for how we should actually think about material scarcity, circular economy, and sustainable design. Grounded in geoscience (USGS classification systems), industrial ecology (OECD material flow projections), and a university-level "Finite Resources" curriculum, not vibes.

1. Reserves Are Not the Same Thing as Resources

This is the "McKelvey box" concept as a nested-box diagram: reserves are the small, certain, economically viable core; resources balloon outward as certainty drops.

Here's the mix-up at the heart of almost every "we're running out" headline: people treat reserves as if they were the total amount of a material on Earth. They're not even close to the same thing.

Geologists and mining agencies (like the U.S. Geological Survey) work with a whole ladder of categories, moving from "we're pretty sure this exists and we could dig it up profitably right now" all the way to "there's probably some out there but we haven't found it yet." Roughly, from most to least certain:

• Reserves – the amount that's been measured, and that can be extracted profitably with today's technology and prices.

• Reserve base – reserves plus material that's identified but doesn't yet meet the minimum quality, depth, or purity thresholds for current mining practice.

• Contingent reserves – deposits that could become economical under assumed future technical or economic conditions.

• Initial reserves – a rough estimate made at the moment a deposit is discovered, before it's been properly surveyed. Highly uncertain, but it's what sets the sale price of a mine and how much investors are willing to put into it.

• Prospective / hypothetical / speculative resources – educated guesses about what might be sitting undiscovered in geologically similar areas.
  

Reserves sit in the small, high-confidence, high-profitability corner of this picture. The "total resource" is the entire picture, most of which is uncertain, uneconomical to extract right now, or simply hasn't been located yet.

The USGS itself defines a reserve as "that part of the reserve base that could be economically extracted or produced at the time of determination," with the reserve base being the larger, in-place pool that reserves are drawn from deliberately not the same as "everything that exists."

A quote often used in geology lectures to describe this uncertainty is that estimating reserves is a bit like wearing a blindfold and trying to describe an elephant you can only partially touch.

Why this matters: a "20 years of reserves left" headline is not a countdown timer to extinction of that material. It's closer to a snapshot of what mining companies have already found, tested, and judged profitable at today's prices. That number moves constantly and usually it moves upward.

Section takeaway: "Reserves" measure what's profitable to mine right now, not what exists on Earth. Confusing the two is the single biggest reason resource-scarcity headlines feel scarier than the underlying science.

2. Reserves Actually Grow Over Time (Yes, Really)

A simple bar chart showing the indium reserve jump (2,800 t → 11,000 t, 2006–2007), paired with a short annotated timeline of the USGS Critical Minerals List size over the years (50 minerals in 2022 → 60 in 2025) to show reserves and criticality lists are dynamic, not fixed.

If reserves are just "what we've confirmed is profitable so far," then it follows that reserves should grow as we explore more, get better at detection, or as prices rise enough to make previously uneconomical deposits worth mining. And that's exactly the pattern we see.

A textbook example: worldwide indium reserves reportedly jumped from about 2,800 tonnes in 2006 to 11,000 tonnes in 2007, a roughly fourfold increase in a single year. Nothing new was created in the ground. What changed was exploration, technology, and economics, the boundary between "reserve" and "not yet a reserve" simply moved.

This is also why the U.S. Geological Survey doesn't publish a critical minerals list once and leave it, it's a living document. The most recent (2025) list contains 60 minerals, up from 50 in 2022, and by law it has to be reviewed at least every three years because criticality, supply risk, and reserve estimates keep shifting with markets, geopolitics, and new discoveries.

Reserve numbers respond to:

• Exploration effort, you can't count what you haven't looked for.

• Prices, a higher market price makes lower-grade or harder-to-reach deposits worth extracting.

• Technology, better extraction and refining methods lower the cost threshold for what counts as "economical."

• Politics and policy, subsidies, export bans, and trade restrictions change what's worth mining domestically vs. importing.

Section takeaway: Reserve figures are a moving economic and technical estimate, not a fixed geological fact. They tend to expand with exploration, price incentives, and better technology — which is a big part of why "running out" predictions have historically been wrong.

3. But "It's Not Running Out" Doesn't Mean "No Problem", Enter the Hubbert Curve

Two paired line charts. (1) A stylized Hubbert-style bell curve of extraction rate over time, with "maximum global rate of production" marked at the peak. (2) A downward-sloping "ore grade over time" chart (e.g., using historical gold content data, g Au/tonne, from the 1800s to 2010s across countries like the US, Australia, South Africa, Canada) to show the declining-quality trend visually next to the extraction-rate curve. None of this means resource extraction is a free lunch. Even if the total amount of a resource keeps expanding on paper, the rate at which we can pull it out of the ground in a given place is not infinite, and it doesn't stay high forever.

This is the idea behind the Hubbert curve, first developed for oil production. The logic: individual wells, mines, or fields ramp up, peak, and decline, and when you stack many of these production curves from a region together, the combined output tends to follow a bell-shaped curve over time: slow start, a peak, then decline as the easiest and cheapest sources are used up.

Real-world oil production data (aggregated from sources like the "gross" and "net" Hubbert curve reconstructions used in peak-oil research) does roughly track this bell shape when plotted from 1900 into projected extraction through the 2100s, although real production has repeatedly been pushed higher and later than early predictions expected, because of new discoveries, deepwater and unconventional extraction (like shale), and improved recovery technology.

There's a second, closely related pattern worth knowing: quality tends to decline over time. The "low-hanging fruit", the richest, most accessible, cheapest-to-process deposits, gets extracted first. What's left afterward is typically lower grade, deeper, or more remote, meaning it takes more energy and money to extract the same unit of material. Gold mining is a clean illustration: ore grades (grams of gold per tonne of rock mined) in major producing countries like the US, Australia, and Canada have trended broadly downward from the 1800s to today, even as total gold reserves and production figures have stayed resilient, because miners now process far more rock to get the same amount of metal.

Section takeaway: Materials not "running out" globally is compatible with production in a specific mine, field, or country peaking and declining. Extraction rate and ore quality both follow predictable curves — the easy, high-quality sources go first, and what's left costs more energy and money to get.

4. The Real Pressure Isn't Geology, It's Demand

A multi-panel dashboard-style graphic:

• Line chart: global material use 2017 vs. projected 2060 (89 Gt → 167 Gt), OECD data.

• Donut or bar chart: global circularity rate, 9.1% (2018) vs. 6.9% (2025), Circularity Gap Report.

• Simple stat callouts: "70% of people will live in cities by 2050" and "chemical production value projected to triple by 2050."

If the ceiling on "how much material exists" keeps moving upward, the more useful question for a sustainability-focused reader isn't "will we run out?", it's "how fast is demand growing, and can supply, recycling, and design keep up?"

And demand is growing fast, on multiple fronts at once:

• Population and living standards. Global population is projected to keep climbing through the mid-century mark, alongside rising life expectancy and living standards, all of which increase demand for housing, food, mobility, and consumer goods.

• Urbanization. Roughly half the world's population lived in cities as of the late 2000s; that share is projected to reach about 70% by 2050, which drives enormous demand for construction materials, infrastructure, and energy systems.

• Chemical and material complexity. The number of chemical substances registered globally has grown by well over 100 million entries since 1970, and the value of chemical production is projected to roughly triple between 2010 and 2050 according to OECD environmental outlook estimates. Products themselves have also gotten more materially complex, a plastic milk carton alone can be made of six or seven layered polymer films, and researchers have catalogued more than 10,000 distinct additive chemicals used across packaging, textiles, food contact materials, toys, and medical devices.

• Overall material use. According to the OECD's Global Material Resources Outlook, global primary materials use is projected to almost double, from around 89 gigatonnes in 2017 to roughly 167 gigatonnes by 2060, driven mainly by population growth and rising incomes in non-OECD economies.

• We're also getting worse at closing the loop, not better. The Circularity Gap Report 2025 found that only about 6.9% of the more than 100 billion tonnes of materials used globally each year come from recycled (secondary) sources, down from 9.1% in 2018. Meanwhile, total material consumption has passed 100 billion tonnes annually for the first time in history.
  

Put together, this is really the core sustainability problem: it's not that the planet's material stock is about to hit zero, it's that our rate of consumption is accelerating faster than our ability to recycle, substitute, or design our way out of dependency on virgin extraction.

Section takeaway: The scarcity story isn't really about geology running dry, it's about consumption accelerating (population, urbanization, product complexity) faster than recycling and substitution can keep pace. That's a demand-and-design problem, which means it's also a solvable problem.

5. What This Actually Means for Sustainable Materials Practice

A simple two-column "before vs. after" or "linear vs. circular" infographic contrasting a linear material flow (extract → use → landfill) with a circular one (extract → use → recover → reuse), annotated with the 6.9% vs. potential 25% circularity figures to make the opportunity gap visually obvious.

This is where it stops being an abstract geology lesson and becomes directly relevant to circular economy, LCA (life cycle assessment), and material-choice work.

1. Stop asking "will we run out?", start asking "what's the energy and impact cost of what's left?" Because ore grades decline and deposits get harder to reach, the environmental footprint (energy use, water use, land disturbance, emissions) of extracting the next tonne of a material is usually higher than the footprint of the last tonne extracted. This is a core reason LCA studies increasingly need to model future extraction scenarios, not just current-average data, a material that looks fine on paper today may carry a rising footprint as reserves shift toward lower-grade sources.

2. Design for disassembly and recyclability now, while material complexity is still (relatively) manageable. Products have become dramatically more materially complex, multi-layer packaging, composite textiles, thousands of plastic additives, cars with dozens of distinct engineered polymer and metal components. Every layer of complexity that isn't designed with end-of-life in mind is complexity that recycling systems will struggle to unpick later. This is precisely why "benign by design" and circular-by-design approaches matter more than end-of-pipe recycling fixes.

3. Push for higher secondary material use, the current global level is very low. At just 6.9% global circularity, there is enormous headroom. Analysts estimate that if all currently recyclable material were actually captured and reused, global circularity could rise from 6.9% to as much as 25% without even needing new material substitutions, just better collection, sorting, and reuse infrastructure. That's a large, achievable near-term lever.

4. Track critical mineral lists and price signals as leading indicators, not doom signals. When a mineral moves onto or up a critical-minerals list (like the USGS's expanded 2025 list of 60 minerals), that's a supply-risk and geopolitical signal, worth watching for substitution and sourcing strategy, but it is not evidence the material is disappearing from the planet.

5. Substitute toward renewable and more abundant materials where function allows. Every non-renewable input in a supply chain, plasticizers, flame retardants, rare-earth magnets, certain alloying metals, is a candidate for evaluation against renewable or more abundant alternatives, provided performance and safety requirements can still be met. This is exactly the kind of trade-off LCA is built to quantify.

Section takeaway: For sustainable materials practice, the practical response to "resources aren't running out" isn't relief, it's a redirection of focus toward energy/impact intensity of extraction, design-for-recycling, and closing the (currently huge) circularity gap.

Key Takeaways

• "Reserves" ≠ "resources." Reserves are the small, economically viable, well-measured slice of a much larger, more uncertain resource base — not a countdown to depletion.

• Reserves tend to grow, not shrink, as exploration, prices, and technology evolve — indium reserves alone quadrupled in a single year (2006–2007).

• Extraction rates and ore quality follow predictable curves (Hubbert-type curves, declining ore grades) — meaning where and how we extract still matters even when total resources aren't disappearing.

• The real driver of pressure on materials is demand growth — population, urbanization, and product complexity — not geological scarcity. Global material use is projected to almost double by 2060.

• The circular economy has enormous untapped potential: only 6.9% of materials used globally come from recycled sources, down from 9.1% in 2018, even though modelling suggests recycling everything recyclable could push that closer to 25%.

• The actionable response isn't panic, it's design: track impact intensity of extraction, design products for disassembly, invest in recovery infrastructure, and substitute toward renewable materials where feasible.
  

References

1. U.S. Geological Survey (USGS), Mineral Commodity Summaries — reserve and reserve base definitions.

2. U.S. Geological Survey, About the 2025 List of Critical Minerals, usgs.gov.

3. Congressional Research Service, U.S. Geological Survey's Critical Minerals List, congress.gov (IF13145, R47982), 2025–2026.

4. OECD (2019), Global Material Resources Outlook to 2060: Economic Drivers and Environmental Consequences, OECD Publishing, Paris.

5. Circle Economy & Deloitte (2025), Circularity Gap Report 2025, circularity-gap.world.

6. Wiesinger, H., Wang, Z., & Hellweg, S. (2021), Deep Dive into Plastic Monomers, Additives, and Processing Aids, Environmental Science & Technology.

7. Wik, A., & Dave, G. (2009), Occurrence and effects of tire wear particles in the environment – A critical review and an initial risk assessment, Environmental Pollution, 157, 1–11.

8. Kümmerer, K. & Reich, M., Modul Finite Resources — Introduction, MSc SusSci Resources, Materials and Chemistry, Leuphana University Lüneburg (lecture materials; USGS reserve classification diagrams, Hubbert curve, gold ore-grade data adapted from David Murphy / theoildrum.com).