Circular Economies

This brief is the economics of circularity and nothing else. The engineering is already published by the Institute in two places and is not restated here: Circular Infrastructure Systems owns the built-environment treatment, and Industrial Ecology owns plant-to-plant material exchange and the methods of material flow and life cycle accounting. What is left, and what is genuinely unoccupied, is the macroeconomics: material productivity as a ratio and its trend, absolute against relative decoupling, extended producer responsibility read as a fiscal instrument, the market structure of secondary materials, who constructs the headline circularity numbers and from what, and the rebound mechanism by which closing a loop can raise total throughput.

The claim under test — that material loops can be closed at economy scale — is three claims wearing one coat. A physical claim that loops can close; an economic claim that they close at prices which clear; and an accounting claim that we can tell when they have. This brief owns the second and third, and the third turns out to be where the argument is weakest. The most-quoted metric in the field is produced by an organisation founded to promote the concept it measures, from national material-flow accounts built to answer a different question, and it now has a peer-reviewed critique from inside its own discipline. Start with the accounting, because every economic claim in the subject is stated in its units.

Established Begin with the denominator, because every circularity success is a share of it. On the UN International Resource Panel's 2024 accounting, global material extraction has tripled in the past five decades and is projected to rise by 60% by 2060 from 2020 levels on current trends, with high-income countries using six times more resources per capita and generating ten times the climate impacts of low-income countries. Established The OECD's independent projection, across 61 materials, has global primary materials use doubling by 2060, with metals and non-metallic minerals growing fastest, while noting that structural change in non-OECD economies and technology improvement partially dampen that growth. Frontier That adverb is the entire decoupling debate. Established Mark the interests and note that they run in opposite directions: the resource panel is an environmental body whose mandate is to argue for resource-use reduction, the OECD is an economic organisation with no such mandate, and they agree on direction and roughly on magnitude, which raises the weight of both. A loop that recovers a fixed fraction of a doubling stock is not closing.

Established The decoupling evidence is the best in the subject and it comes from a systematic review large enough to be treated as a census. Wiedenhofer and colleagues screened over 11,500 papers and reviewed 835 empirical studies of the relationship between GDP, resource use and greenhouse gas emissions; Haberl and colleagues synthesised the same 835. Their findings, in order of how often they are misquoted: relative decoupling is frequent for material use and for greenhouse gas and carbon dioxide emissions; long-term absolute decoupling is rare, and rare for useful exergy in particular; and some industrialised countries have recently achieved absolute decoupling for consumption-based emissions. Established That last one is the strongest pro-green-growth data point in the review and it belongs in any honest account. Established Their conclusion is in their own words: “large rapid absolute reductions of resource use and GHG emissions cannot be achieved through observed decoupling rates.” Established Both papers are open-access and funded by research councils rather than by an advocacy organisation on either side.

Established The critical distinction, and the one that makes or breaks every summary of the above: decoupling of emissions and decoupling of materials are different results with different evidence bases. The consumption-based emissions result is real and recent. There is no equivalent material result. Frontier A great deal of policy argument moves silently from the first to the second, and the review that is usually cited for it says the opposite in the same abstract.

Established Material productivity measured the flattering way is partly a measure of how much extraction a country has offshored. Wiedmann and colleagues built consumption-based material footprint accounts across 186 countries and found that a country's use of non-domestic resources is on average about threefold larger than the physical quantity of the traded goods themselves — because trade statistics record the shipped tonne, not the tonnes moved to produce it. Established For every 10% increase in GDP, the average national material footprint rises about 6%. Apparent resource-productivity gains in wealthy countries substantially reflect externalised extraction rather than efficiency. Established A correction to the paper's supporting information was published in 2023 and is noted here because a decade-old headline result cited without its correction is a citation, not a reading. Frontier The consequence for the framing is direct: GDP per tonne of domestic material consumption is not a measure of circularity, any claim that a rich economy has raised material productivity enough to matter must be made on footprint, and on footprint the elasticity is about 0.6 — improving, and nowhere near the sign change the framing requires.

Established The sharpest owned finding in this slot is that the main economy-wide fiscal instrument for closing loops does almost nothing. Joltreau assembled a panel of 25 EU countries, 1998–2015, four packaging materials — glass, paper and board, plastics, metals — using temporal variation in the weight-based fees producer responsibility organisations actually charge, with dynamic panel methods and controls for material prices, fuel costs, consumption and green-party vote share as a proxy for environmental preference. The finding: the financial incentive produced “very little (though statistically significant) packaging reduction”, and no systematic substitution effects between packaging materials. The fees are too small relative to total packaging costs to enter a design decision. Frontier Read as public finance this is a clean result: a Pigouvian instrument set well below the level at which it changes behaviour, functioning as a hypothecated collection levy rather than as a price signal. That is the honest description of most such schemes and it is very much under-reported. Established Note where the interest runs: the finding cuts against the stated purpose of the instrument the author is evaluating, which is the direction that raises a result's weight.

Frontier On the market side, the loop does not close on price, and the mechanism is not a quality complaint. Zink and Geyer identify limited substitutability between secondary and primary products, and the price dynamics that follow, as the core mechanism: secondary material that does not fully substitute does not displace primary production one for one — it adds to total supply, lowers price, and raises total consumption. Frontier This is the single most important economic idea in the slot, and it means the framing's question is malformed: at economy scale, closing a loop changes prices, changed prices change quantities, and the quantity response is the thing nobody has measured. Frontier The literature has matured rather than resolved — a 2025 revisiting paper by the original author, eight years on, is itself evidence that the effect is considered real and remains unquantified.

Established Where the numbers are best, they show recovery economics failing exactly where material value is lowest. Secondary supply is already 33% of copper demand and 31% of nickel demand as of 2023; recovery from available feedstock runs over 40% for cobalt but 20% for lithium; projections to 2050 put copper near 40% and battery metals at 20–30% of demand; and collection runs 40–50% in Europe and North America, below 5% in developing Asia and Latin America and 1% in Africa. Frontier The diagnostic detail is that lithium-iron-phosphate battery recycling is identified as facing weak economics because of low residual value, with toll-based recycling and regulatory mandates proposed as the remedy — an admission, by an energy-security agency rather than by a critic, that the market price does not support the loop. Established The same source states the case for the loop: recycled nickel, cobalt and lithium carry about 80% lower greenhouse gas emissions than primary, and without more recycling, mining investment to 2040 would need to be about 30% higher. Frontier And in plastics the binding constraint is supply rather than demand: meeting pledged recycled-PET volumes for 2025 would have required the annual recycling growth rate to double against its 2014–2018 trend.

Established Finally, reported recycling rates are measured at the wrong point in the chain, and this is a measurement-convention finding rather than a technology finding, which is why it belongs here. A tonne is typically counted when it is collected for recycling rather than when recycled material leaves a facility, so losses at sorting, contamination rejection and reprocessing yield are not netted; and exported material is counted as recycled by the exporter. Established On the plastics case, the UN environment programme states that only nine per cent of plastics produced are mechanically recycled and projects plastic waste nearly tripling by 2060 under business as usual — and it is the convenor of the global plastics treaty process, arguing for production limits, so mark it. Frontier The Institute's own construction brief has already worked the same critique through one sector's official recovery figure and decomposed it; that decomposition is cited there and not re-derived here. What would settle the general question is publishing rates disaggregated by destination and measured at output rather than at collection, and nobody reports plastics that way either.

Frontier The organising question is whether absolute decoupling of material use is achievable, and the live positions are further apart than the policy literature admits. (a) Green growth: absolute decoupling is achievable and beginning; the honest evidence is the recent absolute decoupling of consumption-based emissions in some industrialised countries; settled by whether the emissions result ever extends to materials, of which there is no sign. (b) It does not happen: long-term absolute material decoupling is rare across 835 studies and the footprint elasticity is about 0.6; settled by a high-income economy sustaining a falling material footprint with rising GDP for a decade, which has not been observed. (c) Sufficiency or degrowth is required: the review's own sentence — that large rapid absolute reductions cannot be achieved through observed decoupling rates — is the strongest empirical support any degrowth position has, and it is a statement about observed rates, not about achievable ones.

Frontier A methodological position sits underneath all three and deserves separating out. The bibliometric half of the review finds that the overwhelming majority of the 835 studies approach the question statistically and econometrically while largely overlooking thermodynamic constraints, and rarely consider the tension between growth and the systemic changes climate targets require. Speculative If that is right, the decoupling literature is not so much wrong as unfit for the question, and the body of biophysically grounded long-run studies that would replace it does not exist.

Speculative On the headline metric, two positions are live and this brief can settle neither. One holds that the circularity-gap figure is sound enough to steer policy by; it is widely cited and built on standard material-flow accounts. The other holds that the concept has limits, and it now has a 2025 peer-reviewed paper behind it in Resources, Conservation and Recycling — a critique from inside the material-flow discipline rather than from its opponents, which raises its weight. Handwave The research pass behind this brief could retrieve that paper's title, authors, journal and licence but not its abstract, so its argument is not characterised here in any direction beyond the claim its title makes. That is a deliberate limit, and it is stated rather than papered over.

Frontier On the fiscal instrument, the disagreement is now about level rather than about principle. The stated purpose of producer responsibility is to create an eco-design incentive; measured across 25 countries and eighteen years it does not, at current fee levels. The design response is eco-modulation — varying the fee by product characteristics — and the identified problem with it is weak environmental linkage between the fee criteria and actual impact, with life-cycle assessment proposed as the basis. Speculative A 2026 conference paper describing the shift from volume-based fee recovery to punitive eco-modulated frameworks frames its own contribution as helping firms minimise their fees through design choices, which is either the instrument working exactly as intended or the instrument being gamed, depending entirely on whether the fee criteria track impact. Frontier And here is a finding shaped like an absence: nobody has estimated the fee level at which an eco-design response begins. One paper establishes that the current level is below it. No paper estimates where it is, and no cross-country study of who ultimately pays the fee — producer or consumer — exists at all.

Frontier On secondary markets the two positions make opposite predictions and both are testable. They will clear once volumes rise: copper at 33% and nickel at 31% of demand are already substantial, and scale economies do the rest. They will not, because substitution is imperfect: the same imperfect substitutability that generates rebound also prevents one-for-one displacement, and the low-residual-value streams need mandates or toll arrangements to move at all. Speculative What would separate them is an elasticity of substitution between secondary and primary material estimated in a real market, which is essentially unmeasured. Frontier A third position cuts across both: loops can be climate-positive while remaining price-negative — about 80% lower emissions for recycled battery metals, and roughly 30% higher mining investment avoided to 2040 — in which case the correct instrument is a carbon price rather than a circularity target.

Speculative Two further positions are held by serious people and are unfalsified. One is that producer responsibility is really a hypothecated collection levy whose incidence falls on consumers, which is implied by the fee magnitudes and which no pass-through study has tested. The other is the strongest deflationary claim available: that the whole programme is a category error — circularity is a property of a materials system, and applying it as an economy-wide policy target produces metrics that cannot be gamed honestly. Handwave Its test would be whether any economy-wide circularity target has ever produced a measurable material outcome, and the honest answer is that nobody has looked in a way that would show it.

Speculative And the national case everyone reaches for is empty. Japan's sound-material-cycle framework is the most-cited national programme with explicit material-productivity targets and a legislated indicator framework, and a credible independent evaluation of whether the targets were met, and what caused the movement, does not appear in the searchable literature. Established That absence is a finding and is stated here plainly rather than filled with the programme's own claims.

Established The first bottleneck is that the standard national metric measures the wrong thing. Domestic material consumption counts tonnes crossing a border as tonnes, so a country that stops smelting and starts importing metal records a productivity gain. Footprint accounting closes that gap and finds non-domestic resource use running about threefold the physical mass of the traded goods. Frontier Every national circularity claim made on domestic material consumption is therefore partly a claim about the geography of extraction, and the two series are not interchangeable.

Frontier The second is that the headline circularity metric is constructed rather than measured. It is built on economy-wide material flow accounts, which are national statistical products designed for a different purpose, and it requires a definition of what counts as cycled that those accounts do not natively supply. Established It is also produced by an organisation whose purpose is to promote the concept it is measuring. That is not a reason to discard it; it is a reason to state the interest and to prefer an independent reconstruction, which nobody has published.

Established The third is arithmetic, and it is the source of the most common misreading in the subject. When extraction grows faster than recovery, the cycled share falls while every recovery number rises. A falling circularity rate is not evidence that recycling is failing. Frontier The Institute's industrial-ecology brief already carries the specific figure and its treatment; this brief's contribution is the denominator arithmetic and the provenance question, not the number.

Handwave And the number itself is absent here for a reason that should be stated rather than implied. The publisher's site was unreachable in the research pass behind this brief, so no circularity-gap percentage and no edition number is quoted anywhere above, including the trend claim that the share has been falling across editions. Established That claim is arithmetically plausible on the extraction growth recorded here and it is not verified here, and the difference between those two statements is the whole of this brief's method.

Frontier The fourth is that rates are measured at collection rather than at output, and exports count as recycled by the exporter. Losses at sorting, contamination rejection and reprocessing yield disappear into the gap between those two measurement points. Speculative How large the gap is nobody can say in general, because the disaggregation that would show it is not published for any major material stream.

Frontier The fifth is prices. The claim that recycled feedstock is often more expensive than virgin is widely asserted and is not verified here: the trade-press price series that carry it were unreachable in the research pass, and the peer-reviewed price-gap literature is thin. Frontier It is carried as a mechanism rather than as a number — imperfect substitution, collection and sorting cost, small-lot markets, and virgin prices that track oil — and a published independent price series for matched grades over a full oil-price cycle would settle it.

Frontier The first dependency is statistical rather than scientific: consumption-based material footprint accounting inside routine national statistics. It exists as a research product across 186 countries and does not exist as an official series anyone updates on a policy timetable. Established Until it does, national material-productivity targets are set in a unit that rewards offshoring, and the elasticity that matters — about 6% of footprint per 10% of GDP — is invisible to the bodies setting the targets.

Frontier Second, an elasticity of substitution between secondary and primary material, estimated in a real market. Everything in the rebound argument and everything in the market-clearing argument turns on it, and it is essentially unmeasured. Speculative The data would come from matched-grade price and volume series with a demand shifter, which is a standard industrial-organisation design applied to a market nobody has treated as an industrial-organisation problem.

Frontier Third, an independent reconstruction of the headline circularity metric from the same accounts by a party with no stake in the answer. That is a data-and-method exercise, not a research programme, and its absence after a decade of the metric's policy adoption is itself informative.

Frontier Fourth, the fee threshold. The eco-design response begins somewhere above current producer-responsibility fee levels and nobody has estimated where. Speculative That single number would convert an instrument currently justified by its purpose into one that could be set by its effect, and it is estimable from the same panel variation that established the current level is too low.

Frontier The highest-value study available is an evaluation of an eco-modulated scheme using the design that produced the null on flat fees. The original panel used temporal variation in the fees actually charged across 25 countries and four materials; several jurisdictions have since introduced modulated fees on dated schedules. Speculative Running the same estimator across the modulation events would answer the field's live question — whether the instrument fails at any level or only at the level it has been set at — and would do it on data that already exists.

Frontier Second, a pass-through study of producer-responsibility fees into shelf prices. No incidence study exists anywhere: nobody knows whether the fee is borne by the producer, which is the instrument's premise, or by the consumer, which would make it a consumption tax with a recycling label. Speculative Fee schedules change on announced dates and scanner price data is routinely available, so this is a conventional tax-incidence design waiting for someone to run it.

Established Third, publish recycling rates disaggregated by destination and measured at output. This is a reporting reform rather than an experiment, and it is the single change that would make every national circularity claim comparable. Frontier The general principle and the sector-specific version are both already stated in the Institute's construction brief; nobody reports plastics that way either, and plastics is where the headline nine-per-cent figure comes from.

Frontier Fourth, an empirical rebound estimate for one specific closed loop. The mechanism is eight years old, has a revisiting paper behind it, and has no magnitude attached anywhere. Speculative A single well-identified case — a material stream where secondary supply expanded sharply against a plausible counterfactual, with primary production and total consumption tracked — would move the most important idea in this slot from argument to measurement.

Speculative Fifth, an independent evaluation of the leading national programme. Japan's framework has legislated indicators and two decades of data, and the evaluation does not exist. Handwave It would not be a difficult study; it would be an unwelcome one, which is a different obstacle and probably the operative one.

Established Producer responsibility as a fiscal instrument has four moving parts and the record shows which one is set wrong. A fee base, almost always weight; a fee level, set by the producer responsibility organisation to recover collection and treatment cost; a modulation rule, if any, varying the fee by product characteristics; and a hypothecation rule directing the proceeds. Established Measured against the fees actually charged across 25 countries and eighteen years, the level is below the point at which it enters a design decision, and the observed outcome is very little packaging reduction with no substitution between materials. Frontier Everything the instrument does well, it does as a collection levy. Everything it is advertised to do, it does not do, because the price is wrong.

Frontier Eco-modulation is the correct response in principle and its identified weakness is specific. It operates through changes in the fees producers pay, and the core problem is weak environmental linkage between the criteria the fee varies on and the impact the fee is meant to price; life-cycle assessment is the proposed basis. Speculative That proposal inherits every boundary-and-allocation problem life-cycle assessment has, which is a methods question owned elsewhere on this map and is noted here only because it lands directly in the fee schedule.

Frontier Secondary-material market structure explains why the fee level is not the only price that is wrong. Secondary supply competes against a primary price that tracks energy and is set in a deep global market, while the secondary product is a heterogeneous, small-lot, quality-variable good with collection and sorting costs sunk before the sale. Established Where residual value is high the market clears without help — copper at 33% of demand, nickel at 31%. Where residual value is low it does not, which is why toll-based arrangements and regulatory mandates are proposed for lithium-iron-phosphate batteries by an agency that would prefer a market answer.

Established Collection is the physical constraint on the whole apparatus and it is distributed extremely unevenly. Rates of 40–50% in Europe and North America, below 5% in developing Asia and Latin America, and 1% in Africa. Frontier No fiscal instrument reaches a stream that is not collected, and the fee-design literature is written almost entirely about jurisdictions in the first band.

Frontier And in plastics the constraint runs the other way from the usual telling: it is supply, not demand, and not quality. Meeting pledged recycled-PET volumes for 2025 would have required the annual recycling growth rate to double against its 2014–2018 trend. Speculative A trade-press account makes the complementary claim that demand for recyclate exists while supply does not and that recyclate quality is not the obstacle; it is an industry publication and is carried at that weight.

The boundary is tight and is stated as a test rather than as a list. If a passage explains design for disassembly, bolted against welded connections, recycled aggregate replacement ratios, structural steel reuse, urban mining, material passports, or a by-product exchange between two plants, it belongs in another brief and not in this one. Those subjects are owned by Circular Infrastructure Systems, which holds the built-environment treatment and the construction-waste recovery decomposition, and by Industrial Ecology, which holds plant-network symbiosis and material flow and life cycle assessment as methods. This brief may cite the construction decomposition once, as an instance of the general measurement principle, and does; it does not re-derive it.

Within this map the other live connections are Zero Carbon Industrial Systems, whose build-out is the demand side of the extraction projections here; Advanced Battery Technologies, where the chemistry choice determines whether the recovery economics work at all; and Autonomous Supply Chains, which owns the reverse-logistics operations that collection rates describe in aggregate. Critical-mineral concentration as a producer and fiscal question belongs to this category's resource-economies slot, which uses the same recycling data for a different purpose.

Outside it: environmental and resource economics, which supplies the Pigouvian framing the fee evidence is read against; public finance, which supplies the incidence question nobody has answered; industrial organisation, which supplies the substitution elasticity the whole rebound argument waits on; and national accounting, which supplies the material-flow accounts every circularity metric is built from and which were not designed to carry them.

Established The institutional finding here is about who owns the numbers. The most-quoted circularity metric is produced by an advocacy organisation; the extraction baseline comes from an environmental body with a reduction mandate and is corroborated by an economic organisation without one; the recycling economics come from an energy-security agency; and the plastics figure comes from the convenor of a treaty process arguing for production limits. Frontier None of that makes the numbers wrong. It means the field's headline statistics are produced almost entirely by parties with positions, and that the one independent thing available — a research-council-funded systematic review of 835 studies — is the piece of evidence this brief leans on hardest for exactly that reason.

Established Two absences are institutional rather than scientific and both are choices. There is no independent reconstruction of the headline metric from the same public accounts, and there is no independent evaluation of the leading national material-productivity programme despite two decades of legislated indicators. Frontier Both would be inexpensive; neither is anyone's job; and each would produce a number some institution would rather not have.

Frontier The instrument-design finding is the one that matters most for policy and it is recorded below as a typed requirement. Producer responsibility fees are set to recover collection cost, not to price behaviour, and the level at which behaviour changes has never been estimated. Established That is not a scientific unknown of the ordinary kind; it is a number that could be estimated from existing panel data by anyone who wanted it, and the bodies that set the fees are the bodies that would be judged by it.

Speculative And the reporting convention is an institutional artefact with a long tail. Counting at collection rather than at output is administratively convenient because collection is where the weighbridge is, and it produces a national statistic that is systematically generous in an unknown direction and magnitude. Frontier Changing it requires a reporting rule rather than a technology, and the jurisdictions that would look worst under the new rule are the ones that would have to adopt it.

Established The distributional fact underneath the whole subject is that the apparent efficiency of rich economies is partly other countries' extraction. High-income countries use about six times more resources per capita and generate about ten times the climate impacts of low-income ones, and non-domestic resource use runs about threefold the physical mass of the goods actually traded. Frontier A national target denominated in domestic material consumption can therefore be met by moving a smelter, and the country whose statistics improve is not the country whose material burden changed.

Established Collection rates make the same point at the other end of the chain. 40–50% in Europe and North America against 1% in Africa is not a statement about technology; it is a statement about where the infrastructure and the fee revenue are. Frontier A circular-economy programme built on producer-responsibility funding reproduces that distribution, because the funding follows the point of sale.

Frontier Counting exported material as recycled by the exporter is the accounting expression of the same asymmetry. The tonne leaves, the statistic improves, and what happens next is recorded in a jurisdiction with weaker reporting. Speculative How much of the reported gap this accounts for is unknown for every major stream, and it is unknown because nobody publishes rates disaggregated by destination.

Frontier And there is an unexamined incidence question with real distributional content. If producer-responsibility fees are passed through to shelf prices, the instrument is a regressive consumption tax hypothecated to waste collection. If they are absorbed, it is a producer levy. Speculative No study anywhere has established which, and the answer determines whether one of the EU's most widely adopted environmental instruments is progressive or regressive.

Established The civilizational statement this brief can defend is arithmetical rather than rhetorical. Extraction has tripled in five decades and is projected to rise a further 60% by 2060 on one accounting and to double on another; national material footprint rises about 6% for every 10% of GDP; and long-term absolute decoupling of material use is rare across 835 empirical studies. Frontier Against that, closing loops at the observed rates changes the composition of throughput far more than its magnitude.

Frontier The rebound argument makes the stronger version of that point and it is the one worth carrying. If a partially substitutable secondary material adds to supply rather than displacing primary production one for one, then closing a loop at economy scale is not neutral with respect to the quantity of material moved — it may increase it. Speculative Nobody has measured the magnitude, which means the most consequential economic claim about circularity is currently a sign without a number.

Speculative The honest terminal position is that the question “has any economy closed its loops?” is not currently answerable, because the metrics that would answer it are produced by advocates, built on accounts designed for another purpose, or measured at the wrong point in the chain. Handwave That is a more useful statement than a verdict assembled from advocacy material, and it is also an indictment of a field that has had twenty years and a great deal of money to build the measurement it argues from.

These horizons track statistical reporting reforms, fee schedules and extraction projections rather than technology:

• 10 yr: Established The dated things are the projections themselves: extraction rising toward the 2060 baselines, and battery-metal secondary supply moving toward the 20–30% of demand projected for 2050 from 20–40% recovery rates today. Frontier Expect eco-modulated fee schedules to spread faster than evaluations of them, and expect the first serious eco-modulation evaluation to be the single most informative publication in this subject. Speculative Expect no published incidence study and no rebound magnitude unless a funder specifically commissions one.
• 25 yr: Speculative Either footprint accounting becomes an official national series and material-productivity targets are restated in it, or it does not and the targets keep rewarding offshoring. Speculative On markets, the discriminating observation is whether low-residual-value streams ever clear without mandates; if they do not by the 2040s, the toll-and-mandate model becomes the default and circularity becomes a regulated utility rather than a market outcome. Handwave Which happens depends on carbon prices and on virgin material prices, neither of which is forecastable at this horizon.
• 50 yr: Speculative If absolute material decoupling is never observed at economy scale, the policy argument resolves into either a sufficiency position or an acceptance that material throughput grows and the environmental target must be met elsewhere — in energy and emissions rather than in tonnes. Speculative The consumption-based emissions result already points that way, and it is the only absolute decoupling anyone has demonstrated.
• 100 / 250+ yr: Handwave Beyond useful forecasting. The only observation with that reach is that material stocks accumulate in buildings and infrastructure for decades before becoming available at all, so an economy's recoverable stock is set by construction decisions taken a lifetime earlier — which is a statement about lags, not a forecast, and the engineering of those stocks belongs to another brief.

• Depends on Nothing on this map. This brief waits on no result another brief produces. Its unresolved questions are a fee level, an incidence estimate, a substitution elasticity and a reporting convention, and every one of them could be settled from data that already exists. No typed depends-on edge is claimed.
• Requires (not on this map) Three things this subject waits on, none of them a result another brief produces. First, producer-responsibility fees set above the eco-design threshold. On a panel of 25 EU countries over 1998–2015 across four packaging materials, using the fees producer responsibility organisations actually charge, the financial incentive produced “very little (though statistically significant) packaging reduction” and no systematic substitution between materials, because the fees are small relative to total packaging costs. The instrument funds collection; it does not price behaviour. Nobody has estimated the level at which a design response begins, and eco-modulation — the proposed fix — has a weak environmental linkage between its fee criteria and actual impact. Second, recycling rates reported at output and by destination. A tonne is typically counted when collected for recycling rather than when recycled material leaves a facility, so sorting losses, contamination rejection and reprocessing yield are never netted, and exported material counts as recycled by the exporter; the headline figure that only nine per cent of plastics produced are mechanically recycled is measured under those conventions like everything else. Third, a measured magnitude for circular-economy rebound: limited substitutability between secondary and primary products means secondary material may add to total supply, lower price and raise total consumption rather than displacing primary production one for one, and eight years after the mechanism was named — with a 2025 revisiting paper by the original author confirming it is still considered live — no empirical magnitude exists for any specific loop. The first two are reporting and fee-setting decisions regulators could make. The third is a number nobody is producing, and without it the central economic claim about closing loops at economy scale has a sign and no size.
• Enables In principle every materials-intensive programme on this map inherits the extraction denominator recorded here, and any policy that assumes secondary supply displaces primary production one for one inherits the rebound question. No typed enabling edge is claimed, because the relationship has never been measured: there is no empirical magnitude for circular-economy rebound anywhere, and no economy-wide circularity target has been shown to produce a measurable material outcome.
• Adjacent Environmental and resource economics, which supplies the Pigouvian framing; public finance, which supplies the unanswered incidence question; industrial organisation, which supplies the substitution elasticity; national accounting, which supplies the material-flow accounts every circularity metric is built on; and within this map Circular Infrastructure Systems and Industrial Ecology, which own the engineering this brief deliberately does not restate.

Established “The global circularity rate is falling, so recycling is failing.” This is denominator arithmetic and it is the most useful thing this brief can correct. Extraction has tripled in five decades and is projected to rise a further 60% by 2060 or to double, depending on the accounting. When the denominator grows faster than recovery, the cycled share falls while every recovery number rises. Frontier A falling share is a statement about extraction growth, not about recovery performance, and reading it the other way has become routine.

Handwave “The circularity gap is about seven per cent.” No such figure appears anywhere in this brief, and its absence is deliberate. The publisher's site was unreachable in the research pass behind this brief, so neither the percentage nor the edition nor the falling-across-editions trend is restated here. Speculative What can be said without it is that the metric is produced by an organisation founded to promote the concept it measures, is built on national material-flow accounts designed for another purpose, and has a 2025 peer-reviewed critique from inside the material-flow discipline. That critique's argument is not characterised here because only its title was retrievable, and a title is not an argument.

Frontier “Rich countries have decoupled growth from resource use.” On domestic material consumption several have improved; on footprint, much of that improvement is externalised extraction, with non-domestic resource use running about threefold the physical mass of traded goods and footprint rising about 6% per 10% of GDP. Established The genuine absolute decoupling result in the literature is about consumption-based emissions in some industrialised countries, and it is not a material result. Conflating the two is the single most common error in the policy literature and the review usually cited for it says so.

Established “Extended producer responsibility drives eco-design.” Measured across 25 countries and eighteen years using the fees actually charged, it produced very little packaging reduction and no substitution between materials. Frontier The fees are set to recover collection cost and are too small to enter a design decision. The instrument works; it is just not the instrument it is described as. Whether eco-modulation changes that is an open question with no evaluation behind it yet.

Frontier “Recycled feedstock is more expensive than virgin.” Widely asserted, mechanistically plausible — collection and sorting costs, small-lot heterogeneous markets, imperfect substitution, virgin prices tracking oil — and not verified here, because the price series that carry it were unreachable and the peer-reviewed price-gap literature is thin. Speculative It is carried as a mechanism, not as a number, and a matched-grade independent price series over a full oil-price cycle would settle it in either direction.

Speculative “Japan has demonstrated economy-scale material productivity gains.” The framework has legislated indicators and two decades of published targets, and no independent evaluation of whether the targets were met, or what moved them, appears in the searchable literature. Established Repeating a programme's own claims about itself is not evidence, and saying that the evaluation is missing is the honest alternative.

Frontier “Closing a loop necessarily reduces material throughput.” Not necessarily, and the argument against is economic rather than technical. Secondary material that only partially substitutes for primary adds to total supply, lowers price and raises total consumption; profit-driven private adoption of circular practices may therefore undermine the intended environmental gains without regulatory intervention. Speculative The mechanism is eight years old and has no measured magnitude for any specific loop, which makes this the largest open question in the subject rather than a settled objection.

Established And one whole class of material is out of scope rather than disputed. Nothing above concerns design for disassembly, urban mining, recycled aggregate replacement ratios, structural reuse, material passports or by-product exchange between plants. Those belong to the Institute's construction and industrial-ecology briefs, which is where the engineering evidence lives and where it is properly measured. Handwave Anyone arguing from a plant-level symbiosis case to an economy-wide circularity conclusion is making exactly the aggregation step this brief exists to question.