No company in this market has share, and the reason is structural rather than competitive. Polypropylene capacity is not invented by the companies that own it; it is bought as a process licence, built in two to three years, and then operated as a spread business. That produces an industry with dozens of credible producers, no pricing power at the top, and a competitive contest that is decided by the propylene cost position rather than by anything a marketing department does.
The numbers make the point quickly. Borouge Group International - formed by combining Borouge with Borealis and acquiring Nova Chemicals, with closing expected at the end of March 2026 - will hold 13.6 Mt of nameplate polyolefin capacity across Europe, the Middle East and North America, making it the fourth-largest polyolefins producer in the world. That 13.6 Mt is polyethylene and polypropylene combined. Set against an 86 Mt PP resin market, the fourth-largest polyolefins company on earth holds a mid-single-digit fraction of this one product. The top of the table is genuinely flat.
LyondellBasell's 2025 results show what the bottom of a cycle does to that structure. The company reported a net loss of USD 738 million for the year and EBITDA of USD 1.1 billion, or USD 2.5 billion excluding identified items - the difference being non-cash asset write-downs, transaction costs, its Cash Improvement Plan and discontinued operations. It generated USD 2.3 billion of operating cash, reinvested USD 1.9 billion and returned USD 2.0 billion to shareholders, and it enters 2026 planning capital expenditure of USD 1.2 billion. Its stated first-quarter operating plan is the most informative line in the whole disclosure: Olefins and Polyolefins Americas at approximately 85%, Olefins and Polyolefins Europe, Asia and International at approximately 75%, Intermediates and Derivatives at approximately 85%. A complex running at 75% is not optimising its sales mix. It is holding a position and waiting.
| Archetype | Examples | Source of advantage | 2026 posture |
|---|---|---|---|
| Integrated feedstock owner | Gulf and North American producers with ethane or associated gas; Chinese refinery-integrated majors | Molecule cost at the wellhead or refinery gate | Running for volume, taking share of a shrinking margin |
| Process licensor and compounder | LyondellBasell (Spheripol, Spherizone); Borealis (Borstar, now inside Borouge Group International) | The grade slate and the catalyst, sold as a licence and a service | Earning fees on other people's capital while their own assets run light |
| PDH-linked Chinese entrant | Private integrated refiners and propane dehydrogenation complexes | Provincial support, cheap capital, domestic propane logistics | Running through negative propylene margins to hold downstream share |
| Merchant converter-facing producer | European and Japanese units without captive propylene | Proximity, service, qualification on specific grades | Exiting, selling, or converting to trade defence |
Polypropylene is one of the most licensed technologies in the chemical industry. A new entrant does not develop a polymerisation process; it buys a licence, signs a catalyst supply agreement, and receives a grade slate that has been qualified with converters over decades. That is why the technology holders have survived three downturns better than the producers who bought from them: the licensor's revenue is a fee on somebody else's capital expenditure, and it does not fall when the spread compresses.
The grade slate is the commercial substance behind the licence. A process that can make a broad melt-flow range, control xylene solubles and deliver consistent impact copolymer is a process whose owner can supply automotive compounds. A process limited to commodity homopolymer is a process whose owner competes on delivered price against a Chinese container. Two plants with identical nameplate capacity are not in the same business.
LyondellBasell will close its remaining 260,000 tonne a year polypropylene unit at Brindisi, Italy, by the end of 2026. The unit is historically significant: it was the first Spheripol polypropylene unit ever built, started in 1982, producing commodity homopolymer for packaging. It is closing because of the pipe next door, not because of the market.
LyondellBasell does not operate a propylene cracker at Brindisi. It took up to 220,000 tonnes a year of propylene by sea from Eni's Versalis cracker on the same site until that unit shut down in April 2025. Since then feedstock has been shipped in from the company's other Italian sites, principally Priolo, and the freight has removed whatever cost position the unit had. Trade unions report the unit was offered for sale at a symbolic price with no buyer, because no buyer can secure propylene without the cracker. Eni has hired an investment bank to sell the idled cracker; if that sale succeeds, a buyer for the polypropylene unit becomes possible again. Two assets, one pipeline, and the entire commercial fate of both decided by the same question.
That is the mechanism by which European polypropylene capacity is disappearing. Announced closures in this industry are rarely decisions about polypropylene. They are decisions about the cracker, transmitted one stage downstream.
Companies that cannot win on cost have two options in 2026, and both are visible at scale. The first is to buy capacity rather than build it. Borouge Group International combines a Gulf feedstock position, a European technology and compounding franchise, and a North American asset base in a single transaction, and does so with investment-grade ratings rather than with a construction programme. LyondellBasell went the other way and sold several European resin and feedstocks plants to a newly formed regional company, Velogy, completing on 1 May 2026, while retaining the market through a narrower network.
The second option is to litigate the import. Brazil's anti-dumping measure on polypropylene resin from South Africa and India is under review, initiated at the petition of Braskem on 24 December 2025, with the final determination opinion scheduled for 28 September 2026. Braskem is not defending a technology position or a grade franchise; it is defending a domestic price that rose 20.6% in five months while Asian prices fell. Trade defence is what competitive strategy looks like when the cost curve has already decided the outcome.
Three variables decide who earns a return in this market over the next eight years, and none of them is market share. The first is the delivered cost of propylene. The second is whether the producer's grade slate reaches compounding and specification work, where the premium is USD 1,325 a tonne over base resin. The third is whether the producer's home market gives it a trade regime it can survive in.
Companies with none of the three are doing one of four things: closing the asset, selling it, merging it into someone else's balance sheet, or asking a government to tax the alternative. Companies with all three are adding capacity in the Middle East and Asia and letting the surplus clear where it lands.
" value_chain_content: "The polypropylene value chain is usually drawn as four boxes: propylene, polymerisation, compounding, conversion. The useful version of that diagram makes a narrower point. Two of those boxes earn a return and two do not, and which two depends almost entirely on whether the business is selling a molecule or a specification.
Global propylene demand is roughly 130 Mt a year, of which about 67% becomes polypropylene. Three production routes supply it. Steam crackers make propylene as a co-product of ethylene, alongside butadiene and aromatics. Refineries make it in fluid catalytic crackers as a side effect of gasoline production. On-purpose plants make nothing else: propane dehydrogenation strips hydrogen from propane and produces propylene directly.
The mix between those routes changed the industry's economics permanently. As US and Middle Eastern crackers shifted to ethane - which yields almost no propylene - co-product supply stopped growing while polypropylene demand kept compounding. The answer was on-purpose capacity, and China industrialised it at extraordinary scale: propane dehydrogenation capacity above 20 Mt a year by 2025, designed largely around imported US propane.
That buildout overshot. Chinese PDH operating rates fell below 75% and margins were negative for much of 2024 and 2025, yet plants kept running, to hold downstream market share and to feed integrated polypropylene lines. This is the single most important structural fact in the chain: a large block of the world's propylene supply is now produced by plants that do not shut down when the spread is negative. Every stage downstream inherits that insensitivity.
A polypropylene plant converts propylene into resin through a catalysed polymerisation in loop, gas-phase or multi-zone reactors. The conversion cost - energy, catalyst, labour, maintenance, depreciation - is largely fixed per tonne. The plant therefore earns the propylene-to-polypropylene spread multiplied by volume, and nothing else. There is no brand, no service contract and no switching cost at this stage for commodity homopolymer.
The grade slate is where the stage begins to differentiate. Homopolymer is the commodity product. Random copolymer adds clarity and flexibility for packaging. Impact copolymer adds low-temperature toughness for automotive and appliances. Each step up in specification narrows the number of plants that can make the grade reliably, and reliability - not capability - is what converters pay for, because a moulding line that stops is more expensive than a tonne of resin.
Under surplus, the spread is set by the marginal plant, and in 2026 the marginal plant is a Chinese PDH-integrated complex that runs regardless of margin. That is why price is a shock absorber rather than a signal in this chain: it can fall a long way before it changes behaviour at the point of production.
Roughly 8 Mt of the world's PP is not sold as resin. It is compounded: blended with talc or glass fibre for stiffness, with elastomers for impact, with flame retardants, stabilisers and pigments, and then sold against a customer's material specification. On this report's numbers that layer carries an average premium of USD 1,325 a tonne over base resin, which at a base resin price of USD 981 is 1.35 times the price of the polymer it is built from.
The premium is not a mixing fee. It is the price of qualification: a compounder has to be approved into an automotive OEM's material system, guarantee properties across a production run measured in years, absorb formulation changes when a filler or additive is restricted, and hold a supply commitment through model lifecycles. That is a service business wearing a chemical company's clothes, and it is the only stage in this chain where a supplier can raise price while the resin index falls.
It is also the stage with the most regulatory exposure, because the additives that make a compound work - flame retardants, processing aids, certain stabilisers - are precisely the substances that restriction programmes target. A compounder's portfolio is therefore a regulatory question as much as a technical one.
Converters buy resin on index-linked or quarterly contract terms and sell film, mouldings and thermoforms to brand owners on annual agreements. That mismatch means converters absorb the timing difference between a resin movement and a shelf price, which is why their margins are inversely cyclical to the producers' and why packaging price changes reach consumers with a lag of two to three quarters.
This stage is where the industry's volume lives and where almost none of its margin does. A BOPP film line, an injection moulder making crates and a thermoformer making dairy tubs are all buying the same tonne and competing on conversion cost, utilisation and freight. Their returns depend on running full, and their pricing power is limited to the narrow window between a resin fall and a customer renegotiation.
Regulation has added a stage that did not exist a decade ago. Collection, sorting, mechanical recycling, and re-compounding now sit alongside the virgin chain, because European packaging law requires minimum recycled content in the plastic part of packaging from 2030 - 35% for plastic packaging outside the contact-sensitive categories, 10% for contact-sensitive packaging made from plastics other than PET, and 30% for contact-sensitive PET.
The bottleneck in that loop is not collection and not sorting technology. It is food-grade polypropylene recyclate: the stream that can legally go back into a pack that touches food. It is scarce, it trades at a premium to virgin resin in exactly the years when virgin resin is cheapest, and it is the reason mono-material PP packaging design is spreading faster than the recycling capacity that is supposed to supply it.
The pattern is consistent across the chain, and it is not the pattern the industry's strategy documents describe. Value accrues at two points: where propylene is cheapest, and where the specification is hardest to replace. It does not accrue in the middle, where a commodity pellet is a commodity pellet and the buyer's only question is the delivered price of the next 5,000 tonnes.
That is why the compounding layer, at roughly 8% of the volume, supplies around a fifth of the value growth in this forecast. It is also why the industry's most repeated strategic phrase - move downstream - is directionally correct and almost always under-costed, because moving downstream means acquiring qualification, not acquiring a converting line.
" investment_activity_content: "The most revealing number in this industry in 2026 is not a capacity figure. It is USD 1.2 billion - LyondellBasell's planned capital expenditure for the year, against USD 1.9 billion actually reinvested in 2025 and a company that generated USD 2.3 billion of operating cash and returned USD 2.0 billion to shareholders. That is the arithmetic of an industry funding distributions and maintenance while deferring growth. Read alongside a 13.6 Mt consolidation in the Gulf and continuing additions in China, it tells you where capital has decided the future of polypropylene is.
Borouge Group International is the formation of a polyolefins major out of three existing ones: Borouge, Borealis, and the acquired Nova Chemicals, closing expected at the end of March 2026, with 13.6 Mt of nameplate capacity across Europe, the Middle East and North America and a position as the world's fourth-largest polyolefins producer. The ratings attached to it - investment grade across all three agencies - are part of the strategy rather than a by-product: consolidation paid for with a strong balance sheet is cheaper than greenfield construction in a market with 120% capacity cover in its largest consuming country.
The Borouge 4 complex shows how the capital is being structured. B4 is a 1.5 Mt ethane cracker with 1.4 Mt of polyethylene capacity, 70% owned by ADNOC and 30% by OMV, with the first plant starting up in the first quarter of 2026 and a progressive ramp through the year. It is integrated into what will become the world's largest single-site polyolefins complex. But the listed entity will not own it: under an asset usage agreement signed in March 2026, Borouge - and subsequently Borouge Group International - operates and markets the volumes for an at-cost utilisation fee, delivering an estimated USD 400 million of cumulative net profit over three years, or about 10% annual accretion post ramp-up, with acquisition of the asset not expected before 2029.
That structure is a template for the decade. The company takes the commercial position and the earnings without the capital charge, and the sovereign owner keeps the asset. If that becomes the standard way polyolefin capacity changes hands, then the industry's investment statistics will increasingly measure construction in the Gulf and earnings in Rotterdam, and the two will no longer reconcile in a single company's balance sheet.
European capital allocation is best described as subtraction. LyondellBasell completed the sale of several European resin and feedstocks plants to a newly formed regional petrochemical and polymer company, Velogy, on 1 May 2026, and is closing the 260,000 tonne a year Brindisi polypropylene unit by the end of 2026 because it lost its propylene supply when the neighbouring Versalis cracker shut in April 2025. Neither decision is a bet on polypropylene demand. Both are decisions about which assets can be supplied at a competitive cost.
What survives in Europe is not commodity polypropylene. It is compounding, application development and specification work, which is where the USD 1,325 a tonne premium sits and where a European producer can charge for qualification rather than for a molecule. That is a coherent strategy and it is a small one: the compound layer is roughly 8% of the volume in this market.
China is the only region still adding polypropylene capacity at scale, and the additions are no longer a response to price. Capacity stands at about 120% of domestic consumption in 2026, exports are annualising near 4.8 Mt, and the April 2026 programme from the industry ministry and six other departments to upgrade outdated petrochemical plants and phase others out by 2029 - with rolling surveys of newly identified facilities from 2027 - is a filter applied to capacity that is already financed and under construction.
Two features of Chinese investment matter to anyone forecasting this market. The first is that the feedstock is largely on-purpose: propane dehydrogenation above 20 Mt a year, coal-to-olefins and refinery integration, none of which is priced off a propylene market that clears. The second is that the investment decision is made by entities whose cost of capital and whose objective function are not those of a listed Western chemical company. A project that would be cancelled at a 75% operating rate elsewhere is completed here, because the operating rate is not the objective.
Three positions can still underwrite a polypropylene investment in 2026, and they are narrow. Feedstock advantage in the Gulf and North America, where ethane or associated gas is available at a cost no other region can match - this is what B4 is. Demand growth with a protected border, principally India, where domestic projects have lagged consumption and the country has become a major import growth centre for polypropylene; investment there is justified by demand rather than by cost, which is the only kind of investment that survives a surplus. And qualification, where a compounder adding a line beside an automotive plant is not competing with a Chinese container at all.
Everything else is a bet that the surplus clears before the depreciation schedule does. On this report's numbers it does not: the conservative path assumes flat real prices to 2033, and it is only 0.88 percentage points below the baseline.
LyondellBasell's MoReTec-1 unit, under construction through 2026, is the clearest example of the one category of Western polyolefin investment that is growing: chemical recycling. Its justification is not that pyrolysis oil is cheaper than naphtha - it is not - but that European packaging regulation creates a legally mandated market for recycled content from 2030, at 35% for plastic packaging outside the contact-sensitive categories and 10% for contact-sensitive packaging made from plastics other than PET.
A legislated offtake is the only demand signal in this industry that a producer can underwrite. That is why the recycling capital is being committed while cracker capital is not, and why the same companies are simultaneously closing polymerisation capacity and building depolymerisation capacity at the same sites. The assets are related; the markets are not.
Investment in this industry is no longer allocated to markets. It is allocated to cost positions and to regulated offtake. That is the assumption embedded in the 3.88% baseline: no new wave of Western capacity, continuing Chinese additions at a decelerating pace, and a compound layer that attracts the only genuinely profitable capital in the chain.
The risk in that assumption is not that it is wrong about today. It is that capital cycles turn faster than forecasts. A company that defers USD 700 million of growth capex in 2026 because the spread is negative is also a company that will have nothing to start up in 2029 if the surplus clears in 2028.
" technology_innovation_content: "Polypropylene has no shortage of technology narratives. It has a shortage of technologies that change the delivered cost of a tonne or the price a customer will pay for one. The distinction matters here more than in most markets, because the polymerisation technology is now mature and sold as a licence, while the compounding and design work at the end of the chain is where the margin actually moves.
Three frontiers are real, in ascending order of how much they matter to the 2033 number: catalyst and process capability, product and composite design, and circularity. Only the second earns a premium today.
Every polypropylene plant in the world runs on a catalyst system and a reactor configuration, and the pair determines what the plant can make. The generations are well understood: earlier Ziegler-Natta systems delivered a broad but coarse product slate; later generations and metallocene systems gave producers far finer control over molecular weight distribution, melt flow and the balance between stiffness and impact strength. Multizone reactor designs - LyondellBasell's Spherizone, and Borealis's Borstar process now inside Borouge Group International - allow bimodal and multimodal grades to be made in a single train, which is how producers reach the grade slate that automotive and appliance compounds require.
The commercial consequence is that a polypropylene plant is not a fungible asset. Two plants with the same nameplate tonnage can have completely different access to the part of the market that pays. A commodity homopolymer line competes with a delivered container from Ningbo; a line that can make consistent high-impact copolymer for interior trim does not.
What has changed is that this capability is no longer a differentiator at the top. It is a licence that any well-financed entrant can buy, which is precisely how Chinese capacity was built so quickly. Process technology has been democratised, and a democratised technology stops earning a return. That is one reason the industry's research spend has shifted downstream.
The technology that moves this forecast is not in the reactor. Roughly 8 Mt of the world's PP is compounded - filled with talc for stiffness, reinforced with glass fibre for strength, modified with elastomers for low-temperature impact, stabilised and flame-retarded - and that layer carries an average premium of USD 1,325 a tonne, or 1.35 times the base resin price.
Three product families are doing the work. Long-glass-fibre and glass-mat thermoplastic PP replaces metal and engineering plastics in structural parts: front-end modules, seat structures, battery housings and underbody shields. Mineral-filled and impact-modified PP does the higher-volume work in interior trim, instrument panels and HVAC ducting, where cost per part governs. High-melt-strength PP makes extrusion, thermoforming and foaming viable, which is how PP enters applications that used to belong to other polymers.
The demand evidence is in the composites data rather than in vendor claims. Global demand for light-vehicle composite materials reached a record 4.9 billion pounds, about 2.22 Mt, in 2024, according to the American Composites Manufacturers Association - a record set in a year when global vehicle production was flat. Usage per vehicle is rising, and battery electric vehicles take more composite per vehicle than combustion vehicles do, because weight saved buys range. Polypropylene is the volume polymer in that mix.
Packaging design is the second half of this frontier and it is less glamorous and more consequential. European law requires minimum recycled content in the plastic part of packaging from 2030: 35% for plastic packaging outside the contact-sensitive categories, 10% for contact-sensitive packaging made from plastics other than PET, 30% for contact-sensitive PET and 30% for single-use plastic beverage bottles. A multi-material laminate cannot meet a recycled-content quota and cannot easily be sorted. An all-PP pack can. The technology that resolves this is not a new polymer; it is seal-layer, barrier and stiffness design that lets a pack be built from one material family - and it is the reason mono-material PP formats are spreading faster than the recycling capacity meant to supply them.
Chemical recycling - pyrolysis of mixed plastic waste back to an oil that can be cracked, and depolymerisation routes back to monomer - is the most capital-intensive frontier and the smallest in tonnage. Its economics do not work on a virgin-resin comparison; pyrolysis oil is not cheaper than naphtha. It works because European packaging law creates a legally mandated demand for recycled content, and because brand owners will pay for a mass-balanced recycled claim on a pack.
Two constraints limit how fast this scales. The first is scale economics: a pyrolysis plant needs a very large secured waste stream at a single site to make sense, which excludes most local and specialty streams. The second is what the industry calls mass balance. A plant producing a few thousand tonnes of pyrolysis oil feeds it into a cracker handling millions of tonnes, and the recycled content has to be allocated across output by an accounting convention. The convention can be defensible; the marketing claim built on it frequently is not.
Mechanical recycling remains far larger and cheaper for the streams it can handle. The binding constraint there is not technology but food-grade polypropylene recyclate - the stream legally allowed back into a pack that touches food. It is scarce, it trades at a premium to virgin resin in exactly the years virgin resin is cheapest, and it is the real bottleneck in the 2030 compliance story.
Advanced process control, inferential sensing and predictive maintenance are genuinely deployed in this industry, and they pay back quickly because they need no new molecule. But in a market with structural surplus, the return on a process improvement accrues to the customer rather than the producer: a yield gain raises industry output and lowers industry price, because price is set by the marginal producer's cash cost. Process innovation in a trough is therefore defensive. It keeps a plant running; it does not raise a margin.
The technologies that will matter to 2033 are not the ones with the best press release. They are the ones that let a supplier charge for a specification in a region that cannot compete on feedstock: compound formulations qualified into vehicle programmes, mono-material packaging designs that satisfy a recycled-content quota, and pyrolysis capacity with a secured waste stream.
Polymerisation innovation is now a licence fee. Compounding and design innovation is a margin. That asymmetry is the whole technology story of this market, and it is why the compound layer supplies around a fifth of the value growth in this forecast from roughly 8% of the volume.
" market_risk_content: "The risks in this market are unusually easy to rank, because one of them dominates and the others are mostly mechanisms for it to get worse. Structural surplus is the risk. Trade defence, feedstock geopolitics, the European cost position, recycled-content scarcity and substitution are the channels through which the surplus reaches price.
China's polypropylene capacity is running at about 120% of its own consumption in 2026, and the country accounts for a little under 40 Mt of the 86 Mt global market. When the largest producer by a wide margin can cover its domestic demand with a fifth to spare, the surplus is exported and the export price becomes the world price.
The arithmetic of the imbalance is straightforward. Restoring operating rates to anything like their previous cycle levels would require global capacity growth of roughly 1.4 Mt a year through the mid-2030s; the current build is closer to 4 Mt a year. That gap is not a forecasting assumption - it is what the announced project schedule implies, and it is why this report's price path is essentially flat while volume grows 3.24% a year.
The transmission is through price, and it is already visible: Chinese raffia fell 4.6% between February and July 2026 while the US contract rose 5.5% and the European contract 2.5%. Price is the shock absorber, and in a market where a large block of supply does not respond to negative margins, the absorber compresses for everyone with a higher cost base.
As Chinese volume lands in new destinations, the destinations respond with duties - and each duty relocates the tonne rather than removing it. Brazil's anti-dumping measure on polypropylene resin from South Africa and India is under review at Braskem's petition, initiated on 24 December 2025, with the evidence phase closing on 30 June 2026 and the final determination opinion scheduled for 28 September 2026.
Brazil shows the mechanism cleanly: its CFR price rose 20.6% between February and July 2026 while Asian prices fell. That is what a protected border does to a buyer, and it is also why the world does not have one price. The consequence for a supplier is that market access becomes a licensing question rather than a cost question - a company's realisable price depends on which duty regime its cargo can enter.
The strategic risk is second-order and worse. Each new measure teaches exporters to route around it, which means cargoes move to the next open market and depress the price there. Trade defence protects a domestic price; it does not restore industry utilisation, and it makes the geographic map of price more volatile than the underlying surplus.
Chinese propylene supply now depends heavily on propane dehydrogenation - over 20 Mt a year of capacity - designed largely around imported US propane. That is a long supply chain with concentrated points of failure: export terminals in one basin, a shipping route, and a feedstock whose price moves with US heating demand and with LNG economics rather than with Chinese polymer demand.
The exposure runs both ways. A disruption or a price spike at the propane end raises the cost of the marginal polypropylene producer, which supports price but compresses the spread on integrated Chinese complexes already running at sub-75% rates and negative margins. Conversely, cheap propane keeps the export machine running regardless of polymer economics, because the feedstock is bought on long-term contracts and the alternative to producing is not producing at all.
This is the risk that most often surprises a forecast, because it can move in either direction within a quarter and it is not visible in the demand data.
Europe's polypropylene capacity does not close because polypropylene became uneconomic in isolation. It closes because the cracker next door closed. LyondellBasell's 260,000 tonne a year Brindisi unit is the worked example: it lost up to 220,000 tonnes a year of propylene when Eni's Versalis cracker on the same site shut in April 2025, and its substitute supply now comes by sea from Priolo at a freight cost the unit cannot absorb.
The forward risk is that the announced programme understates the real one. When a cracker leaves an integrated cluster, the derivative units around it lose their pipeline and start buying merchant molecule at merchant prices plus freight. Some of the next wave of polypropylene closures will therefore arrive with no announcement, as a quiet operating-rate decision rather than a press release. Borealis has already signalled that it expects further polyethylene and polypropylene closures in 2028 and 2029 on current overcapacity.
From 1 January 2030 - or three years from the implementing act, whichever is later - European packaging law requires minimum recycled content in the plastic part of packaging: 35% for plastic packaging outside the contact-sensitive categories, 10% for contact-sensitive packaging made from plastics other than PET, 30% for contact-sensitive PET, and 30% for single-use plastic beverage bottles. Polypropylene packaging sits mostly in the 35% and 10% categories.
The risk is not that the rule will not apply. It is that the recycled polypropylene capable of meeting it - food-grade material authorised for contact use - is the scarcest stream in the system, and it trades at a premium to virgin resin in precisely the years when virgin resin is cheapest. A converter facing a 35% obligation and a thin recyclate market has three options: pay the premium, redesign the pack to a material with better recyclate availability, or move the format to fibre. All three are demand risks for polypropylene, and two of them are permanent.
Packaging is about 43% of this market's value and automotive about 18%. Between them they carry 61% of the demand, and they are exposed to the two most different cycles in the economy: consumer staples volumes, which are stable and slow, and vehicle production, which is cyclical, regional and being reshaped by powertrain mix.
Fibre-based food-contact packaging, including PFAS-free formats now being specified in Europe, is taking share in foodservice and dry-food applications that PP thermoforms used to serve. Mono-PE and mono-PET structures take share in flexible packaging. None of these substitutions is large in any one year, and all of them are one-directional. A market whose largest segment grows at 3.27% rather than at packaging volume growth is a market losing share inside its own biggest application.
The conservative path - USD 120.3 billion by 2033 at 3.00% - is not a stress case. It is the baseline with the surplus clearing two years later than assumed, which is what happens if China's 2029 phase-out list goes unenforced and non-Chinese closures stall. That single deferral is worth USD 8.5 billion of 2033 value.
The optimistic path is worth USD 8.2 billion above the baseline and requires the opposite: enforcement in China plus faster exit elsewhere. The band is therefore narrow in percentage terms and material in absolute terms, and it is decided by supply behaviour rather than by demand. That is unusual for a consumer-facing material and it is the single thing a reader of this report should carry away.
" regulatory_landscape_content: "The regulation that actually reprices polypropylene in the 2026-2033 window is packaging law, not carbon law. That is the opposite of what most material written about European chemicals assumes, and getting the order right changes what a producer should spend money on.
The Carbon Border Adjustment Mechanism's covered sectors at entry into force are cement, iron and steel, aluminium, fertilisers, electricity and hydrogen. Organic chemicals and polymers - polypropylene included - are outside the mechanism. A European polypropylene producer therefore carries the emissions trading cost of making the resin while an importer of the same resin carries nothing.
That asymmetry is a meaningful part of why European contract polypropylene trades USD 580 a tonne above Chinese raffia, and it is the industry's strongest policy complaint. It is also the reason any extension of CBAM to organic chemicals is watched more closely by producers than by their customers: extension would remove the distortion and would raise the cost of every European manufacturer downstream of imported chemistry.
The practical consequence for a reader is that a polypropylene strategy built on a carbon border adjustment assumption is built on a regulation that does not yet exist for this product. The regulation that does exist is about what packaging is made of.
Regulation (EU) 2025/40 sets minimum recycled content in the plastic part of packaging, calculated as an average per manufacturing plant and year, per packaging type and format. From 1 January 2030 - or three years from the date of entry into force of the relevant implementing act, whichever is later - the thresholds are:
| Packaging category | From 2030 | From 2040 |
|---|---|---|
| Contact-sensitive packaging, PET as the major component (excluding single-use beverage bottles) | 30% | 50% |
| Contact-sensitive packaging, plastics other than PET (excluding single-use beverage bottles) | 10% | 25% |
| Single-use plastic beverage bottles | 30% | - |
| All other plastic packaging | 35% | - |
Polypropylene sits overwhelmingly in the second and fourth rows, which is why the obligation is more demanding for this polymer than the headline 30% suggests. The regulation also excludes categories where recycled content would compromise safety - immediate packaging of medicinal products, contact-sensitive packaging for medical devices and in vitro diagnostics, and contact-sensitive food packaging for infants, young children and special medical purposes - and that exclusion list is a reminder of how tightly the rule is bound to food-contact law.
The regulation also amends the Single-Use Plastics Directive, which is the mechanism by which packaging formats have been progressively restricted by material and application rather than by product. For a polymer whose largest single end use is packaging, a mandate on the composition of packaging is a demand instrument of the first order, and it operates in two directions at once: it favours mono-material polypropylene designs that can be sorted and recycled, and it raises the cost of any pack that cannot demonstrate compliant recycled content.
The binding constraint is not collection and not sorting. It is food-grade polypropylene recyclate - material authorised to go back into a pack that touches food. That stream is thin, it trades at a premium to virgin resin in exactly the years virgin resin is cheapest, and its scarcity is the reason the 35% threshold is more demanding in practice than it looks on the page.
This is also where the chemical recycling investment comes from. Pyrolysis and depolymerisation produce a recycled feedstock that can be allocated to packaging output under a mass-balance convention, and the resulting claim is what brand owners are buying. The regulation does not mandate chemical recycling; it mandates an outcome, and chemical recycling is one of the few routes that can deliver that outcome at food-contact quality.
European chemicals regulation bites polypropylene not in the polymer but in the additives that make a grade usable. Flame retardants, processing aids, certain stabilisers and pigments are the substance groups most exposed to restriction programmes, and the direction of travel for more than five years has been toward broad, group-based restrictions rather than substance-by-substance decisions.
For a compounder this is a portfolio question rather than a compliance cost, because substitution work has to be done years before a restriction applies and re-qualification inside an automotive programme is slow. The practical effect in the forecast is to raise the value of an already-qualified compound and to raise the barrier to entry - both of which support the USD 1,325 a tonne premium rather than threatening it.
The most consequential polypropylene regulation in the world in this decade is being written in Beijing, and it is industrial rather than environmental. In April 2026 the industry ministry and six other departments set out a programme to upgrade some outdated petrochemical plants and phase others out by 2029, with lists compiled by local governments and annual rolling surveys of newly identified facilities from 2027.
Two features of that programme determine whether it changes the forecast. The first is that it operates on plants that already exist, which makes it a quality filter rather than a brake on committed capacity. The second is that enforcement is delegated to provincial authorities whose incentives have historically favoured running the asset. A reader should treat the 2029 date as a policy intention with a real chance of slipping, and the conservative scenario in this report assumes exactly that.
Brazil's anti-dumping measure on polypropylene resin is under review, initiated at Braskem's petition on 24 December 2025, with the evidence phase closing on 30 June 2026, disclosure of the essential facts scheduled for 19 August 2026 and the final determination opinion for 28 September 2026. The measure covers resin from South Africa and India under two customs codes, homopolymer and copolymer.
That timetable matters beyond Brazil because it is the pattern: as Chinese volume is redirected, destination markets open reviews, and each review changes the delivered price available to a specific exporter in a specific market. For a producer, trade remedies are now a regulatory exposure with a calendar, not a background risk.
The regulatory effect on this forecast is a mix effect, not a volume effect. Nothing in the current rulebook reduces the tonnage of polypropylene the world will use by 2033; packaging regulation changes what the tonne has to contain and which formats survive, and carbon regulation changes where the tonne is made. That is why Europe's growth rate is 2.08% rather than negative, and why the 2030 recycled-content deadline is modelled as a cost and a design constraint rather than as a demand shock.
The one regulation that could change the volume line is not European. It is whether Beijing's 2029 phase-out is enforced. That is the single policy variable that moves this market's supply side, and it is being decided by a government that has every incentive to protect downstream polymer users from the price consequences of enforcing it.
" faq_question_1: "How large is the polypropylene and polypropylene composites market in 2025?" faq_answer_1: "USD 95.0 billion at first sale in 2025, on 86.0 million tonnes of polypropylene resin, reaching USD 128.8 billion by 2033 at 3.88 per cent. Volume grows 3.24 per cent a year; realised price contributes almost nothing. The 8.0 million tonne compound layer supplies USD 10.6 billion of that total from 9.3 per cent of the volume." faq_question_2: "What is happening to Chinese polypropylene capacity?" faq_answer_2: "China's PP capacity now covers about 120 per cent of its own consumption. Exports ran 1.6 million tonnes in January to April 2026 against 1.0 million tonnes a year earlier, a 4.8 million tonne annualised pace against 3.1 million tonnes for all of 2025. Global capacity would need to grow roughly 1.4 million tonnes a year to restore operating rates; the current build is closer to 4 million." faq_question_3: "Why does European polypropylene trade so far above Chinese material?" faq_answer_3: "On 2026 monthly contract assessments the EU contract sits USD 580 a tonne above Chinese raffia, which is 1.78 times, before freight. European crackers have closed while Chinese capacity kept building. Brindisi is the clearest case: LyondellBasell closes 260,000 tonnes a year of PP at the end of 2026 because Eni's Versalis cracker next door shut in April 2025 and feedstock now comes from Priolo." faq_question_4: "Who are the leading polypropylene producers?" faq_answer_4: "Read the list by layer. Cracker- and refinery-integrated producers supply the resin: LyondellBasell, ExxonMobil, SABIC, Borealis, Braskem, Reliance, Formosa, Sinopec, PetroChina and INEOS. The compound layer, glass-fibre and mineral-filled grades, is a separate and far more fragmented business. No single firm holds a double-digit global share of resin." global_schema: organization: "@id": MIWS-root#organization "@type": Organization url: "https://www.pheonixresearch.com/" name: Pheonix Research website: "@id": MIWS-root "@type": WebSite url: "https://www.pheonixresearch.com/" name: Pheonix Research api: "@id": MIWS-root#api "@type": WebAPI url: "https://www.pheonixresearch.com/" graph: node_id: MIMR-node_194ae89 graph_node_endpoint: "https://graph.pheonixresearch.com/api/v1/query/live/node/MIMR-node_194ae89" graph_snapshot: "https://graph.pheonixresearch.com/api/v1/read/live/graph" graph_exclusions: "https://graph.pheonixresearch.com/api/v1/read/live/exclusions" graph_meta: "https://graph.pheonixresearch.com/api/v1/read/live/meta" graph_bundle: "https://graph.pheonixresearch.com/api/v1/read/live/bundle" discovery: discovery_json: "https://www.pheonixresearch.com/.well-known/pheonix-discovery.json" llms_txt: "https://www.pheonixresearch.com/llms.txt" sitemap: "https://www.pheonixresearch.com/sitemap.xml" --- # Global Polypropylene and Polypropylene Composites Market Report, Size & Forecast 2025-2033: A USD 95 Billion Material Growing at 3.9% With 120% Capacity Cover in China - **Market Size (Base Year):** 95.0 - **Forecast Value:** 128.8 - **CAGR:** 3.88% - **Coverage:** Global ## Market Report ## What this market is, and where the boundary sits Polypropylene is the second-largest commodity polymer in the world and the one with the weakest structural defence. It is also the polymer whose international price is currently being set by one country's capacity programme. Both facts make the boundary question more important than usual: a number that mixes resin with converted parts will show a market several times larger and will hide the only mechanism that matters. This report covers two layers and states both. The first is **polypropylene resin** - homopolymer, random copolymer and impact copolymer - valued at realised producer price. The second is the **compounding premium** on the roughly 8 Mt of PP that is not sold as resin but as an engineered compound: glass-fibre reinforced, mineral-filled, impact-modified, flame-retardant. Converted parts are excluded. Film, mouldings and packaging are the destination of this market, not its perimeter, and folding them in would import converter labour and brand-owner margin into a materials forecast. The route to the total is one line long. Global propylene demand is roughly 130 Mt a year, of which about 67% is consumed making polypropylene. That is 87.1 Mt of propylene bound into PP. Net of process losses and off-specification material, PP resin output is about **86.0 Mt** in 2025. Weighted across regions at realised prices - Asia Pacific USD 925 a tonne, Europe USD 1,290, North America USD 1,000, the Middle East and Africa USD 940, Latin America USD 1,020, the Former Soviet Union USD 880 - that volume is worth **USD 84.4 billion**. The compound premium adds **USD 10.6 billion** on 8.0 Mt at an average USD 1,325 a tonne above base resin, which makes a compound cost 1.35 times the resin it is built from. The market in this report is therefore **USD 95.0 billion in 2025**, and each component of that sentence can be checked against a different source. Two cross-checks are worth stating because they were not reverse-engineered from each other. China's polypropylene capacity now stands at roughly 120% of its domestic consumption, and China is a little under 40 Mt of PP demand - about 46% of the 86 Mt global total. That is consistent with China's 46% share of world chemical sales, a figure reported independently. The 8 Mt of compounded PP is Pheonix's estimate, of which roughly 1.8 Mt is glass-fibre reinforced and the balance mineral-filled, impact-modified or otherwise specified. It is the least externally verifiable number on the page, and it is included because excluding it would leave the only part of the chain with pricing power outside the market. ## The number that matters is 120 per cent China's PP capacity is running at about **120% of domestic consumption** in 2026. That single ratio explains more about this market than any demand forecast. When a country's polymer capacity exceeds its own consumption by a fifth, the surplus does not stay home. The export numbers have moved quickly. China shipped **1.6 Mt of PP in January-April 2026** against 1.0 Mt in the same months of 2025. Annualised, that is a **4.8 Mt** export year against **3.1 Mt** for the whole of 2025 - and before the capacity surge of the last few years, Chinese PP exports were a rounding error in the monthly trade data. China is now exporting at a scale comparable with established exporters such as South Korea or Thailand, and the volume is landing in North and Southeast Asia, Africa and Latin America. The mechanism is upstream, and it is not a demand story. Chinese propylene supply is dominated by on-purpose production: **propane dehydrogenation capacity above 20 Mt a year** by 2025, built largely around imported US propane, plus coal-to-olefins and refinery-integrated capacity. PDH operating rates fell **below 75%** and margins were negative for much of 2024 and 2025 - and plants kept running, because an integrated PDH-PP complex that stops loses its share of the downstream market and its position in the provincial industrial plan. Propylene supply in China is price-insensitive. Polypropylene supply inherits that. The second consequence is that Chinese capacity additions no longer respond to price either. Additions continue into 2026 at several million tonnes, against a global market that would need capacity growth of roughly **1.4 Mt a year** to restore operating rates anywhere near their previous cycle levels - the current build is closer to **4 Mt a year**. That gap, not the demand line, is what the forecast is built on. ## Price is where the argument is, not volume Monthly contract assessments for 2026 tell the story in five lines. | Assessment (USD per tonne) | | February 2026 | | July 2026 | | Change | | US PP homopolymer, injection, contract | | 992 | | 1,047 | | +5.5% | | EU PP homopolymer, contract | | 1,292 | | 1,324 | | +2.5% | | China PP raffia, CFR | | 780 | | **744** | | **-4.6%** | | Southeast Asia PP raffia, CFR | | 845 | | 815 | | -3.6% | | Brazil PP homopolymer, CFR | | 889 | | 1,072 | | +20.6% | Read the first and third lines together. Between February and July 2026 the US contract rose 5.5% and the European contract 2.5%, while Chinese raffia *fell* 4.6%. On July assessments the European contract sat **USD 580 a tonne above** Chinese raffia - European PP costs **1.78 times** the Chinese benchmark before freight. That gap is not a tariff. It is the price of energy, of a smaller and older cracker fleet, and of a carbon cost that imports do not carry. This is what a surplus looks like when it is exported rather than absorbed. The marginal tonne in Asia is priced by Chinese export volume at USD 744 to 815, and that tonne sets a ceiling for every producer with a higher cost base. A European producer does not compete with another European producer on grade or service; it competes with a delivered container. The assessments show the consequence: European contract prices have been broadly flat for six months while Chinese prices fell, so the European premium is widening precisely as European capacity closes. Brazil is the exception that proves the rule. Its CFR price rose 20.6% in five months, on a closed domestic market, a live anti-dumping review and import volumes that are pressuring local utilisation. Price is set by the trade regime in Brazil and by the export machine in Asia, and almost nowhere by local demand growth. ## Forecast, 2025-2033 Every rate below was computed, not asserted. Value is modelled as volume times realised price, with the two layers modelled separately: resin volume grows faster than resin price, and the compound layer grows faster than both. | Scenario | | 2025 (USD bn) | | 2033 (USD bn) | | CAGR | | What has to be true | | **Baseline** | | 95.0 | | **128.8** | | **3.88%** | | Chinese additions continue at a slower pace; European closures complete; compound volume grows 5.2% a year | | Conservative | | 95.0 | | 120.3 | | 3.00% | | Surplus persists, price flat in real terms, European closures slip into the 2030s | | Optimistic | | 95.0 | | 137.0 | | 4.68% | | China's phase-out programme bites before 2029 and non-Chinese closures accelerate | Absolute growth across the window is **USD 33.8 billion**. Of that, resin contributes USD 27.6 billion and the compound layer USD 6.2 billion, so roughly a fifth of all value growth comes from 8% of the volume. **3.88% is a position.** Resin volume grows **3.24%** a year, from 86.0 Mt to 111.0 Mt. Realised price contributes almost nothing: USD 981 a tonne to USD 1,009, an eighth of the volume rate, because a market running at 120% capacity cover in its largest producing country does not regain pricing power on schedule. Anyone forecasting high single digits is forecasting either a supply response that has not been announced or a price recovery that has not begun. ## What is actually driving the number Three forces set the trajectory, and only one of them is visible in demand statistics. **China is still building.** Capacity at 120% of consumption, exports annualising at 4.8 Mt, and an April 2026 programme from the industry ministry and six other departments to upgrade some outdated petrochemical plants and phase others out by 2029, with rolling surveys of newly identified facilities from 2027. That is a real instrument. It is also a 2029 instrument applied to capacity already committed, which makes it a filter rather than a brake. **Europe is exiting, and the exits cascade.** The clearest case is LyondellBasell's **260,000 tonne a year** polypropylene unit at Brindisi, due to close by the end of 2026. It is the first Spheripol polypropylene unit ever built, started in 1982. It lost its propylene supply when Eni's Versalis cracker on the same site shut in April 2025; it had taken up to 220,000 tonnes a year from that cracker, and now hauls feedstock by sea from Priolo at a freight disadvantage it cannot recover. The unit was offered for sale at a symbolic price with no buyer willing to take it without a secured propylene source. Polypropylene capacity does not close because polypropylene demand fell. It closes because the cracker next door closed. **Compounding earns what polymerisation cannot.** The compound layer grows **5.20%** a year, from 8.0 Mt to 12.0 Mt, and its premium per tonne widens from USD 1,325 to USD 1,400. That premium does not move with naphtha or propane, which is the entire point of it: a compound is sold against a specification, and the specification is what the customer is paying for. ## Regional outlook Regional bases are built from the propylene-to-PP volume route, allocated by regional demand and priced off the regional assessments above. Compound premium is allocated by the region where the compound is made, not where the part is assembled. | Region | | Resin volume 2025 (Mt) | | 2025 (USD bn) | | 2033 (USD bn) | | CAGR | | **Asia Pacific** | | 47.3 | | 48.52 | | **68.20** | | **4.35%** | | Europe | | 9.5 | | 14.38 | | 16.96 | | 2.08% | | North America | | 9.4 | | 11.52 | | 14.52 | | 2.94% | | Middle East and Africa | | 8.2 | | 8.35 | | 12.96 | | 5.65% | | Latin America | | 7.4 | | 8.19 | | 11.61 | | 4.46% | | Former Soviet Union | | 4.2 | | 4.02 | | 4.55 | | 1.56% | | **Total** | | **86.0** | | **95.0** | | **128.8** | | **3.88%** | Asia Pacific grows 4.35% and supplies the majority of absolute growth because it is already 51% of the base. The Middle East and Africa posts the fastest rate at 5.65% from a base one-sixth the size - the only region where both demand growth and feedstock advantage point the same way. Europe's 2.08% is not a demand forecast; it is what happens when a region's volume grows 1.4% a year while its price premium stops widening. ## End-use outlook | End use | | 2025 (USD bn) | | 2033 (USD bn) | | CAGR | | What moves it | | Rigid packaging and containers | | 22.80 | | 29.50 | | 3.27% | | Recycled-content quotas raise cost per tonne and favour mono-material PP | | Flexible film and BOPP | | 18.05 | | 24.09 | | 3.67% | | Food packaging growth, partly displaced by paper and mono-PE laminates | | Automotive, including compounds | | 17.10 | | **24.60** | | **4.65%** | | Metal replacement, battery housings, thermal management | | Construction | | 8.55 | | 10.95 | | 3.14% | | Pipe, geotextile, insulation; tied to European and Chinese building volumes | | Electrical and electronics | | 6.65 | | 9.14 | | 4.06% | | Flame-retardant grades, appliance housings | | Household and appliances | | 8.55 | | 10.95 | | 3.14% | | Volume-linked, low cyclicality, heavy Chinese export exposure | | Healthcare and medical | | 4.75 | | 6.96 | | 4.89% | | Syringes, IV components, sterilisation-resistant grades | | Agriculture, fibre and other industrial | | 8.55 | | 12.61 | | 4.98% | | Woven bags, ropes, nonwovens, geosynthetics | | **Total** | | **95.0** | | **128.8** | | **3.88%** | | Rounding against the regional total | Packaging is 43% of the market and grows below the average; automotive is 18% and grows above it. That mix shift is the quiet structural story of the next eight years. It also means the end use that carries the most regulatory risk is the one that grows slowest, which is the opposite of how the industry usually tells the story. ## The demand side in detail **Packaging regulation is now a polymer-selection rule.** Regulation (EU) 2025/40 requires minimum recycled content in the plastic part of packaging from 1 January 2030: 30% for contact-sensitive packaging made from PET as the major component, **10% for contact-sensitive packaging made from plastics other than PET**, 30% for single-use plastic beverage bottles and **35% for all other plastic packaging**. PP sits overwhelmingly in the second and fourth of those categories, and the obligation dates from 2030 or three years from the implementing act, whichever is later. From 2040 the targets rise to 50% for contact-sensitive PET and 25% for other contact-sensitive plastics. Two effects follow and they push in opposite directions. Mono-material PP designs gain, because a pack that is all PP can meet a recycled-content quota and still be sorted, while a multi-material laminate cannot. And cost per tonne rises, because food-grade PP recyclate is the scarcest stream in the system. The first supports volume, the second suppresses value growth - which is one reason rigid packaging is modelled at 3.27% rather than the 5% that packaging volume growth alone would suggest. **Automotive is where PP earns its margin.** Global demand for light-vehicle composite materials reached a record 4.9 billion pounds, about 2.22 Mt, in 2024, according to the American Composites Manufacturers Association, and that growth came while vehicle production was flat. Polypropylene is the volume polymer in that mix. Battery electric vehicles take more composite per vehicle than combustion vehicles, so the mix shift helps even when unit volumes do not. **Substitution is real and it runs one way.** Fibre-based food-contact packaging, including the PFAS-free formats now being specified in Europe, takes share in foodservice and dry-food packs where PP thermoforms used to sit. In Latin America the mechanism is different: thermoformed PP is a growth market while import competition and anti-dumping duties decide who supplies it. ## Companion analysis This report sits inside the [global chemicals market](https://www.pheonixresearch.com/chemicals-materials/industrial-chemical/market-report/global-chemical-market-2025-2033/), which sizes the USD 5.55 trillion perimeter that polypropylene is 1.7% of. Its feedstock is sized in the [global propylene market](https://www.pheonixresearch.com/chemicals-materials/industrial-material/market-report/global-propylene-market/), and its olefin sibling in [ethylene](https://www.pheonixresearch.com/chemicals-materials/industrial-chemical/market-report/global-ethylene-market-2025-2033/), where the same surplus is measured in tonnes rather than in price. On the demand side, the [global material and packaging market](https://www.pheonixresearch.com/packaging/rigid-thermoformed-packaging/market-report/global-material-and-packaging-market-2025-2033/) carries the packaging aggregates this resin flows into, [thermoformed packaging in Peru](https://www.pheonixresearch.com/packaging/rigid-thermoformed-packaging/market-report/peru-thermoformed-packaging-market/) is a live example of PP competing with duties at its border, and [PFAS-free food-contact packaging in Europe](https://www.pheonixresearch.com/packaging/paper-fibre-based-packaging/market-report/europe-pfas-free-food-contact-packaging-market/) is the substitution threat in its most concrete form. ## What would change the forecast Two variables carry almost all the variance. The first is whether China's 2029 upgrade and phase-out programme is enforced plant by plant. If it is, the export machine slows from the supply side and the optimistic path opens. If it is not, 3.00% is not a stress case - it is the base case with the surplus clearing two years later. The second is the pace of non-Chinese closure. Borealis has already signalled that it expects further polyethylene and polypropylene closures in 2028 and 2029 on current overcapacity. Every tonne that leaves Europe or Japan tightens the Asian balance slightly and raises the floor. A closure programme that runs two years faster than announced is worth the difference between 3.88% and 4.68%. ## Table of Contents OverviewTwo layers, one perimeter, and the route to USD 95.0 billion The 120 per cent ratioCapacity cover in China and the mechanism that turns it into exports Price, not volume2026 contract assessments and a USD 580 gap between Europe and China Forecast, 2025-2033Baseline USD 128.8bn at 3.88%; conservative and optimistic paths What is actually driving the numberChina building, Europe exiting, compounding earning Regional outlookAsia Pacific 4.35%; the Middle East and Africa 5.65%; Europe 2.08% End-use outlookPackaging 43% growing below average; automotive 18% growing above it The demand side in detailRecycled-content quotas, automotive composites and the substitution threat Companion analysisChemicals, propylene, ethylene and the packaging cluster What would change the forecastChina’s 2029 enforcement and the pace of non-Chinese closure Competitive LandscapeNobody has share; capacity is bought as a licence and lost with the pipe next door Value ChainPropylene, polymerisation, compounding and the specification premium Investment ActivityBuying capacity instead of building it, and where the build still happens Technology InnovationCatalyst, grade slate and mono-material design Market RiskSurplus, trade defence, feedstock geopolitics and recyclate scarcity Regulatory LandscapeWhat PPWR actually requires, and why polymers are not in CBAM ## Competitive Landscape No company in this market has share, and the reason is structural rather than competitive. Polypropylene capacity is not invented by the companies that own it; it is bought as a process licence, built in two to three years, and then operated as a spread business. That produces an industry with dozens of credible producers, no pricing power at the top, and a competitive contest that is decided by the propylene cost position rather than by anything a marketing department does. The numbers make the point quickly. Borouge Group International – formed by combining Borouge with Borealis and acquiring Nova Chemicals, with closing expected at the end of March 2026 – will hold 13.6 Mt of nameplate polyolefin capacity across Europe, the Middle East and North America, making it the fourth-largest polyolefins producer in the world. That 13.6 Mt is polyethylene and polypropylene combined. Set against an 86 Mt PP resin market, the fourth-largest polyolefins company on earth holds a mid-single-digit fraction of this one product. The top of the table is genuinely flat. LyondellBasell’s 2025 results show what the bottom of a cycle does to that structure. The company reported a net loss of USD 738 million for the year and EBITDA of USD 1.1 billion, or USD 2.5 billion excluding identified items – the difference being non-cash asset write-downs, transaction costs, its Cash Improvement Plan and discontinued operations. It generated USD 2.3 billion of operating cash, reinvested USD 1.9 billion and returned USD 2.0 billion to shareholders, and it enters 2026 planning capital expenditure of USD 1.2 billion. Its stated first-quarter operating plan is the most informative line in the whole disclosure: Olefins and Polyolefins Americas at approximately 85%, Olefins and Polyolefins Europe, Asia and International at approximately 75%, Intermediates and Derivatives at approximately 85%. A complex running at 75% is not optimising its sales mix. It is holding a position and waiting. Archetype Examples Source of advantage 2026 posture Integrated feedstock owner Gulf and North American producers with ethane or associated gas; Chinese refinery-integrated majors Molecule cost at the wellhead or refinery gate Running for volume, taking share of a shrinking margin Process licensor and compounder LyondellBasell (Spheripol, Spherizone); Borealis (Borstar, now inside Borouge Group International) The grade slate and the catalyst, sold as a licence and a service Earning fees on other people’s capital while their own assets run light PDH-linked Chinese entrant Private integrated refiners and propane dehydrogenation complexes Provincial support, cheap capital, domestic propane logistics Running through negative propylene margins to hold downstream share Merchant converter-facing producer European and Japanese units without captive propylene Proximity, service, qualification on specific grades Exiting, selling, or converting to trade defence The licence is the real asset Polypropylene is one of the most licensed technologies in the chemical industry. A new entrant does not develop a polymerisation process; it buys a licence, signs a catalyst supply agreement, and receives a grade slate that has been qualified with converters over decades. That is why the technology holders have survived three downturns better than the producers who bought from them: the licensor’s revenue is a fee on somebody else’s capital expenditure, and it does not fall when the spread compresses. The grade slate is the commercial substance behind the licence. A process that can make a broad melt-flow range, control xylene solubles and deliver consistent impact copolymer is a process whose owner can supply automotive compounds. A process limited to commodity homopolymer is a process whose owner competes on delivered price against a Chinese container. Two plants with identical nameplate capacity are not in the same business. Brindisi is the whole industry in one site LyondellBasell will close its remaining 260,000 tonne a year polypropylene unit at Brindisi, Italy, by the end of 2026. The unit is historically significant: it was the first Spheripol polypropylene unit ever built, started in 1982, producing commodity homopolymer for packaging. It is closing because of the pipe next door, not because of the market. LyondellBasell does not operate a propylene cracker at Brindisi. It took up to 220,000 tonnes a year of propylene by sea from Eni’s Versalis cracker on the same site until that unit shut down in April 2025. Since then feedstock has been shipped in from the company’s other Italian sites, principally Priolo, and the freight has removed whatever cost position the unit had. Trade unions report the unit was offered for sale at a symbolic price with no buyer, because no buyer can secure propylene without the cracker. Eni has hired an investment bank to sell the idled cracker; if that sale succeeds, a buyer for the polypropylene unit becomes possible again. Two assets, one pipeline, and the entire commercial fate of both decided by the same question. That is the mechanism by which European polypropylene capacity is disappearing. Announced closures in this industry are rarely decisions about polypropylene. They are decisions about the cracker, transmitted one stage downstream. Consolidation and litigation are the two strategies left Companies that cannot win on cost have two options in 2026, and both are visible at scale. The first is to buy capacity rather than build it. Borouge Group International combines a Gulf feedstock position, a European technology and compounding franchise, and a North American asset base in a single transaction, and does so with investment-grade ratings rather than with a construction programme. LyondellBasell went the other way and sold several European resin and feedstocks plants to a newly formed regional company, Velogy, completing on 1 May 2026, while retaining the market through a narrower network. The second option is to litigate the import. Brazil’s anti-dumping measure on polypropylene resin from South Africa and India is under review, initiated at the petition of Braskem on 24 December 2025, with the final determination opinion scheduled for 28 September 2026. Braskem is not defending a technology position or a grade franchise; it is defending a domestic price that rose 20.6% in five months while Asian prices fell. Trade defence is what competitive strategy looks like when the cost curve has already decided the outcome. What competition is actually about Three variables decide who earns a return in this market over the next eight years, and none of them is market share. The first is the delivered cost of propylene. The second is whether the producer’s grade slate reaches compounding and specification work, where the premium is USD 1,325 a tonne over base resin. The third is whether the producer’s home market gives it a trade regime it can survive in. Companies with none of the three are doing one of four things: closing the asset, selling it, merging it into someone else’s balance sheet, or asking a government to tax the alternative. Companies with all three are adding capacity in the Middle East and Asia and letting the surplus clear where it lands. - **Structure:** moderately_consolidated - **Tier 1 Players:** 8 - **Intensity:** high ## Value Chain The polypropylene value chain is usually drawn as four boxes: propylene, polymerisation, compounding, conversion. The useful version of that diagram makes a narrower point. Two of those boxes earn a return and two do not, and which two depends almost entirely on whether the business is selling a molecule or a specification. Stage one: propylene, and why its supply stopped responding to price Global propylene demand is roughly 130 Mt a year, of which about 67% becomes polypropylene. Three production routes supply it. Steam crackers make propylene as a co-product of ethylene, alongside butadiene and aromatics. Refineries make it in fluid catalytic crackers as a side effect of gasoline production. On-purpose plants make nothing else: propane dehydrogenation strips hydrogen from propane and produces propylene directly. The mix between those routes changed the industry’s economics permanently. As US and Middle Eastern crackers shifted to ethane – which yields almost no propylene – co-product supply stopped growing while polypropylene demand kept compounding. The answer was on-purpose capacity, and China industrialised it at extraordinary scale: propane dehydrogenation capacity above 20 Mt a year by 2025, designed largely around imported US propane. That buildout overshot. Chinese PDH operating rates fell below 75% and margins were negative for much of 2024 and 2025, yet plants kept running, to hold downstream market share and to feed integrated polypropylene lines. This is the single most important structural fact in the chain: a large block of the world’s propylene supply is now produced by plants that do not shut down when the spread is negative. Every stage downstream inherits that insensitivity. Stage two: polymerisation, a spread business with a fixed cost base A polypropylene plant converts propylene into resin through a catalysed polymerisation in loop, gas-phase or multi-zone reactors. The conversion cost – energy, catalyst, labour, maintenance, depreciation – is largely fixed per tonne. The plant therefore earns the propylene-to-polypropylene spread multiplied by volume, and nothing else. There is no brand, no service contract and no switching cost at this stage for commodity homopolymer. The grade slate is where the stage begins to differentiate. Homopolymer is the commodity product. Random copolymer adds clarity and flexibility for packaging. Impact copolymer adds low-temperature toughness for automotive and appliances. Each step up in specification narrows the number of plants that can make the grade reliably, and reliability – not capability – is what converters pay for, because a moulding line that stops is more expensive than a tonne of resin. Under surplus, the spread is set by the marginal plant, and in 2026 the marginal plant is a Chinese PDH-integrated complex that runs regardless of margin. That is why price is a shock absorber rather than a signal in this chain: it can fall a long way before it changes behaviour at the point of production. Stage three: compounding, where the specification is sold Roughly 8 Mt of the world’s PP is not sold as resin. It is compounded: blended with talc or glass fibre for stiffness, with elastomers for impact, with flame retardants, stabilisers and pigments, and then sold against a customer’s material specification. On this report’s numbers that layer carries an average premium of USD 1,325 a tonne over base resin, which at a base resin price of USD 981 is 1.35 times the price of the polymer it is built from. The premium is not a mixing fee. It is the price of qualification: a compounder has to be approved into an automotive OEM’s material system, guarantee properties across a production run measured in years, absorb formulation changes when a filler or additive is restricted, and hold a supply commitment through model lifecycles. That is a service business wearing a chemical company’s clothes, and it is the only stage in this chain where a supplier can raise price while the resin index falls. It is also the stage with the most regulatory exposure, because the additives that make a compound work – flame retardants, processing aids, certain stabilisers – are precisely the substances that restriction programmes target. A compounder’s portfolio is therefore a regulatory question as much as a technical one. Stage four: conversion, and who absorbs the timing Converters buy resin on index-linked or quarterly contract terms and sell film, mouldings and thermoforms to brand owners on annual agreements. That mismatch means converters absorb the timing difference between a resin movement and a shelf price, which is why their margins are inversely cyclical to the producers’ and why packaging price changes reach consumers with a lag of two to three quarters. This stage is where the industry’s volume lives and where almost none of its margin does. A BOPP film line, an injection moulder making crates and a thermoformer making dairy tubs are all buying the same tonne and competing on conversion cost, utilisation and freight. Their returns depend on running full, and their pricing power is limited to the narrow window between a resin fall and a customer renegotiation. The new fifth stage: the recycled-content loop Regulation has added a stage that did not exist a decade ago. Collection, sorting, mechanical recycling, and re-compounding now sit alongside the virgin chain, because European packaging law requires minimum recycled content in the plastic part of packaging from 2030 – 35% for plastic packaging outside the contact-sensitive categories, 10% for contact-sensitive packaging made from plastics other than PET, and 30% for contact-sensitive PET. The bottleneck in that loop is not collection and not sorting technology. It is food-grade polypropylene recyclate: the stream that can legally go back into a pack that touches food. It is scarce, it trades at a premium to virgin resin in exactly the years when virgin resin is cheapest, and it is the reason mono-material PP packaging design is spreading faster than the recycling capacity that is supposed to supply it. Where the value actually accrues The pattern is consistent across the chain, and it is not the pattern the industry’s strategy documents describe. Value accrues at two points: where propylene is cheapest, and where the specification is hardest to replace. It does not accrue in the middle, where a commodity pellet is a commodity pellet and the buyer’s only question is the delivered price of the next 5,000 tonnes. That is why the compounding layer, at roughly 8% of the volume, supplies around a fifth of the value growth in this forecast. It is also why the industry’s most repeated strategic phrase – move downstream – is directionally correct and almost always under-costed, because moving downstream means acquiring qualification, not acquiring a converting line. - **Model:** vertically_integrated - **Distribution:** distributor_led - **Supply Complexity:** high ## Investment Activity The most revealing number in this industry in 2026 is not a capacity figure. It is USD 1.2 billion – LyondellBasell’s planned capital expenditure for the year, against USD 1.9 billion actually reinvested in 2025 and a company that generated USD 2.3 billion of operating cash and returned USD 2.0 billion to shareholders. That is the arithmetic of an industry funding distributions and maintenance while deferring growth. Read alongside a 13.6 Mt consolidation in the Gulf and continuing additions in China, it tells you where capital has decided the future of polypropylene is. The largest deal of the cycle buys capacity instead of building it Borouge Group International is the formation of a polyolefins major out of three existing ones: Borouge, Borealis, and the acquired Nova Chemicals, closing expected at the end of March 2026, with 13.6 Mt of nameplate capacity across Europe, the Middle East and North America and a position as the world’s fourth-largest polyolefins producer. The ratings attached to it – investment grade across all three agencies – are part of the strategy rather than a by-product: consolidation paid for with a strong balance sheet is cheaper than greenfield construction in a market with 120% capacity cover in its largest consuming country. The Borouge 4 complex shows how the capital is being structured. B4 is a 1.5 Mt ethane cracker with 1.4 Mt of polyethylene capacity, 70% owned by ADNOC and 30% by OMV, with the first plant starting up in the first quarter of 2026 and a progressive ramp through the year. It is integrated into what will become the world’s largest single-site polyolefins complex. But the listed entity will not own it: under an asset usage agreement signed in March 2026, Borouge – and subsequently Borouge Group International – operates and markets the volumes for an at-cost utilisation fee, delivering an estimated USD 400 million of cumulative net profit over three years, or about 10% annual accretion post ramp-up, with acquisition of the asset not expected before 2029. That structure is a template for the decade. The company takes the commercial position and the earnings without the capital charge, and the sovereign owner keeps the asset. If that becomes the standard way polyolefin capacity changes hands, then the industry’s investment statistics will increasingly measure construction in the Gulf and earnings in Rotterdam, and the two will no longer reconcile in a single company’s balance sheet. Europe: divestment is the investment strategy European capital allocation is best described as subtraction. LyondellBasell completed the sale of several European resin and feedstocks plants to a newly formed regional petrochemical and polymer company, Velogy, on 1 May 2026, and is closing the 260,000 tonne a year Brindisi polypropylene unit by the end of 2026 because it lost its propylene supply when the neighbouring Versalis cracker shut in April 2025. Neither decision is a bet on polypropylene demand. Both are decisions about which assets can be supplied at a competitive cost. What survives in Europe is not commodity polypropylene. It is compounding, application development and specification work, which is where the USD 1,325 a tonne premium sits and where a European producer can charge for qualification rather than for a molecule. That is a coherent strategy and it is a small one: the compound layer is roughly 8% of the volume in this market. China: capital already committed, policy applied afterwards China is the only region still adding polypropylene capacity at scale, and the additions are no longer a response to price. Capacity stands at about 120% of domestic consumption in 2026, exports are annualising near 4.8 Mt, and the April 2026 programme from the industry ministry and six other departments to upgrade outdated petrochemical plants and phase others out by 2029 – with rolling surveys of newly identified facilities from 2027 – is a filter applied to capacity that is already financed and under construction. Two features of Chinese investment matter to anyone forecasting this market. The first is that the feedstock is largely on-purpose: propane dehydrogenation above 20 Mt a year, coal-to-olefins and refinery integration, none of which is priced off a propylene market that clears. The second is that the investment decision is made by entities whose cost of capital and whose objective function are not those of a listed Western chemical company. A project that would be cancelled at a 75% operating rate elsewhere is completed here, because the operating rate is not the objective. Where new investment is actually justified Three positions can still underwrite a polypropylene investment in 2026, and they are narrow. Feedstock advantage in the Gulf and North America, where ethane or associated gas is available at a cost no other region can match – this is what B4 is. Demand growth with a protected border, principally India, where domestic projects have lagged consumption and the country has become a major import growth centre for polypropylene; investment there is justified by demand rather than by cost, which is the only kind of investment that survives a surplus. And qualification, where a compounder adding a line beside an automotive plant is not competing with a Chinese container at all. Everything else is a bet that the surplus clears before the depreciation schedule does. On this report’s numbers it does not: the conservative path assumes flat real prices to 2033, and it is only 0.88 percentage points below the baseline. The circularity capex, and what it is really for LyondellBasell’s MoReTec-1 unit, under construction through 2026, is the clearest example of the one category of Western polyolefin investment that is growing: chemical recycling. Its justification is not that pyrolysis oil is cheaper than naphtha – it is not – but that European packaging regulation creates a legally mandated market for recycled content from 2030, at 35% for plastic packaging outside the contact-sensitive categories and 10% for contact-sensitive packaging made from plastics other than PET. A legislated offtake is the only demand signal in this industry that a producer can underwrite. That is why the recycling capital is being committed while cracker capital is not, and why the same companies are simultaneously closing polymerisation capacity and building depolymerisation capacity at the same sites. The assets are related; the markets are not. What the capital cycle is telling the forecast Investment in this industry is no longer allocated to markets. It is allocated to cost positions and to regulated offtake. That is the assumption embedded in the 3.88% baseline: no new wave of Western capacity, continuing Chinese additions at a decelerating pace, and a compound layer that attracts the only genuinely profitable capital in the chain. The risk in that assumption is not that it is wrong about today. It is that capital cycles turn faster than forecasts. A company that defers USD 700 million of growth capex in 2026 because the spread is negative is also a company that will have nothing to start up in 2029 if the surplus clears in 2028. - **Trend:** stable - **Capital Intensity:** high - **Recent M&A:** yes ## Technology & Innovation Polypropylene has no shortage of technology narratives. It has a shortage of technologies that change the delivered cost of a tonne or the price a customer will pay for one. The distinction matters here more than in most markets, because the polymerisation technology is now mature and sold as a licence, while the compounding and design work at the end of the chain is where the margin actually moves. Three frontiers are real, in ascending order of how much they matter to the 2033 number: catalyst and process capability, product and composite design, and circularity. Only the second earns a premium today. Frontier one: catalyst and process, now table stakes Every polypropylene plant in the world runs on a catalyst system and a reactor configuration, and the pair determines what the plant can make. The generations are well understood: earlier Ziegler-Natta systems delivered a broad but coarse product slate; later generations and metallocene systems gave producers far finer control over molecular weight distribution, melt flow and the balance between stiffness and impact strength. Multizone reactor designs – LyondellBasell’s Spherizone, and Borealis’s Borstar process now inside Borouge Group International – allow bimodal and multimodal grades to be made in a single train, which is how producers reach the grade slate that automotive and appliance compounds require. The commercial consequence is that a polypropylene plant is not a fungible asset. Two plants with the same nameplate tonnage can have completely different access to the part of the market that pays. A commodity homopolymer line competes with a delivered container from Ningbo; a line that can make consistent high-impact copolymer for interior trim does not. What has changed is that this capability is no longer a differentiator at the top. It is a licence that any well-financed entrant can buy, which is precisely how Chinese capacity was built so quickly. Process technology has been democratised, and a democratised technology stops earning a return. That is one reason the industry’s research spend has shifted downstream. Frontier two: compounding and design, where the premium is The technology that moves this forecast is not in the reactor. Roughly 8 Mt of the world’s PP is compounded – filled with talc for stiffness, reinforced with glass fibre for strength, modified with elastomers for low-temperature impact, stabilised and flame-retarded – and that layer carries an average premium of USD 1,325 a tonne, or 1.35 times the base resin price. Three product families are doing the work. Long-glass-fibre and glass-mat thermoplastic PP replaces metal and engineering plastics in structural parts: front-end modules, seat structures, battery housings and underbody shields. Mineral-filled and impact-modified PP does the higher-volume work in interior trim, instrument panels and HVAC ducting, where cost per part governs. High-melt-strength PP makes extrusion, thermoforming and foaming viable, which is how PP enters applications that used to belong to other polymers. The demand evidence is in the composites data rather than in vendor claims. Global demand for light-vehicle composite materials reached a record 4.9 billion pounds, about 2.22 Mt, in 2024, according to the American Composites Manufacturers Association – a record set in a year when global vehicle production was flat. Usage per vehicle is rising, and battery electric vehicles take more composite per vehicle than combustion vehicles do, because weight saved buys range. Polypropylene is the volume polymer in that mix. Packaging design is the second half of this frontier and it is less glamorous and more consequential. European law requires minimum recycled content in the plastic part of packaging from 2030: 35% for plastic packaging outside the contact-sensitive categories, 10% for contact-sensitive packaging made from plastics other than PET, 30% for contact-sensitive PET and 30% for single-use plastic beverage bottles. A multi-material laminate cannot meet a recycled-content quota and cannot easily be sorted. An all-PP pack can. The technology that resolves this is not a new polymer; it is seal-layer, barrier and stiffness design that lets a pack be built from one material family – and it is the reason mono-material PP formats are spreading faster than the recycling capacity meant to supply them. Frontier three: circularity, which regulation pays for Chemical recycling – pyrolysis of mixed plastic waste back to an oil that can be cracked, and depolymerisation routes back to monomer – is the most capital-intensive frontier and the smallest in tonnage. Its economics do not work on a virgin-resin comparison; pyrolysis oil is not cheaper than naphtha. It works because European packaging law creates a legally mandated demand for recycled content, and because brand owners will pay for a mass-balanced recycled claim on a pack. Two constraints limit how fast this scales. The first is scale economics: a pyrolysis plant needs a very large secured waste stream at a single site to make sense, which excludes most local and specialty streams. The second is what the industry calls mass balance. A plant producing a few thousand tonnes of pyrolysis oil feeds it into a cracker handling millions of tonnes, and the recycled content has to be allocated across output by an accounting convention. The convention can be defensible; the marketing claim built on it frequently is not. Mechanical recycling remains far larger and cheaper for the streams it can handle. The binding constraint there is not technology but food-grade polypropylene recyclate – the stream legally allowed back into a pack that touches food. It is scarce, it trades at a premium to virgin resin in exactly the years virgin resin is cheapest, and it is the real bottleneck in the 2030 compliance story. What digitalisation is worth, honestly Advanced process control, inferential sensing and predictive maintenance are genuinely deployed in this industry, and they pay back quickly because they need no new molecule. But in a market with structural surplus, the return on a process improvement accrues to the customer rather than the producer: a yield gain raises industry output and lowers industry price, because price is set by the marginal producer’s cash cost. Process innovation in a trough is therefore defensive. It keeps a plant running; it does not raise a margin. What this means for who earns the margin The technologies that will matter to 2033 are not the ones with the best press release. They are the ones that let a supplier charge for a specification in a region that cannot compete on feedstock: compound formulations qualified into vehicle programmes, mono-material packaging designs that satisfy a recycled-content quota, and pyrolysis capacity with a secured waste stream. Polymerisation innovation is now a licence fee. Compounding and design innovation is a margin. That asymmetry is the whole technology story of this market, and it is why the compound layer supplies around a fifth of the value growth in this forecast from roughly 8% of the volume. - **Innovation:** low - **Patent Activity:** low - **Maturity:** mature ## Market Risk The risks in this market are unusually easy to rank, because one of them dominates and the others are mostly mechanisms for it to get worse. Structural surplus is the risk. Trade defence, feedstock geopolitics, the European cost position, recycled-content scarcity and substitution are the channels through which the surplus reaches price. Risk one: capacity cover above 100% in the largest market China’s polypropylene capacity is running at about 120% of its own consumption in 2026, and the country accounts for a little under 40 Mt of the 86 Mt global market. When the largest producer by a wide margin can cover its domestic demand with a fifth to spare, the surplus is exported and the export price becomes the world price. The arithmetic of the imbalance is straightforward. Restoring operating rates to anything like their previous cycle levels would require global capacity growth of roughly 1.4 Mt a year through the mid-2030s; the current build is closer to 4 Mt a year. That gap is not a forecasting assumption – it is what the announced project schedule implies, and it is why this report’s price path is essentially flat while volume grows 3.24% a year. The transmission is through price, and it is already visible: Chinese raffia fell 4.6% between February and July 2026 while the US contract rose 5.5% and the European contract 2.5%. Price is the shock absorber, and in a market where a large block of supply does not respond to negative margins, the absorber compresses for everyone with a higher cost base. Risk two: trade defence fragments the market without shrinking it As Chinese volume lands in new destinations, the destinations respond with duties – and each duty relocates the tonne rather than removing it. Brazil’s anti-dumping measure on polypropylene resin from South Africa and India is under review at Braskem’s petition, initiated on 24 December 2025, with the evidence phase closing on 30 June 2026 and the final determination opinion scheduled for 28 September 2026. Brazil shows the mechanism cleanly: its CFR price rose 20.6% between February and July 2026 while Asian prices fell. That is what a protected border does to a buyer, and it is also why the world does not have one price. The consequence for a supplier is that market access becomes a licensing question rather than a cost question – a company’s realisable price depends on which duty regime its cargo can enter. The strategic risk is second-order and worse. Each new measure teaches exporters to route around it, which means cargoes move to the next open market and depress the price there. Trade defence protects a domestic price; it does not restore industry utilisation, and it makes the geographic map of price more volatile than the underlying surplus. Risk three: on-purpose propylene is a single-basin bet Chinese propylene supply now depends heavily on propane dehydrogenation – over 20 Mt a year of capacity – designed largely around imported US propane. That is a long supply chain with concentrated points of failure: export terminals in one basin, a shipping route, and a feedstock whose price moves with US heating demand and with LNG economics rather than with Chinese polymer demand. The exposure runs both ways. A disruption or a price spike at the propane end raises the cost of the marginal polypropylene producer, which supports price but compresses the spread on integrated Chinese complexes already running at sub-75% rates and negative margins. Conversely, cheap propane keeps the export machine running regardless of polymer economics, because the feedstock is bought on long-term contracts and the alternative to producing is not producing at all. This is the risk that most often surprises a forecast, because it can move in either direction within a quarter and it is not visible in the demand data. Risk four: Europe’s closure cascade is not finished Europe’s polypropylene capacity does not close because polypropylene became uneconomic in isolation. It closes because the cracker next door closed. LyondellBasell’s 260,000 tonne a year Brindisi unit is the worked example: it lost up to 220,000 tonnes a year of propylene when Eni’s Versalis cracker on the same site shut in April 2025, and its substitute supply now comes by sea from Priolo at a freight cost the unit cannot absorb. The forward risk is that the announced programme understates the real one. When a cracker leaves an integrated cluster, the derivative units around it lose their pipeline and start buying merchant molecule at merchant prices plus freight. Some of the next wave of polypropylene closures will therefore arrive with no announcement, as a quiet operating-rate decision rather than a press release. Borealis has already signalled that it expects further polyethylene and polypropylene closures in 2028 and 2029 on current overcapacity. Risk five: recycled content is a demand rule, and the supply does not exist yet From 1 January 2030 – or three years from the implementing act, whichever is later – European packaging law requires minimum recycled content in the plastic part of packaging: 35% for plastic packaging outside the contact-sensitive categories, 10% for contact-sensitive packaging made from plastics other than PET, 30% for contact-sensitive PET, and 30% for single-use plastic beverage bottles. Polypropylene packaging sits mostly in the 35% and 10% categories. The risk is not that the rule will not apply. It is that the recycled polypropylene capable of meeting it – food-grade material authorised for contact use – is the scarcest stream in the system, and it trades at a premium to virgin resin in precisely the years when virgin resin is cheapest. A converter facing a 35% obligation and a thin recyclate market has three options: pay the premium, redesign the pack to a material with better recyclate availability, or move the format to fibre. All three are demand risks for polypropylene, and two of them are permanent. Risk six: concentration in packaging and automotive Packaging is about 43% of this market’s value and automotive about 18%. Between them they carry 61% of the demand, and they are exposed to the two most different cycles in the economy: consumer staples volumes, which are stable and slow, and vehicle production, which is cyclical, regional and being reshaped by powertrain mix. Fibre-based food-contact packaging, including PFAS-free formats now being specified in Europe, is taking share in foodservice and dry-food applications that PP thermoforms used to serve. Mono-PE and mono-PET structures take share in flexible packaging. None of these substitutions is large in any one year, and all of them are one-directional. A market whose largest segment grows at 3.27% rather than at packaging volume growth is a market losing share inside its own biggest application. What the risk set is worth in the forecast The conservative path – USD 120.3 billion by 2033 at 3.00% – is not a stress case. It is the baseline with the surplus clearing two years later than assumed, which is what happens if China’s 2029 phase-out list goes unenforced and non-Chinese closures stall. That single deferral is worth USD 8.5 billion of 2033 value. The optimistic path is worth USD 8.2 billion above the baseline and requires the opposite: enforcement in China plus faster exit elsewhere. The band is therefore narrow in percentage terms and material in absolute terms, and it is decided by supply behaviour rather than by demand. That is unusual for a consumer-facing material and it is the single thing a reader of this report should carry away. - **Substitution Risk:** high ## Regulatory Landscape The regulation that actually reprices polypropylene in the 2026-2033 window is packaging law, not carbon law. That is the opposite of what most material written about European chemicals assumes, and getting the order right changes what a producer should spend money on. Carbon border adjustment does not cover polymers The Carbon Border Adjustment Mechanism’s covered sectors at entry into force are cement, iron and steel, aluminium, fertilisers, electricity and hydrogen. Organic chemicals and polymers – polypropylene included – are outside the mechanism. A European polypropylene producer therefore carries the emissions trading cost of making the resin while an importer of the same resin carries nothing. That asymmetry is a meaningful part of why European contract polypropylene trades USD 580 a tonne above Chinese raffia, and it is the industry’s strongest policy complaint. It is also the reason any extension of CBAM to organic chemicals is watched more closely by producers than by their customers: extension would remove the distortion and would raise the cost of every European manufacturer downstream of imported chemistry. The practical consequence for a reader is that a polypropylene strategy built on a carbon border adjustment assumption is built on a regulation that does not yet exist for this product. The regulation that does exist is about what packaging is made of. Recycled content is now a design mandate Regulation (EU) 2025/40 sets minimum recycled content in the plastic part of packaging, calculated as an average per manufacturing plant and year, per packaging type and format. From 1 January 2030 – or three years from the date of entry into force of the relevant implementing act, whichever is later – the thresholds are: Packaging category From 2030 From 2040 Contact-sensitive packaging, PET as the major component (excluding single-use beverage bottles) 30% 50% Contact-sensitive packaging, plastics other than PET (excluding single-use beverage bottles) 10% 25% Single-use plastic beverage bottles 30% – All other plastic packaging 35% – Polypropylene sits overwhelmingly in the second and fourth rows, which is why the obligation is more demanding for this polymer than the headline 30% suggests. The regulation also excludes categories where recycled content would compromise safety – immediate packaging of medicinal products, contact-sensitive packaging for medical devices and in vitro diagnostics, and contact-sensitive food packaging for infants, young children and special medical purposes – and that exclusion list is a reminder of how tightly the rule is bound to food-contact law. The regulation also amends the Single-Use Plastics Directive, which is the mechanism by which packaging formats have been progressively restricted by material and application rather than by product. For a polymer whose largest single end use is packaging, a mandate on the composition of packaging is a demand instrument of the first order, and it operates in two directions at once: it favours mono-material polypropylene designs that can be sorted and recycled, and it raises the cost of any pack that cannot demonstrate compliant recycled content. What the recycled-content rule does to the polypropylene chain The binding constraint is not collection and not sorting. It is food-grade polypropylene recyclate – material authorised to go back into a pack that touches food. That stream is thin, it trades at a premium to virgin resin in exactly the years virgin resin is cheapest, and its scarcity is the reason the 35% threshold is more demanding in practice than it looks on the page. This is also where the chemical recycling investment comes from. Pyrolysis and depolymerisation produce a recycled feedstock that can be allocated to packaging output under a mass-balance convention, and the resulting claim is what brand owners are buying. The regulation does not mandate chemical recycling; it mandates an outcome, and chemical recycling is one of the few routes that can deliver that outcome at food-contact quality. Additive restriction is a compound-portfolio problem European chemicals regulation bites polypropylene not in the polymer but in the additives that make a grade usable. Flame retardants, processing aids, certain stabilisers and pigments are the substance groups most exposed to restriction programmes, and the direction of travel for more than five years has been toward broad, group-based restrictions rather than substance-by-substance decisions. For a compounder this is a portfolio question rather than a compliance cost, because substitution work has to be done years before a restriction applies and re-qualification inside an automotive programme is slow. The practical effect in the forecast is to raise the value of an already-qualified compound and to raise the barrier to entry – both of which support the USD 1,325 a tonne premium rather than threatening it. China: the only supply-side regulation that matters The most consequential polypropylene regulation in the world in this decade is being written in Beijing, and it is industrial rather than environmental. In April 2026 the industry ministry and six other departments set out a programme to upgrade some outdated petrochemical plants and phase others out by 2029, with lists compiled by local governments and annual rolling surveys of newly identified facilities from 2027. Two features of that programme determine whether it changes the forecast. The first is that it operates on plants that already exist, which makes it a quality filter rather than a brake on committed capacity. The second is that enforcement is delegated to provincial authorities whose incentives have historically favoured running the asset. A reader should treat the 2029 date as a policy intention with a real chance of slipping, and the conservative scenario in this report assumes exactly that. Trade remedies as a regulatory variable Brazil’s anti-dumping measure on polypropylene resin is under review, initiated at Braskem’s petition on 24 December 2025, with the evidence phase closing on 30 June 2026, disclosure of the essential facts scheduled for 19 August 2026 and the final determination opinion for 28 September 2026. The measure covers resin from South Africa and India under two customs codes, homopolymer and copolymer. That timetable matters beyond Brazil because it is the pattern: as Chinese volume is redirected, destination markets open reviews, and each review changes the delivered price available to a specific exporter in a specific market. For a producer, trade remedies are now a regulatory exposure with a calendar, not a background risk. What the regulatory net effect is worth The regulatory effect on this forecast is a mix effect, not a volume effect. Nothing in the current rulebook reduces the tonnage of polypropylene the world will use by 2033; packaging regulation changes what the tonne has to contain and which formats survive, and carbon regulation changes where the tonne is made. That is why Europe’s growth rate is 2.08% rather than negative, and why the 2030 recycled-content deadline is modelled as a cost and a design constraint rather than as a demand shock. The one regulation that could change the volume line is not European. It is whether Beijing’s 2029 phase-out is enforced. That is the single policy variable that moves this market’s supply side, and it is being decided by a government that has every incentive to protect downstream polymer users from the price consequences of enforcing it. - **Complexity:** high - **Approval Pathway:** standards_and_mandate_based ## Frequently Asked Questions **Q: How large is the polypropylene and polypropylene composites market in 2025?** USD 95.0 billion at first sale in 2025, on 86.0 million tonnes of polypropylene resin, reaching USD 128.8 billion by 2033 at 3.88 per cent. Volume grows 3.24 per cent a year; realised price contributes almost nothing. The 8.0 million tonne compound layer supplies USD 10.6 billion of that total from 9.3 per cent of the volume. **Q: What is happening to Chinese polypropylene capacity?** China's PP capacity now covers about 120 per cent of its own consumption. Exports ran 1.6 million tonnes in January to April 2026 against 1.0 million tonnes a year earlier, a 4.8 million tonne annualised pace against 3.1 million tonnes for all of 2025. Global capacity would need to grow roughly 1.4 million tonnes a year to restore operating rates; the current build is closer to 4 million. **Q: Why does European polypropylene trade so far above Chinese material?** On 2026 monthly contract assessments the EU contract sits USD 580 a tonne above Chinese raffia, which is 1.78 times, before freight. European crackers have closed while Chinese capacity kept building. Brindisi is the clearest case: LyondellBasell closes 260,000 tonnes a year of PP at the end of 2026 because Eni's Versalis cracker next door shut in April 2025 and feedstock now comes from Priolo. **Q: Who are the leading polypropylene producers?** Read the list by layer. Cracker- and refinery-integrated producers supply the resin: LyondellBasell, ExxonMobil, SABIC, Borealis, Braskem, Reliance, Formosa, Sinopec, PetroChina and INEOS. The compound layer, glass-fibre and mineral-filled grades, is a separate and far more fragmented business. No single firm holds a double-digit global share of resin.