Chemical Industry Radar / September 2026 / V3.1
Capital moves before the cycle turns
German investment gains remain selective. The wider operating recovery still has to earn its capital.
33 source records · 46 candidates · Ten evidence figures · Observation dates and company plans kept explicit.
- 33 sources
- 46 candidates
- 26 claims
- 10 figures
Test the claim
What does September actually establish?
Claim
“The recovery is self-sustaining.”
Stronger order values and a Q2 production rebound support the case. Weak July volumes and higher September energy costs leave the conclusion unproven.
Counter-claim
“Germany no longer attracts investment.”
Commissioned capacity in Ludwigshafen and new battery research in Krefeld contradict a blanket claim. They do not yet establish broad industrial renewal.
The September verdict
Selective capital progress; weak operating economics.
- Supports
- German capability is still attracting investment in specific assets, research and products.
- Counts against
- The evidence does not show a broad self-sustaining volume recovery or a general restoration of competitive delivered costs.
- Test
- Watch physical output, source-consistent utilization, delivered energy costs and executed German growth investment.
The German chemical industry ends September with more evidence of selective investment and less evidence of a broad operating recovery. The mood index slips to 3.3 out of 10, from 3.4 in August. BASF has commissioned a larger plant in Ludwigshafen. LANXESS has opened a battery laboratory. Yet physical production remains weak, wholesale energy has become more expensive, and the listed-company basket has lost ground. These developments belong in the same account. An industry can improve individual assets while its overall economics deteriorate.
September also brought the possibility of a BASF–Evonik transaction. The industrial question is what any eventual combination would change about products, capital discipline, operating costs and the location of investment. At the cutoff, the public record established a non-binding approach, with different descriptions of the state of discussions. It did not establish a transaction, a price, realized synergies or a new German investment program. S13 S14 S15
This edition therefore tests two propositions: that the recovery has become self-sustaining, and that Germany has ceased to attract chemical investment. Neither survives intact. The more useful conclusion is narrower: capital is being allocated selectively within a production system that still struggles to earn its costs.
Reporting window: September 1–30, 2026, Europe/Berlin. Evidence cutoff: September 30, 23:59:59. September publication dates often describe July activity, Q2 accounts or Q3 surveys. The equity basket is provisional through September 29, its last common valid close. Later releases are excluded from the economic assessment.
Act I The state of the industry
01 A small index decline with a clear composition
The September score is 3.3. Its unrounded weighted total is 3.275, compared with August’s 3.425. The exact arithmetic decline is 0.150 points; the displayed monthly change is −0.1. Those extra decimal places explain the accounting. They do not imply that an editorial assessment measures industrial conditions to three decimal places.
The same ten components and weights remain in force. Production and utilization, orders and demand, energy and feedstocks, and financial expectations deteriorate. Investment and innovation improve. Profitability, employment and productivity, trade competitiveness, and regulation remain unchanged. Holding a score is an explicit judgment about the available evidence; it does not certify that the underlying economy stood still.
Exhibit 01S32
The recovery loses a little ground
Mood index, 0–10 · October 2025–September 2026
- September 3.3 / 10
- August 3.4 / 10
- All eleven earlier published scores unchanged.
Inspect exact values and definitions
| Edition | Mood index | Method |
|---|---|---|
| October 2025 | 3.0 | legacy |
| November 2025 | 2.5 | legacy |
| December 2025 | 2.3 | legacy |
| January 2026 | 2.8 | legacy |
| February 2026 | 3.1 | legacy |
| March 2026 | 1.8 | legacy |
| April 2026 | 1.7 | legacy |
| May 2026 | 1.6 | legacy |
| June 2026 | 2.6 | legacy |
| July 2026 | 3.2 | V2 |
| August 2026 | 3.4 | V3 |
| September 2026 | 3.3 | V3.1 (unchanged V3 index) |
Why this form. A single line preserves the exact monthly sequence and familiar archive grammar. Challenger considered. Horizon graph: Compression into bands would obscure this short, low-valued series.
The series from October 2025 through August 2026 is frozen. Earlier editions retain their original methods and vintages. July’s V2 and August’s V3 transitions remain marked; the historical line is a record of published assessments, not a retrospectively harmonized statistical index. V3.1 strengthens the evidence, date controls, source distinctions and publication assets. It changes neither the V3 weights nor the score formula. S32
Exhibit 02S01 · S06 · S07 · S10 · S32
Operating pressure outweighs selective progress
Change in weighted index points · August to September
- Exact total: 3.425 → 3.275
- Operating and market effects: −0.250
- Investment and innovation offsets: +0.100
August 3.425 → September 3.275. Weight × score change; the endpoints reconcile exactly. 4 components unchanged.
Inspect exact values and definitions
| Component | Weight | August | September | Contribution change | Confidence |
|---|---|---|---|---|---|
| Production & utilization | 15% | 2.5 | 2.0 | −0.075 | 0.91 |
| Orders & demand | 10% | 5.5 | 5.0 | −0.050 | 0.85 |
| Profitability & revisions | 10% | 7.0 | 7.0 | 0.000 | 0.65 |
| Energy & feedstocks | 15% | 1.5 | 1.0 | −0.075 | 0.94 |
| Investment & German location | 15% | 1.5 | 2.0 | +0.075 | 0.80 |
| Employment & productivity | 10% | 3.0 | 3.0 | 0.000 | 0.66 |
| Trade & competitiveness | 10% | 3.5 | 3.5 | 0.000 | 0.87 |
| Innovation | 5% | 5.0 | 5.5 | +0.025 | 0.78 |
| Regulation & policy | 5% | 3.0 | 3.0 | 0.000 | 0.72 |
| Financial expectations | 5% | 6.0 | 5.0 | −0.050 | 0.78 |
Why this form. Additive steps reconcile every score change to the published total. Challenger considered. Tornado sensitivity chart: A tornado chart would imply a sensitivity experiment; these are additive accounting changes, not parameter sweeps.
| Component | Weight | Aug → Sep | Contribution | Reason |
|---|---|---|---|---|
| Production & utilizationConfidence 0.91 | 15% | 2.5 → 2.0 | ▼−0.075 | July chemical output fell 2.9% month on month. Cefic reports Q3 utilization at 72.9% in its separate Eurostat series. S01S02S05 |
| Orders & demandConfidence 0.85 | 10% | 5.5 → 5.0 | ▼−0.050 | July value orders fell 0.7% month on month despite +7.4% year on year; no verified September chemical-specific survey balance. S01S32 |
| Profitability & revisionsConfidence 0.65 | 10% | 7.0 → 7.0 | ■0.000 | Carry August earnings assessment; September supplies selective guidance, not a fresh sector earnings season. S20S32 |
| Energy & feedstocksConfidence 0.94 | 15% | 1.5 → 1.0 | ▼−0.075 | Wholesale power mean rises to €144.82/MWh; TTF monthly close proxy rises; exposure remains plant-specific. S06S07 |
| Investment & German locationConfidence 0.80 | 15% | 1.5 → 2.0 | ▲+0.075 | Commissioned BASF expansion earns a modest upgrade; one project and domestic consolidation cannot establish broad investment recovery. S10S11S16S20 |
| Employment & productivityConfidence 0.66 | 10% | 3.0 → 3.0 | ■0.000 | July productivity +0.8% YTD and unit labor costs +0.8%; training and site-transfer news cannot resolve net sector employment. S01S16S25 |
| Trade & competitivenessConfidence 0.87 | 10% | 3.5 → 3.5 | ■0.000 | Nominal exports +1.8% YTD, but US and China exports negative; output lags world and China. S01S03 |
| InnovationConfidence 0.78 | 5% | 5.0 → 5.5 | ▲+0.025 | German battery R&D and substitution products broaden capability; recycling and AI remain stage-qualified. S18S19S24S26 |
| Regulation & policyConfidence 0.72 | 5% | 3.0 → 3.0 | ■0.000 | PVC investigation is an initiated process; infrastructure and alliance activity have no measured sector cost dividend yet. S21S23S28S29 |
| Financial expectationsConfidence 0.78 | 5% | 6.0 → 5.0 | ▼−0.050 | Fixed basket weakens through common September 29 close; deal target return is not a factory-cycle signal. S07S13S14 |
Confidence is reported separately: the weighted evidence-confidence assessment is 0.815. It is neither a probability of recovery nor a confidence interval, and it never changes the component weights. Fresh wholesale prices support a high-confidence assessment of the benchmark. Incomplete September earnings and employment evidence justify more caution about company-wide margins and jobs.
02 The quarterly rebound meets the monthly data
VCI’s September quarterly release records a 3.0% quarter-on-quarter increase in chemical production excluding pharmaceuticals in Q2. Its utilization series moved only from 73.1% to 73.2%. The July release then shows chemical production falling 2.9% month on month and 2.2% year on year. A quarterly rebound and subsequent weakness can coexist; comparing them without their frequencies invites a false contradiction. S01 S02
Orders require a second distinction. July chemical order values were 7.4% above a year earlier but 0.7% below June. The VCI table identifies this as a value series. It cannot, on its own, establish stronger physical demand. Higher prices, product mix, inventory timing and disrupted competing supply can all support nominal orders while the amount produced remains weak.
Exhibit 03S01
Value orders and physical output diverge
Year-to-date change, % · January–July 2026
- Petrochemicals: orders +3.6%, output −5.9%
- Polymers: orders +2.4%, output −4.3%
- Orders measure value; production measures physical activity.
Different concepts: the gap between the bars is not an estimated volume conversion rate.
Inspect exact values and definitions
| Subsector | Orders YTD | Output YTD | Orders m/m | Output m/m | Orders y/y | Output y/y |
|---|---|---|---|---|---|---|
| Inorganic | +12.3% | +1.3% | +9.0% | −4.4% | +15.8% | −0.5% |
| Petrochemicals | +3.6% | −5.9% | +2.6% | −4.2% | +14.1% | −1.8% |
| Polymers | +2.4% | −4.3% | −0.5% | −4.2% | +7.0% | −5.2% |
| Fine & specialty | +3.6% | −0.4% | −2.9% | −1.2% | +3.4% | −0.1% |
| Detergents & care | +2.1% | −3.2% | −3.4% | −2.7% | −1.0% | −5.9% |
Why this form. Paired zero-based bars compare direction while labels distinguish value and volume. Challenger considered. Slopegraph: Slope connections could imply a common measure transitioning from orders into production; these are different constructs.
The subsector evidence is uneven. Inorganics combine positive year-to-date orders and output. Petrochemicals and polymers show positive order values alongside contracting production. That pattern is consistent with price support and temporary sourcing effects, but it does not identify their separate contributions. No estimated “conversion rate” is calculated by subtracting two series with different economic meanings.
The latest chemical-specific ifo balances numerically verified for this edition remain those carried in August’s radar. They are not relabeled as September observations. Broad manufacturing surveys and improved national forecasts provide context, but neither substitutes for a missing chemical-sector balance. The absence of a fresh observation reduces confidence; it is not scored as a zero.
03 Germany’s idle capacity needs two labels
Cefic’s report, posted on September 14 and carrying an August document date, reports German chemical utilization at 72.9% in Q3, down from 73.6% in its Q2 series. The EU27 rises from 74.4% to 75.0%. Germany therefore weakens within that particular cross-country survey series while several European peers improve. S05
Exhibit 04S05
German utilization moves against the EU trend
Capacity utilization, % · Q2 and Q3 2026 · Cefic/Eurostat
- Germany: 73.6% → 72.9%
- EU27: 74.4% → 75.0%
- Q2 → Q3 within Cefic/Eurostat; no VCI splice.
Source-specific survey series. VCI’s Q2 value of 73.2% is not spliced into this comparison.
Inspect exact values and definitions
| Region | Q2 2026 | Q3 2026 | Change, pp |
|---|---|---|---|
| Germany | 73.6% | 72.9% | −0.7 |
| Belgium | 71.4% | 71.8% | +0.4 |
| France | 76.5% | 77.7% | +1.2 |
| Netherlands | 74.0% | 76.1% | +2.1 |
| Poland | 73.4% | 75.8% | +2.4 |
| Spain | 81.8% | 80.6% | −1.2 |
| EU27 | 74.4% | 75.0% | +0.6 |
Why this form. Connected points preserve the within-source quarterly change and country level. Challenger considered. Slopegraph: A slopegraph preserves change but crossing country labels impose a larger mobile reading burden.
VCI’s Q2 value of 73.2% and Cefic’s Q3 value of 72.9% are not joined into a new time series. Their definitions and source constructions have not been reconciled. VCI cites an 83% long-run benchmark; August’s radar used 80.4% from its source. Cefic uses an EU benchmark of 81.3% and contains an inconsistent shortfall in its prose: 75.0% is 6.3 percentage points below 81.3%, not 6.9. This edition preserves the reported levels, calculates differences independently, and records the disagreement instead of silently choosing a convenient baseline.
The robust conclusion does not require a contested decimal: German chemical capacity remains far from normal utilization. The precise size of the historical shortfall depends on the series and benchmark selected.
Act II The mechanism behind the weak recovery
04 Energy pressure extends beyond the monthly average
German–Luxembourg day-ahead electricity averaged €144.82/MWh in September, 14.2% above the frozen August benchmark of €126.853. Across 2,880 quarter-hours, 645 exceeded €200/MWh: 22.4% of the month, or 161.25 hours. August’s corresponding share was 6.0%. The September maximum was €740.01/MWh, while 143 intervals were negative. S06 S32
Exhibit 05S06 · S32
High prices occupy more of the month
September day-ahead price, €/MWh · 2,880 quarter-hours
- Mean: €144.82/MWh
- 22.4% of intervals above €200
- August comparable share: 6.0%
Inspect exact values and definitions
| Statistic | September | August benchmark |
|---|---|---|
| Mean day-ahead price | €144.82/MWh | €126.853/MWh |
| Change in mean | +14.2% | — |
| Maximum | €740.01/MWh | — |
| Minimum | −€19.00/MWh | — |
| Quarter-hours above €200/MWh | 645 of 2,880 (22.4%) | 178 of 2,976 (6.0%) |
| Hours above €200/MWh | 161.25 | — |
| Negative quarter-hours | 143 | — |
Why this form. A price-duration curve reveals the costly tail without arbitrary histogram bins. Challenger considered. Quantile function plot: An ascending quantile function is valid but makes the share of expensive intervals less immediate than the descending duration curve.
For a continuous process, the expensive tail can matter as much as the occasional cheap interval. A plant cannot necessarily move a reaction, shutdown cycle or customer delivery into the cheapest quarter-hour. Flexible loads, storage, on-site generation and procurement contracts change the exposure. They also mean that an unweighted day-ahead mean is not the electricity bill of a representative chemical factory.
An illustrative sensitivity makes the boundary concrete. The benchmark increase from August is €17.97/MWh. A fully exposed operation using one additional MWh per tonne would face €17.97 more electricity cost per tonne, before contracts, hedges, fees, taxes, operational changes or pass-through. At two MWh, that arithmetic becomes €35.93. These are conditional calculations, not estimates of actual sector costs or margins.
05 Feedstock and financing add separate constraints
The September daily-close TTF futures proxy averaged €75.44/MWh and ended at €72.36. The Brent proxy averaged $102.26 per barrel. Henry Hub averaged $2.95/MMBtu. These are distinct front-contract proxies, with different units and contract rolls. The chart rebases each to August 31 = 100; it does not equate the gas and oil cost of production. S07
Exhibit 06S07
Gas and oil retain the cost pressure
Front-contract close proxies · August 31 = 100
- TTF average €75.44/MWh
- Brent average $102.26/barrel
- Henry Hub average $2.95/MMBtu; different units.
Distinct contracts and units: the rebased lines compare proportional moves, not production costs.
Inspect exact values and definitions
| Series | September mean | Last close | Last date | Index, Aug 31 = 100 |
|---|---|---|---|---|
| TTF | €75.44/MWh | €72.36/MWh | 2026-09-30 | 103.6 |
| Henry Hub | $2.954/MMBtu | $3.026/MMBtu | 2026-09-30 | 103.1 |
| Brent | $102.26/bbl | $103.53/bbl | 2026-09-30 | 114.4 |
Why this form. Common rebasing compares proportional changes while preserving distinct contract units. Challenger considered. Braided line chart: Braiding creates unnecessary reading cost with only three series.
INEOS reported indefinite suspension of its three large acetyl plants in Hull, with two already stopped when the September 22 statement appeared. This is a material European operating signal. Hull is in the United Kingdom, and the announcement is an idling decision rather than proof of permanent dismantling. INEOS’s much larger claims about international energy and carbon-footprint ratios are company assertions whose full measurement basis was not independently established here. They are not generalized into a German industry ratio. S22
Financing moved in an unfavorable direction as well: the ECB raised its key rates by 25 basis points, effective September 16. The deposit facility rate became 2.50%. This changes the macroeconomic financing environment, although a particular company’s debt maturity, hedge book and credit spread determine the timing of the effect. No mechanical earnings deduction is attached to the policy move. S08
06 Temporary sourcing support can coexist with structural weakness
VCI’s quarterly account attributes part of the spring improvement to inventory behavior and disruptions affecting competing supply. Such substitution can raise European orders or prices without establishing durable end-market growth. The eventual test is whether customers keep buying, plants operate more intensively and producers retain viable margins after alternative supply returns. S02
The alternative explanation deserves space. A delayed recovery could still be developing: customer orders may lead production, one weak month can reflect maintenance or logistics, and a capital project commissioned today reflects a decision made earlier. September’s evidence does not identify a single causal explanation for July output. It does, however, place a demanding burden of proof on a claim that the recovery is already broad and self-sustaining.
Low water adds a physical constraint to the cost discussion. The joint autumn forecast identifies low-water conditions since mid-July as an impediment to chemical production. Transport bottlenecks can interrupt otherwise profitable production and complicate inventory management. A company can therefore face weak demand, high energy and logistical disruption simultaneously; the data do not permit their effects to be added as separate quantified losses. S09
Act III The system in which German chemicals compete
07 A common vintage changes the global comparison
On the common January–July reporting window, chemical output excluding pharmaceuticals grew 1.0% globally and 3.8% in China. It fell 2.7% in Germany, 2.0% in the EU27 and 2.1% in the United States. Germany trails the world by 3.7 percentage points and China by 6.5 points. The US result also prevents a simplistic account in which cheaper feedstock guarantees volume growth. S03
Exhibit 07S03
Germany trails a slowing world
Chemical output, % year to date · January–July 2026
- China +3.8% · World +1.0%
- Germany −2.7% · EU27 −2.0%
- Common January–July window; chemicals excluding pharma.
Germany trails the world by 3.7 percentage points and China by 6.5.
Inspect exact values and definitions
| Region | Output, % YTD | Latest observation |
|---|---|---|
| China | +3.8% | 2026-07 |
| Spain | +2.8% | 2026-07 |
| World | +1.0% | 2026-07 |
| South Korea | +0.2% | 2026-07 |
| India | 0.0% | 2026-07 |
| France | 0.0% | 2026-07 |
| EU27 | −2.0% | 2026-07 |
| United States | −2.1% | 2026-07 |
| Germany | −2.7% | 2026-07 |
| Brazil | −3.1% | 2026-07 |
| Italy | −3.3% | 2026-07 |
| Japan | −3.8% | 2026-07 |
| Netherlands | −5.0% | 2026-07 |
| Singapore | −9.3% | 2026-07 |
Why this form. Position provides direct regional comparison with a visible zero. Challenger considered. Barcode strip: A barcode loses readable country-level values in this small sample.
The global comparison measures production growth. It says nothing directly about the profitability of the incremental Chinese tonne, the age of installed assets, carbon intensity, or the share of economic value retained locally. Germany’s competitiveness question is therefore broader than matching a foreign growth rate. It concerns which products and processes can sustain domestic value added under the prevailing cost structure.
For the forward view, VCI’s September international report forecasts chemical production growth of 1.3% globally in 2026, including −1.5% in Germany and +3.5% in China. These are forecasts, kept separate from the observed July-vintage figures. A slower global cycle would limit the help Germany can expect from external demand even if domestic conditions improve. S04
08 The trade surplus does not settle the competitiveness question
German chemical exports reached €83.672 billion in January–July, up 1.8% year on year; imports were €59.800 billion, down 0.5%. The resulting €23.872 billion surplus is substantial. Its geographical composition is less reassuring than the aggregate: exports to the US and China remained below their year-earlier values, while the EU and India provided support. S01
Exhibit 08S01
Exports hold up unevenly
Nominal chemical trade · January–July 2026
- Exports €83.672bn
- Imports €59.800bn
- Balance €23.872bn; partner groups overlap.
| Partner | Exports € m | Imports € m | Balance € m | Exports y/y | Imports y/y |
|---|---|---|---|---|---|
| World | 83,672 | 59,800 | 23,872 | ▲+1.8% | ▼−0.5% |
| EU27 | 49,862 | 37,193 | 12,669 | ▲+3.5% | ▲+5.8% |
| United States | 5,177 | 4,464 | 713 | ▼−7.7% | ▼−4.5% |
| China | 3,603 | 3,532 | 71 | ▼−1.7% | ▲+2.5% |
| India | 1,366 | 976 | 390 | ▲+5.0% | ▲+4.0% |
| Japan | 1,662 | 1,502 | 160 | ▼−4.5% | ▼−13.8% |
| South Korea | 1,101 | 838 | 263 | ▲+0.7% | ▲+7.5% |
Partner groups overlap: the EU27 total contains individual member states, so rows must not be summed.
Inspect exact values and definitions
| Partner | Exports, € m | Imports, € m | Balance, € m | Exports y/y | Imports y/y |
|---|---|---|---|---|---|
| World | 83,672 | 59,800 | 23,872 | +1.8% | −0.5% |
| EU27 | 49,862 | 37,193 | 12,669 | +3.5% | +5.8% |
| United States | 5,177 | 4,464 | 713 | −7.7% | −4.5% |
| China | 3,603 | 3,532 | 71 | −1.7% | +2.5% |
| India | 1,366 | 976 | 390 | +5.0% | +4.0% |
| Japan | 1,662 | 1,502 | 160 | −4.5% | −13.8% |
| South Korea | 1,101 | 838 | 263 | +0.7% | +7.5% |
Why this form. A table preserves euro levels, growth rates and balances without mixed axes. Challenger considered. Parallel coordinates plot: Parallel coordinates require multiple unfamiliar axes and obscure exact euro balances and overlapping groups.
A surplus can persist while domestic capacity contracts, particularly if imports weaken too or prices lift export values. Conversely, more imported intermediates can support competitive domestic downstream production. The table therefore separates export growth, import growth and the balance. EU totals and individual countries overlap; adding all displayed rows would manufacture a larger trade total.
Eurostat’s euro-area trade release reports a wider chemical contribution to the aggregate surplus. That broader category and geography include activities outside this radar’s German chemicals-excluding-pharma definition. It is context, not corroboration of an identical series. S31
Trade policy advanced on one identifiable file. A proceeding concerning certain PVC imports from China, Korea, Mexico and Taiwan was initiated in September, according to the German government’s trade agency, with a notice dated September 23. INEOS welcomed it on September 30. Initiation opens an investigation; it does not establish dumping, a final duty or an immediate margin benefit. The legal notice itself was blocked during retrieval, so this edition attributes the procedural summary to GTAI and does not offer a legal interpretation. S29 S30
09 Domestic investment is a classification problem
BASF’s modernized acid-chloride and chloroformate plant in Ludwigshafen is the strongest counterexample to a blanket retreat narrative. It is operating, has approximately 30% more capacity and represents an investment described as a low triple-digit million-euro amount. That is sufficient evidence for a modest improvement in the German-location component. It is not sufficient evidence for a sector-wide capex turn. S10
Exhibit 09S10 · S11 · S13 · S16 · S17 · S18 · S24
German investment needs a stage test
Selected September milestones · geography and maturity
- BASF Ludwigshafen: operating expansion
- Evonik: domestic transfers planned
- India MDI: feasibility; recycling pilot: planned 2028.
BASF SCF
Ludwigshafen
- Stage
- Operating expansion
- German capture
- Yes
Approximately 30% more capacity; low triple-digit million-euro investment S10
Evonik cosmetics
Hamburg → Essen
- Stage
- Planned domestic transfer
- German capture
- Consolidation
50 employees at originating site; no net German job-loss estimate S16
Evonik chlorosilanes
Bitterfeld → Rheinfelden
- Stage
- Planned domestic transfer
- German capture
- Consolidation
40 employees at originating site; no foreign relocation asserted S16
LANXESS battery lab
Krefeld-Uerdingen
- Stage
- Operating R&D
- German capture
- Yes, R&D
Cell testing and LFP precursor development; no battery-factory output S18
Evonik biotech
Slovenská Ľupča
- Stage
- Construction
- German capture
- EU, outside Germany
Groundbreaking; no current incremental production counted S17
BASF MDI
Dahej, India
- Stage
- Feasibility
- German capture
- Outside Germany
Land secured; investment decision pending S11
Covestro pyrolysis
Fraunhofer UMSICHT partnership
- Stage
- Pilot planned 2028
- German capture
- Potential German capability
2,000 t/year planned pilot; no commercial recovery rate assumed S24
BASF–Evonik
Group transaction
- Stage
- Non-binding proposal
- German capture
- Undetermined
No synergies, integration or German capex credited S13
Stage and location are classified from disclosed milestones. No capital expenditure or output value is invented for undisclosed projects.
Inspect exact values and definitions
| Project | Place | Stage | German capture | Source | Meaning |
|---|---|---|---|---|---|
| BASF SCF | Ludwigshafen | Operating expansion | Yes | S10 | Approximately 30% more capacity; low triple-digit million-euro investment |
| Evonik cosmetics | Hamburg → Essen | Planned domestic transfer | Consolidation | S16 | 50 employees at originating site; no net German job-loss estimate |
| Evonik chlorosilanes | Bitterfeld → Rheinfelden | Planned domestic transfer | Consolidation | S16 | 40 employees at originating site; no foreign relocation asserted |
| LANXESS battery lab | Krefeld-Uerdingen | Operating R&D | Yes, R&D | S18 | Cell testing and LFP precursor development; no battery-factory output |
| Evonik biotech | Slovenská Ľupča | Construction | EU, outside Germany | S17 | Groundbreaking; no current incremental production counted |
| BASF MDI | Dahej, India | Feasibility | Outside Germany | S11 | Land secured; investment decision pending |
| Covestro pyrolysis | Fraunhofer UMSICHT partnership | Pilot planned 2028 | Potential German capability | S24 | 2,000 t/year planned pilot; no commercial recovery rate assumed |
| BASF–Evonik | Group transaction | Non-binding proposal | Undetermined | S13 | No synergies, integration or German capex credited |
Why this form. An evidence ledger keeps location, stage and interpretation together. Challenger considered. Parallel categories plot: Parallel categories imply a reconcilable cohort or population flow; these are deliberately selected heterogeneous projects.
Evonik’s planned transfers provide a different type of evidence. Cosmetics activity moves from Hamburg to Essen; chlorosilanes move from Bitterfeld to Rheinfelden. The originating sites employ roughly 50 and 40 people respectively. These are significant local changes, but the disclosed destination is elsewhere in Germany. Ninety affected employees cannot be reported as ninety net German jobs lost, and domestic concentration cannot automatically be recorded as production moving abroad. S16
The foreign projects need equally precise labels. Evonik’s biotechnology groundbreaking in Slovakia is a European construction milestone. BASF’s proposed MDI project in Dahej has land and an advanced feasibility process, with the final investment decision still pending. German ownership does not turn either into German domestic investment; equally, feasibility is not completed capital expenditure. S11 S17
The practical implication is to maintain an investment ledger with location, stage, capacity effect and expenditure status. A list of press-release amounts is not a measure of domestic industrial renewal. Projects with undisclosed values should stay unquantified rather than disappearing from the assessment or receiving invented estimates.
10 Innovation has several clocks
LANXESS’s battery laboratory in Krefeld-Uerdingen is operating research infrastructure. Its work on iron oxides, phosphates and cell testing supports a German capability position, without constituting a commissioned battery-materials factory. The CCPD tire-protection announcement is another form of progress: commercial production is planned for Q4. Product testing and registration statements do not by themselves establish universal environmental safety or commercial adoption. S18 S19
Covestro’s rigid-foam recycling work is earlier in the scale-up process. The disclosed target is a 2,000-tonne-per-year pilot plant in 2028. Recovered feedstock quality and modeled product footprints are promising company-reported results; collection, contamination, process energy, yields, certification and customer acceptance still determine commercial economics. This edition records the milestone without treating the target capacity as current recycled output. S24
Greensand’s commercial opening in Denmark creates a different option: infrastructure for permanent CO₂ storage. The initial phase is described as capable of storing up to 400,000 tonnes annually, with a much larger future expansion ambition. Capacity is not verified annual throughput, and the initial CO₂ supply is mainly from Danish biomethane producers. German chemical decarbonization would still require capture projects, transport arrangements and a viable commercial chain. S23
A hydrogen transport-capacity reservation for Wacker’s Burghausen site likewise advances one link in a chain. It does not establish delivered hydrogen volumes, price or carbon intensity. The question for the next edition is which additional contractual and operating steps close the distance between infrastructure access and usable industrial input. S21
AI appears this month in molecular modeling and laboratory research. BASF’s personal-care research and the DECHEMA program illustrate areas of application. They do not provide an audited estimate of faster commercialization, plant-wide productivity or improved margins. The highest-value digital watchlist remains concrete: experimental reproducibility, integration into laboratory records, process-control validation, cybersecurity and accountable decisions when models fail. No major sector-wide cyber incident was independently verified in this research; that is a coverage limitation, not a finding that none occurred. S26 S27
Act IV Diagnosis at company and industry level
11 Corporate transformation earns credit only when it changes outcomes
BASF is managing several distinct decisions: domestic plant modernization, international investment options, agricultural-business separation and a possible transaction involving Evonik. Its September IPO update targets Agricultural Solutions readiness in mid-2027; it does not announce a completed listing. The strategic risk is execution across several large programs. That is an analytical inference about managerial and financial demands, not a claim that any program has already failed. S10 S11 S12 S13
The public takeover record also contains a disagreement worth retaining. BASF described exploratory talks; Evonik confirmed receipt of a non-binding approach while saying that no talks were currently taking place. RAG-Stiftung disclosed contact in relation to its 43.8% holding. The statements establish interest and process uncertainty. Neither wording permits a completed-deal assumption. Unconfirmed price reports and speculative synergies remain outside the index. S13 S14 S15
Wacker’s Capital Markets Day emphasizes selective growth and capital returns. The disclosed ambitions include annual capex of €300–400 million and an EBITDA margin of 15% ±2 percentage points. Management also described Q3 EBITDA as expected to be around Q2’s level, with different conditions across chemicals and solar polysilicon. These are forward statements and targets, not September realized results. Their significance is the emphasis on selective allocation under constrained returns. S20
For LANXESS, the month contains a useful tension between new technical capability and poor equity performance. For Evonik, portfolio concentration, biotechnology investment and takeover interest need separate assessments. For INEOS, idling and carbon-storage infrastructure show that defensive operating choices and long-run investment can occur within the same group. A single favorable or unfavorable headline is an inadequate company diagnosis.
12 Markets offer a provisional, heterogeneous signal
The August basket is retained without additions or deletions. Using August 31 to the last common valid close on September 29, its mean adjusted-close return is −1.93%, the median is −1.95%, and five of fifteen names rise. LANXESS falls 16.39%; Fuchs rises 6.66%; Evonik rises 4.37%. The basket is provisional because complete September 30 values were unavailable across the membership. S07 S32
Exhibit 10S07 · S32
The market rally narrows
Adjusted-close return, % · August 31–September 29 · provisional
- Basket mean −1.93%
- 5 of 15 names positive
- Common September 29 close; provisional.
Provisional: last common valid close Sep 29. Unweighted, adjusted-close returns; the DAX is a performance index and the chemicals ETF a proxy.
Inspect exact values and definitions
| Company | Symbol | Aug 31 close | Sep 29 close | Return |
|---|---|---|---|---|
| Fuchs pref. | FPE3.DE | 40.26 | 42.94 | +6.66% |
| Qiagen | QIA.DE | 37.24 | 39.12 | +5.05% |
| Evonik | EVK.DE | 19.01 | 19.84 | +4.37% |
| Sartorius pref. | SRT3.DE | 249.20 | 260.00 | +4.33% |
| Symrise | SY1.DE | 92.20 | 93.50 | +1.41% |
| Bayer | BAYN.DE | 49.04 | 48.88 | −0.33% |
| Merck | MRK.DE | 139.00 | 137.40 | −1.15% |
| Beiersdorf | BEI.DE | 79.12 | 77.58 | −1.95% |
| Gerresheimer | GXI.DE | 28.12 | 27.44 | −2.42% |
| Henkel pref. | HEN3.DE | 76.04 | 73.82 | −2.92% |
| Brenntag | BNR.DE | 62.84 | 59.28 | −5.67% |
| BASF | BAS.DE | 52.98 | 49.94 | −5.74% |
| Wacker | WCH.DE | 93.85 | 88.05 | −6.18% |
| K+S | SDF.DE | 16.51 | 15.18 | −8.06% |
| LANXESS | LXS.DE | 14.95 | 12.50 | −16.39% |
| DAX performance index | — | — | — | −3.27% |
| Euro STOXX 600 Chemicals ETF proxy | — | — | — | −1.53% |
Why this form. Direct labels expose return dispersion and deal-target outperformance. Challenger considered. Barcode strip: A barcode would hide individual exposures and the LANXESS tail.
The same-date DAX performance index declines 3.27%; the European chemicals ETF proxy declines 1.53%. These benchmarks are informative but not identical instruments: the DAX is a performance index, and adjusted-close provider series are not independently constructed total-return indices. No claim of a precisely measured sector alpha follows from the comparison.
The basket is also deliberately broader than basic chemicals. Pharmaceuticals, life science, consumer products and packaging-related exposures remain present because changing membership would break continuity. Their business mix can produce very different returns from an upstream commodity producer. The mean is an unweighted descriptive statistic, not a recommended portfolio or a proxy for German chemical operating profit.
| Company / unchanged security | Return to Sep 29 | Interpretive boundary |
|---|---|---|
| BASF | -5.74% | Group exposure; no German plant-profit inference |
| Bayer | -0.33% | Group exposure; no German plant-profit inference |
| Evonik | +4.37% | Group exposure; no German plant-profit inference |
| LANXESS | -16.39% | Group exposure; no German plant-profit inference |
| Wacker | -6.18% | Group exposure; no German plant-profit inference |
| K+S | -8.06% | Group exposure; no German plant-profit inference |
| Brenntag | -5.67% | Group exposure; no German plant-profit inference |
| Henkel pref. | -2.92% | Group exposure; no German plant-profit inference |
| Symrise | +1.41% | Group exposure; no German plant-profit inference |
| Fuchs pref. | +6.66% | Group exposure; no German plant-profit inference |
| Merck | -1.15% | Group exposure; no German plant-profit inference |
| Sartorius pref. | +4.33% | Group exposure; no German plant-profit inference |
| Beiersdorf | -1.95% | Group exposure; no German plant-profit inference |
| Gerresheimer | -2.42% | Group exposure; no German plant-profit inference |
| Qiagen | +5.05% | Group exposure; no German plant-profit inference |
13 Labor and productivity remain an evidence gap
July’s sector productivity and unit labor-cost series both stand 0.8% above their year-earlier year-to-date levels. Such aggregates do not identify whether productivity improved through technology, mix, hours, restructuring or plant selection. LANXESS’s intake of 123 apprentices provides a concrete training signal, but it cannot resolve net hiring across the sector. S01 S25
The distinction matters for policy and company decisions. A plant consolidation may improve utilization at the receiving site while imposing concentrated costs on workers and suppliers near the departing site. A better group margin can coexist with a weaker local industrial ecosystem. Employment analysis therefore needs destination jobs, exits, redeployment, hours and skill requirements, not just announced headcounts. The employment component stays at 3.0 with lower confidence until those data improve.
14 Cycle, structure and capability require different evidence
Cycle
Value orders and prices hold up better than physical output.
Evidence that would change itRepeated volume growth, customers retaining orders after supply normalization.
Structural competitiveness
Energy disadvantage, low utilization and selective capex persist.
Evidence that would change itCompetitive delivered costs, durable returns and several executed German growth investments.
Capability
Battery R&D, product substitution, recycling, biotech and storage advance at different stages.
Evidence that would change itQualification, throughput, offtake and funded scale-up; project-specific productivity gains.
Policy transmission
Trade investigation and infrastructure access advance processes.
Evidence that would change itEffective measures, actual take-up and measurable industrial cost or supply-security benefits.
Demand recovery is primarily tested through volumes, order persistence and utilization. Structural competitiveness requires relative costs, investment returns and location decisions over longer periods. Capability formation involves research, skills, infrastructure and scale-up milestones whose payoffs may be years away. The three interact, but they should not be collapsed into a single recovery story.
This also frames the policy assessment. The Critical Chemicals Alliance provides an institutional mechanism for identifying critical capabilities and obstacles. September did not provide a verified sector-wide cost reduction attributable to that mechanism. PFAS and REACH remain material monitoring topics, but this edition does not claim a new universal September restriction without a verified legal source. Operational support should be evaluated through effective dates, eligibility, take-up and actual cost incidence. S28
Act V Decisions and the tests for October
15 Three scenarios, with observable failure conditions
The August scenario taxonomy is retained: temporary substitution support, a genuine volume recovery, and reversal as exceptional support fades. August’s 55/20/25 probability allocation remains in the prior-edition record. September does not assign new numerical probabilities because the evidence does not support a calibrated probability update. The scenarios below are conditional planning cases, not forecasts with estimated confidence bands.
Scenario A · conditional
Temporary substitution support
- What would be observed
- Prices and selected orders benefit; utilization remains below 76%.
- Falsification
- Orders fade or persistent physical growth broadens.
- Decision implication
- Avoid extrapolating exceptional margins into long-lived capex.
Scenario B · conditional
Genuine volume recovery
- What would be observed
- Positive physical output across several subsectors; utilization moves above the prior 76% upside trigger.
- Falsification
- Volumes fail to persist, or delivered costs prevent profitable production.
- Decision implication
- Advance qualified expansion with verified customer commitments.
Scenario C · conditional
Reversal after normalization
- What would be observed
- Pricing support fades before energy and utilization improve.
- Falsification
- Broad output recovery and sustained reduction in delivered costs.
- Decision implication
- Protect liquidity, critical skills and options for restarting viable capacity.
The central planning case remains selective activity under difficult operating economics. An upside case requires breadth: several subsectors, sustained volumes, improving utilization and relief that reaches actual plant costs. A downside case becomes more credible if temporary pricing or sourcing support fades before operating costs normalize. Corporate deal activity can accompany any of the three.
16 What leaders can decide now
For operating teams, test actual exposure to the expensive hours, including process constraints and procurement coverage. Use the price-duration curve as a prompt to inspect the contract and load profile. The public benchmark alone cannot establish the savings from a shutdown, a hedge or a flexibility investment.
For finance and strategy teams, separate maintenance, efficiency, compliance, expansion and market-entry capex. Require a location-specific bridge from spending to cash flow and a clear stage designation. Compare the business case under ordinary competing supply as well as disrupted supply; otherwise temporary pricing can be capitalized into a long-lived asset decision.
For innovation teams, attach the next irreversible milestone to every promising project: customer qualification, continuous pilot operation, validated yield, certified footprint, offtake or a funded scale-up decision. Publication counts and launch announcements are useful evidence of activity but weak substitutes for those milestones.
For policymakers, evaluate delivered industrial costs and the continuity of critical capabilities. An investigation, alliance or transport reservation can be progress without yet changing a plant’s operating economics. Local effects deserve their own accounting: moving production within Germany preserves a different set of capabilities from closing production and replacing it with imports.
17 A three-, six- and twelve-month watchlist
Three months
Watch. August/September output releases; Q3 earnings; common-date market breadth; industrial procurement costs.
Test of the diagnosis. Production positive month on month and year on year on repeated releases; utilization first above 75%, then 76%; compare wholesale power with the prior €100/MWh watch level.
Primary owner · Operations / CFO
Six months
Watch. New German final investment decisions; customer offtake; consolidated-site operation; policy take-up.
Test of the diagnosis. Several independent growth projects execute; location and net employment effects become measurable.
Primary owner · Strategy / workforce / policy
Twelve months
Watch. Commercial scale-up, delivered hydrogen and carbon-management contracts, capacity and return outcomes.
Test of the diagnosis. Pilot claims progress to sustained throughput; domestic value added and viable returns follow spending.
Primary owner · Board / innovation / capital allocation
October’s most useful tests are deliberately capable of disproving the current diagnosis. Repeated positive physical-output readings would weaken the claim that the rebound remains mainly nominal. Verified broad German growth investment would weaken the view that domestic gains are isolated. Lower wholesale prices without lower delivered industrial costs would show that the transmission mechanism remains incomplete.
The index should respond to those developments under the same weights. It should not rise because a transaction becomes more prominent, a target date gets closer or a missing data point is filled with a plausible number. The German chemical industry still possesses investable capabilities. September’s evidence asks whether enough of them can operate profitably, at scale, in Germany.
Appendix Sources and evidence boundaries
Thirty-three source records support the analysis and continuity audit. Forty-six candidate records cover operating data, energy, trade, companies, labor, innovation, infrastructure, policy and research gaps. Numerical observations, company statements, forecasts and editorial judgments are distinguished in the data and claim ledger. All ten figures have exact-value fallbacks and reproducible inputs.
The source register below is part of the article. Dynamic URLs, particularly the VCI barometer and market endpoints, may later change. The package records the edition’s extracted values and acquisition details. No full third-party press release or report is republished.
- S01
VCIChemie-Barometer September 2026
Official statisticsClaims C01, C02, C03, C06, C25
Vintage: July activity/trade; August producer prices
- S02
- S03
VCIWorld Chemistry Report September 2026
Official statisticsClaims C05
Vintage: Country vintages explicit; common sample January–July 2026
- S04
- S05
CeficChemical Trends Report Q2 2026
Primary statementClaims C07
Vintage: Document says August; Q3 survey utilisation, earlier activity vintages
- S06
Fraunhofer ISE Energy-Charts / SMARDGermany–Luxembourg day-ahead prices
Official statisticsClaims C01, C04
Vintage: September 1–30; 2,880 quarter-hours
- S07
Yahoo Finance chart endpointFixed basket, benchmarks and energy futures daily history
Market dataClaims C11, C12
Vintage: Equity common close September 29; energy closes September 30
- S08
European Central BankMonetary policy decisions
Primary statementClaims C18
Vintage: Rates effective September 16
- S09
ifo / Joint Economic ForecastJoint autumn forecast
Primary statementClaims C21
Vintage: 2026–2028 forecasts; July/September context
- S10
BASFModernized SCF plant commissioned in Ludwigshafen
Primary statementClaims C01, C08
Vintage: Commissioned; capacity approximately +30%
- S11
BASFMDI project in India advances to feasibility stage
Primary statementClaims C22
Vintage: Land secured; final investment decision pending
- S12
BASFAgricultural Solutions IPO preparations
Primary statementClaims C22
Vintage: Readiness targeted mid-2027; no completed IPO
- S13
- S14
Evonik / EQSNon-binding approach from BASF
Primary statementClaims C10
Vintage: Ad-hoc; no binding offer; issuer says no current talks
- S15
RAG-Stiftung / EQSStatement regarding potential Evonik takeover
Primary statementClaims C10
Vintage: Release September 25; body dateline September 26; 43.8% holding
- S16
EvonikBusiness concentration in Germany
Primary statementClaims C09
Vintage: Transfers planned for 2027; Hamburg/Bitterfeld sites
- S17
EvonikBiotechnology expansion groundbreaking
Primary statementClaims C23
Vintage: Slovenská Ľupča; construction milestone
- S18
LANXESSBattery development laboratory
Primary statementClaims C13
Vintage: Krefeld-Uerdingen operating R&D infrastructure
- S19
LANXESSVulkanox CCPD tire protection product
Primary statementClaims C13
Vintage: Commercial production planned Q4; company test claims
- S20
Wacker via pressetext / Reuters distributionCapital Markets Day 2026: Refocus
Primary statementClaims C19
Vintage: Management ambitions, guidance and planned capex; not realized results
- S21
Bavarian Ministry of Economic AffairsHydrogen transport reservation for Burghausen
Primary statementClaims C23
Vintage: Transport reservation; future supply chain
- S22
Primary statementClaims C16
Vintage: Two idled; third planned within days; indefinite suspension
- S23
INEOSGreensand commercial opening
Primary statementClaims C15
Vintage: Up to 0.4 Mt/year initial capacity; 4–8 Mt expansion aspiration
- S24
CovestroSmart pyrolysis for rigid foam recycling
Primary statementClaims C14
Vintage: Pilot planned for 2028; 2,000 tonnes/year target
- S25
LANXESS123 apprentices begin training
Primary statementClaims C25
Vintage: Training intake; not net job creation
- S26
BASFIFCSS 2026 research presentations
Primary statementClaims C24
Vintage: Molecular modeling and peptide research; no audited productivity effect
- S27
DECHEMADECHEMA FORUM 2026 program
Primary statementClaims C24
Vintage: Conference September 15–17; program not proof of factory deployment
- S28
European CommissionCritical Chemicals Alliance
Primary statementClaims C26
Vintage: Background institution; no verified September implementation dividend
- S29
Germany Trade & InvestPVC anti-dumping investigation initiated
Government summaryClaims C17
Vintage: Notice dated September 23; investigation, no final duties
- S30
INEOSStatement on PVC investigation
Primary statementClaims C17
Vintage: Industry position; allegations distinct from findings
- S31
EurostatEuro area international trade July 2026
Official statisticsClaims C26
Vintage: July all goods; chemicals category includes pharmaceuticals
- S32
Michael SchymuraChemical Industry Radar August 2026
Prior editionClaims C01, C11
Vintage: Frozen previous published edition and supplied V3 ZIP
- S33
Michael SchymuraChemical Industry Radar archive
Prior edition
Vintage: Website audit only; not September economic evidence
Claim ledger26 material claims with status, sources and limitations
| ID | Claim | Status | Limitation | Sources |
|---|---|---|---|---|
| C01 | September mood index 3.3/10; August 3.4; exact weighted total 3.275 versus 3.425. fig-02 | editorial judgment | Judgment-based scores; arithmetic precision is not measurement precision. | S01S06S10S32 |
| C02 | July chemical output −2.9% month on month and −2.2% year on year. fig-03 | observed | Published September; excludes pharmaceuticals. | S01 |
| C03 | July value orders +7.4% year on year but −0.7% month on month. fig-03 | observed | Nominal order values do not identify volume recovery. | S01 |
| C04 | September wholesale power averaged €144.82/MWh; 22.4% of quarter-hours exceeded €200. fig-04 | derived | Unweighted wholesale day-ahead benchmark, not delivered industrial cost. | S06 |
| C05 | China chemical output +3.8%, Germany −2.7%, world +1.0% in January–July. fig-05 | observed | Output rates; not value added, profits or domestic retention. | S03 |
| C06 | Germany exported €83.672bn and imported €59.800bn of chemicals in January–July. fig-06 | observed | Nominal values; overlapping partner groups are not additive. | S01 |
| C07 | Cefic’s Q3 German utilization estimate is 72.9%, versus EU27 75.0%. fig-07 | observed survey | Separate Eurostat series; not spliced into VCI Q2 73.2%. | S05 |
| C08 | BASF commissioned a Ludwigshafen expansion with approximately 30% more SCF capacity. fig-08 | company disclosure | Selected plant, not German sector capex total. | S10 |
| C09 | Evonik plans transfers from Hamburg to Essen and Bitterfeld to Rheinfelden. fig-08 | company plan | 90 affected origin-site employees are not a net employment-loss count. | S16 |
| C10 | The BASF–Evonik situation remained non-binding at the cutoff. | company disclosure | Conflicting descriptions of talks remain visible; no transaction price or synergy credited. | S13S14S15 |
| C11 | The unchanged 15-name basket returned -1.93% on average; 5/15 rose through September 29. fig-09 | derived proxy | Provisional common close; not a tradable index or pure chemicals basket. | S07S32 |
| C12 | The TTF daily-close futures proxy averaged €75.44/MWh in September. fig-10 | derived proxy | Continuous front-contract series may roll; not physical gas delivery or settlement. | S07 |
| C13 | LANXESS opened a battery research laboratory and announced CCPD production for Q4. | company disclosure / plan | Research operation and planned production are different maturity stages. | S18S19 |
| C14 | Covestro targets a 2,000-tonne annual pilot plant for 2028. | company plan | Not achieved commercial recycling capacity; manufacturer impact estimates remain unverified. | S24 |
| C15 | Greensand entered commercial operation with initial capacity up to 0.4 Mt CO₂/year. | company disclosure | Capacity is not annual realized storage; initial feedstock mainly Danish biogenic CO₂. | S23 |
| C16 | INEOS reported indefinite idling of three Hull acetyl plants, with two stopped at announcement. | company disclosure | UK operating decision; not counted as a German closure. Quoted cost/emissions ratios unverified. | S22 |
| C17 | The PVC anti-dumping proceeding concerns China, Korea, Mexico and Taiwan. | government-reported process | Initiation, not a finding of dumping or an imposed final tariff; EUR-Lex full text blocked. | S29S30 |
| C18 | The ECB raised key rates 25 basis points, effective September 16. | official decision | Macro financing context; no plant-specific financing impact estimated. | S08 |
| C19 | Wacker announced €300–400m annual capex ambitions and a 15±2% margin ambition. | management plan | Targets and selective investment strategy; not realized earnings improvement. | S20 |
| C20 | Quarterly chem-only production increased 3.0% in Q2 while utilization was 73.2% in VCI’s series. | observed | Quarterly bounce does not resolve later July weakness; distinct utilization series. | S02 |
| C21 | German growth forecasts improved while low water and weak private investment remain constraints. | forecast/context | Macroeconomic forecast is not chemical demand already realized. | S09 |
| C22 | BASF’s India MDI proposal is in feasibility; agricultural IPO readiness is targeted for mid-2027. | company plan | Neither a completed investment nor completed IPO. | S11S12 |
| C23 | Hydrogen transport reservation and Slovak biotech groundbreaking advance infrastructure and capability. | official / company milestone | Neither proves delivered German low-cost hydrogen or new German output. | S17S21 |
| C24 | September AI research and laboratory programs provide capability evidence without audited productivity effects. | research/program evidence | No sector AI productivity estimate is available. | S26S27 |
| C25 | 123 LANXESS apprentices started; July sector productivity and unit labor costs both rose 0.8% YTD. | observed / company intake | Training starts are not net hiring; productivity can rise with lower hours. | S01S25 |
| C26 | EU critical-chemicals institutions and trade-surplus headlines do not establish a German competitiveness recovery. | interpretation | Different geography, scope and implementation stage. | S28S31 |
Methods and continuityIndex contract, evidence clocks, transformations, disagreements and reproduction
This is a V3.1 publication and evidence upgrade with the V3 judgment-based index unchanged. Germany is the primary geography, chemicals excluding pharmaceuticals the principal statistical scope. Company groups and the fixed equity basket are broader and explicitly labeled. Language follows the existing English series.
Index contract
Mood = Σ weight × component score. Each score is an editorial judgment on a 0–10 scale, using direction, level, breadth, persistence and relevance to German chemicals. The August package does not supply deterministic normalization thresholds. None have been invented for September. Weights sum to one and are never rescaled for missing data. September scores, reasons, confidence and source IDs are stored in radar.json and data/radar-data.json.
| Component | Weight | Aug | Sep | Contribution change | Reason |
|---|---|---|---|---|---|
| Production & utilization | 15% | 2.5 | 2.0 | -0.075 | July chemical output fell 2.9% month on month. Cefic reports Q3 utilization at 72.9% in its separate Eurostat series. |
| Orders & demand | 10% | 5.5 | 5.0 | -0.050 | July value orders fell 0.7% month on month despite +7.4% year on year; no verified September chemical-specific survey balance. |
| Profitability & revisions | 10% | 7.0 | 7.0 | +0.000 | Carry August earnings assessment; September supplies selective guidance, not a fresh sector earnings season. |
| Energy & feedstocks | 15% | 1.5 | 1.0 | -0.075 | Wholesale power mean rises to €144.82/MWh; TTF monthly close proxy rises; exposure remains plant-specific. |
| Investment & German location | 15% | 1.5 | 2.0 | +0.075 | Commissioned BASF expansion earns a modest upgrade; one project and domestic consolidation cannot establish broad investment recovery. |
| Employment & productivity | 10% | 3.0 | 3.0 | +0.000 | July productivity +0.8% YTD and unit labor costs +0.8%; training and site-transfer news cannot resolve net sector employment. |
| Trade & competitiveness | 10% | 3.5 | 3.5 | +0.000 | Nominal exports +1.8% YTD, but US and China exports negative; output lags world and China. |
| Innovation | 5% | 5.0 | 5.5 | +0.025 | German battery R&D and substitution products broaden capability; recycling and AI remain stage-qualified. |
| Regulation & policy | 5% | 3.0 | 3.0 | +0.000 | PVC investigation is an initiated process; infrastructure and alliance activity have no measured sector cost dividend yet. |
| Financial expectations | 5% | 6.0 | 5.0 | -0.050 | Fixed basket weakens through common September 29 close; deal target return is not a factory-cycle signal. |
The independently recomputed total is 3.275; Decimal ROUND_HALF_UP to one decimal produces 3.3. The previous total is 3.425 and previous published value 3.4. Exact accounting change −0.150; displayed change −0.1. Historical October 2025–August 2026 values are copied exactly and are not backcast. Method changes before September are marked. Confidence 0.815 is a separately weighted editorial evidence assessment, not a probability, inferential interval or score weight.
A score is necessarily contestable. Moving any single 15%-weight component by half a point changes the total by 0.075; the corresponding effect for a 10% component is 0.050 and for 5% is 0.025. This sensitivity is a mechanical illustration, not a confidence band. It explains why the small monthly decline should be read as mild deterioration rather than a precisely measured shock.
Evidence clocks and cutoff
Event window 2026-09-01T00:00:00+02:00 to 2026-09-30T23:59:59.999999+02:00. Acquisition on October 1 supports a September retrospective; records are admitted by publication/market observation date, not download time. Q2 accounts, July output/trade and August producer prices retain their observation labels. Source S33 is a post-window website audit only. Company planned dates remain future plans.
Data transformations
Electricity: all 2,880 DE-LU day-ahead quarter-hours in local September; no missing or duplicate timestamps; intervals are 900 seconds. Unweighted arithmetic mean, minimum, maximum, count below zero and count strictly above €200/MWh. Counts convert to hours at 0.25 hours each. Sorted duration curve uses the observed intervals; no smoothing. August's average and threshold count are frozen prior-edition observations, not a revised series. Wholesale benchmarks exclude plant hedges, grids, taxes and procurement terms.
Markets: exactly the prior 15 symbols, no additions/deletions. Adjusted-close percentage change from August 31 to September 29, the last valid date common to every constituent and the equity ETF. All 15 retained, 5 positive; equal-weight mean −1.931514%, median −1.946412%. No September 30 value is imputed. Benchmarks use the same endpoints. DAX is a performance index; the ETF and provider-adjusted stock series differ in construction. Basket is broader than chemicals and has no portfolio recommendation status.
Energy futures: TTF=F EUR/MWh, NG=F USD/MMBtu, BZ=F USD/barrel. 21 valid September closes for each. Continuous front-contract proxies, not spot or delivered industrial prices; contract-roll effects are possible. Rebased chart = 100 × close / August 31 close. All dates converted using exchangeTimezoneName. Every observation after September 30 is excluded, including an October 1 FX bar in the acquisition response. FX is not used in any published conversion or ratio. Raw response archives are not bundled; the necessary frozen observations are included.
Global output: use January–July for every displayed region; exclude earlier country vintages from that comparison. World/China/Germany percentage-point gaps are arithmetic differences of growth rates, not contributions to global growth. Forecasts from S04 remain separate. Trade balances are export minus import values in €million; partner groups overlap. Value orders, physical output and producer prices are not deflated or combined into a synthetic real order series.
Disagreements, missingness and source limits
- Utilization: VCI Q2 Germany 73.2% and Cefic/Eurostat Q3 Germany 72.9% are different source series. Cefic Q2 is 73.6%. Preserve both, never splice. August's 73.2% vintage label and 80.4% benchmark are frozen; September explains the new official Q2 labeling and 83% VCI benchmark.
- Cefic reports EU Q3 75.0% and long-run 81.3%; correct difference 6.3pp. Its prose also says 6.9pp. Use independent arithmetic and disclose.
- BASF describes exploratory talks; Evonik says it has received a non-binding approach and no current talks. Neither is replaced by a fabricated reconciled state. RAG release timestamp September 25 differs from September 26 body dateline.
- No numerically verified September chemical-specific ifo balance; August survey levels are explicitly carried and not presented as September.
- No complete month-end equity basket; provisional September 29 cutoff explicit in chart, table, prose and data.
- Company environmental performance and international cost-ratio claims are not independent sector facts. Research/registration is not universal safety, planned capacity is not realized output, and commercial storage capacity is not annual captured volume.
- September PFAS/legal implementation, net sector employment, a comprehensive capex census, fertilizer-specific operational results and verified sector-wide AI productivity effects remain incomplete. These gaps are not filled with plausible estimates.
- Full EUR-Lex PVC notice could not be retrieved. Procedural claim is attributed to GTAI and triangulated with INEOS; no legal advice or tariff assumption.
Reproduction and rights
Run python code/reproduce.py for offline arithmetic and lineage checks. Charts additionally require Python 3.12, numpy and matplotlib; run python code/render_charts.py. Website rebuilding requires Node 22+ and marked; the delivered HTML itself has no external runtime or font dependency. Carousel source uses @oai/artifact-tool; PDF and PNG pages are derived from that same slide composition.
Necessary factual extracts and original analysis are supplied with source links. Full third-party articles/PDFs are excluded. S06 identifies CC BY 4.0 data attribution. No broad redistribution licence is asserted for Yahoo, VCI or issuer content. The code/acquisition-urls.csv file records source endpoints; source PDFs may be overwritten upstream. data/input-checksums.json records downloaded inputs used locally. No confidential company information was used.
Recorded disagreements
- utilization
- Keep VCI and Cefic/Eurostat separately; no splice.
- BASF/Evonik talks
- Preserve issuer wording discrepancy; non-binding status is common ground.
- Cefic EU gap arithmetic
- 75.0 versus 81.3 is 6.3pp; reject 6.9pp prose figure.