China’s factory inflation is not a demand boom

China’s factory inflation is not a demand boom

Research cutoff: 14 October 2026, 00:07 Toronto / 04:07 UTC. This daily edition covers developments since October 13 at 22:15 UTC, during Asia-Pacific trading. The four-month horizon remains February 13, 2027. U.S. CPI and the IEA report are still ahead; neither is treated as released.

The decision

Hold one March 2027 SOFR future, SR3H27; add zero. The existing simulated long remains at its original assumed entry of 95.475. The new exact-contract mark is 95.480, observed at 03:58:03 UTC. Stop 95.100, target 95.850 and February 12 time exit are unchanged. There is no new assumed fill.

The strongest new macro finding is the gap between Chinese upstream price pressure and consumer inflation. It argues against interpreting every commodity-price rise as a demand boom. Copper has a genuine new supply-risk development, but its higher futures price offsets much of that improvement. The cheaper March wheat/corn spread moves from Avoid to Wait; the current-price economics, not an inherited entry band, drive that upgrade.

What changed—and what did not

China’s NBS released September prices at 01:30 UTC October 14: producer prices rose 5.0% year-on-year, while consumer inflation was 0.9% and core consumer inflation 1.0%, unchanged. Producer purchasing prices rose 6.9%, faster than factory selling prices. Production-material prices rose 6.5%, but consumer-goods factory prices fell 0.1%. These are different baskets, so their differences are not a measured economy-wide profit-margin change. They are nevertheless consistent with uneven cost pressure rather than uniformly stronger final demand. NBS producer prices · Consumer prices · NBS interpretation

Our inference: a manufacturer unable to pass input costs through can lose margin without selling more. For copper, this strengthens the need to distinguish mine supply and available inventories from a broad Chinese consumption thesis. It does not prove industrial demand is falling, and it does not forecast today’s U.S. CPI.

Singapore also supplied new evidence. MAS held the appreciation rate of its exchange-rate policy band and narrowed its 2026 core inflation forecast to 1.5–2.0%. That is a policy pause despite cost pressure, not a rate cut. Singapore’s framework and domestic conditions differ from America’s; expected easing in inflation from mid-2027 also lies beyond our February horizon. MAS October 14 statement

Reuters reported that Escondida supervisors announced a Wednesday strike after mediation failed. This is a new development since our last cutoff, but neither actual participation nor lost copper output is established here. An announced supervisors’ strike is not equivalent to the entire mine shutting. Reuters, October 13 late report

Saudi official reporting corroborated the previously reported incident at an Aramco administrative building. It does not quantify an oil-production outage or establish that all operations are unaffected. No second attack is inferred from a second report. Saudi Ministry of Energy via SPA

SOFR: retain the small policy-path position

What the price means

At 95.480, March three-month SOFR implies about 4.520% compounded overnight funding for its future reference quarter. December was 95.745 at 03:39:20 UTC. The 26.5-basis-point difference is half a basis point narrower than our prior, separately timed observations. It describes two forward quarters, not a literal count of Fed hikes or guaranteed carry.

The dated September Fed median was 4.1% at year-end in both 2026 and 2027; the September target range was 3.75–4.00%. Williams’s October 13 remarks left room for another hike and then a possible pause. Those remain the central support for the trade—not new overnight announcements. Policy-rate projections and compounded repo funding are different benchmarks, with basis and risk premia between them. Fed decision · Projections · Williams reporting

The intended February 12 exit precedes the contract’s March 17–June 16 reference period and June 15 last trade. This is a repricing trade, not a final-settlement forecast realized within four months. One contract is $25 per basis point, with a current 0.005 tick worth $12.50. CME mechanics

Underwriting at the new reference

For comparability with alternatives, a hypothetical new unit from 95.480 includes $32 full round-trip friction: $7 commission plus two adverse ticks. The existing position is not re-entered at this price. We retain the prior scenario weights because no new U.S. inflation release justifies manufacturing an overnight conviction upgrade.

Ten probability points shifted from the favorable branch to an ordinary stop-first exit at 95.100 leave +$143. Shifting ten points instead to the severe-gap branch leaves +$80.50. Combining those two changes makes −$107. Doubling full friction leaves +$298.50. Stops do not cap discontinuous losses.

One fresh contract’s buffered quote-to-stop study is $1,219.50, about 0.122% NAV. The existing ledger retains its original entry-based risk of $1,191.375; current mark-to-stop risk with only future exit costs is $1,203.50. These use different starting points, not conflicting risk estimates. The existing $5,000 modeled margin reserve remains an assumption, not an exact broker initial-margin quote.

Hold, no add. Two contracts fit mechanically, but doubling the same inflation-path exposure requires new supportive evidence. A contained core CPI could preserve the case without making an add automatically attractive at the subsequent price. Broad sequential core pressure plus guidance validating materially higher 2027 rates would justify an exit review before the numeric stop. Headline energy inflation alone is not the whole test.

From the original entry, the unchanged modeled life-to-exit expectation remains +$355.50 after full lifecycle costs; from the new mark, expected future P&L is +$346.50 after future exit friction. Neither is an actual return. No fresh CFTC observation is used to boost conviction.

Copper: more supply risk, less entry cushion

The exact March micro last rose to $6.877/lb at 04:00:20 UTC, versus 6.839 previously. December was 6.7685 at 04:00:48. Separately observed books put the March premium over December at 10.60–10.85 cents/lb. Deferred COMEX metal still costs more; a London prompt shortage does not automatically translate into the same U.S. deliverable shortage.

Newly retrieved October 13 LME data show cash $14,741/t, three-month $14,625.50 and stocks 229,650 tonnes. Stocks fell 1,625 tonnes, but the cash premium narrowed to $115.50 from $164.50. This mixed signal replaces the older physical reference. Its publication clock was not verified, so it is dated data newly incorporated—not claimed as an overnight release. Westmetall’s compiled LME table

We modestly raise the favorable scenario weight from 35% to 40%, taking five points from the base branch for Escondida risk. We do not reduce the adverse-demand branch merely because upstream prices rose. Centinela talks and operating impact remain relevant; threatened losses are not counted as realized shortages.

Before ranking, one March micro at the observed 6.8765 ask with a 6.35 structural stop risks $1,654.81 after the 1.25 buffer and $9.50 friction, or 0.165% NAV. Two exceed the preferred starter range. One contract represents 2,500 lb. CME specification

Expected net is +$149.25, versus just +$74.25 without the probability revision. A 3.32-point shift from favorable to adverse erases the revised edge. Wait: review actual Chilean operating information and the next inventory/physical-premium update. At 6.80 with this same supported tree, expected net would be +$340.50 and survive either a five-point bull-to-bear or ten-point base-to-bear stress separately. That is a new reassessment reference, not a standing order or a permanent entry veto. Stronger verified supply consequences could also justify entry at a higher price.

March micro copper last trades February 24; the February 12 intended exit avoids an automatic November roll. A 6.35 study stop and reversal of physical tightness without stronger demand define invalidation.

Oil: the curve matters more than a mismatched headline quote

Exact December Micro WTI last was $93.69 at 04:02:18 UTC; March was $90.37 at 03:15:07. Books were 93.67/93.68 and 90.33/90.37. Their prompt-premium bounds narrowed to $3.30–3.35 from $3.36–3.45. These are separately observed books, not a synchronized spread execution.

The latest verified Gulf survey is still October 13: 329,059 barrels/day shut in. Recovery supports normalization, but delivered global availability and Saudi operating risk matter more than another arbitrary shut-in threshold. MMA survey

For one 100-barrel micro short, retain 35/40/15/10% probabilities:

One micro stays below the 0.50% thesis ceiling under the illustrated gap; two do not. The tail risk also exceeds the usual starter band, so low ordinary-stop risk is not a reason to force larger size. Five points from slow normalization to the severe gap make December EV −$14.50. For March the same change leaves +$21; raising both normalization endpoints $1 then makes −$54.

Wait for the October 14 IEA assessment and useful Saudi production/export evidence. The report’s launch date is verified, not its conclusions. Better delivered supply and inventory rebuilding could justify a short; quantified renewed losses would challenge it. December and March micro last trades are November 19 and February 19 respectively; the study exits precede them. IEA launch

Agriculture and gold: feasible contracts, different economic answers

All following scenarios are analyst judgments for February 12 exits, not promises of stop execution. Full friction and whole-contract feasibility are included before ranking.

Gold — Wait. April 1oz gold is now an exact observed small contract: last 4,256.50 at 03:48:43 UTC, book 4,254.25/4,255.25. It replaces the prior MGC proxy. Eight units fit with $60 total friction. One 10oz April MGC at its separate 4,254 ask would risk $2,559, slightly above the normal starter ceiling. Ten probability points from favorable to adverse make the eight-ounce model −$214. Reassess CPI breadth and real-rate response; higher inflation is not automatically bullish gold. Gold would partly duplicate the SOFR position’s easier-rate exposure. April avoids a December roll before the February exit. CME 1oz specifications

Corn — Wait. March last was 496.50 at 04:00 UTC. This is genuinely after the October 13 Crop Progress release. That release—corn harvest 33% versus 39% normal—is carried context, not new information twice. One standard now fits at 0.245% NAV, but a 3.61-point favorable-to-adverse shift removes its edge. Reconsider when harvest catch-up reaches marketed supply and exports/basis no longer offset it. Study invalidation remains 535 and a fundamental tightening of the balance. USDA report

Wheat/corn — upgrade Avoid to Wait. March wheat last was 680.50 at 04:02:03 UTC. Buying the 680.75 ask and selling corn at 496.25 gives 184.50, cheaper than the prior 188.75 close-based study. Both legs are 5,000 bushels, with $64 package friction. Two packages fit the normal starter range, but more size does not improve the distribution. Five points from the 230 branch to 150 leave just +$23.50; doubling costs then makes −$40.50. Planting and emergence delays do not establish permanent yield loss. Review relative export and fieldwork evidence before entry; February 12 avoids the March delivery period.

Soy oil/meal — Avoid re-entry. Fresh March standard oil was 68.62 at 04:01:55 UTC and meal 359.90 at 03:25:05. For three smaller pairs, the oil ask is 68.62 and the micro-compatible meal sale is conservatively rounded to 359.80. These are standard-contract proxies, not micro fills. The package metric per pair, 60×oil−10×meal, is $519.20. Retaining 35/40/25% outcomes at 870/480/53.60 gives −$89.10 after $61.20 friction for three pairs; adverse-scenario buffered risk is $1,807.20. Negative economics, not missing quote precision, reject it. October 15 NOPA crush and oil stocks are the next useful joint test.

Liquid-market challengers and the rest of the board

The overnight technology rebound improves a potential Nasdaq sale price but does not prove a broad earnings contraction. These December studies exit or are re-underwritten November 20, not held blindly to February.

Full friction is $9.50 MES, $8 MNQ and $38.25 ZN. The exact last clocks were 04:01:52, 04:01:57 and 04:02:11 UTC respectively; MNQ last was 30,878 and ZN 104.8125. Wait all three. Current economics remain vulnerable to plausible path changes. Bank/technology results and U.S. inflation are concrete near-term reassessment points. March MES is also listed and was 7,951.75 at 03:59:13; its deferred premium needs a separate February model rather than pretending carry is free.

Natural gas — Wait. No fresh gas quote was verified. Carry NGZ26 3.475 and NGH27 2.774 from October 13 09:58 UTC, a $0.701 winter premium. One December 1,000-MMBtu micro proxy study to a 2.90 stop risks $727.75 including $9 friction. The EIA October 8 report showed 3,500 Bcf storage, 68 Bcf above the five-year average; that dated surplus challenges immediate scarcity. Require weather-adjusted tightening and durable LNG pull in the October 15 review, not a generic winter story. A December trade needs a November exit or separately justified roll. EIA storage

Euro short — Wait. The softer overnight dollar and limited-Fed-tightening view counter euro-area fiscal/energy risks. No new exact micro-euro quote was established. The October 5 standard-contract proxy at 1.1242 is retained with its actual date, not used as a fresh sale. Reassess relative policy after U.S. CPI. Smaller contracts exist; the weak divergence thesis is the main reason to wait.

INR — context watch only. No new exchange-listed implementation or policy change was verified. The dated October 10 dollar-window policy context does not become a new futures trade.

Retired March MES setup, soybean directional and silver long — Avoid new entry. No new evidence warrants reopening the obsolete MES plan or reversing the previous negative silver-long economics. No refreshed silver/soybean directional quote is claimed. The active six-asset-class challenger scan above is daily monitoring, not a repeated full monthly universe scan.

Next 72 hours

BLS schedule · Federal Reserve calendar · EIA oil holiday schedule · Gas schedule · NOPA release calendar

Chile operating statements and physical-metal updates are event-driven; no unverified exact release hour is invented.

Simulated book and source conventions

The only open position is +1 SR3H27, historically assumed bought at 95.475 on October 13 at 22:16:53 UTC, using the exact 95.470 reference from 22:06:51 plus one adverse tick. This edition records no new assumed trade and no real order.

The mark-only ledger update at October 14 04:07:26 UTC uses the 03:58:03 exact last of 95.480. Cash remains $1,001,814.375; open gross P&L is +$12.50; NAV is $1,001,826.875. The $25 increase from the prior valuation is a mark change, not realized profit. Cumulative closed-trade net P&L remains $1,817.875. No pending entry orders; margin reserve $5,000; original entry-based open risk $1,191.375. All portfolio limits remain respected.

Native prices above were obtained from read-only IBKR market data, with exchange-month identity and last-trade clocks retained. Bid/ask fields are observation-time studies, not independently timestamped prints. Scenarios are judgments; older observations, proxies and unavailable markets are identified. No fresh positioning estimate was verified or invented, and no unobserved intraday stop path is simulated.

Research and an illustrative paper portfolio only. Futures can lose more than initial margin; stops and modeled gap scenarios do not bound every possible loss.