A wider trade gap did not break the bond rebound

A wider trade gap did not break the bond rebound

Research cutoff: October 6, 2026, 12:13 Toronto / 16:13 UTC

The first important release after the prior cutoff was not bullish for the U.S. external balance. The August goods-and-services deficit widened to $105.6 billion from a revised $92.8 billion in July. Exports rose 1.4% to $315.2 billion, but imports increased 4.3% to $420.8 billion. The real-goods deficit widened 8.2% to $114.7 billion. Imports of goods increased by $17.2 billion, including $6.2 billion more capital goods and $2.4 billion more semiconductors. That mix is a potential drag from net exports, but it is not a clean recession signal: part of the deterioration came from stronger investment-related imports. BEA, released October 6 at 08:30 EDT

The market response matters more than the headline. The cash ten-year yield fell toward 5.25% from 5.31% on Monday while Brent dropped roughly 2%; the S&P 500 rose about 0.8%. Lower oil reduced the immediate inflation premium enough to outweigh the wider trade deficit. This is not evidence that fiscal or issuance risk has disappeared. It means the near-term duration signal is still dominated by energy inflation, positioning and auction clearing. Associated Press, October 6 at 11:00 EDT WSJ market update, October 6 at 08:58 EDT

Decision ledger

No real orders were placed. There is no new paper fill in this occurrence.

Corn: the mark worsened, but the position still fits

December corn traded at 502.50 cents at 15:33 UTC, with a 502.25/502.50 bid/ask, 496.00 low and 503.25 high. That is 2.75 cents above the prior research mark and implies an indicative −$100 open P&L from the 500.50 assumed entry before costs. March corn was 516.75 cents, leaving 14.25 cents of December–March carry. Farmbucks exact CBOT corn chain

The carry and large September stocks still describe an adequately supplied nearby balance. The countercase is meaningful: harvest was only 23% complete versus a 27% five-year average, crop condition slipped three points to 54% good/excellent, and a yield cut of roughly two bushels per acre could absorb the stocks cushion before other balance-sheet changes. Those are carried facts from Monday’s report, not new midday evidence. The next decisive balance-sheet update is the October 9 WASDE and Crop Production report at noon Eastern. USDA WASDE schedule

At the current 502.50-cent reference, one contract loses $1,500 at the 532.50 stop, makes $1,125 at 480 and makes $2,125 at 460. A deliberately cautious 35% / 45% / 20% tree for those outcomes has a $406.25 gross mean and $341.25 after the canonical $65 allowance. Holding the target probability at 20%, continuation value turns negative only when the stop probability rises to about 48%. A ten-cent stop overshoot would cut the cautious-tree result to about $166 after costs.

That positive continuation value does not justify adding. It rose mechanically because the market moved against the short, not because evidence improved. One contract’s buffered current-reference stop risk is about $1,956, or 0.195% of indicative NAV; two contracts would exceed the ordinary starter range. Hold one and add zero. Stops do not cap losses during gaps or daily-limit moves.

Rates: the signal is real, but the fill is not

Exact December ten-year note futures were 104.41 at 13:36:50 UTC after trading as high as 104.625. That is a genuine dated retreat below the 104.50 watch zone. A later vendor page, however, showed 104.55 while carrying an inconsistent month code. The clean exact quote is therefore stale relative to the later rebound indication. Entering now at the earlier dip would be retroactive execution, not the authorized nearest-price convention.

The economics remain interesting. A hypothetical short at 104.50 with a 105.265625 stop and 101.50 target risks about $1,054 per contract under a conservative gap-and-cost buffer, or roughly 0.105% of NAV. One or two whole contracts fit. A judgmental 25% target / 35% 104.25 / 30% stop / 10% 106.00 gap tree has about $420 net value, but shifting ten probability points from the target to the gap erases it. That fragility matters when oil is falling, equities are rallying and auction demand is untested.

Treasury will sell $58 billion of three-year notes today, followed by $39 billion of ten-year notes Wednesday and $22 billion of thirty-year bonds Thursday. Recent weak auctions make the clearing level more important than the existence of buyers: primary dealers ensure participation, but a large tail can transmit volatility across rates and equities. Reuters, October 6

Decision: Wait. Require a fresh exact-contract rejection after auction evidence. The earlier retreat is evidence, not a fill.

Oil and products: a crude surplus is not a product surplus

November WTI was $88.81 at 15:40 UTC. The curve remained backwardated, but the November–December premium narrowed to roughly $0.90 from about $1.23 at the prior settlement. Recovering Gulf exports and reserve-release plans argue for lower futures. The physical counterweight remains strong: the latest EIA balance put crude inventories at 427.3 million barrels, about 2% above their five-year average, while gasoline sat 7% below and distillates 14% below. Total petroleum stocks fell about seven million barrels. A modest crude surplus cannot be generalized into comfortable refined-product supply. EIA weekly petroleum status

One Micro WTI contract fits easily: an illustrative $92 stop from $88.81 reserves about $418 after a $15 allowance and the 1.25 buffer. Sizing is not the veto. The veto is the payoff distribution. Shipping, insurance, refinery and product bottlenecks preserve an upside-gap tail larger than the nearby downside payoff. The most recent physical-flow release also leaves important country and grade detail for mid-October, so a clean normalization claim is premature.

Decision: Avoid a new outright short. This is an economic rejection, not a missing-price delay.

Challengers

Gold rose to $4,190.10 as yields and the dollar eased. One MGC with an illustrative $4,090 stop fits starter risk, but the entry depends on the same duration rally already under test at Treasury auctions. Wait rather than pay up for a correlated expression.

Copper held near $6.651/lb while Chinese exchanges remained closed. Price strength without reopened Chinese physical and inventory evidence is insufficient. One MHG fits, but the prior simplified tree had only a few dollars of edge before a one-cent move overwhelmed it. Wait for October 8 price discovery.

Equities were stronger: ESZ26 reached 7,893.00 and Nasdaq futures were near 31,597, both close to session highs. Lower oil and yields explain the relief rally, but record-level pricing and the unresolved auction sequence make a late long unattractive. The existing 7,730 pullback/reclaim study remains a research condition, not a standing order.

December euro futures rebounded to 1.1295. That move conflicts with reopening the closed euro short, while France-specific fiscal stress remains better expressed in rates than in broad FX. Soybeans, wheat, livestock and softs were screened; none combined cleaner fresh evidence, whole-contract risk and scenario value than the existing corn position.

Next 72 hours

• October 6, 13:00 Toronto: $58 billion U.S. three-year Treasury auction.

• October 7, 10:30 Toronto: EIA weekly petroleum report; product inventories matter more than crude alone.

• October 7, 13:00 Toronto: $39 billion U.S. ten-year reopening.

• October 7, 14:00 Toronto: Federal Reserve September meeting minutes.

• October 8, 08:30 Toronto: initial jobless claims.

• October 8, 13:00 Toronto: $22 billion U.S. thirty-year reopening.

• October 9, 12:00 Toronto: USDA WASDE and Crop Production, the next corn decision point.

The four-month regime view is unchanged: persistent inflation and policy divergence remain the leading path, but falling oil and resilient equities show why outright duration shorts need a cleaner trigger. The current corn contract will not be carried blindly through that horizon; it exits by November 20, and any March expression requires a fresh decision.

*All prices are research references, not broker fills. Standard-contract proxies are labelled where used for micro sizing. Scenario probabilities are analyst judgments, not market-implied facts. Simulated results exclude unmodeled gap and daily-limit losses.*