Bessent Bids the Long Bond, Gold Runs the Table
We grade last week's calls after an activist Treasury capped the long end and gold blew through our target, then reset the book around a policy-flattened curve, a re-armed oil fade and a yen short whose catalyst is finally lining up.
Scorecard: Last Week's Calls
Leading with the wins.
Long gold, $4,500 three-month target, add toward $4,200: Correct, and the standout. Spot ran from about $4,385 on 17 August to roughly $4,640 on 24 August, clearing the target; the $4,200 add never triggered (CNBC, 24 August; TradingEconomics, 24 August).
Long silver: Correct. Silver rose from about $65 to roughly $68.6 to $69.6 an ounce, up around 6% on the week (Kitco, 24 August; JM Bullion, 24 August).
Do not chase the equity record, AI and semis neutral: Correct. The S&P 500 fell to 7,652.86 on 24 August from about 7,745, a tech and chip-led slide, with Micron off 5.8% (CNBC, 24 August). Sitting out paid again.
Soft-dollar bias: Correct. DXY eased to about 98.8 from roughly 99.5, pressured by the Treasury's buyback move (TradingEconomics, 24 August).
Small constructive GBP above 1.34, cautious gilts: Correct. Cable held above 1.36 near a six-month high; the 10-year gilt stayed elevated around 5.05% (Yahoo Finance, 21 August; TradingEconomics, 20 August).
Neutral-to-small-long AUD, buy dips toward 0.685: Correct and tracking. AUD/USD rose to about 0.717, never testing the dip-buy, after a hawkish hold confirmed in the 19 August minutes (Yahoo Finance, 21 August; RBA, 19 August).
Structurally long Indian IT and financials, base case Nifty bases in the 24,200 zone, trim only above 96.5 USD/INR: Correct on the base case. The Nifty held the zone, closing 24,252.00 on 21 August; the rupee sat about 95.7, short of 96.5 (5paisa, 21 August; Yahoo Finance, 21 August).
Overweight IG, underweight HY, scale in only past 3.25 to 3.5% OAS: Correct and untriggered. HY OAS was 2.75% on 20 August, still historically tight (FRED/ICE BofA, 20 August).
Energy equity as a hedge only: Correct. Energy outperformed a falling broad index as crude held up while tech sold off (CNBC, 21 August).
Oil fade paused between $85 and $92, re-enter seller above $93, base case Brent $85 to $92: Tracking correct. Brent oscillated between the high $80s and a mid-week high near $94.5, tripping the above-$93 sell trigger, then eased to about $92 on 24 August (Forbes, 24 August; Reuters, 24 August).
Middle East, hold the call tail, do not chase spot: Correct and tracking. Attacks continued but no fresh escalation shock landed; the premium held rather than exploded (Al Jazeera, 20 August).
Now the misses, plain.
Cut the outright 10-year add, own front-end duration, run a 2s10s or 5s30s steepener: Mixed, and the steepener leg was wrong. Front-end duration worked as the 2-year held about 4.19% and the 10-year stayed near 4.71%, but the Treasury's 19 August buyback rallied the 30-year from 5.31% to about 5.20%, flattening the curve against the steepener rather than steepening it (FRED, 20 August; CNBC, 19 August).
Own the ECB hike via front-end, fade EUR/USD above 1.16: Split. The hike thesis is intact on 2.9% final HICP and a strong flash PMI, but the EUR/USD fade above 1.16 was offside as the pair rose to about 1.168 on broad dollar weakness (Eurostat, 19 August; Yahoo Finance, 21 August).
Short USD/JPY toward 150, contrarian, offside: Still offside, not worse. The pair sat about 159, unchanged on the week and well short of the 161 cut line, but the September BoJ hike that would validate the trade is now roughly 84% priced after 1.9% July CPI (TradingEconomics, 24 August; CNBC, 21 August).
Thresholds: Brent above $93 triggered and re-armed the fade; gold above $4,500 triggered the trim zone. The 10-year did not reach 4.85% or fall below 4.55%; HY OAS stayed far inside 3.25%; DXY held between 98 and 102; USD/INR held below 96.5; USD/JPY held below 161.
Hit rate: ten clean or tracking wins, one split, two misses. The wins were gold, silver, the soft dollar, sterling, Australia, India, credit discipline, not chasing stocks, and the oil fade re-arming on cue. The misses were the steepener, run over by an activist Treasury, and the EUR/USD fade, run over by that same dollar move.
Key Levels Dashboard
Brent crude, ~$92/bbl, up ~1% on the week. Hit about $94.5 mid-week, then eased on profit-taking ahead of US Iran sanctions (Reuters, 24 August; TradingEconomics, 24 August).
WTI crude, ~$85/bbl, up ~1% on the week. Traded about $85.6 on 24 August as the sanctions plan landed (Reuters, 24 August).
Gold spot, ~$4,640/oz, up ~6% on the week. Highest since mid-May on the Treasury buyback and debasement bid (CNBC, 24 August; TradingEconomics, 24 August).
Silver spot, ~$68.6/oz, up ~6% on the week. Held above $68, up more than 18% on the month (Kitco, 24 August; JM Bullion, 24 August).
Federal Reserve funds target, 3.50-3.75%, unchanged. Minutes on 19 August showed a hawkish 9-3 hold; next decision 16-17 September (Federal Reserve, 19 August).
ECB deposit rate, 2.25%, unchanged. Next meeting 10 September with fresh projections (ECB).
BoE Bank Rate, 3.75%, unchanged. Next decision 17 September (Bank of England).
RBA cash rate, 4.35%, unchanged. Minutes on 19 August confirmed a hawkish hold; next decision 28-29 September (RBA, 19 August).
RBI repo rate, 5.25%, unchanged. Neutral stance held since 5 August (RBI).
Latest US jobs print, July -23k, unemployment 4.1%, unchanged. Next report 5 September (BLS, 7 August).
S&P 500, ~7,653, down ~1.2% on the week. Chip-led selloff on 24 August, Micron off 5.8% (CNBC, 24 August).
US 2-year Treasury, ~4.19%, up ~4bp on the week. Front end anchored to the September hold (FRED, 20 August).
US 10-year Treasury, ~4.71%, roughly flat on the week. Held near 4.7% after the buyback capped a mid-week backup (FRED, 20 August; TradingEconomics, 24 August).
US 30-year Treasury, ~5.20%, down ~11bp on the week. Fell after Treasury doubled long-bond buybacks; August auction had cleared 5.216%, highest since 2001 (CNBC, 19 August).
DXY, ~98.8, down ~0.7 on the week. Under pressure after the buyback and on fiscal worry (TradingEconomics, 24 August).
EUR/USD, ~1.168, up ~0.9% on the week. Broke above 1.16 on dollar weakness (Yahoo Finance, 21 August).
GBP/USD, ~1.364, up ~0.7% on the week. Near a six-month high after firm UK data (Yahoo Finance, 21 August).
AUD/USD, ~0.717, up ~1.4% on the week. Firm after the hawkish RBA minutes (Yahoo Finance, 21 August).
USD/INR, ~95.7, rupee down ~0.3% on the week. Pressured by oil, still short of 96.5 (Yahoo Finance, 21 August).
USD/JPY, ~159.0, roughly flat on the week. September BoJ hike now about 84% priced (TradingEconomics, 24 August).
Nifty 50, ~24,252, roughly flat on the week. Held the 24,200 support zone as IT steadied (5paisa, 21 August).
US HY OAS, ~2.75%, wider ~4bp on the week. Still deep inside the scale-in trigger (FRED/ICE BofA, 20 August).
Executive Summary
The week turned on a policy decision, not a data point. On 19 August the Treasury said it would at least double the size of its long-dated buyback operations: "The current maximum size of $2 billion per operation will be at least $4 billion per operation. This change is effective September 9, 2026 and will be in effect for the remainder of this refunding quarter (through November 4, 2026)," covering the 10-to-20-year and 20-to-30-year sectors (US Treasury, 19 August). Yields fell on the announcement, the 10-year shedding 6 basis points to 4.647% and the 30-year giving up 9 to 5.196% (Quartz, 19 August). This was Bessent's second market intervention of the month after the 1 August joint yen operation, and it reframes the whole rates picture: the long end is no longer a clean macro instrument, it is a managed one. That is the honest correction to last week's steepener, which assumed the long end would stay heavy on supply and term premium. Supply is still there, total public debt topped $40 trillion on 19 August, but now there is a large, price-insensitive buyer leaning against it (Reuters, 19 August).
Gold read the same news the opposite way and won. A Treasury actively suppressing yields, a $40 trillion debt stock and a soft dollar are the debasement trade in three lines, and spot cleared our $4,500 target to about $4,640, its best level since mid-May (CNBC, 24 August; TradingEconomics, 24 August). That is the position that did the work again, exactly as it did the week before, and it is the one we press.
The macro read did not change. July CPI stayed benign, the FOMC minutes confirmed a hawkish 9-3 hold with sympathy for a hike beyond the three dissenters, and CME FedWatch put the September meeting at a 68.4% probability of a hold as of 20 August (Federal Reserve, 19 August; Beansprout/CME FedWatch, 20 August). The equity tape sagged on chip weakness into Nvidia's 26 August print, vindicating the don't-chase call once more (CNBC, 24 August). And the Middle East kept simmering rather than boiling: transits stayed depressed, attacks continued, and on 24 August the Treasury unveiled "Operation Economic Outcast," sanctioning more than 60 entities and threatening secondary measures on Iran's enablers, China included (CNBC, 24 August; NPR, 24 August). Oil spiked toward $94.5 then faded, which is why our above-$93 fade re-armed and worked.
The forward path is a contest between two policy hands and one metal. Bessent caps the long end; Warsh speaks at Jackson Hole on 28 August and has signalled he may not use rate hikes to fight inflation; gold sits above every fiat argument. We position front-end long, steepener off, gold pressed, oil sold into strength, and the yen short held into a BoJ hike that is finally lining up.
Signal-Filter Takeaways for Clients
The variable that moved this week was Treasury supply management, not the Fed and not the data. An activist Treasury flattens the long end when it leans on buybacks, which is why our steepener lost and our front-end duration held. Trade the policy hand, not the term-premium story we told last week.
Gold is not a hedge that occasionally works; it is the cleanest expression of a debased-fiat, managed-yield world, and it beat both stocks and bonds again. Keep it core and take some profit into strength rather than chasing.
Oil is a fade, not a chase, even with sanctions escalating. Flows persist, OPEC+ is still adding, and every spike toward the mid-$90s has met selling. Sell strength, do not buy breakouts.
01 Middle East: Sanctions Replace Strikes as the Lever
What happened. The pressure shifted from the water to the wire. On 24 August the Treasury launched "Operation Economic Outcast," sanctioning more than 60 corporations, individuals and vessels across the UAE, Hong Kong, China, Singapore, Switzerland and Europe, with Bessent vowing to "sever every economic lifeline that sustains the tyrannical regime" and signalling China is not exempt (CNBC, 24 August; NPR, 24 August). Trump billed it as Iran's "economic D-Day." On the water, the standoff held: transits ran 73 in the week to 16 August, down from 91, and five commercial vessels were attacked in the prior week, one fatally, though the most recent confirmed strikes clustered around 13 to 15 August (Al Jazeera, 20 August). Oil flows nearly tripled to about 6.1 million bpd during the now-expired US-Iran memorandum, still far below pre-war levels (Al Jazeera, 20 August).
Why it matters. The escalation is now financial, which is slower-burning and harder to price than a strike. Sanctions on Chinese enablers risk blowback into the Trump-Xi meeting due in late September and could remove discounted Iranian barrels, a mixed signal for crude. The market read it as bearish this week, taking profits, because sanctions take time and flows keep moving.
Scenarios
Base case, 55%: sanctions bite slowly, attacks continue at low intensity, Brent holds the high $80s to low $90s.
Downside, 30%: a strike lands or China retaliates on the trade truce, Brent back toward $100.
Upside, 15%: a face-saving de-escalation or Iran-Oman route deal, Brent to the low $80s.
What to watch. Whether the sanctions name major Chinese banks, the Trump-Xi meeting timing, and any confirmed strike inside the next week.
House view. Hold the small long-dated call tail against a shock, do not chase spot, and treat the sanctions news as a slow drip rather than a spike. Medium conviction.
02 Energy Markets: The Spike That Rolled Over
What happened. Brent pushed to about $94.5 mid-week, its highest since late July, on the sanctions rhetoric, then rolled back to roughly $92 by 24 August as traders booked profit and judged that flows would persist (TradingEconomics, 24 August; Reuters, 24 August). WTI tracked to about $85.6. The pattern held: every push toward the mid-$90s met sellers, and the above-$93 line we drew last week armed the fade.
Why it matters. This is a supplied market with a chokepoint that will not fully clear and a Treasury Secretary trying to talk energy prices down. The upside is capped by persistent covert flows and OPEC+ additions, the downside floored by attack risk. The mid-$80s remains the gravity point once the acute premium bleeds.
Scenarios
Base case, 55%: Brent oscillates $85 to $93 as supply and risk premium offset.
Downside for prices, 25%: flows normalise further, OPEC+ barrels bite, Brent to the low $80s.
Upside for prices, 20%: a strike or a China clash, Brent through $95 toward $100.
What to watch. The 6 September OPEC+ meeting for any change to the flagged Q4 pause, and Kpler and Lloyd's List transit counts.
House view. Contrarian and disciplined: we sell strength above $93 toward the mid-$80s, small size, given persistent flows and OPEC+ supply. Low conviction, and it worked this week.
03 United States: The Treasury Takes the Wheel
What happened. Two events, one direction. First, the 19 August FOMC minutes showed a fractured 9-3 hold with hawkish sympathy well beyond the three dissenters, participants noting "policy tightening would likely be necessary if inflation did not decline," and Warsh floating a cut from eight meetings to six (Quartz, 19 August; PNC, 19 August). Second, and bigger, the Treasury doubled long-bond buybacks the same day, rallying the 30-year from a 5.216% auction, its highest since 2001, back toward 5.20% and pulling the 10-year to 4.647% before it settled near 4.71% (CNBC, 19 August; FRED, 20 August). Total federal debt crossed $40 trillion (Reuters, 19 August). CME FedWatch held September at a 68.4% probability of no change as of 20 August (Beansprout/CME FedWatch, 20 August).
Why it matters. This is the flip within the flip. Last week we owned the steepener because supply and term premium were pushing the long end up. Now the Treasury is leaning directly on that long end, which is a flattening force, not a steepening one. Our front-end call was right; the steepener structure was the wrong way to express it. We keep the front-end duration and cut the steepener rather than fight a buyer with a printing press behind it.
Scenarios
Base case, 55%: Fed holds in September, front end anchors near 4.15 to 4.20% on the 2-year, Treasury caps the long end, the curve stays rangebound to flatter.
Downside, 25%: sticky oil and fiscal fear overwhelm the buybacks, the 10-year retests 4.85%.
Upside, 20%: buybacks plus soft data pull the 10-year toward 4.55%.
Market implications
Asset class: Front-end duration over long duration; steepener off, because the Treasury is flattening the long end.
Currency and flows: A managed long end plus a soft front end keeps the dollar heavy, which fits our soft-dollar bias.
Sector rotation: Long-duration equities got no relief; the S&P slid on chips into Nvidia.
Entry and exit: Hold front-end duration; add long-end only above 4.85%; do not run the steepener while buybacks are live through 4 November.
What to watch. Warsh at Jackson Hole on 28 August, July PCE on 26 August, and Nvidia the same evening.
House view. Flip flagged: we keep front-end duration and cut the steepener, because the driver of the long end is now Treasury supply management, not term premium we can trade cleanly. Medium conviction.
04 Europe: Growth Holds, the Hike Stays Live, the Fade Fails
What happened. Final July HICP confirmed 2.9%, with core revised up to 2.5% and energy at 10.3%, and the August flash composite PMI rose to 52.1, its strongest since November, manufacturing at a four-and-a-half-year high (Eurostat, 19 August; S&P Global, 21 August). That keeps the 10 September ECB hike genuinely live. The one thing that broke was the currency call: EUR/USD rose above 1.16 to about 1.168 on the dollar's buyback-driven slide, running over our fade (Yahoo Finance, 21 August).
Why it matters. The rate thesis strengthened while the FX expression failed, and the reason was made in Washington, not Frankfurt. We separate the two: keep owning the hike via front-end euro rates and European financials, but stop fading EUR/USD at 1.16 while the dollar is being actively softened.
Market implications
Asset class: Own the ECB hike via front-end euro rates.
Currency and flows: Neutral EUR; drop the 1.16 fade, the dollar leg has changed.
Sector rotation: Overweight European financials into a live hike; the Stoxx 600 recovered into the weekend on the strong PMI.
Entry and exit: No fresh EUR short above 1.16; revisit only on a dollar turn.
What to watch. ECB commentary into 10 September and the 1 September flash August HICP.
House view. Long the ECB hike via front-end rates, overweight European financials, EUR fade dropped. Medium conviction, with the FX leg retired.
05 United Kingdom: Sterling Bid, Gilts Heavy, Data Firm
What happened. The data ran firm and mixed. August flash composite PMI rose to 52.5, the fastest since April, and July CPIH picked up to 3.1% with core CPI at 2.6% and services easing to 3.4% (S&P Global, 21 August; ONS, 21 August). But July retail sales fell 0.5% and public borrowing came in at £1.8 billion, about £2.3 billion above the OBR path, keeping the fiscal question live (ONS, 21 August). Cable held above 1.36 near a six-month high, and the 10-year gilt stayed elevated around 5.05% (Yahoo Finance, 21 August; TradingEconomics, 20 August).
Why it matters. Firm activity and sticky services inflation argue against BoE cuts and support sterling, while the fiscal slippage and a 5.156% gilt auction keep us cautious on duration. The pound is the cleaner expression than the bond.
Market implications
Asset class: Cautious on gilts given the fiscal path and the global long-end backdrop.
Currency and flows: Constructive GBP above 1.34.
Sector rotation: No fresh call.
Entry and exit: Add cable dips toward 1.34, fade toward 1.37.
What to watch. The 17 September BoE decision and any budget-date announcement.
House view. Constructive GBP above 1.34, cautious gilts. Low conviction, unchanged.
06 Australia: Hawkish Hold, Confirmed Again
What happened. The 19 August minutes detailed how close the August hold was and reinforced the tightening bias, with the Board warning it would raise "further if upside risks materialise" and flagging oil and Hormuz as an external inflation risk (RBA, 19 August; investingLive, 19 August). AUD/USD rose to about 0.717, never near our dip-buy.
Why it matters. A central bank still leaning hawkish with a firm labour market underpins the currency and validates buying dips rather than chasing. Nothing this week weakened it.
Market implications
Asset class: Rates biased to stay restrictive.
Currency and flows: Neutral-to-small-long AUD.
Sector rotation: No fresh call.
Entry and exit: Buy AUD dips toward 0.69, trim toward 0.725.
What to watch. The 28-29 September decision, November hike pricing, and July CPI on 26 August.
House view. Neutral-to-small-long AUD, buy dips. Medium conviction, unchanged.
07 India: The Index Steadies, the Base Case Holds
What happened. The Nifty stopped falling and held the 24,200 support zone, closing 24,252.00 on 21 August with IT steadying and Infosys leading some sessions higher (5paisa, 21 August). The rupee sat about 95.7, still short of the 96.5 trim line, with oil the swing factor (Yahoo Finance, 21 August).
Why it matters. This is last week's shaky position turning the corner. The base case we set, that the index bases in the 24,200 zone as earnings season ends, is playing out, and IT firming takes pressure off the structural long.
Scenarios
Base case, 60%: the Nifty consolidates 24,200 to 24,700 as the tape steadies.
Downside, 25%: oil above $95 pushes the rupee toward 96.5 and forces a trim.
Upside, 15%: oil rolls over, IT rebounds, the index reclaims 24,700.
Market implications
Asset class: Structurally long large-cap IT exporters and financials.
Currency and flows: Watch the rupee against 96.5; oil is the swing factor.
Sector rotation: Add financials on weakness; IT now an add on the turn, not just a hold.
Entry and exit: Stagger entries, trim only if USD/INR breaks 96.5.
What to watch. USD/INR versus 96.5, Brent, and the late-September RBI meeting.
House view. Structurally long large-cap Indian IT exporters plus financials, trim only above 96.5. Medium conviction, off the short leash after the tape steadied.
08 Precious Metals and Safe-Haven Assets
What happened. Gold cleared our $4,500 target to about $4,640, its best since mid-May, and silver ran above $68, up more than 18% on the month (CNBC, 24 August; Kitco, 24 August). The catalyst was explicit: the Treasury's doubled buyback and a $40 trillion debt stock revived the debasement trade, pushing yields and the dollar down and bullion up, "as concerns over US debt management and fiscal sustainability persisted following the Treasury's unexpected ramp-up in longer-dated debt buybacks" (TradingEconomics, 24 August).
Why it matters. This is the thesis paying in full. When the Treasury manages yields lower and the dollar softens, real assets win, and gold beat both stocks and bonds again. The target is met, so we take some profit into strength and reset higher rather than chase.
Market implications
Asset class: Long gold, long silver, taking partial profit into strength.
Currency and flows: A soft dollar and managed yields underpin bullion.
Sector rotation: No fresh call.
Entry and exit: Trim gold above $4,700, add toward $4,400; add silver on pullbacks.
What to watch. Warsh at Jackson Hole, July PCE, and real yields.
House view. Long gold, target raised to $4,800 over three months, bank partial profit above $4,700, add toward $4,400; long silver. Medium-High conviction, up from Medium after the target was hit.
09 Sovereign and Credit Conditions
What happened. HY OAS was 2.75% on 20 August, wider by about 4 basis points but still historically tight, while the Treasury's buyback pulled the risk-free long end lower (FRED/ICE BofA, 20 August; CNBC, 19 August). The all-in yield story improved at the margin as the 30-year fell, but spreads still price a soft landing.
Why it matters. Tight spreads into a managed-yield backdrop leave little cushion if the buyback support fades after 4 November. We stay up in quality. The scale-in trigger is nowhere close.
Market implications
Asset class: Overweight IG, underweight HY.
Currency and flows: No fresh call.
Sector rotation: Favour higher-quality, shorter-spread-duration credit.
Entry and exit: Begin HY scale-in only past 3.25 to 3.5% OAS.
What to watch. Whether spreads widen as the buyback program winds toward its 4 November end.
House view. Overweight IG, underweight HY, no scale-in yet. Medium conviction, unchanged.
10 Other Flashpoints
Japan and the yen: July CPI hit 1.9% headline and 1.8% core, the highest this year, and September BoJ hike odds jumped to about 84% (CNBC, 21 August; TradingEconomics, 24 August). USD/JPY held about 159, short of our 161 cut line. The short is offside but its catalyst is finally lining up; we hold, small, contrarian. Medium conviction.
AI concentration and semis: Nvidia reports after the close on 26 August, the next big test; chips led the 24 August selloff, Micron off 5.8% (CNBC, 24 August). Neutral stance intact.
Jackson Hole: Warsh delivers his first keynote as Chair on 28 August, having signalled hikes may not be his tool of choice against inflation. The framework signal matters more than usual with forward guidance dropped.
US-China: The Iran sanctions plan threatens secondary measures on Chinese enablers ahead of a Trump-Xi meeting due late September, a live trade-truce risk (CNBC, 24 August).
US midterms: 3 November, a rising fiscal catalyst that overlaps the buyback window; no new development this week.
Cyber and Ukraine-Russia: No genuinely new flashpoint this week worth adding.
Upcoming Events and Catalyst Calendar
26 August: US July PCE. The Fed's preferred gauge before September.
26 August: Nvidia Q2 earnings, after the close. The single biggest test of the AI-capex trade.
26 August: Australia July CPI. Feeds the RBA's hawkish-hold debate.
27-29 August: Jackson Hole symposium, Warsh keynote on 28 August. His framework signal 19 days before the FOMC.
28-29 August: US Q2 GDP second estimate. Confirms the growth path.
1 September: Euro-area flash August HICP. First read on whether energy pass-through persists.
5 September: US August payrolls. The next read after July's -23k.
6 September: OPEC+ meeting. Watch for any change to the flagged Q4 pause.
9 September: Treasury's expanded long-bond buybacks begin. The mechanism now capping the long end goes live.
10 September: ECB decision and projections. Our owned September-hike call is graded here.
11 September: US August CPI. The inflation read before the FOMC.
16-17 September: FOMC. Hold priced near 68%; the long-end reaction is the risk.
17 September: BoE decision. Our sterling and gilt views are tested.
17-18 September: BoJ meeting. A hike validates the short-yen thesis.
28-29 September: RBA decision. Hawkish bias in focus.
Late September or early October: RBI decision, and a possible Trump-Xi meeting.
3 November: US midterms.
4 November: Treasury buyback program review and next Quarterly Refunding.
Recommendations (Staged)
Immediate, gold: Stay long, raise the target to $4,800 over three months, bank partial profit above $4,700, add toward $4,400. Treat the $4,500 target as met, not as a reason to chase.
Immediate, rates: Keep front-end duration (2-5 year) where the September hold is priced; cut the steepener while Treasury buybacks flatten the long end through 4 November. Add long-end only above a 4.85% 10-year.
Immediate, oil: Sell Brent strength above $93 toward the mid-$80s, small size; hold the long-dated call tail against a shock; cover the fade if a strike lands.
Near-term, Europe: Hold the ECB-hike-via-front-end position and European financials into 10 September; retire the EUR/USD fade above 1.16 while the dollar is being softened.
Near-term, UK: Keep the constructive GBP tilt above 1.34, add dips toward 1.34, stay cautious on gilts.
Near-term, Australia: Keep neutral-to-small-long AUD, buy dips toward 0.69, trim toward 0.725.
Structural, India: Stay long large-cap IT exporters and financials, add IT on the turn, trim only if USD/INR breaks 96.5.
Structural, credit: Overweight IG over HY; begin HY scale-in only past 3.25 to 3.5% OAS; watch the 4 November buyback wind-down.
Structural, Japan: Hold the short USD/JPY toward 150 as a contrarian position, small, into the 17-18 September BoJ meeting; cut it if the hike fails and the pair breaks above 161.
Thresholds That Change the Calls
10-year above 4.85% despite buybacks: add long-end duration to the front-end book.
10-year below 4.55%: reduce front-end risk, the long-end thesis is normalising.
A confirmed Hormuz strike or blockade tightening: re-arm the oil long, drop the fade.
Confirmed flow normalisation or a China clash that removes Iranian barrels: reassess the fade both ways.
Brent above $93: hold the sell-strength fade.
HY OAS past 3.25 to 3.5%: begin HY scale-in.
Gold above $4,700: trim; toward $4,400, add.
DXY reclaiming 102: revisit long-dollar; below 98, press the short.
USD/INR breaks 96.5: trim the India tilt.
USD/JPY above 161 with no BoJ move: cut the short.
Markers of Concern
Levels are a dated snapshot, mostly 21 to 24 August close or intraday; markets move.
Scenarios are probabilities, not forecasts.
The long end is now managed by Treasury buybacks that end 4 November, so the flattening force is policy and can reverse; a correct macro call does not guarantee a correct duration call.
Gold has run far and fast; a stronger dollar or a hawkish Warsh at Jackson Hole could force a sharp pullback in the trim zone.
The Iran sanctions plan could escalate into a US-China clash that reprices both oil and equities faster than any single strike would.