Research reports
Each weekday morning the research loop takes the standing macro map’s weakest assumption and pressure-tests it against primary sources. The memo states what the evidence changed, the call at the margin, and the observations that would flip it.
August 2026
- August 31, 2026
The weakest assumption fails in a specific way: the ETF era created a large, custodian-controlled Bitcoin demand channel that exchange netflow cannot fully see. That supports downgrading the raw pulse, but it does not prove that every exchange-flow print is useless.
- August 29, 2026
The weakest assumption is not that EM markets reopened; it is that reopening proves durable, broad solvency. The evidence supports a real but liquidity-sensitive window with two-speed access, not a clean normalization.
- August 28, 2026
The weakest assumption does not survive the latest evidence: current positions and flows no longer point to a single synchronized risk bundle. The risk is real in volatility and parts of gold and semiconductors, but the evidence supports a narrower action than blanket avoidance.
- August 27, 2026
The weakest assumption fails in its strongest form: recent evidence does not show that duration has become a uniformly bad hedge, but it does show that long-end duration carries meaningful inflation, supply, and positioning risk.
- August 26, 2026
The single weakest assumption is that a recent spot-dollar pullback is already a durable regime break. Current official evidence does not clear that bar: price has softened, but policy remains relatively dollar-supportive, cross-border capital is still flowing into US securities,
- August 25, 2026
Yes. Move from passive CONSIDER to targeted ACT/HEDGE, but do not make a single-node packaging bet. The weakest assumption was the ranking claim that one bottleneck explains all accelerator lead times and that front-end capacity is irrelevant. Current primary disclosures instead
- August 24, 2026
The weakest assumption is that a larger 65+ population automatically creates a broad, near-term consumer-demand trade. Current data supports demand reallocation into age-linked essentials and care, not a generalized consumer beta.
- August 22, 2026
The weakest assumption fails in its broad form: multifamily refinancing stress is real, but current evidence does not show a generalized bank-solvency event.
- August 21, 2026
The weakest assumption is not whether electricity demand is rising; it is whether the headline pipeline becomes contractual, buildable load that earns regulated or contracted returns. Current evidence supports changing the action now, but only to a selective grid-bottleneck tilt.
- August 20, 2026
The weakest assumption is not that new nuclear plants can arrive before AI load; it is that nuclear's near-term physical limits make it irrelevant to near-term power decisions. That is too strong, but only at the margin.
- August 19, 2026
The weakest assumption fails in its strongest form: long global semiconductors is still labeled the most crowded trade, but the reported crowding has fallen to 52-53% from 82-83%, while futures and flow evidence show substantial de-risking.
- August 18, 2026
The weakest assumption was that dollar softness is already becoming a funding disorder. The evidence does not support that escalation: the New York Fed says the broad dollar was little changed in Q2 2026, and the Fed and Treasury did not intervene in FX markets [120.2-3]. Current
- August 17, 2026
The weakest assumption is the binary description of China's capital account as effectively closed. China's own foreign-exchange authority reports meaningful RMB use in capital-account transactions, so the marginal diversification channel is real; it still has not become a reserve
- August 15, 2026
Maintain action=avoid. The single weakest assumption of the prior thesis - that Hormuz disruption would normalize "soon" enough to vindicate the bear case - has now been retested by 167 consecutive days of effective closure and falsified in the wrong direction: Brent rose to $88.
- August 14, 2026
The fiscal and term-premium risk is real, but the weakest link - an already broad, supply-constrained capex regime with immediate funding stress - is not confirmed.
- August 13, 2026
The weakest assumption is that grid demand evidence automatically becomes durable, high-return earnings across the whole supply chain. Current evidence supports demand, but only selected operators have demonstrated conversion from orders and backlog into cash and margin.
- August 11, 2026
The weakest assumption is not that crowding exists; it is that the crowding is one synchronized short-real-rates/long-duration-risk factor. The latest flow and positioning evidence rejects that stronger version.
- August 10, 2026
The weakest assumption is treating the absence of a new broad default wave as evidence that EM debt risk is largely gone. The current evidence supports a narrower conclusion: no broad domino is the base case, but the nonbank funding channel remains fragile and country-concentrate
- August 8, 2026
The weakest assumption is that the labor-supply shock is already a broad, identifiable scarcity shock. The latest flow data show marginal tightening, but they do not yet separate aging-driven scarcity from weaker demand.
- August 7, 2026
The weakest assumption is that a positive 2026 software-AI productivity signal is already broad and large enough to offset the buildout's resource costs. The current official data do not pass that test.
- August 6, 2026
The weakest assumption is that oral delivery automatically converts GLP-1 efficacy into durable, superior economics. Current evidence supports oral adoption, but not that economic conclusion: pills are gaining users while efficacy gaps, gastrointestinal discontinuation, and net-p
- August 5, 2026
The weakest assumption is using a clean historical market tape to infer that common-dependency cyber risk is not investable or operationally actionable. The evidence supports a narrow risk upgrade, not a broad macro move.
- August 4, 2026
The weakest assumption was that one market-cycle label could guide every insurance and reinsurance decision. Current evidence confirms real, capital-supported softening in property catastrophe, but not a universal soft market: casualty remains loss-experience-sensitive and underl
- August 3, 2026
The weakest assumption was that a geopolitical energy shock automatically becomes persistent, broad inflation. Current data do not support that conclusion yet: July euro-area inflation shows a real but modest move in underlying measures, while expectations and wage pressure are s
- August 1, 2026
The weakest assumption - that 2026 CRE maturity pressure is already becoming a broad bank-credit event - is not supported by current data. The risk is real and concentrated in CMBS workouts, office, retail, and weaker multifamily exposures, but realized bank losses and systemwide