North American
Portfolio

The S&P 500 finished July broadly unchanged, but the index endpoint concealed a much more volatile and highly dispersed month. The index declined marginally, recording its first negative July since 2014 and a second consecutive monthly loss, while the Nasdaq Composite fell 3.2%. From its intra-month peak, the S&P 500 declined 3.4% and the Nasdaq drew down 7.0%.

Dispersion, rather than broad market direction, was the defining feature. Equal-weighted equities outperformed the capitalisation-weighted index as leadership shifted away from technology and semiconductors toward energy, financials and other value-oriented sectors. The semiconductor complex fell 22.1% as investors reassessed the timing and magnitude of returns from record AI capital expenditure. Importantly, spending levels remained strong; the adjustment occurred primarily in the valuation multiples attached to continued capex acceleration, scarcity and exceptional margins.

The Federal Reserve held the funds rate at 3.50–3.75% for a fifth consecutive meeting, although three policymakers dissented in favour of a 25 basis point increase. Softer inflation and employment data were offset by the Fed’s continued focus on price stability. Longer-term Treasury yields subsequently moved higher while the two-year yield eased, producing a bear steepener that supported financials and placed further pressure on long-duration growth equities.

Energy was the strongest sector, gaining 12.6%, as renewed US–Iran hostilities pushed Brent crude toward US$86 per barrel. Financials advanced 5.6%, while information technology declined 2.0% and industrials fell 2.7%. The rotation reinforced the importance of maintaining diversified exposure across the AI infrastructure and energy transformation value chains rather than relying exclusively on semiconductor leadership.

The VIX ended July at 15.99, below its June close, despite briefly reaching 20.66 following the FOMC meeting. Subdued index volatility alongside elevated sector and single-stock dispersion created a more favourable environment for selective volatility-carry, skew and stock-specific option strategies than for broad index hedging. For Australian investors, the 1.5% appreciation in the Australian dollar against the US dollar created an equivalent headwind for unhedged USD portfolio returns.

North American Energy Transition Q2 2026: The Interconnection Bottleneck Becomes a Policy Story

The defining shift this quarter was not demand growth that has been the story since 2024 but the migration of the constraint from capital availability to physical delivery. Washington increasingly treated the interconnection bottleneck as a national policy priority rather than a market imbalance that would resolve on its own.

FERC’s 18 June intervention was the quarter’s most consequential policy event. Rather than pursuing a multi year rulemaking, FERC used Section 206 show cause authority to give all six RTOs and ISOs 60 days to justify or rewrite their large load interconnection tariffs, alongside a 30 day resource adequacy reporting requirement. Under the emerging framework, data centres would bear their own interconnection costs, helping to protect ratepayers while reducing the procedural delays that have made grid connection the longest lead time item in data centre development.

This represents a genuine regime change in how large loads not only generation are regulated at the federal level. Implementation will not be frictionless. State regulators, including NARUC, have argued that federal standardisation could limit states’ ability to respond to regional conditions and protect affordability. Litigation and jurisdictional risk are therefore likely to run alongside implementation through the third and fourth quarters.

Co location is becoming the dominant workaround, and it is reshaping capital allocation. On site and behind the meter generation is projected to account for approximately 30% of new data centre capacity in 2026, up from near zero a year earlier, with some forecasts suggesting it could reach 50% as hyperscalers secure direct generation partnerships. This creates a structural tailwind for gas turbine OEMs, SMR developers, storage providers and distributed power platforms.

The same shift also introduces stranded asset risk for utilities. Capacity planned to serve large load customers may never be required on the public grid if hyperscalers increasingly bypass traditional interconnection pathways. The key investment distinction is therefore moving from simple exposure to power demand toward exposure to the assets and technologies that shorten time to energisation.

The gas versus renewables mix has moved more decisively toward gas than consensus expected six months ago. Natural gas’ share of planned data centre capacity increased from 11.1% to 18.1% between 2024 and 2026. Non renewable additions rose approximately 71% from 2025 to 2026, while renewable growth flattened to around 2%. One important driver is economics: natural gas interconnection costs are estimated at roughly one tenth those of solar and offshore wind.

For the portfolio thesis, this complicates the view that AI demand will pull clean energy deployment forward in a straight line. In the near term, AI demand is accelerating gas deployment more rapidly, while decarbonisation appears as a second order effect through storage, SMRs, efficiency and distributed generation rather than through bulk renewable grid supply alone.

The issue has also become a rates and inflation story. Data centre driven electricity demand has been estimated to add approximately 0.1 percentage points to core inflation in both 2026 and 2027, concentrated in PJM states, alongside a 2.3% year on year rise in national retail electricity prices. This creates a new macro transmission channel linking AI infrastructure demand, regional power inflation and Federal Reserve policy.

Q3 watch list: RTO compliance filings due in mid August; additional state level pushback or litigation; and whether the PJM emergency capacity auction structure under which technology companies fund new plants directly is replicated elsewhere. These developments should provide the clearest read through on which infrastructure companies capture the co location capex cycle and which utilities face the greatest stranded asset risk.

Reading the Market’s Second Layer

A calm index over a violent rotation.

On the surface, July looked relatively contained. The S&P 500 finished almost unchanged and the VIX closed at 15.99, below its June close of 16.45. Beneath that calm endpoint, however, the S&P 500 declined 3.4% from its intra-month peak and the Nasdaq drew down 7.0%, while sector and single-stock performance diverged sharply.

We don’t only watch the index. We track equity risk on three levels: what the options market charges for protection on the index, what it charges on the individual companies inside it, and how options dealers are positioned — because their hedging either cushions the market or accelerates it.

The first two levels have precise, published measures. The VIX prices thirty-day volatility on the S&P 500 as a whole. Its newer companion, the Cboe S&P 500 Constituent Volatility Index (VIXEQ), applies the same calculation to the individual stocks inside the index, weighted by their size — in effect, what the market charges to insure the average large American company rather than the basket. The two can diverge dramatically, and the gap between them is itself an index: the Cboe Dispersion Index (DSPX). When constituent volatility (VIXEQ) is high but index volatility (VIX) is low, the arithmetic permits only one explanation — stocks are expected to move a lot, just not together. The degree to which they move together is correlation, and it can be read directly from these three numbers.

July again demonstrated why index volatility alone can provide an incomplete picture of portfolio risk. Energy gained 12.6% and financials rose 5.6%, while semiconductors fell 22.1% and information technology declined 2.0%. The equal-weighted S&P 500 outperformed the capitalisation-weighted index as value and cyclicals offset the repricing of long-duration growth. The subdued index therefore reflected offsetting moves across sectors rather than an absence of risk.

The third layer: how dealers are positioned

Price alone doesn’t tell us how a shock will behave once it starts — whether it will be absorbed or amplified. For that we watch options dealers’ aggregate gamma exposure (GEX): the degree to which dealers must buy into a rally or sell into a decline to stay hedged, as a mechanical consequence of the options positions they’ve written to the market. When dealer gamma is positive, their hedging flow leans against the market and dampens moves — a “pinning” effect. When it turns negative, their hedging flow leans with the market and can accelerate a move in either direction. This is not a market view; it is a structural fact about who holds which options, and it changes day to day.

The month’s clearest index-volatility event came on FOMC day, when the S&P 500 reached its monthly low and the VIX briefly rose to 20.66. The spike was meaningful but short-lived: the VIX finished the month below where it began even as the Nasdaq and semiconductor complex sustained material losses. This was consistent with a concentrated repricing of AI-related exposures and a rotation across sectors, rather than a persistent market-wide liquidation. Dealer gamma remained an important timing input around the event, helping distinguish a temporary acceleration in hedging flows from a durable deterioration in broad market risk.

Implied volatility and where the yield comes from

The yield side of the programme lives inside the same data. July’s dispersion kept option premiums on many individual holdings materially richer than index volatility, particularly across semiconductors, AI infrastructure and other high-beta exposures. Selective covered calls could therefore capture more income per unit of upside surrendered than index-level writing, while elevated single-stock skew improved the economics of collars and put spreads around the portfolio’s most valuation-sensitive positions.

The distinction between cheap index volatility and expensive stock-specific volatility was particularly important in July. Broad index hedges offered limited protection against a momentum reversal concentrated in semiconductors and AI-linked equities, because gains in energy, financials and defensive sectors dampened the index decline. Where portfolio risk was concentrated, stock-specific collars and puts provided a more direct hedge than simply increasing broad S&P 500 protection.

At the same time, low index volatility preserved relatively inexpensive protection against a future rise in correlation. We therefore continued to separate two risks: stock-specific repricing, managed through targeted collars, puts and premium harvesting; and systemic risk, managed through selectively purchased index protection before volatility expands. Dealer positioning remains the timing mechanism that determines when hedging flows are likely to absorb a shock and when they may amplify it.

July was therefore a stock-picker’s and structure-selector’s market. The opportunity lay in harvesting rich single-stock volatility and managing risk at the source, while retaining inexpensive index insurance against the possibility that dispersion eventually gives way to a broader correlation shock.