Matryoshka market: Nested convictions in equities

• Geopolitical resolution is the master assumption. Equity markets are pricing a framework agreement by September, consistent with the prediction market consensus and the pronounced kink in VIX term structure at Q3. That bet has logic: the global (political) cost of ongoing conflict and mid-term elections make a deal valuable to the US administration by autumn. But a genuine agreement requires Iran to provide credible security guarantees for the Strait of Hormuz. Tail and downside risks are visible in options but not in equity valuations. In our downside scenario, equities could correct by 25% and 20%, respectively, in Europe and the US. Indeed, if resolution fails and oil remains structurally elevated, every subsequent layer inherits a more adverse starting point: margins compress, central banks are constrained and the stagflation risk premium returns not as a tail scenario but as a working assumption.
• The earnings story is concentrated exposure to some sector tailwinds, not broad-based resilience. The S&P 500 has delivered double-digit EPS growth for six consecutive quarters, and European corporates surprised with Q1 growth of +11.5% y/y. Record margins should be read as limited remaining buffer. Beyond the headline figure, approximately 17pps of S&P 500 EPS growth in Q2 2026 originate from the IT sector alone – growing at an estimated +66% y/y and representing roughly 31% of total index earnings. Excluding technology, the rest of the S&P 500 expands at around +12%, in line with a 6% nominal GDP world. Europe tells a parallel story: exclude energy, the direct beneficiary of elevated crude, and STOXX 600 EPS growth falls from +11.5% to approximately +7%. Furthermore, a 20% equity correction would additionally impair US consumption by an estimated 1-1.5pp of GDP through the wealth-effect channel – a feedback loop markets are pricing as exogenous.
• Non-cyclical index composition concentrates rate sensitivity rather than reducing it and the pending mega-IPO wave deepens the crowding. The shift in major equity indices away from cyclicals toward technology, healthcare, utilities and high-dividend defensives is read as structural insulation. Non-cyclical sectors divide into bond-proxy profiles (utilities, healthcare) and long-duration growth profiles (large-cap technology): both are materially sensitive to real interest rates, through yield competition and discount rate mechanics respectively. The index has not removed rate risk, it has hidden it in less visible form. The pending mega-IPO wave deepens this structural vulnerability. SpaceX (USD1.75trn), OpenAI (USD850bn) and Anthropic (USD850bn) – combined USD3.45trn – would represent approximately 4-5% of S&P 500 market capitalization at index inclusion, making their combined entry the largest supply event in US equity market history. All three would land in or adjacent to technology, further crowding the most rate-sensitive index segment.
• The real equity risk premium is a breakeven, not a buffer. The negative nominal equity risk premium (ERP) is rationalized two ways: equities offer inflation protection that nominal bonds do not, and dominant technology constituents carry a structural growth premium that justifies a negative nominal spread, by analogy with long-duration bonds whose value sits in distant cash flows. On the real ERP measure (i.e. earnings yield minus real risk-free rate) the picture is thinner still, sitting at approximately 2.5%. That is not a buffer, it is a breakeven. A rise in real rates of 50–75bps eliminates it entirely. At that level, the reallocation from equities into investment-grade credit becomes a mechanical trade for liability-matched institutional investors. The multiple stays compressed until either real rates fall or earnings growth widens the premium. In a scenario where geopolitics and growth have simultaneously deteriorated, neither condition arrives quickly.
• Technical positioning amplifies fundamental moves non-linearly while the mega-IPO pipeline adds further risk. The 2025–2026 rally has been amplified by structural technical tailwinds: CTA trend-following models near maximum long, option dealers suppressing intraday volatility and FINRA margin debt near cycle highs. Each is mechanically reversible, and the reversal is self-reinforcing rather than gradual. A 5-7% fundamental-driven decline can cascade into a 15–20% technical drawdown before positional overhang clears. The mega-IPO pipeline introduces another amplifier poorly captured by conventional risk models. Passive vehicles must rebalance to new index weights on a defined schedule, selling existing holdings irrespective of valuation or momentum. These rebalancing flow would represent some of the largest single-session institutional selling pressure seen outside of crisis conditions, enough to breach the technical thresholds that convert a managed correction into a self-sustaining cascade without any prior fundamental deterioration. A lockup expiry cliff at approximately 180 days post-listing adds a second, later-dated supply wave.
• In the adverse scenario, all three policy levers are simultaneously constrained. Since 2010, equity pricing has reflected not only fundamental value but policy optionality – the assumption that a geopolitical deal, fiscal stimulus or central bank rate cuts are available in any deterioration. The underpriced risk is not the absence of any single lever but the correlation of constraints across all three. The geopolitical channel has been partially activated - ceasefire negotiations are themselves a policy response. The fiscal channel has structural limits: Europe‘s substantial post-2022 commitments to energy security and industrial policy did not fully offset structural competitiveness damage from sustained energy repricing. The central bank channel requires either sovereign market dysfunction or a banking stress event to activate – high-bar triggers in an inflationary environment. 

As of May 2026, the S&P 500 and Euro Stoxx 600 sit within a few percent of their all-time highs. Implied volatility has spiked on event days but mean-reverted quickly. It seems that equity markets are looking through the war in Iran – one of the most geopolitically significant events in the Middle East in a generation. However, the market is not ignoring the war. It is pricing a specific outcome: resolution over the summer. That outcome is also holding a set of convictions on other key issues from interest rates to earnings to economic policies. In that sense equity markets can be seen as a Matryoshka doll: it is not a deceptive object, but the outer vivid layer hides how many layers remain and what sits at the center.

The assumption of a resolution makes sense but it comes with risks. The options market's implied volatility term structure shows a pronounced kink around Q3 2026: elevated near-term volatility fading sharply into year-end, consistent with pricing a binary resolution event. Confirming this conviction, the prediction market for an Iran ceasefire or framework agreement by September sits above 60%. Implicitly, what market participants and analysts are expecting is that the political cost of the ongoing crisis will lead to a resolution – one way or the other. Surging inflation, fuel shortages, the potential social and economic risks coupled with the perspective of US mid-term elections make a deal by autumn “inevitable”. This is not an irrational bet. But it is a specific, time-bound, politically-conditioned bet. If it was to break down, it would have more consequences down the line.

The critical issue is the concession structure required for a genuine agreement. A durable deal requires Iran to move simultaneously on two fronts: a verifiable freeze or rollback of its nuclear program, and some form of security guarantees for the strait of Hormuz. A partial framework agreement is plausible and may be sufficient to sustain equity markets temporarily. But a nominal settlement that unravels, a spoiler event – proxy escalation, third-party strike – or a US administration that claims resolution without genuine Iranian compliance, would leave the problem intact. The probability-weighted payoff from being long equities in a world where a September resolution fails is underrepresented in current levels. If the conflict persists and oil remains structurally elevated, the stagflation risk premium returns. In our downside scenario, equities could correct by 25% and 20%, respectively, in Europe and the US. Brent reprices higher, energy input cost pass-through hits margins and inflation persistence that follows constrains central bank flexibility. Every subsequent layer of analysis would inherit a more adverse starting point.

The earnings picture entering 2026 has been much better than feared. The S&P 500 index has been recording double-digit earnings growth for over six consecutive quarters now, and the trend is expected to continue throughout 2026. On the other side of the Atlantic, the STOXX Europe 600 surprised with Q1 earnings jumping by +11.5% y/y (Figure 3). Europe, often dismissed as structurally lagging, is increasingly demonstrating that earnings resilience is becoming a cyclical force of its own. Revisions have been net positive for two consecutive quarters. Growth forecasts have held – US GDP is tracking above 2%, European growth has recovered modestly from the 2024-25 soft patch and inflation has remained below the 3.5-4% threshold that historically triggers meaningful demand destruction. Corporate margins are at or near record highs. The market reads this as durable fundamental support: a solid earnings floor. Inflation is treated as a manageable pass-through phenomenon that is fueling turnover and not hurting volume significantly. The implicit assumption is that margin resilience reflects a relative pricing power.

Record margins are a sign of limited remaining buffer. When margins are at historical peaks, the distribution of outcomes is skewed: the scope for positive earnings surprise is narrow, while the scope for compression if costs rise or volumes soften is substantial. Markets are treating elevated margins as a baseline from which to project forward earnings. They should instead be read as a vulnerability indicator – the starting point from which any adverse shock produces a larger-than-average revision cycle.

Behind the headline: one sector is doing the heavy lifting. Aggregate earnings growth can provide valuable insights, though it is important to consider the nuances to make a fully informed assessment. Figure 2 (see pdf) breaks down the S&P 500's current 12-month forward EPS trajectory, comparing it to the historical relationship between corporate earnings and nominal GDP growth. This relationship has shown remarkable stability across cycles from 2012 to 2025. The conclusion is clear: if we exclude the technology sector, the remaining 69% of S&P earnings align with the regression line, with an estimated growth rate of around 12% year-on-year, in line with the 6% nominal GDP growth rate. This indicates a stable and predictable performance, consistent with market cycles. The macro is performing its intended function. The addition of technology has led to a 29% increase in EPS growth, a level that would typically require an 8-10% nominal GDP to maintain. The index is not being priced according to the economic conditions of the market it is intended to represent. It is being priced for one sector's ability to defy the economy it inhabits.

The arithmetic behind this discrepancy matters. It is estimated that approximately 17pps of headline S&P EPS growth in Q2 2026 originate from the IT sector, accounting for roughly 31% of total index earnings. This sector is projected to expand at an estimated rate of +66% year-on-year. This pace is not indicative of the broader earnings cycle, but rather a capex monetization thesis, concentrated in a select group of names, that is meaningfully ahead of any macro anchor. The remaining components of the index are expanding at a rate that aligns with the broader economic environment. However, this implies limited potential for further growth and limited capacity to offset deficiencies in other components of the index.

This reframing of the asymmetry argument provides greater precision. The concern is not that margins will compress uniformly across the index; it is that a single sector will decelerate. A moderation from +66% to a still-exceptional +30% IT earnings growth – the kind of deceleration that would accompany any credible concern about AI capital spending returns, hyperscaler spending fatigue or monetization delays – would have a materially negative effect on the headline S&P EPS number, with limited offset from the rest of the market. The index's earnings recovery, which is currently supported by a broad range of stakeholders, is, in essence, a highly concentrated exposure to a single factor: the continued compounding of AI capex monetization at its current rate. The "record margin" story is, arithmetically, a single-sector story – strip out IT and the index is operating in line with a 6% nominal GDP world.

One underappreciated risk is the circular reference in the current macro configuration. US equity markets and US consumption are not independent variables. Household wealth effects from equity performance materially influence consumer spending, which in turn supports corporate revenues and earnings. A 20% equity market correction – the order of magnitude in our downside scenario – would impair US consumption by an estimated 1-1.5pp of GDP. That growth impairment feeds back into earnings revisions, which would extend the equity correction. Markets are pricing equities as though consumption is an exogenous growth driver but it is partly endogenous to equity performance itself.

Q1 2026 corporate results and earnings revisions paints a two-speed market.  While both regions show improving forward EPS trajectories into 2026, sector dynamics are playing a huge role in the diverging performance across the Atlantic. While the continued earnings expansion of the S&P 500 remains largely concentrated within the technology sector (EPS: +55% growth in Q1) – primarily driven by AI-related capex acceleration and hyperscaler monetization dynamics – the STOXX Europe 600 (historically weaker and more volatile) has recently shown a meaningful improvement in market sentiment and earnings expectations, despite Europe’s historically more fragile and cyclical macro backdrop. After delivering the worst earnings growth in seven quarters in Q4 2025 (-2.0% y/y), European corporates are heading to the best quarter in three years, recording a double digit growth of +11.5% y/y not seen since Q1 2023 (Figure 3). Yet, this noteworthy growth is fundamentally explained by one single sector: energy, registering an EPS growth of around +50% in Q1. Excluding this sector, the STOXX-600’s average EPS growth rate is around +7%. The escalation of geopolitical tensions in the Middle East has sustained elevated crude prices and widened upstream profitability, allowing major oil & gas companies to deliver a substantial expansion in operating margins and cash-flow generation, ultimately translating into significantly stronger P&L performance across the sector. Furthermore, consensus earnings expectations for the upcoming quarters of 2026 have materially improved as well, supported by the strong momentum in the European energy complex. By the end of 2025, EPS estimates for Q2 and Q3 growth pointed to +3.1% and +6.6%, respectively, while current estimates point to double-digit figures of +12.8% and +14.2%. This upward revision reflects not only a sustained period of elevated crude prices following more than three months of geopolitical escalation in the Middle East, but also the increasing likelihood that heightened energy market volatility will persist well into 2026, supporting structurally stronger upstream earnings and reinforcing the positive revisions momentum across the sector.

European Q1 2026 earnings dynamics point to a market that is resilient on the surface but highly differentiated at the sector level. The entire universe of European corporates have reported aggregated EPS growth of +10.3% y/y, achieved despite an exceptionally subdued top-line environment, with revenues expanding only +0.3%, underscoring the extent to which European corporates have leaned on pricing discipline, cost-out programs and operating leverage to preserve margins in a still-fragile macro and sentiment backdrop. Sector dispersion has been a key feature: Energy stands out as the primary earnings engine, benefitting from the sustained geopolitical tensions and the elevated crude dynamics, while marine transport delivered outsized profitability amid high bunker oil prices and limited pricing power. At the same time, several traditionally cyclical, industrial-heavy segments – including chemicals, machinery, paper & forest products and automobiles – emerged as notable underperformers, reflecting a combination of structurally elevated fixed-cost bases (particularly energy-linked input costs in chemicals and paper) and higher demand elasticity, which together amplified margin pressure over the period.

Unlike the US market where earnings expansion remains heavily concentrated in a narrow set of tech-driven growth engines, Europe’s earnings resilience appears materially broader across the real economy. This reduces concentration risk at the index level, but simultaneously leaves the region more exposed to cyclical and commodity-sensitive dynamics rather than to a single dominant secular growth theme. The divergence is particularly evident in the composition of earnings leadership. In Europe, upgrade momentum has been driven predominantly by energy and other macro-sensitive sectors, reflecting the region’s higher operating leverage to commodity prices and geopolitical developments. By comparison, the US earnings profile remains structurally more durable and less dependent on commodity-linked upside, with growth continuing to be anchored by technology-intensive industries. As shown in Figure 5, semiconductors (+39.1% y/y) and technology hardware (+22.4%) recorded the strongest revenue growth rates of any US sector in Q1, reinforcing the extent to which AI-related capital expenditure and digital infrastructure investment continue to dominate the US corporate earnings cycle. Paper & forest products stood out as the clearest earnings casualty of the quarter in the US, caught between deteriorating pricing dynamics, structurally elevated energy and freight costs, and weakening end-market demand, a combination that effectively erased operating leverage across the industry.

Non-cyclical composition concentrates rate sensitivity. The S&P 500 and Euro Stoxx 600 carry a significantly higher weight in non-cyclical sectors than in prior cycles. Non-cyclical sectors can be divided into two groups: bond-proxy profiles (utilities, healthcare, high-dividend energy stocks) and long-duration growth profiles (large-cap Technology). Both are materially sensitive to real interest rates – the former through yield competition and leverage cost, the latter through discount rate effects on terminal value. The index shift away from cyclicals has not reduced rate sensitivity but it has concentrated it in a less visible form. In a rising real rate environment, the "defensive" parts of the index are among the most exposed and the earnings quality argument for holding them dissolves precisely when it is most needed.

The anticipated listings of OpenAI and Anthropic, alongside the already-discussed SpaceX entry, represent a structurally distinct event from the ordinary IPO cycle. Together, these three companies carry private-market valuations of USD1.75trn (SpaceX) and USD850bn ( for each OpenAI and Anthropic) – a combined USD3.45trn, roughly equivalent to the entire current weight of the financial sector in the S&P 500.  This trend signals continued risk appetite and is expected to deepen index liquidity, attract fresh capital inflows and sustain the momentum dynamic that has supported equity returns in early 2026. Their combined entry into capitalization-weighted indices would be the largest supply event in US equity market history, and their sectoral concentration compounds the risk. All three would land in, or adjacent to, technology – the index segment already identified as carrying the most concentrated rate sensitivity. The simultaneous addition of three megacap AI-and-infrastructure names does not diversify the index; it deepens a crowding that already exists.

The mechanics are straightforward but their scale is not. Passive and semi-passive vehicles – which now account for over 55% of US equity AUM – are rules-bound buyers at index inclusion. They have no discretion on timing or price: they must acquire the weight the index assigns, funded by proportional sales of every other existing constituent. For additions of this size, the required reallocation is not a rounding error. It is a structural flow event, one that persists for the adjustment period and leaves the index systematically lighter in every other sector. At current index weights, a combined entry at this scale would represent 4-5% of S&P 500 market capitalization on day one of full inclusion – forcing capital redistribution that would suppress every incumbent constituent's relative performance during the rebalancing window – regardless of fundamentals. The AI premium that has inflated Technology valuations through 2025 and into 2026 would, paradoxically, be the mechanism through which the rest of the index is diluted. Anticipation of these listings partially prices in the demand side, the supply-side redistribution impact on incumbents remains largely absent from current consensus.

The real equity risk premium is a breakeven, not a buffer. The negative nominal equity risk premium is rationalized two ways. Equities are real assets whose cash flows index to inflation, while nominal bonds do not – in an inflationary regime, the inflation-protection argument compresses the ERP markets are willing to accept. The structural growth premium of dominant technology constituents is treated as deserving of a negative nominal premium, by analogy with long-duration bonds where value sits in distant cash flows. On the real ERP measure – earnings yield minus real (inflation-adjusted) risk-free rate – the picture looks more supportive, sitting at approximately 2.5%. This is thin by historical standards but positive, and the market treats it as sufficient justification for current institutional equity weights.

The inflation-hedge argument is valid, but only while real rates remain contained. It breaks down precisely when it is most needed: in a regime where inflation is persistent and central banks respond by keeping real rates elevated. In that environment, the inflation-hedge and the rate-headwind operate on the same asset simultaneously, and the rate headwind dominates for non-cyclical, long-duration equity profiles – which is most of the index. The competitive pressure from credit is more immediate than equity multiples suggest. Investment-grade credit at BBB spreads of approximately 120bps, or high-yield at approximately 180bps, delivers real returns comparable to equities, with materially lower volatility, senior claim in the capital structure, and defined maturity.

The 2.5% real ERP is not a buffer; it is a breakeven. A rise in real rates of 50-75bps eliminates it. At that point, the reallocation from equities into investment-grade credit is a mechanical, structurally-driven trade – not a sentiment event, and not reversible until real rates fall or earnings growth widens the premium again. The de-rating that follows is structural rather than cyclical, driven by the rate level and not by risk appetite. The multiple stays compressed until the underlying arithmetic turns; in a scenario where geopolitics and growth have also deteriorated, neither condition arrives quickly.

Technical positioning amplifies fundamental moves non-linearly. The 2025-2026 rally has been supported by structural technical tailwinds as well as by fundamentals. Commodity Trading Advisors (CTA) trend-following models are near maximum long exposure, reflecting the sustained upward trend in major indices. Options dealer positioning – driven by concentrated retail options activity at round-number strikes – reinforces trend directionality. FINRA margin debt (i.e. funds that investors have borrowed from brokers) has recovered from its 2022 lows to levels near cycle highs. Markets read these conditions as healthy indicators of broad participation and sustainable momentum.

Each of these tailwinds is mechanically reversible, and the reversal dynamic is self-reinforcing rather than gradual. CTA momentum models do not taper when trend signals reverse, they flip. The transition from maximum long to maximum short can occur within days, concentrating supply at precisely the moment other sellers are also emerging; a relatively modest move can initiate selling that amplifies into a substantially larger one. Margin debt creates a third channel: leveraged holders facing margin calls cannot wait for fundamental improvement, and the supply they generate is price-indifferent. The combined effect is a non-linear response to adverse fundamentals. A 5-7% fundamental-driven decline can cascade into a 15-20% technical drawdown before the positional overhang clears. Recovery requires not just fundamental stabilization but a full clearing of technical positions – CTAs rebuilding long exposure, margin debt being reduced etc. – a process that typically extends over weeks to months, well beyond what fundamental analysis alone would project.

The mega-IPO pipeline introduces a technical amplifier that operates through a distinct channel that is poorly captured by conventional risk models. Large index inclusions generate predictable, time-compressed, price-insensitive flows: passive vehicles must rebalance to the new index weights on a defined schedule, selling existing holdings in proportion to their current weight irrespective of valuation or momentum. This is mechanical selling driven by the arithmetic of capitalization weighting. For listings at the scale now under discussion, the rebalancing flow would represent some of the largest single-session institutional selling pressure seen in US markets outside of crisis conditions.

The interaction with existing technical positioning creates an adverse compounding effect. CTA momentum models hold existing large-cap technology names near maximum long. In such rebalancing-driven supply event in precisely those names (forced, scheduled, and price-indifferent), the ordinary self-reinforcing dynamics take hold: dealers sell into weakness, CTAs follow trend signals, margin calls cascade. The IPO supply event would not need to be the proximate cause of a broader market dislocation, it only needs to be large enough to move prices through the technical thresholds that convert a controlled rebalancing into a self-sustaining cascade. The lockup expiry cliff – typically 180 days post-listing, at which point insiders and pre-IPO institutional holders become sellers – adds a second, later-dated supply wave that extends the technical overhang well beyond the initial listing date. Markets are not pricing either wave.

The policy safety net is thinner than markets assume. A defining feature of equity behavior since 2010 has been the implicit assumption that policy will intervene when markets deteriorate sufficiently. This has been validated repeatedly: QE programs, fiscal stimulus, emergency rate cuts and geopolitically-motivated pivots have all been deployed in ways that supported equities. The result is that pricing reflects not just fundamental value but policy optionality. In 2026, markets appear to be pricing at least one of three possible interventions as available in any adverse path: a geopolitical deal (the TACO put), fiscal stimulus or central bank rate cuts. The implicit assumption is that whichever lever activates, the response will be sufficient to prevent a full cascade.

The underpriced risk is not the absence of any single put, it is the correlation of constraints across all three. The geopolitical channel has been partially activated: the ceasefire negotiations themselves are the response. But a full activation requires concessions exceeding prior pivots (e.g. verifiable nuclear rollback from Iran plus a reparations framework for Iran) and an administration that claims a partial settlement as a complete victory would exhaust this option without resolving the underlying risk. The fiscal channel has structural limits that Europe has already demonstrated. Substantial fiscal commitments to energy security, defense and industrial policy since 2022 provided some support, but they could not offset the structural competitiveness damage from sustained energy-price repricing, not for lack of fiscal support, but because energy-intensive sector damage is not curable by spending alone.

The central bank channel is the most powerful but the most constrained in this specific adverse scenario. Monetary easing at scale requires either dysfunction in sovereign bond markets or a banking-sector stress event – high-bar triggers. In an environment of persistent inflation, the direct consequence of sustained oil-driven energy costs, the central bank cannot cut pre-emptively without material credibility cost. The put is available, but its strike is deep out of the money; the cost of waiting for conditions to deteriorate enough to activate it is borne by equity holders in the interim. By the time intervention arrives, the technical amplifier dynamic described in the previous section has already moved prices well past the level early action would have stabilized.

Our six layers of analysis converge on two primary variables that govern the outcome space: whether the geopolitical situation resolves in the near term, and whether the real rate environment remains contained. The four combinations of these variables define materially distinct regimes for equity markets.