Tech selloff stirs bubble fears in US stock market

Market bubble concerns persist even as stocks rise into the last session of Q2 and H1: BofA’s Bubble Risk Indicator is at 0.91 for the PHLX Semiconductor Sector and 0.82 for the Technology Select Sector, while the Buffett Indicator sits at 218% (near the 219% record). Valuation is stretched with the S&P 500 price-to-sales at 3.22 vs a 1.84 long-term average and the S&P 500 trading at 20.2x expected 12-month earnings (vs 25.2x at the dotcom peak), though AAII sentiment is not at euphoric extremes (bull-bear spread 8.8% vs 6.5% average). Net: solid earnings support the rally and leadership is broadening (equal-weight vs cap-weight gap ~3 points), but investors are urged to stay diversified amid worries about debt-funded AI spending and a hawkish Fed.

Analysis

The key market implication is not simply that tech is expensive; it’s that the index is increasingly hostage to a narrow set of balance sheets whose capex is now being treated like a quasi-utility investment cycle. That creates a lagged risk: even if the AI story keeps earnings elevated near term, any sign of slower incremental spend would hit the suppliers first — semis, networking, and power infrastructure — before the mega-cap buyers fully re-rate. In other words, the upside is still in the ecosystem, but the durability of returns is being questioned at the capital-allocation layer, which is where bubble regimes tend to break.

The broadened rally is the main reason to avoid aggressive outright shorts. If participation continues to improve, the index can grind higher even while leadership compresses, which argues for relative-value rather than naked beta bets. The most vulnerable exposure is concentrated growth ownership via QQQ/SMH where multiple compression can happen without a recession; the cleaner beneficiaries of rotation are equal-weight and cyclicals with less duration embedded in expectations.

Catalyst timing matters: over days, strong labor data can keep the soft-landing narrative alive and punish vol sellers; over 1-3 months, earnings commentary on AI payback periods and capex budgets is the real test; over 6-18 months, the structural issue is whether margin assumptions baked into mega-cap tech are sustainable once financing costs and depreciation catch up. The contrarian point is that this may be more of a valuation-surface correction than a full bubble unwind — unless breadth rolls over and financing spreads widen, the market can stay stretched longer than fundamentals suggest.

Key Decisions for Investors

• Tactically underweight QQQ vs RSP for the next 4-8 weeks; the trade monetizes further leadership broadening while reducing exposure to multiple compression in crowded AI winners. Falsifier: if QQQ reasserts leadership and the equal-weight spread widens back out, cover quickly.
• Buy a 1-3 month SOXX put spread rather than outright shorting the index; the best risk/reward is on semis if AI capex sentiment cools, but downside can be fast if guidance or order commentary softens. Falsifier: another round of upward capex revisions from hyperscalers.
• Long CBOE on a 2-6 week horizon as a convexity/dispersion beneficiary if market fragility rises; this is a cleaner way to express a volatility pickup than chasing VIX products outright. Falsifier: realized vol stays suppressed and index breadth continues improving.
• If already long AI infrastructure names, rotate from the most capex-sensitive names into the better capitalized platform beneficiaries; avoid adding into weakness until the next earnings cycle clarifies payback periods. Watch for any order-book or backlog deceleration as the trigger to de-risk.