Capex boom threatens to crowd out buybacks, key equity demand driver

S&P 500 net buybacks hit a record $270 billion in Q1, while gross buybacks rose to about $300 billion, suggesting AI-driven capex is not broadly crowding out shareholder returns. Deutsche Bank said the bulk of spending increases is concentrated in five hyperscalers, but earnings growth from AI infrastructure suppliers and continued buybacks across the rest of the index remain supportive. Rising IPO and secondary issuance could modestly pressure net repurchases, though strong inflows and earnings growth should offset much of the supply.

Analysis

The important second-order effect is that AI capex is becoming a capital-allocation filter, not just an earnings driver. The hyperscalers are effectively trading near-term per-share optics for strategic control of compute supply, while the rest of the market continues to behave like a mature cash-return regime; that bifurcation should widen valuation dispersion between “AI landlords” and “AI picks-and-shovels.” The surprising nuance is that buyback support is not disappearing — it is migrating downstream into the suppliers and adjacent infrastructure names that are now monetizing the buildout, which means the market can absorb higher issuance without a broad de-rating if earnings breadth stays intact.

The risk is not that buybacks vanish, but that the marginal dollar of equity supply rises faster than passive and retail demand can digest in a 1-2 quarter window. That matters most for newly listed or secondary-heavy names where float expansion can suppress momentum even if fundamentals are strong; the first-order winners of AI spending may be the last to see it in share price if positioning is crowded. If issuance stays elevated into late summer while rates remain sticky, factor leadership could rotate from mega-cap growth to balance-sheet-strong infrastructure beneficiaries and selective value/cash-return names.

A more contrarian read is that the market may be underestimating how much of the S&P’s buyback engine is now concentrated in fewer, higher-quality cash generators outside the hyperscalers. That concentration actually makes the market more resilient than the headline issuance numbers suggest, but it also means any earnings miss from a handful of mega-caps could have outsized impact on both index-level buybacks and passive flows. The setup favors relative trades over outright beta: own the companies capturing AI capex with recurring repurchase capacity, and fade names where issuance is dilutive before the incremental revenue is visible.

Key Decisions for Investors

• Long SMH vs short a basket of hyperscaler equities on a 3-6 month horizon: use it as a relative-value way to express that AI infrastructure vendors should enjoy stronger buyback capacity and cleaner EPS torque than the capex-heavy platform names. Risk: if hyperscalers re-accelerate monetization, the basket can re-rate quickly.
• Buy AMZN/MSFT/GOOGL on 6-12 month weakness only, not strength; these remain strategic winners but near-term per-share returns are likely capped by capex intensity and dilution optics. Prefer scaling in after post-spend pullbacks rather than chasing breakouts.
• Long ORCL vs short high-secondary-risk recent IPOs over 1-3 months: ORCL has AI exposure without the same magnitude of issuance overhang as fresher listings. The trade benefits if the market rewards credible cash flow over story-driven supply.
• Pair long QQQ with short a recent IPO/secondary basket for 1-2 quarters if issuance keeps running hot: the index should be cushioned by earnings breadth and buybacks, while fresh supply can pressure newly listed names. Tighten risk if rates fall sharply, which would improve appetite for new supply.
• Add SMCI selectively on pullbacks with a tactical stop, but treat it as a high-beta operating leverage trade rather than a capital-returns story; upside is strong if AI spending broadens, but execution and multiple risk remain high.