1 Number Every Investor Needs to Look For When Nvidia, Alphabet, and Amazon Report Earnings, Whether You Own the Stocks or Not

The article highlights that “other income” from unrealized investment gains is inflating S&P 500 earnings—about $69B combined from Nvidia ($16B other income), Alphabet (nearly $38B), and Amazon ($16B) last quarter, roughly 10% of index earnings. Stripping these gains would reduce estimated earnings growth from ~29% to ~15%, even as consensus still targets +23.6% y/y S2 earnings growth and +25.1% for the full year. It warns that the second-quarter contribution could swing negative (e.g., SpaceX/Intel drawdowns), potentially masking weaker underlying operating earnings and raising valuation risk.

Analysis

The key setup is not that mega-cap AI leaders have weak businesses; it’s that reported EPS is being amplified by a mark-to-market tailwind that can disappear faster than operating growth can slow. That matters because index-level valuation is being justified off headline earnings, so a decline in those marks would compress the market multiple even if revenue trends stay intact. The reflexive loop works both ways: rising public/private AI valuations mechanically lower the apparent P/E, which supports flows into the same names that created the effect.

The second-order loser is the cap-weighted index itself, especially QQQ and any systematic strategies that screen on near-term earnings momentum. If public comps or late-stage private AI assets roll over, GAAP earnings revisions can gap lower over 1-3 months and create a bigger de-rating than the underlying operating miss would imply. By contrast, companies with cleaner cash earnings and less exposure to investment marks should hold up better in a risk-off tape; NDAQ could see some offset from higher volatility volume, but that is an underpowered hedge versus a broad multiple reset.

Contrarian view: the market may be overreacting to the accounting noise because the operating story at the largest AI platforms is still improving, and the marks are a byproduct of that ecosystem rather than the thesis itself. The real vulnerability is timing: as soon as AI-beta softens or a private funding round comes in below last marks, the reported EPS support can reverse abruptly. That is a 1-3 month catalyst if public tech sells off; structurally, it means investors should stop treating headline earnings growth as durable without stripping out non-operating gains.

Key Decisions for Investors

• Buy 1-3 month QQQ put spreads on strength; use a 5%-7% downside target and cover if cap-weighted tech keeps outperforming after earnings season.
• Run a relative-value hedge: short NVDA/GOOG/AMZN basket against long RSP over the next 4-8 weeks; thesis is that headline EPS quality deteriorates faster at cap-weighted mega-cap tech than at the equal-weight index.
• For existing long exposure to AMZN/GOOGL/NVDA, use collars into quarterly prints; the risk is not operational collapse but a sudden removal of the earnings tailwind from investment marks.
• Set an alert on private-AI valuation resets and public comps for late-stage AI holdings; if secondary pricing or funding rounds print below prior marks, expect a fast de-rating in reported index EPS expectations.