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

The article flags that a large share of recent “earnings growth” for Nvidia, Alphabet, and Amazon is driven by “other income” (largely unrealized investment gains): combined other income of about $69B last quarter contributed ~10% of S&P 500 earnings, implying S&P 500 earnings growth could fall from ~29% to ~15% without these gains. With analysts still projecting S&P 500 earnings growth of 23.6% in Q2, 25.1% for the full year, and ~24.9% over five years, the piece warns that GAAP beat optics may reverse if investment valuations fall—potentially creating a downside feedback loop for valuations.

Analysis

The market is likely overstating how much of the current index-level EPS strength is repeatable. For the megacaps, the real issue is not that operating momentum has vanished; it’s that a meaningful slice of reported earnings is now behaving like a levered equity book, which means the S&P 500’s “earnings growth” can decelerate even if core demand stays intact. That matters most for cap-weighted ETFs and factor products: SPY/QQQ can look cheaper on headline P/E just as the numerator is getting inflated by mark-to-market gains.

Winners are the companies holding large minority stakes and the passive holders exposed to their index weight; losers are investors underwriting the index multiple as if this were pure operating growth. The second-order effect is relative performance: names with cleaner cash earnings and less balance-sheet/mark-to-market noise, such as MSFT/META/AVGO, should be rewarded if investors rotate toward earnings quality. The real vulnerability is that a broad risk-off tape would hit both the underlying holdings and the unrealized gains simultaneously, creating a double-counted earnings reversal.

The near-term catalyst is the next round of quarterly prints: the market may initially reward another “beat,” but the risk is that consensus gets anchored to an inflated GAAP run-rate and then has to reset when the equity portfolio marks go the other way. Over 1-3 months, a 5-10% drawdown in the relevant private/public comps would be enough to mechanically shave reported EPS and compress multiples. Over 6-18 months, this becomes a credibility issue: if investors conclude index earnings quality is deteriorating, the re-rating could come through lower forward multiples rather than lower operating estimates alone.

Key Decisions for Investors

• Prefer a relative-value hedge: long RSP / short QQQ for the next 1-3 months. If headline EPS breadth keeps being driven by mark-to-market gains, cap-weighted tech should lag equal-weight on a cleaner operating-earnings basis. Stop if QQQ leadership broadens beyond the same few mega-caps on confirmed operating upside.
• Initiate a small tactical short in SPY or QQQ via put spreads into the next megacap earnings cycle. This is a hedge against an index-level multiple reset, not a call to short individual core businesses. Risk/reward is best if implied vol is still below the realized volatility in the index.
• Avoid chasing AMZN/GOOG/NVDA on GAAP beats alone until the market distinguishes operating EPS from investment income. For long-only portfolios, wait for post-earnings price action and focus on core revenue/margin commentary; if the stock rallies only on financial marks, fade the move.
• If long mega-cap tech already, pair with long MSFT or META versus the names most exposed to mark-to-market noise. The goal is to keep AI exposure while reducing sensitivity to a reversal in private/public portfolio valuations.
• Set a watch item on the next print: if reported EPS growth is >5 pts above non-GAAP/core operating growth, treat that as a falsifier for any bullish broad-market valuation thesis and reduce index exposure.