← News·MarketsMarkets

AI stocks drove S&P 500 to 94th-percentile earnings quarter, but 2027 deceleration risk sharpens

With roughly 90% of S&P 500 (^GSPC) companies having reported, about 80% logged year-over-year EPS growth in the second quarter, placing the period in the 94th percentile for that metric, per Bank of America research.…

NM
NewsMV Markets Desk
3 min read
11 August 2026Markets desk
Share this dispatch

Key takeaways

  • About 80% of S&P 500 companies logged year-over-year EPS growth in Q2, placing the quarter in the 94th percentile per Bank of America.
  • The index is on track for four consecutive quarters of EPS growth above 20%, a sequence that has occurred only 10 times since 1936.
  • AI-related stocks posted median EPS growth of 28% versus 12% for the median non-AI name, which BofA called the index's primary growth engine.
  • JPMorgan raised its year-end S&P 500 target to 8,000 from 7,800, roughly 3% above Friday's close, citing earnings strength and hyperscaler spending.
  • Consensus forecasts show EPS growth decelerating in 2027, and BofA notes such above-trend-but-slowing periods have historically produced weaker equity returns.

With roughly 90% of S&P 500 (^GSPC) companies having reported, about 80% logged year-over-year EPS growth in the second quarter, placing the period in the 94th percentile for that metric, per Bank of America research. BofA strategists led by Savita Subramanian note the index is now on track for four consecutive quarters of EPS growth exceeding 20%, a sequence that has occurred only 10 times since 1936. The AI trade is doing the heavy lifting, and the question on the tape is how long that holds.

The AI spread inside the index

Ten of eleven S&P 500 sectors are on pace for positive year-on-year earnings growth this quarter. That breadth sounds reassuring. It isn't the whole story. Median EPS growth for AI-related stocks came in at 28%, against 12% for the median non-AI name, a gap BofA called the index's primary growth engine. Apollo Global Management's Torsten Sløk framed the same dynamic around margins: expansion in the tech sector has been steady, while the rest of the market has seen little to none. "The AI capex boom is so far only showing up in the sellers' margins, not the buyers'," Sløk wrote. The longer the S&P 493 waits for return on investment, he argued, the bigger the downside risks to a market this concentrated in the AI trade.

The capex math and what it means for the setup

JPMorgan Chase's equity strategy team, led by Dubravko Lakos-Bujas, lifted its year-end S&P 500 target to 8,000 from 7,800 on Monday. If reached, that would represent roughly 3% appreciation from Friday's close. The team cited broad earnings strength and hyperscaler capital spending as the central theme. AI spending from the hyperscalers is now expected to reach $900 billion by year-end, after those companies raised their own projections. Against total S&P 500 expected capex of $1.5 trillion, the AI industry accounts for roughly 59% of index-wide capital expenditures.

What to watch: the 2027 deceleration and beat premiums

The setup carries a visible seam. Companies that beat both EPS and sales estimates saw shares outperform by just 0.8 percentage points the following session, a figure BofA put at slightly above half the historical average. The strategists read that as evidence much of the good news was already priced in. Consensus forecasts show EPS growth decelerating in 2027, and BofA noted that years of above-trend but slowing earnings growth have historically produced weaker equity returns. The next hard data point comes from hyperscaler capital expenditure guidance during third-quarter earnings calls.

Related reading

Categoryearnings

Filed via finance.yahoo.com

Keep reading

More from the markets desk

Frequently asked

How much are the hyperscalers expected to spend on AI this year?

AI spending from the hyperscalers is now expected to reach $900 billion by year-end, accounting for roughly 59% of the S&P 500's total expected capex of $1.5 trillion.

Why does concentration in the AI trade pose a risk?

Apollo's Torsten Sløk argues the AI capex boom is only showing up in sellers' margins, not buyers', so the longer the S&P 493 waits for return on investment, the bigger the downside risks for a market this concentrated.

How did the market react to companies beating estimates?

Companies that beat both EPS and sales estimates outperformed by just 0.8 percentage points the next session, slightly above half the historical average, which BofA read as a sign much good news was already priced in.

What is the next key data point to watch?

The next hard data point comes from hyperscaler capital expenditure guidance during third-quarter earnings calls.

How broad was the earnings growth across sectors?

Ten of eleven S&P 500 sectors are on pace for positive year-on-year earnings growth this quarter, though AI-related names drove the bulk of the gains.