Pulse24 Original
8,700. That's Goldman's S&P 500 Target for 2027, Even as It Sees Earnings Growth Slow by More Than Two Thirds.
November 14, 2026
Goldman Sachs says the S&P 500 climbs to 8,700 by the end of 2027, a roughly 14% gain from current levels. The catch is in the bank's own numbers, which show earnings growth slowing by more than two-thirds to get there.
8,700. That's the level Goldman Sachs says the S&P 500 reaches by the end of 2027, a call the bank's chief global equity strategist Peter Oppenheimer and his team published Friday. It implies a roughly 14% gain from Wednesday's close near 7,619, or about 16% once dividends are counted. The timing is almost funny: the index spent the same week falling, not rising, closing Friday at 7,561.67 after a selloff in chip stocks pulled the Nasdaq lower.
[[IMG1]]
Oppenheimer's own explanation is the interesting part. "Equities have de-rated sharply," he and his team wrote, "with more attractive valuations, positive but slower nominal economic and earnings growth, and lower bond yields, inflation and oil prices, we expect equities to continue to rise, but with high volatility." Translated: stocks got expensive, then came back down to earth a bit, and now have room to climb again, just more slowly and with bigger swings along the way.
What Changed
The headline number is the 8,700 target. The more consequential number sits underneath it: Goldman's economists see S&P 500 earnings growth slowing from 36% in 2026 to 11% in 2027, a drop of more than two-thirds. That's not a small revision, and it comes largely from one place: the AI buildout. AI spending has been behind roughly half of America's economic growth since last year, and Goldman now expects that specific engine to lose power as capital spending growth on AI data centers peaks sometime in 2027.
Slowing capex growth doesn't mean spending falls outright, just that the rate of increase stops accelerating. For the suppliers that sell into that buildout, the distinction barely matters. Goldman's team wrote that cooling hyperscaler capex growth will "weaken the support for some of the picks and shovels of the AI-related capex boom," language that lands days after chip stocks lost $1.4 trillion in a single session, their worst drop since the pandemic crash. Investors have already started pricing a version of this slowdown. Goldman is effectively telling them the direction of that repricing is correct, just early.
Why It Matters
A bank telling clients stocks go up 14% next year sounds bullish until you read the fine print. Oppenheimer's own words, "risks are becoming more two-sided," read like a hedge dressed up as a forecast. The S&P hitting 8,700 and earnings growth cutting by two-thirds aren't contradictory in Goldman's model: cheaper valuations and lower bond yields are supposed to do the work that faster earnings growth used to do. If bond yields don't cooperate, the math gets harder. And right now they're not cooperating. The 10-year yield touched 5.17% this week, its highest since October, not far removed from the same week a weak Treasury auction briefly pushed the 10-year yield to 5.02%. Goldman's entire case for 8,700 assumes yields head lower from here, not higher.
The regional calls tell a similar story about where Goldman thinks the safer bets are. It's overweight the US and Japan, neutral on emerging markets and broader Asia, and cautious on Europe, where it sees the STOXX 600 reaching just 700. Goldman also expects Brent crude to fall from around $100 a barrel at the end of this year to $78 by the end of 2027, a decline the bank says should help cool inflation and give central banks room to ease off their recent hawkish turn. That last point cuts against the Fed's own signals lately; Fed officials have pointed to three straight months of 0.3% core inflation readings as reason enough to debate a bigger December hike, the opposite of the lower-rate world Goldman is underwriting its 2027 target on.
What to Watch Next
Nvidia reports earnings Tuesday, and given how much of Goldman's capex-peaking thesis rests on hyperscaler spending plans, that report and the guidance attached to it will be an early test of the call. A print that reaffirms aggressive 2027 capex guidance from Nvidia's biggest customers would undercut the peak-capex argument before the ink is dry. Dial that guidance back instead, and the Goldman thesis gets a boost.
Bond yields are the other thing to track closely. Goldman's framework needs yields to fall meaningfully from here, not just stabilize. If the Fed delivers a hawkish hold or an actual hike in December and the 10-year holds above 5%, the valuation support Oppenheimer is counting on simply isn't there, and 8,700 starts to look like a number that requires everything to break the bank's way at once.
The Pulse24 Take
Goldman isn't wrong that stocks have gotten cheaper this year; a year of flat-to-falling prices alongside rising earnings will do that to a multiple. But there's a difference between a forecast built on one clean driver and one that needs several things to go right at the same time: earnings deceleration has to be gentle, bond yields have to fall, oil has to cooperate, and the AI capex slowdown has to be orderly rather than abrupt. History suggests slowdowns in capital spending cycles are rarely that orderly; they tend to overshoot in both directions once they start. The more useful way to read this note isn't as a prediction that the S&P hits 8,700. It's as a checklist of what has to happen for that number to be right, and investors watching chip earnings and Treasury auctions over the next few months now have Goldman's own thesis to measure against.
How we read the data
Curious how we get from raw data to a take like this? Our Trader's Toolkit walks through the tools we lean on.
Explore the Toolkit