Wall Street Expects S&P 500 Profit Growth to Collapse From 35% to 15% in 2027

S&P 500 earnings growth is projected to decelerate sharply to 15% in 2027 from an estimated 35% in 2026, although the 2027 forecast remains above the 35-year median annual growth rate of 10%. Hyperscaler AI capital expenditure is expected to rise from more than $800 billion this year to $1.1 trillion in 2027, but its growth rate would slow from nearly 100% to 37%, pressuring incremental revenue growth for chip and data-center infrastructure suppliers. Higher rates following the Fed's 25bp September increase and valuation compression—S&P 500 forward P/E at 19x versus 22x in January—add downside risk, particularly if hyperscalers reduce 2027 spending guidance.

Analysis

The relevant market risk is not the lower index-level growth rate but dispersion: AI-capex beneficiaries are priced on incremental orders, so a deceleration in spend growth can drive earnings-estimate cuts even if absolute budgets remain high. Semis and data-center hardware with high revenue concentration—NVDA, AVGO, ANET, VRT, ETN and GLW—face the greatest 1-3 month guidance sensitivity; cloud platforms with internally funded capex (MSFT, GOOGL, AMZN, META) retain strategic flexibility and may gain negotiating leverage over suppliers as capacity bottlenecks ease.

A normalization in earnings growth should also widen the valuation gap between cash-generative quality and levered, long-duration AI infrastructure. Higher financing costs disproportionately pressure data-center developers (DLR, EQIX) and power-intensive buildouts where returns depend on sustained utilization and cheap debt; utilities and electrical equipment suppliers may be more resilient because grid interconnection lead times preserve backlog visibility. The apparent breadth deterioration raises the probability that index downside is masked until a mega-cap miss forces passive de-risking.

Consensus likely overstates the binary importance of a headline hyperscaler budget cut. A shift from accelerator purchases toward networking, inference optimization, power, and owned data-center capacity can rotate rather than eliminate spending. Conversely, supplier guidance that merely confirms large budgets may not be enough: current multiples require order visibility extending into 2028, making book-to-bill, lead-time commentary and cancellation language more important than aggregate capex targets.

Key Decisions for Investors

• Reduce beta in high-multiple AI hardware ahead of upcoming hyperscaler reports; use a 1-3 month pair of long MSFT/GOOGL versus short SMH, with sizing capped because a renewed accelerator supply squeeze would reverse the spread.
• Initiate a selective long ETN and VRT versus short ANET only after earnings confirm backlog conversion and no material order pushouts. Electrical/power exposure has a more durable grid constraint, while networking is more exposed to architecture shifts; target 10-15% relative return over 6-12 months, invalidated by falling backlog or margin guidance.
• Buy 3-month SPY put spreads rather than outright index shorts into the guidance window, financed opportunistically after volatility dips. The catalyst is a capex or demand-guide reset from a mega-cap; exit if equal-weight participation materially improves and forward S&P earnings revisions stabilize.
• For financials, favor MS over BCS and STT on a 6-12 month horizon: an earnings slowdown and risk-asset volatility can revive trading and hedging demand, while asset-servicing and more rate-sensitive earnings face pressure if market breadth and fund flows weaken. Reassess on a sustained decline in VIX and trading revenues.