9-21-26 What Happens When Earnings Estimates Fall Short?

The market’s earnings boom is heavily dependent on hyperscaler and data-center investment: consensus AI capital spending is projected to rise from $421 billion to $754 billion and roughly $1.1 trillion in 2027. A reversal in that spending could remove the associated revenue cycle, reduce earnings estimates, and force a market repricing.

Current earnings expectations appear vulnerable to mean reversion: S&P estimates are above the long-term trend channel, while consensus implies roughly 30% earnings growth next year followed by 11%, 10%, and 8% growth. The speaker warned that slower growth or a demand shock could leave companies with excess inventory, require discounting, and compress margins and valuation multiples.

The episode contrasted resilient U.S. corporate profitability—margins were cited at 13.8% despite higher rates, inflation, and tariffs—with the risk that unusually high technology-company margins could revert if supply and demand change. That margin risk is particularly important because the current earnings strength is concentrated in hyperscalers and the Magnificent Seven.

The Federal Reserve’s 25-basis-point hike was unanimous at 12–0, and its dot plot showed 16 of 18 officials expecting at least one more hike this year, including four projecting two more. The speakers said additional tightening could make corporate conditions more difficult, suppress demand, and pressure markets over the next one or two months, although the historical three-month selloff pattern is only an average rather than a guarantee.

The speakers emphasized that rate hikes are a demand-management tool rather than a direct solution to oil-driven inflation or higher goods costs such as an iPhone: they raise short-term borrowing costs and can eventually reduce oil consumption, employment, and broader demand. Rate-sensitive sectors such as utilities and real estate have historically underperformed during hiking cycles and tended to outperform during cutting cycles.