S&P 500 Earnings Break Historical Trend Amid AI Uncertainty
U.S. corporate earnings have exceeded a century-long growth channel, but the surge is driven by buybacks and AI capex rather than pure profit expansion.
Standard & Poor's 500 Index earnings per share have exited a stable historical channel that persisted for nearly nine decades. This upward deviation marks a significant structural break from the 6.5% annualized growth rate observed between 1935 and 2019. The current trajectory surpasses previous economic booms, including the post-war period and the late-1990s technology surge, positioning the market in uncharted territory relative to long-term fundamentals.
Deutsche Bank Research analyst Jim Reid identifies artificial intelligence as the primary driver behind this sustained upside breakout. However, the firm cautions that the metric relies on EPS rather than aggregate corporate profits. Over the past two decades, corporate stock buybacks have reduced the share count, artificially inflating per-share figures. Consequently, raw S&P 500 earnings have actually grown faster than EPS over the last 60 years, suggesting the current record highs may reflect financial engineering as much as operational performance.
Buybacks Distort Historical Growth Metrics
The divergence between total profits and per-share earnings is a critical factor in assessing the sustainability of the current rally. Reid notes that the EPS breakout is heavily influenced by the mechanical effect of share count reduction. While investors often interpret rising EPS as a sign of improved profitability, this specific metric can decouple from actual cash flow generation. The data indicates that the current surge aligns with a period of aggressive capital return strategies, which complicates the comparison to earlier eras of organic growth.
This distinction requires investors to differentiate between genuine productivity gains and balance sheet optimization. If the recent EPS increase is primarily driven by buybacks, the underlying economic expansion may be narrower than the headline figures suggest. The analysis highlights that while the trend line is breaking upward, the composition of that growth has shifted from broad-based profit accumulation to concentrated per-share value, a nuance often overlooked in standard market commentary.
AI Capex Drives Future Earnings Trajectory
The forward-looking uncertainty centers on whether artificial intelligence will sustain this earnings breakout. Reid argues that the current capex boom in AI infrastructure is lifting corporate results across the index. If this investment cycle translates into durable productivity improvements, the current deviation from the historical channel could become the new norm. However, if the gains are temporary and reliant on continued heavy spending, the market may be witnessing a bubble rather than a structural shift.
Investors must choose between two opposing interpretations of the data. One view posits that AI enables a sustainable earnings trajectory unmatched by previous technological revolutions. The alternative view suggests a temporary boost from capital expenditure that will eventually normalize. As reported in materials from GN markets/earnings (en-US), this decision point will influence global capital allocation strategies in the coming quarters, with the outcome determining whether the current high-growth phase is a permanent structural change or a transient anomaly.
Market Reaction to Structural Break
The market is currently pricing in the optimism of the AI-driven growth scenario. Prices reflect an expectation that the earnings breakout will continue, ignoring the historical precedent of mean reversion. This positioning creates a risk premium for any signs of deceleration in AI-related capital expenditure. The disconnect between the measured EPS growth and the broader economic context leaves investors exposed to a potential correction if the underlying profit growth does not match the per-share metrics.
The situation demands a re-evaluation of valuation models that assume historical stability. The exit from the 1935-2019 channel implies that standard deviation bands are no longer reliable predictors of earnings performance. Market participants are now forced to weigh the tangible benefits of AI integration against the speculative nature of the current capex cycle, a challenge that defines the current investment landscape.
