November 10, 2026, (Inside AI) — A quiet but consequential split has emerged in the cost of capital between the small group of companies driving the artificial intelligence boom and the rest of the corporate world. According to a preliminary analysis by Bain & Company, the five leading AI-related companies now enjoy a weighted average cost of capital (WACC) of roughly 7.5%, while the rest of the S&P 500 faces about 8.4%. That gap of nearly 100 basis points may look modest, but it changes which investments create value and which destroy it.
The divergence stems from two forces. First, equity markets have priced AI leaders for extraordinary future growth, pushing their cost of equity down to as low as 7.5%. For other companies, less favorable valuations mean investors demand returns of 9.5% to 10%, a gap of 200 basis points or more. Second, many AI leaders carry investment-grade balance sheets that let them borrow at rates only slightly above Treasury yields. Companies outside the boom face wider credit spreads and higher financing costs.
The timing matters. Long-term Treasury yields have climbed sharply, with 30-year yields reaching their highest levels in more than two decades. That rise has lifted borrowing costs across the economy, but not evenly. The result is a two-track capital market in which the AI elite operate under conditions that resemble the cheap-money era while everyone else adjusts to a more expensive reality.
Why an 8% Return No Longer Cuts It
The practical effect is easiest to see through return on invested capital (ROIC). A business earning 8% ROIC creates economic profit when its cost of capital is 7.5%. The same business destroys value when its cost of capital is 8.4%. Operating performance has not changed, but the verdict on whether the business is worth running has flipped.
Read: Citigroup Downgrades US Equities to Neutral on AI Concerns
That arithmetic explains why Bain argues companies outside the AI boom must now prioritize margins and asset productivity over raw growth. The firm's analysis of two identical strategic options, one adding a percentage point of revenue growth and the other adding a percentage point of operating margin, found that at a 7% cost of capital, faster growth creates nearly twice as much value as margin improvement. At 9%, the relationship reverses and margin improvement wins.
For most non-AI companies, the analysis suggests the tipping point has already arrived. Growth still matters, but only when it generates attractive returns without consuming disproportionate capital. Better pricing, product mix, and working capital discipline can lift cash flow without large new investment.
The gap also has a second-order effect that deserves more attention than it has received. As AI leaders increasingly tap debt markets to fund massive data center and chip buildouts, their borrowing advantage could widen the WACC gap further. Equity currently accounts for most of their capital structure, so the cost of equity drives today's difference. But a sustained shift toward debt-financed infrastructure would make their credit ratings more valuable, not less, and could push the spread beyond 100 basis points. For non-AI companies, that means the competitive disadvantage may still be in its early stages rather than at its peak.
The Portfolio Question Most Boards Avoid
Higher capital costs also force a harder look at existing business units. Divisions that once earned attractive economic profits may now create far less value, or none at all. Bain recommends that leaders reassess capital committed to these units and consider reducing investment, releasing working capital, improving asset utilization, or pursuing divestiture. The goal is to free capital for opportunities that can clear the higher hurdle rate.
The stakes are large because the gap is not merely a financing curiosity. It reflects how public markets now assign value to two different kinds of corporate futures. AI leaders are priced for cash flows extending far into the future, which mechanically lowers the return investors demand today. Everyone else is priced for nearer-term uncertainty, which raises it. That pricing difference then feeds back into real decisions about factories, research budgets, and hiring.
Bain's preliminary estimates carry caveats. The analysis covers only five leading AI-related companies, and the WACC figures are approximations rather than audited numbers. The firm also notes that the precise growth-versus-margin trade-off will differ by company depending on the returns it can generate on invested capital. Still, the direction of the finding aligns with a broader pattern: capital is no longer uniformly cheap, and the penalty for capital-intensive, long-dated projects has grown.
Read: UBS CEO Says AI Investors Priced Perfection Before Correction
For executives outside the AI boom, the message is not to stop investing. It is to invest more selectively, measure returns against a higher bar, and treat capital discipline as a competitive weapon rather than a constraint. The AI boom has captured most of the attention. The more consequential story for the rest of the economy may be the quiet repricing of capital itself.
Shamil has a low tolerance for AI that only works in controlled demos. He covers agentic systems and generative AI with the skepticism of someone who has seen too many “revolutionary” tools quietly disappear six months after launch.
