Thoughts on the Market

2026-10-13 · Hosted by Mike Wilson · Morgan Stanley

Executive Summary

Morgan Stanley's Ariana Salvatore examined the growing political debate over who pays for the electricity infrastructure needed to power the AI investment boom. A newly introduced bipartisan bill, the American Affordability and Jobs Act, would reform federal permitting while requiring data centers — not residential ratepayers — to cover transmission costs tied to their projects.

Key Stories & Changes

1. American Affordability and Jobs Act

  • Bipartisan legislation introduced in Congress reforms federal permitting and requires data centers to cover transmission costs associated with their own projects

  • Builds on the voluntary Ratepayer Protection Pledge, under which hyperscalers already committed to securing their own power and infrastructure upgrade costs rather than passing them to residential consumers

  • The bill would shift that voluntary framework toward a federal requirement

  • Path forward remains uncertain given policy differences and a limited legislative calendar after the November elections

2. Widening US Data Center Power Shortfall

  • Morgan Stanley's Stephen Byrd team raised its projected US data center power shortfall to 57 gigawatts through 2028, up from a prior estimate of 38 gigawatts

  • Estimate is *before* accounting for new solutions that could accelerate power access

  • Permitting reform could reduce delays, but won't resolve bottlenecks in equipment, labor, or grid capacity on its own

3. Two Post-Election Policy Catalysts to Watch

  • Whether Congress reaches agreement on permitting reform to accelerate energy infrastructure construction

  • Whether lawmakers formalize a cost-sharing requirement for data centers, which could reshape where projects get built and which power solutions (e.g., on-site generation, alternatives) developers pursue

1. Power Infrastructure Becoming the New AI Investment Bottleneck

As the scale of AI capital spending has become well understood, the emerging constraint is shifting to the physical grid: permitting speed, transmission cost allocation, and raw power availability are increasingly as important to the pace of AI monetization as chip supply or model capability.

2. Policy Risk Now a First-Order AI Investment Variable

With a bipartisan bill specifically targeting data center cost-sharing, electricity affordability is emerging as a rare point of cross-party agreement, meaning AI infrastructure economics are now directly exposed to legislative and regulatory outcomes in a way they weren't earlier in the cycle. ---

Sentiment Analysis

Overall Market Sentiment: Cautiously Watchful

The tone was analytical rather than alarmed, framing power policy as an increasingly important but still-uncertain variable for AI investors to monitor post-election.

Risk Factors Highlighted

Widening power shortfall: The projected US data center power gap grew from 38GW to 57GW through 2028, raising execution risk for the AI buildout.

Legislative uncertainty: Substantial policy differences and a limited post-election legislative calendar make permitting reform passage far from guaranteed.

Rising transmission cost burden: A shift from voluntary pledges to federal cost-sharing requirements could raise the cost of new data center projects and shift where they are built.

Non-permitting bottlenecks: Equipment, labor, and grid capacity constraints will persist even if permitting reform passes.

This episode was covered in today's [The Market Signal — 2026-10-13](https://marketsignal.beehiiv.com/p/the-market-signal-2026-10-13), a cross-source synthesis of multiple podcast reports.

Keep Reading