Morgan Stanley sounds the credit risk alarm: AI capital expenditure intensity approaches its peak in 2027, and high interest rates continue to squeeze junk bond borrowers.
Morgan Stanley expects tech capital expenditure to approach its peak in 2027, with investment-grade issuance setting another record at US$2.4 trillion; high interest rates reinforce quality differentiation, favoring investment grade over loans over high-yield bonds.
Morgan Stanley sounds the credit risk alarm: AI capital expenditure intensity approaches its peak in 2027, and high interest rates continue to squeeze junk bond borrowers.
In its latest 2027 US credit strategy outlook, Morgan Stanley said the business cycle remains intact, hyperscaler capital expenditure will grow substantially, and capital intensity will approach a peak in 2027; the credit market will write larger checks, with investment-grade bond issuance setting another record, but supply will be more programmatic, shorter in tenor, and already better reflected in spreads.
At the macro level, Morgan Stanley expects US GDP growth of 2.5% in 2027 (Q4 year-over-year), rising to 2.7% in 2028, with tech capital expenditure combined with resilient consumption supporting a healthy backdrop. The contribution of business fixed investment to growth will rise from 1.1 percentage points in 2027 to 1.4 percentage points in 2028; the combined contribution of AI will reach 1.2 percentage points in 2027. Morgan Stanley expects Fed policy to tighten further, but does not believe the economy will overheat in its base case. Inflation will be sticky in the near term, but disinflation is expected to resume in the second quarter of 2027.
The supply side is the main theme of Morgan Stanley's outlook. Investment-grade issuance is expected to set another record at US$2.4 trillion, up 12% year-over-year, driven by AI capital expenditure and active M&A, with a greater tilt toward the short end. AI-related US dollar investment-grade issuance is expected to be about US$370 billion, covering hyperscalers, AI-enabling semiconductors, REITs and data centers, plus about US$100 billion of chip financing. Morgan Stanley expects hyperscaler issuance to be more front-end loaded relative to this year, while data center bond issuance is expected to decline modestly.
On spreads and allocation, Morgan Stanley's base case is for spreads to tighten and excess returns to be positive, with investment grade outperforming leveraged finance. Strategically, Morgan Stanley removed its underweight recommendation on hyperscalers and instead added selectively at the long end; it remains cautious on the front end given the heavy supply of short-duration corporate bonds and structured products; it expects the yield curve to flatten and prefers utilities and business development companies (BDCs).
In leveraged credit, Morgan Stanley believes fundamentals remain healthy but the distribution of outcomes is wider. High interest rates are pressuring free cash flow and interest coverage ratios for sub-investment-grade borrowers, and earnings growth has passed its peak. Morgan Stanley forecasts a high-yield default rate (including distressed exchanges) of 4.5% over the next 12 months and 4.75% for loans; it prefers loans over high yield and BBs over single-Bs.
On credit derivatives, Morgan Stanley prefers to express a "de-compression" trade through CDX indices, going long investment grade and short high yield; within the AI ecosystem, it believes the risk-reward of hyperscalers is better than that of AI-enabling semiconductors.
On risks, Morgan Stanley flags two tails: first, energy-driven inflation forcing the Fed into more aggressive tightening; second, AI monetization falling short of expectations, weakening credit fundamentals and asset values.
Overall, Morgan Stanley believes the 2027 credit market will climb over a higher but more predictable supply peak, and with valuations adjusted and yields still attractive, investment-grade bonds enter the new year in a better allocation posture.
Related Articles

US Stock Market Move | Shares of XPeng (XPEV.US) fell more than 8% after HSBC sharply cut its target price.

Is US energy permitting reform achieving a major breakthrough? Bernstein: Geothermal and power grid equipment are expected to be the biggest winners; AI data center construction may benefit.

US Stock Market Move | Alphabet Inc. Class C (GOOGL.US) rises over 2%; adjusting HBM procurement approach saves about US$50 billion
US Stock Market Move | Shares of XPeng (XPEV.US) fell more than 8% after HSBC sharply cut its target price.

Is US energy permitting reform achieving a major breakthrough? Bernstein: Geothermal and power grid equipment are expected to be the biggest winners; AI data center construction may benefit.

US Stock Market Move | Alphabet Inc. Class C (GOOGL.US) rises over 2%; adjusting HBM procurement approach saves about US$50 billion






