Morgan Stanley's latest estimate: By 2028, AI capital expenditures will drive tech giants to increase debt by $1 trillion

Morgan Stanley predicts that the AI spending gap will create over $1.5 trillion in debt demand. Morgan Stanley's latest research report points out that by 2028, global data center construction will require approximately $2.9 trillion in investment. However, relying solely on the internal cash flow of hyperscale cloud service providers can cover at most $1.4 trillion, leaving a financing gap of $1.5 trillion that will profoundly impact the global credit market landscape. This capital expenditure wave driven by AI is not only enormous in scale but will also bring significant macroeconomic effects. Morgan Stanley's economists expect that between 2025 and 2026, investments related to data center construction and power infrastructure will contribute up to 40 basis points to U.S. real GDP growth. At the same time, the credit market will become an important channel to fill the financing gap, with data center financing becoming a key theme of long-term interest for credit investors. AI Investment Reaches Historic New Heights Generative AI and technological diffusion are rapidly reshaping the global economy. However, the potential of this transformation relies on massive capital expenditures, with data centers being at the core. Morgan Stanley predicts that by 2028, total global spending on data centers will reach approximately $2.9 trillion, with about $1.6 trillion allocated for hardware (such as chips and servers) and another $1.3 trillion for building data center infrastructure, including real estate, construction costs, and maintenance expenses. On an annual basis, by 2028, investment demand related to data centers will exceed $900 billion. In comparison, the total capital expenditures of all companies in the S&P 500 index in 2024 are expected to be around $950 billion, highlighting the astonishing scale of AI investment. Hyperscale Cloud Service Providers' Spending Soars but Still Faces Huge Gaps In recent years, capital expenditures related to AI and data centers have begun to grow rapidly. Spending by hyperscale cloud service providers alone has increased from about $125 billion two years ago to approximately $200 billion in 2024, with the market widely expecting it to exceed $300 billion in 2025. However, Morgan Stanley analysts point out that although the internal operating cash flow of these tech giants has been a major source of funding, the surge in investment demand, coupled with the need to consider cash reserves and shareholder returns, makes it difficult to meet all future expenditures solely with self-funding. It is expected that by 2028, only about $1.4 trillion will be covered by corporate funds, leaving a significant financing gap of $1.5 trillion. The Credit Market Will Become a Key Force in Filling the Financing Gap To fill this gap, Morgan Stanley believes that the credit market will play a central role in the future. Whether in public or private markets, broad credit channels will increasingly play a crucial role in bridging this funding gap. The current market environment is also conducive to the development of this trend, with ample dry powder in the credit market and current real yield levels being attractive for long-term investors, such as insurance companies, sovereign wealth funds, pension funds, university endowments, and high-net-worth individuals The investment preferences of these "sticky" funds are highly aligned with the demand for scalable, high-quality, and diversified assets in AI infrastructure investment, laying a solid foundation for cross-cycle capital mobilization. Morgan Stanley also provided specific forecasts on the structure of major financing channels. Unsecured corporate bonds issued by the technology sector are expected to provide about $200 billion. Asset-backed securities (ABS) based on data centers and commercial mortgage-backed securities (CMBS) are expected to provide about $150 billion. The financing scale of the asset-based private credit market is expected to reach about $800 billion. Other funding sources, such as sovereign wealth funds, private equity, venture capital, and bank loans, total about $350 billion. Among these, private credit is viewed by Morgan Stanley as the most promising main funding channel. This type of capital is at the intersection of asset management scale expansion and a high-interest-rate environment, and it is best suited to meet the complex, global, and customized financing needs associated with AI infrastructure construction. However, Morgan Stanley acknowledges that there are inevitably many assumptions and some speculation when making scale forecasts for the above financing channels. For example, the investment of sovereign funds is difficult to quantify, and over a long cycle, the forms of financing may also change (such as transitioning from asset-backed financing to asset securitization), which means that the statistics for certain financing methods may carry a risk of "underestimation." Nevertheless, Morgan Stanley emphasizes that the credit market will play an increasingly important role in supporting the diffusion of AI-driven technologies. While there will inevitably be winners and losers in the market, data center financing will remain a long-term investment theme for credit investors. Risk Warning and Disclaimer The market carries risks, and investment should be cautious. This article does not constitute personal investment advice and does not take into account individual users' specific investment goals, financial situations, or needs. Users should consider whether any opinions, views, or conclusions in this article align with their specific circumstances. Investment based on this is at one's own risk