The burgeoning artificial intelligence (AI) boom holds immense potential to reshape global economies, driving unprecedented productivity gains and technological advancement. However, beneath the surface of innovation and market enthusiasm, critical questions are emerging regarding the financing structures underpinning this revolution and the potential for financial risks to accumulate unnoticed. The identity of the financiers and the points at which vulnerabilities are building may prove to be as crucial to the long-term economic outlook as the technology itself.
Joao Gomes, a distinguished professor of finance at the Wharton School and Senior Vice Dean of Research, Centers, and Academic Initiatives, has voiced a compelling argument that the Federal Reserve, the central bank of the United States, appears overly preoccupied with its inflation-fighting mandate, potentially at the expense of adequately monitoring and mitigating broader financial stability risks. His perspective underscores a growing concern among some financial experts about the blind spots in current regulatory frameworks, particularly in an era characterized by rapid technological disruption and the expansion of less-transparent financial sectors.
The Federal Reserve’s Balancing Act: Inflation vs. Stability
The Federal Reserve operates under a dual mandate from Congress: to achieve maximum employment and maintain price stability. Following a period of persistently high inflation, reaching levels not seen in decades during 2021-2022, the Fed embarked on an aggressive campaign of interest rate hikes, elevating the federal funds rate from near zero to over 5% within a relatively short span. This decisive action, aimed at cooling an overheating economy and bringing inflation back down to its 2% target, has largely dominated public discourse and the central bank’s operational focus.
However, Gomes contends that while inflation control is undeniably vital, the singular emphasis on this aspect might be diverting attention from the complex and evolving landscape of financial stability. Financial stability, broadly defined as the resilience of the financial system to absorb shocks and continue functioning, is not explicitly part of the Fed’s dual mandate but has increasingly been recognized as a de facto third pillar of its responsibilities, particularly after the devastating 2008 global financial crisis. Post-crisis reforms, such as the Dodd-Frank Act, empowered the Fed and other agencies to play a more active role in systemic risk monitoring. Gomes’s critique suggests that the institutional memory and proactive measures developed after 2008 might be receding, or at least not adequately adapting to new forms of risk.
The Opaque World of Private Credit: A Growing Concern
Central to Gomes’s concerns is the burgeoning role of private credit, an asset class that has experienced explosive growth over the past decade. Private credit refers to non-bank lending, typically provided by private funds to companies, often those that are middle-market, highly leveraged, or in niche sectors, and which may find traditional bank financing less accessible or flexible. Unlike publicly traded bonds or bank loans, private credit transactions are generally bilateral, illiquid, and largely unregulated.
The Ascent of Private Credit
The growth of private credit has been remarkable. Data from Preqin, a leading alternative assets data provider, indicates that the global private credit market’s assets under management (AUM) surged from approximately $400 billion in 2010 to over $1.5 trillion by 2023, with projections suggesting it could exceed $2 trillion by 2027. This expansion has been fueled by several factors: banks facing stricter capital requirements post-2008 have scaled back some riskier lending activities, institutional investors (pension funds, endowments) seeking higher yields in a low-interest-rate environment, and borrowers valuing the speed and flexibility offered by private lenders.
Lack of Transparency and Regulatory Gaps
The opacity of private credit markets presents a significant challenge for financial regulators. Unlike traditional banks, which are subject to stringent oversight, capital requirements, and stress tests, private credit funds operate with far less public disclosure. This lack of transparency makes it difficult for bodies like the Federal Reserve to accurately assess the true extent of leverage within the system, the quality of underlying assets, or the interconnectedness of various financial entities. If a downturn were to occur, particularly in a sector like AI where valuations can be highly speculative, widespread defaults in private credit portfolios could trigger unanticipated contagion, impacting institutional investors and potentially spilling over into broader financial markets. The inability to precisely map these exposures creates a "shadow banking" risk that Gomes highlights as a critical vulnerability.
Echoes of History: Lessons from Past Bubbles
Gomes draws crucial parallels between the current AI investment frenzy and two distinct, yet instructive, historical episodes: the dot-com bubble of the late 1990s and the 2008 housing crisis. These comparisons serve as potent reminders of how speculative investment, unchecked financial innovation, and regulatory blind spots can coalesce into systemic risk.
The Dot-Com Frenzy: Speculation and Overvaluation
The late 1990s witnessed an unprecedented surge in investment in internet-related companies. Many of these firms, often with nascent business models, minimal revenue, or even non-existent profits, commanded astronomical valuations based on speculative future potential rather than tangible earnings. The belief that "this time is different" permeated the market. When the bubble burst in early 2000, trillions of dollars in market capitalization evaporated, leading to widespread investor losses and a significant, albeit contained, economic downturn. Gomes’s analogy suggests that current AI valuations, particularly for some early-stage companies, might similarly be detached from fundamental economic realities, relying heavily on future promise rather than present-day profitability. The enthusiasm for generative AI, for instance, has driven up valuations for chip manufacturers, software developers, and data center operators, sometimes to levels that raise questions about sustainability.
The 2008 Housing Crisis: Systemic Risk and Shadow Banking
The 2008 global financial crisis offered a more catastrophic lesson in systemic risk. It stemmed from a combination of loose lending standards in the subprime mortgage market, the securitization of these risky loans into complex financial instruments (like collateralized debt obligations, CDOs), and their widespread distribution across the financial system. Critically, much of this activity occurred outside the traditional banking sector, in what became known as the "shadow banking" system. Regulators struggled to track the exposures and interconnectedness, leading to a rapid and devastating cascade of failures when the underlying assets (subprime mortgages) began to default. Gomes posits that the opaque nature of private credit today bears a resemblance to aspects of the shadow banking system pre-2008, where risks were allowed to fester and amplify away from the direct scrutiny of central banks and prudential regulators.
AI’s Unfolding Productivity Puzzle and Economic Impact
While the immediate focus is on financial risks, Gomes also touches upon the profound potential of AI to impact long-term productivity. The optimists argue that AI could usher in a new era of economic growth, akin to the industrial revolution or the advent of the internet, by automating tasks, enhancing decision-making, and fostering entirely new industries. This "supply-side" boost could theoretically lead to higher living standards, increased wealth, and even help combat inflationary pressures in the long run by making production more efficient.
Anticipating Transformative Productivity Gains
Major technology companies and consulting firms project significant economic benefits from AI. For example, PwC estimated in 2017 that AI could contribute up to $15.7 trillion to the global economy by 2030, with a 14% boost to global GDP. More recent analyses from McKinsey and others continue to highlight AI’s potential to automate a substantial portion of current work tasks, freeing up human capital for more creative and complex endeavors, thereby driving productivity growth. Investment in AI infrastructure, research, and development has soared, with venture capital funding for AI startups reaching record highs in recent years, despite some recent cooling.
The Lag Effect and Investment Dynamics
However, Gomes points out that the realization of these productivity gains is neither guaranteed nor instantaneous. There is often a significant lag between technological innovation and its widespread economic impact. For instance, it took decades for electricity to fully transform manufacturing and society. Similarly, AI’s full productivity potential may only materialize after substantial investments in complementary technologies, workforce retraining, and organizational restructuring. In the interim, massive investments are being poured into an ecosystem where the immediate returns on capital are often unclear, and the path to profitability for many AI ventures remains nebulous. This disjunction between current investment levels and uncertain future productivity creates a fertile ground for misallocation of capital and potential asset bubbles, especially if speculative financing methods are prevalent.
The Imperative for Enhanced Regulatory Foresight
Gomes’s ultimate call to action is for the Federal Reserve and other financial regulators to develop better models and data for understanding the complex and evolving financial risks surrounding AI. The traditional models used by central banks often rely on historical data and established economic relationships, which may not adequately capture the dynamics of rapidly evolving technological sectors or the unique characteristics of new financial instruments like private credit.
Challenges in Risk Assessment
Assessing risk in the AI sector is particularly challenging. It involves evaluating highly specialized technology, often developed by startups with unique intellectual property and unproven business models. The valuation of these companies can be highly subjective, based on projected future revenue streams that are difficult to quantify. Furthermore, the rapid pace of AI development means that risks can emerge and evolve quickly, making it difficult for regulators to keep pace. The interconnectedness of AI investments, from chip manufacturers to cloud providers to application developers, also means that a shock in one area could propagate widely.
Calls for Proactive Regulatory Frameworks
A proactive approach would involve a multi-pronged strategy. This could include:
- Enhanced Data Collection: The Fed and other agencies need better, more granular data on private credit markets, including borrower profiles, loan terms, and portfolio performance, potentially requiring new reporting requirements for private funds.
- Developing New Analytical Tools: Investing in economic models that can incorporate technological disruption and assess non-traditional financial flows and risks.
- Cross-Agency Coordination: Fostering stronger collaboration among domestic regulators (e.g., the Fed, Treasury, SEC, FSOC) and international bodies to share information and synchronize oversight.
- Scenario Planning and Stress Testing: Conducting stress tests that specifically account for shocks originating in the tech sector or private credit markets, rather than just traditional banking risks.
- Focus on Systemic Risk: Explicitly integrating financial stability considerations more prominently into monetary policy discussions and regulatory mandates.
Broader Implications for the Global Economy
The debate ignited by Gomes’s observations extends beyond the purview of the Federal Reserve, touching upon broader implications for the global financial system and economic policy. The challenge lies in striking a delicate balance: fostering innovation and allowing capital to flow to transformative technologies like AI, while simultaneously safeguarding against the build-up of systemic risks that could trigger another financial crisis.
If regulators fail to adequately monitor the opaque financing of the AI boom, the consequences could be severe. A sudden correction in overvalued AI assets, exacerbated by illiquid private credit markets, could lead to significant write-downs for institutional investors, trigger broader market instability, and potentially dampen future innovation by making capital more scarce. The globalized nature of finance means that such shocks could easily reverberate across borders, impacting economies worldwide.
Ultimately, the AI revolution presents both an unprecedented opportunity and a formidable challenge for economic policymakers. The call for the Federal Reserve to broaden its focus beyond immediate inflation concerns and rigorously scrutinize emerging financial stability risks, particularly those hidden within the rapidly expanding and less transparent corners of the financial system, is a timely and critical one. Ensuring the long-term health and stability of the economy requires not only harnessing the power of new technologies but also meticulously managing the financial architecture that supports them.
About the Author
