Kim Young-ik
The author is an adjunct professor at Hanyang University’s Institute for Future Education.
Financial crises always appear in new forms, which is why people insist that “this time is different.” In 1929, the stock market collapsed. In 2000, the technology bubble burst. In 2008, the financial system itself broke down. Today, the global economy stands at another turning point shaped by AI. While appearances change, the underlying structure repeats. Technology evolves, but human behavior and the dynamics of credit remain largely the same.
Traders react in the S&P 500 pit at the Chicago Mercantile Exchange, September 15, 2008. Global markets plummeted on Monday after investment bank Lehman Brothers filed for bankruptcy protection, rival Merrill Lynch agreed to be taken over and the Federal Reserve threw a life line to the battered financial industry. [REUTERS/YONHAP]
In the 1920s, electricity enabled round-the-clock factory operation, while automobiles transformed logistics and mobility. Mass production lowered costs and raised productivity, supporting economic growth and driving stock prices higher. Yet stock prices reflected expectations of future gains far beyond actual earnings. Margin trading allowed investors to buy large amounts of stock with limited capital. Rising prices attracted more borrowing, reinforcing the cycle.
When markets began to weaken in October 1929, margin calls triggered forced selling. Prices fell further, leading to more liquidation and a rapid collapse. The Dow Jones Industrial Average plunged from 380 in August 1929 to 43 in June 1932, a decline of 89 percent. The crisis extended beyond markets. Banks failed, money supply contracted, and credit across the economy deteriorated. Firms struggled to secure funding, consumption weakened and unemployment surged. The episode marked not just a market correction but a collapse of credit itself.
The dot-com bubble of 2000 began differently, driven by technological optimism. The internet transformed information flows and created new industries. Productivity improved, with annual labor productivity growth rising to 2.9 percent between 1996 and 2000 from 1.5 percent in the prior period. Optimism dominated markets, but investors incorporated future potential into present valuations too aggressively.
Valuations surged to extreme levels. The price-to-earnings ratio of the S&P 500 approached 50, while companies without profit models attracted large investments. Initial public offerings and venture capital expanded rapidly, with capital moving faster than the real economy. The bubble burst when the Federal Reserve raised its policy rate from 4.7 percent in May 1999 to 6.5 percent by March 2000. The Nasdaq fell 78 percent, and many firms disappeared. The internet remained, ultimately becoming core infrastructure. The problem was not the technology but that markets had priced its future too quickly.
The global financial crisis of 2008 differed again. This time, financial structure was the source of instability. Low interest rates drove housing prices higher and fueled household debt. That debt was repackaged into mortgage-backed securities and collateralized debt obligations, spreading risk across the system. Financial institutions believed risk had been reduced and increased leverage accordingly.
However, the system proved both complex and fragile. As housing prices fell, delinquencies rose, and asset values deteriorated sharply. Trust between banks collapsed, and funding markets froze. This was not simply an asset price adjustment but a breakdown of the financial system itself.
Today, AI represents a new platform reshaping industry and productivity. Yet it requires massive capital. Investment in data centers, semiconductors and power infrastructure is rising rapidly, based largely on expectations of future returns rather than current profits. In effect, an unrealized future is already determining present investment and asset prices.
Credit is expanding again, particularly in private credit markets tied to AI investment. This creates a structure that combines valuation expansion with hidden credit accumulation. In this sense, 2026 may reflect both the valuation excess of 2000 and the credit vulnerability of 2008.
Signs of risk are already emerging. Capital is concentrated in a small number of AI-related firms, creating a fragile market structure. Rising oil prices and geopolitical tensions sustain inflation and limit interest rate cuts, placing pressure on long-term AI investments while weighing on both equities and bonds. Meanwhile, strains are emerging in private credit markets, where some funds are facing liquidity pressure and restricting redemptions, echoing earlier structured finance risks.
The key issue is not price but credit. While the current phase may appear as a price adjustment, a deeper shift could occur if credit stress intensifies. The possibility of simultaneous valuation and credit collapse cannot be dismissed.
Stock market data is displayed on a screen in the dealing room of Hana Bank’s headquarters in central Seoul on the afternoon of March 31, as tensions in the Middle East escalate. The won-dollar exchange rate stood at 1,530.1 won per dollar, up 14.4 won from the previous day’s daytime closing price at 3:30 p.m. [NEWS1]
A recurring pattern is evident. Innovation attracts capital, capital expands credit, credit drives prices higher, and rising prices foster overconfidence that eventually leads to collapse. The critical variable is not technology but human psychology. Bull markets create conviction, which justifies leverage and reinforces the belief that conditions have fundamentally changed.
This time may indeed differ. AI and the scale of investment could produce new outcomes. Yet history suggests that while forms change, responses to crisis remain similar. Preparing for adverse scenarios may be the most realistic strategy.
For financial institutions, survival matters more than returns. Leverage may appear efficient during expansion, but it becomes a primary vulnerability in downturns. Liquidity is essential. As seen in 2008, asset values matter little if they cannot be converted into cash.
Corporate stress often emerges through unstable funding structures before demand declines. During expansion, growth dominates decision-making, but when conditions shift, cash flow becomes decisive. Much of today’s AI investment is based on expectations, making it vulnerable to higher interest rates or delayed demand.
Reducing leverage, securing liquidity, preserving cash flow, understanding structural risks and maintaining patience are essential principles. In periods of uncertainty, sustainability is more important than growth.
This article was originally written in Korean and translated by a bilingual reporter with the help of generative AI tools. It was then edited by a native English-speaking editor. All AI-assisted translations are reviewed and refined by our newsroom.
댓글
0