AI Capital Expenditure Is Draining Market Liquidity: A Quiet “Reverse QE”
TechFlow Selected TechFlow Selected
AI Capital Expenditure Is Draining Market Liquidity: A Quiet “Reverse QE”
When capital becomes scarce, people must rigorously evaluate its most efficient uses, and the cost of capital—i.e., the market interest rate—rises accordingly.
Navigating Web3 tides with focused insightsBy: plur daddy
Translated by: AididiaoJP, Foresight News
We are facing a fundamental shift in the market landscape driven by an artificial intelligence (AI) capital expenditure cycle that is causing a shortage of financial capital.
This will have profound implications for asset prices—especially given that capital has been abundant for an extended period. The Web 2.0 and SaaS business models that fueled market prosperity throughout the 2010s required minimal capital, allowing vast amounts of excess funds to flow into speculative assets across the board.
Yesterday, while reflecting on current market conditions, it suddenly clicked for me. This may be the most insightful piece I’ve written in a long time. Below, I’ll break down the underlying mechanics step by step.
The AI capex cycle bears similarities to government fiscal stimulus—a helpful analogy for understanding how it operates.
Under fiscal stimulus, governments issue long-dated bonds, which the private sector absorbs. The government then deploys the proceeds into real economic activity, generating a multiplier effect as money circulates through the economy. Thanks to this multiplier, the net impact on financial asset prices is positive.
In the AI capex cycle, hyperscale tech firms raise capital by issuing debt or selling Treasury securities (and other assets), again with the private sector absorbing the duration. The firms then deploy those funds into projects. These funds likewise circulate through the real economy and generate a multiplier effect, positively impacting financial asset prices.
So long as idle capital remains available in the economy, this process runs smoothly. It’s highly effective—and broadly lifts markets. That’s precisely what happened over the past few years: AI capex acted like additional macroeconomic stimulus, boosting both the real economy and financial markets. But here’s the problem: once idle capital is exhausted, every dollar allocated to AI must be drawn from elsewhere—triggering a fierce competition for capital. When capital becomes scarce, investors must rigorously prioritize its most efficient uses, pushing up the cost of capital (i.e., market interest rates).
Let me reiterate: when capital is scarce, asset classes diverge sharply. The most speculative assets suffer disproportionately—just as they benefited disproportionately during periods of capital abundance but scarce productive investment opportunities. Viewed this way, AI capex functions as a kind of “reverse quantitative easing,” exerting a negative rebalancing effect on portfolios.
Fiscal stimulus rarely faces this dilemma because the Federal Reserve typically acts as the buyer of last resort for Treasuries, thereby avoiding crowding out other capital uses.
Here, the term “capital” is interchangeable with “liquidity.” The word “liquidity” is confusing, however, because it carries multiple meanings.
Think of capital—or liquidity—as water. You need a higher water level in the bathtub to lift financial assets (those floating rubber ducks). There are several ways to achieve this: increase total water volume (e.g., rate cuts or QE), unclog the inflow pipe (e.g., current reverse repo operations, which function as “pipe unclogging”), or reduce the drain rate.
Most current discussions about liquidity in the economy focus narrowly on money supply. Yet money demand is equally critical. What we’re seeing now is excessive demand—hence the crowding-out effect.
Media reports indicate that the world’s wealthiest investors—including the Saudi sovereign fund and SoftBank Group—are nearly tapped out. Over the past decade, global investors have been “well-fed,” holding large positions across asset classes. Let’s consider what this implies: when Sam Altman reaches out asking them to honor prior funding commitments, unlike in earlier years when capital was plentiful, they now must first sell some assets to raise cash. What do they sell? Likely their least-conviction holdings: recent underperformers like Bitcoin, SaaS stocks facing industry disruption, or underperforming hedge fund shares. Those hedge funds, in turn, must liquidate assets to meet redemptions. Falling asset prices erode market confidence, tighten financing conditions, and trigger further selling—creating cascading effects across financial markets.
Compounding matters, Trump selected Walsh. This is especially concerning because he believes the current problem is too much money—whereas the reality is precisely the opposite. That’s why market developments have accelerated since his nomination.
I’ve long sought to understand why memory chipmakers—such as SNDK and MU—have vastly outperformed other equities. Surging product prices certainly contribute. But more importantly, these companies’ current and near-term earnings are exceptionally strong—even though everyone knows those earnings are cyclical and will inevitably decline. As the cost of capital rises, so does the discount rate. Long-duration, speculative assets reliant on distant future cash flows come under pressure, while assets delivering cash flow in the near term gain favor.
In this environment, cryptocurrencies—acting as a sensitive barometer of liquidity—naturally suffer severe damage. That’s why their recent decline appears bottomless.
Highly speculative retail favorites struggle to sustain gains—even sectors with improving fundamentals face headwinds.
With capital demand exceeding supply, yields on both sovereign and credit bonds are rising.
Blind optimism and indiscriminate long positioning are no longer viable.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
@plur_daddy
