There is a specific kind of silence that falls over a trading desk when a political leader says something not just wrong, but mathematically impossible. It is the silence of risk models breaking.
Donald Trump recently claimed that the United States has brought in “$18 trillion” from tariffs. He didn’t say billions. He said trillions. He framed this figure not as a projection, but as money in the bank: assets realized, debts paid, a windfall of historic proportions.
This is a lie.
It is not a “spin,” or an “exaggeration,” or “alternative math.” It is a complete fabrication, detached from any government dataset, Treasury report, or account ledger in existence. The actual cumulative customs duties collected by the government during the peak of the trade war (2018-2019) were roughly $70-80 billion annually. Even if you sum every dime of tariff revenue collected since the founding of the Republic, you do not reach $18 trillion.
For the average voter, this might play as standard campaign bluster. But for the market. Specifically for the capital-intensive, supply-chain-dependent tech sector. This detachment from economic reality is a flashing red signal. It introduces a “Competence Discount” into equity valuations. If the person driving the world’s largest economy believes he has $18 trillion that does not exist, how can he effectively manage the delicate, razor-thin margins of the global semiconductor trade?
The market is no longer pricing in a policy shift. It is pricing in a hallucination. And for tech investors, that is a terrifying variable to solve for.
The Arithmetic of Absurdity
To understand why the $18 trillion figure is so jarring to institutional investors, one has to look at the scale. In finance, specific numbers matter. Scale matters. When a number violates the basic laws of arithmetic, it indicates a breakdown in governance logic.
The Scale Gap
The total Gross Domestic Product (GDP) of the United States. The value of every single thing produced, sold, and serviced in the country in a year is roughly $27 trillion.
For the U.S. government to have collected $18 trillion in tariffs, it would have had to effectively confiscate two-thirds of the entire U.S. economy’s output solely through border taxes.
Let’s look at the actual import data. The United States imports roughly $3.8 trillion worth of goods and services annually. To generate $18 trillion in revenue over a standard four-year term, the government would need to collect $4.5 trillion per year.
If total imports are only $3.8 trillion, this means the tariff rate would need to be greater than 100%. Specifically:
A 118% tariff on all imports: from bananas to iPhones to crude oil, would not generate revenue. It would stop trade immediately. Imports would fall to zero. The revenue collected would be zero. This is the Laffer Curve in its most brutal form. You cannot tax a transaction that no longer happens.
The claim fails the most basic sanity check available to a first-year economics student. It suggests a fundamental misunderstanding of the relationship between tax rates, trade volume, and revenue.
The Real Numbers
According to the Bureau of Economic Analysis (BEA) and the Treasury Department, actual customs duties (tariffs) collected annually look like this:
- Pre-2018: ~$30-40 billion/year
- 2018-2020 (Trade War Peak): ~$70-85 billion/year
- 2023-2024 (Baseline): ~$77-80 billion/year
- 2025 (The “Reciprocal” Spike): ~$240 billion (Year-to-Date)
Even with the massive escalation in tariffs seen throughout 2025, where monthly collections hit a record $30 billion in November, the total is still roughly 1.3% of the claimed $18 trillion. The government is not just missing the target; they are in a different galaxy.
The gap between “$240 billion” and “$18 trillion” is not a rounding error. It is a factor of 75. It is the difference between the distance to the grocery store and the distance to the moon.
The “Chaos Premium” in Tech Valuations
Why does this matter for your portfolio? Why should a holder of NVDA, AAPL, or TSM care about a politician’s loose relationship with math?
Because markets hate randomness. And blatantly false rhetoric is the ultimate generator of randomness.
When a policymaker proposes a controversial policy (e.g., “The government will raise the corporate tax rate to 28%”), the market can model it. Analysts open their spreadsheets, adjust the tax line, lower EPS estimates by 4%, and the stock price re-rates. It is rational. It is calculable.
But when a policymaker says, “The Treasury has $18 trillion in the bank” (when the vault is empty), there is no model for that. It signals that the inputs into the policy-making machine are broken. If the inputs are fake, the outputs: future regulations, trade restrictions, capital controls, become completely unpredictable.
This unpredictability forces institutional allocators to increase the Discount Rate.
Discounting the Future
Tech stocks are “long-duration assets.” Their value is derived heavily from cash flows expected 10, 15, or 20 years in the future (e.g., the AI revolution of the 2030s).
The formula for the present value of a stock is roughly:
Where is the discount rate (risk-free rate + equity risk premium).
When governance credibility collapses, the “Equity Risk Premium” spikes. Investors demand a higher return to hold assets in a jurisdiction where the rules of reality might change overnight. A 1% increase in the discount rate can crash the valuation of high-growth tech stocks by 20-30%.
They aren’t selling Nvidia because they think AI is over. They are selling because the risk of holding assets in a chaotic regulatory environment has just gone up.
Why Tech is the Casualty
The technology sector is uniquely exposed to this specific brand of “Policy Hallucination.” Unlike a domestic utility or a real estate investment trust, Big Tech is:
- Capital Intensive: It relies on massive, long-term bets (fabs, data centers).
- Hyper-Globalized: supply chains cross borders dozens of times.
- Fragile: A single disruption in neon gas or lithography lenses stops the whole machine.
The Semiconductor Choke Point
The semiconductor supply chain represents a finely tuned watch. It relies on the free flow of IP from the United States, chemicals from Japan, lithography from the Netherlands, and fabrication in Taiwan.
If American trade policy is based on the belief that a 100% tariff will generate trillions in revenue, the world could see policies that inadvertently shatter this supply chain.
- Nvidia (NVDA): Relies on TSMC for 100% of its leading-edge GPUs. A trade war sparked by a misunderstanding of tariff mechanics could lock Nvidia out of its own manufacturing base.
- Apple (AAPL): While diversifying, Apple remains tethered to mainland China. Retaliatory tariffs: which are inevitable when one side operates on false math, would destroy Apple’s gross margins instantly.
- AI Infrastructure: The build-out of U.S. data centers depends on imported power equipment (transformers, HVDC cables). Tariffs on these items raise the cost of AI compute, slowing the entire generative AI rollout.
The credibility of the “Chips Act”
The U.S. is currently trying to reshore semiconductor manufacturing via the CHIPS Act. This requires tens of billions of dollars in subsidies and tax breaks.
If the executive branch believes it has an imaginary $18 trillion surplus, it may recklessly cut funding for these critical programs, assuming the “tariff windfall” will cover it. When the windfall never arrives, the fabs currently under construction in Arizona and Ohio could be left stranded, half-built monuments to bad math. Intel and TSMC need 5-year certainty, not 5-minute improvisations.
Perception vs. Reality: The Governance Discount
The most damaging aspect of the $18 trillion lie is not the number itself, but what it says about the feedback loop of the leadership.
Effective governance requires a feedback loop. You implement a policy, you measure the data, you adjust. If the leadership is insulated from data. If they genuinely believe failures are massive successes. There is no mechanism for correction.
For the market, this is the definition of “Uninvestable.”
Investors are already seeing the “Governance Discount” applied to other markets. Emerging markets with erratic leaders trade at 8x or 9x earnings. Stable markets like the U.S. typically trade at 18x-22x earnings.
The premium U.S. stocks enjoy. The reason the S&P 500 is the envy of the world, is based on the assumption of Rational Governance. It is based on the rule of law, the sanctity of contracts, and the reliability of Treasury data.
When the likely head of state stands up and fabricates economic data on a scale exceeding the GDP of China, that “Rational Governance” premium starts to erode. Investors ask: “If he believes this, what else does he believe?”
- Does he believe debt defaults are cost-free?
- Does he believe interest rates are optional?
- Does he believe the dollar’s reserve status is immutable?
Conclusion: The Cost of Denial
The $18 trillion claim is bullshit. There is no other word for it that maintains the necessary precision. It is a number pulled from the ether, defensible by no model, supported by no reality.
But for your portfolio, the truth of the number matters less than the source of the number. The market is a machine that processes information. When the source of the most critical information: U.S. economic policy-making corrupted by fantasy, the machine malfunctions.
Volatility is the price of uncertainty. When the Commander-in-Chief creates uncertainty on an $18 trillion scale, volatility is the only guaranteed outcome.
For tech investors, the lesson is clear: Defensive posturing is back. The era of “efficient markets” assumes rational actors. The market is entering an era where that assumption no longer holds. In a world where policy is driven by imaginary numbers, real cash flows are the only safe harbor left. Ignore the rhetoric. Watch the discount rates. And never, ever assume that the guys in charge have done the math.
🦋 Discussion on Bluesky
Discuss on Bluesky