The US effective tariff rate went from 2.3% at the end of 2024 to 15.8% by mid-2026 — the most dramatic reversal in American trade policy since the Smoot-Hawley Act of 1930 helped push the world into the Great Depression. The goals were loudly stated. The consequences were what economics has always said they would be. This letter is about learning to read what trade policy actually produces, not what it intends.
Not investment advice. Not a recommendation to buy or sell. Research and long-horizon thinking only. Consult a qualified financial advisor before making any investment decision. Figures cited are sourced from the Tax Foundation, J.P. Morgan Global Research, the WTO, the Penn Wharton Budget Model, and the Yale Budget Lab, current as of June 2026.
There is a simple but powerful discipline in economic thinking that most policy debates skip entirely: asking not just what a policy is designed to do, but what it will actually produce once all the parties it affects respond to it. The economist Thomas Sowell, in his foundational work on economic reasoning, calls this moving past "stage one thinking" — the assumption that a policy achieves its stated goal and then stops, as if the world does not react. Every tariff in history has been designed with the clearest, most patriotic-sounding intentions: protect jobs, rebuild industry, reduce dependence on rivals, restore national strength. Stage one looks good. Stage two is where the economics happens.
The 2025-2026 tariff episode is one of the sharpest and most compressedly documented tests of this principle in living memory. Within twelve months, the United States went from the 2.4% effective tariff rate of late 2024 to 15.8% by mid-2026 — a rate not seen since 1947. The goals were explicit: reduce trade deficits, bring manufacturing home, rebalance a global economy the administration argued had taken advantage of American openness for decades. The result was a measurable economic story. Factory employment in the United States declined. Consumer prices rose faster than expected. Three thousand new trade and industrial policy measures were introduced globally in 2025 alone — three times the annual level recorded a decade earlier. That is stage two beginning to arrive.
"The first lesson of economics is scarcity. The first lesson of politics is to disregard the first lesson of economics." — Thomas Sowell
The Smoot-Hawley Tariff Act of 1930 is the defining case study in what happens when stage-one thinking becomes law. It raised average US tariffs to nearly 20% — the highest in well over a century — on the explicit premise that domestic manufacturers would capture demand previously going to imports, and that American workers would keep their jobs. The premise was plausible. The result was not: within five months, the unemployment rate had risen to double digits, and it never fell below that level for any month during the entire decade of the 1930s. Trading partners retaliated. Global trade collapsed. The policy designed to protect American workers accelerated the very spiral it was meant to prevent.
In a 2026 interview at Stanford's Hoover Institution, Sowell — at 95, still sharp and still in print — applied this exact parallel directly to today's tariff regime. He cited the Smoot-Hawley passage in his own Basic Economics by name, against a tariff schedule including an average of 39% on Chinese goods and 25% on Canadian and Mexican imports, and noted that the logic being offered in 2025-26 was recognizable: plausible at stage one, historically expensive at stage two. The parallels are uncomfortable not because the situation is identical — it isn't — but because the mechanism is: import taxes raise prices, invite retaliation, and redirect trade rather than eliminating the deficit that motivated them in the first place.
On February 20, 2026, the US Supreme Court ruled 6-3 that the International Emergency Economic Powers Act — the legal basis for roughly 61% of the year-to-date tariff increase as of late 2025 — does not authorize tariffs. The ruling immediately invalidated a significant portion of the tariff framework that had been in place for most of 2025. The administration responded by shifting to Section 232 (national security tariffs) and Section 122, imposing a 10% universal baseline tariff and keeping China's effective rate at punishing levels through alternative legal authorities.
The ruling did not end the tariff era — it reshuffled the legal instruments producing it. Tax Foundation's April 2026 model estimates that the remaining permanent Section 232 tariffs alone will reduce long-run US GDP by 0.3% before foreign retaliation is accounted for, falling to $697 billion in net revenue over the decade once the negative economic feedback loops are modeled. For investors, the February ruling introduced a new layer of policy-path uncertainty: the administration's trade strategy is now being rebuilt on different statutory foundations than it started on, mid-implementation, with full legal resolution still pending.
Chinese goods are not disappearing from the US market — they are arriving via Vietnam, Mexico, and other intermediary jurisdictions that repackage or finish production locally. A new USTR Section 301 investigation covering 16 major manufacturing hubs is targeting exactly this mechanism, with additional tariffs potentially arriving in late 2026. The game is still being played; the routing has changed.
The administration has signaled tariffs on pharmaceuticals could reach toward 200% by mid-to-late 2026. Large-cap pharma companies have been building US inventory and beginning domestic manufacturing buildouts, but mid-cap and specialty drug makers with less balance sheet flexibility face genuine near-term cost exposure once inventory buffers run out, estimated between mid-2026 and early 2027 for most names.
GM absorbed $3.1 billion in tariff costs in 2025 below expectations; Toyota, Nissan, and Honda have all announced expanded US domestic manufacturing rather than absorbing ongoing tariff costs indefinitely. This is precisely the supply chain restructuring the tariffs were designed to induce — but it takes 3-5 years to materialise in volume, not the quarters investors are pricing.
The Turnberry deal locks EU exports to the US at a 15% tariff rate and signals a broader pattern: bilateral deals that institutionalise higher tariffs as the new baseline rather than restoring pre-tariff free-trade conditions. For investors, this means the tariff regime is increasingly permanent infrastructure rather than a temporary negotiating tactic — supply chain and pricing decisions should be modelled on the new baseline, not a return to 2024 conditions.
The case this letter makes — that tariffs are producing stage-two consequences at odds with their stage-one intentions — is itself the mainstream economic consensus, not a contrarian position. The Tax Foundation, J.P. Morgan, the WTO, the Penn Wharton Budget Model, and Yale Budget Lab all arrive at similar findings through independent methodologies. That agreement does not make the finding wrong, but investors should be aware that "economists say tariffs are costly" has been true in every tariff episode in modern history, and has not historically predicted when tariff policies reverse — politics, not economics, drives that timing, and politically the 2026 tariff regime retains significant domestic support despite measurable economic costs.
The strongest honest counter-argument to this letter's framing is the supply chain resilience argument, and it deserves a genuine hearing. The efficiency-maximising global supply chains of 2010-2024 were not resilient — the pandemic, the Suez Canal blockage, and the semiconductor shortage demonstrated that vividly. If tariff-driven supply chain restructuring produces genuinely more robust domestic production capacity over a 5-10 year horizon, some of the near-term GDP cost may be a legitimate price paid for reduced systemic fragility. Whether that tradeoff actually materialises depends on whether the policy is sustained long enough for manufacturers to complete investment cycles rather than reversing once the political cost of consumer price inflation rises — an open question in June 2026.
The lesson that Smoot-Hawley wrote in 1930 and that this letter's 2026 data confirms is not that trade policy cannot achieve legitimate goals — it can, over long enough horizons, with consistent enough application. The lesson is narrower and more practical: the gap between what a policy says it will do and what it produces once markets, trading partners, and supply chains respond to it is where investment risk actually lives. Plausible at stage one. Expensive at stage two. Understanding which stage you are in is, in the end, what economic thinking is for.
Long-horizon thinking on capital, technology, and the forces shaping the next decade of wealth creation. Written from first principles. Not consensus. Not noise.