Trump may be raising your taxes with his tariffs but he could actually cut inflation with them, too, SF Fed says

San Francisco Fed researchers Regis Barnichon and Aayush Singh find historical evidence (1870–1913 and the interwar period) of a strong negative correlation between tariffs and inflation — historically a one percentage-point tariff rise associated with a 0.6 percentage-point decline in inflation — and argue the large tariff increase in 2025 (average U.S. tariff roughly 15% versus under 3% at end-2024) could depress inflation while raising unemployment via demand- and uncertainty-driven channels. They caution, however, that modern U.S. exposure to imported inputs (imports ≈ $3.2 trillion in 2024) and structural differences since the early 20th century mean historical results may not fully apply today.

Analysis

Market structure: A tariff-driven demand shock can reallocate pricing power toward long-duration assets and domestic incumbents with low import exposure. Expect retail, discretionary and import-reliant supply chains to lose margin while utilities, long-duration growth and government bonds gain; commodities (industrial metals, crude) are likely to see downward price pressure within 1–6 months as global demand softens.

Risk assessment: Tail risks include rapid retaliatory tariffs, a supply-chain squeeze that reverses into cost-push inflation, or fiscal stimulus that offsets demand weakness; each would flip asset correlations in weeks. Immediate window (days) will be dominated by volatility spikes; short-term (0–3 months) sees disinflation and equity repricing; long-term (3–24 months) depends on whether input share (currently high) turns the tariff into persistent inflation.

Trade implications: Position for disinflation + weaker consumption: favor duration (long Treasuries/TLT) and long-duration growth (QQQ) while underweight retail/discretionary (XRT/XLY) and industrial commodities; use 2–6 month option structures to exploit elevated IV on retail. Pair trades: go long TLT and short XRT or long QQQ vs short XLY to capture relative safety/growth vs cyclicals within a 3–6 month horizon.

Contrarian angles: Consensus overlooks that higher tariff rates can be inflationary if imported input share remains large—so a reversal risk exists if CPI prints accelerate >0.4pp month-over-month. Historical parallels (Smoot‑Hawley era) are imperfect; the market may be overpricing persistent disinflation—create asymmetric positions that profit from both further disinflation and a sudden re-inflation shock (e.g., long TLT with hedges).

Key Decisions for Investors

• Establish a 2.5% portfolio notional long position in TLT (or 10y Treasury futures) over the next 2 weeks; target price if 10y yield falls from ~3.6% to 2.8–3.2% (approx. +6–12% TLT), stop-loss if yield >4.0%.
• Construct a dollar-neutral pair: long QQQ 1.5% notional and short XRT 1.5% notional (equal dollars) for 3–6 months to capture rotation from discretionary to large-cap growth; unwind if unemployment rate does not rise by ≥0.2ppt in 2 months or CPI falls <0.3pp YoY.
• Buy 3-month put spread on XRT (sell =25-delta, buy =10-delta) sized 1% notional to hedge consumer weakness and exploit elevated IV; target 200–300% pay-off if retail ETF drops 15–25%.
• Increase USD exposure to 1–2% notional (UUP) and underweight EM equities by 2–4% if risk assets sell off >5% intraday or oil falls >10% in 30 days, indicating global demand shock.
• Monitor three near-term data/catalysts: monthly CPI/PCE prints and 60-day tariff legislative developments. If CPI prints decelerate by ≥0.5pp cumulative over two months, add another 1–2% duration; if tariffs are rolled back, close duration and rotate into industrial cyclicals (XLI) within 30 days.