Politicians love to tell voters that China pays America's tariffs. The New York Federal Reserve has a different story: American households paid $831 per year in additional costs during peak tariff periods between 2018-2020. Every dollar collected went straight from American wallets to the U.S. Treasury — with foreign exporters paying exactly zero.

Key Takeaways

  • The bottom income quintile faced effective tariff tax rates of 4.4% versus 3.2% for top earners
  • Washing machine tariffs triggered 12% price increases across all brands — including American-made Whirlpool and GE
  • Each protected steel job cost consumers $900,000 annually while destroying 1.4 jobs in steel-consuming industries

The Mechanism Everyone Misunderstands

Here's what actually happens when a tariff hits. Walmart imports Chinese electronics. Before those products touch American soil, Walmart writes a check to U.S. Customs and Border Protection for the tariff amount. Not Beijing. Not Shenzhen manufacturers. Walmart.

The 2018-2020 trade war created the perfect natural experiment. The U.S. slapped tariffs on $350 billion worth of Chinese imports — rates from 7.5% to 25%. Peterson Institute economists tracked what happened next: 92-98% of those costs showed up in American consumer prices within 30-60 days.

But the price increases exceeded the tariff amounts. When Trump imposed 20% tariffs on washing machines in January 2018, University of Chicago researchers documented something curious. Prices rose $86-92 per unit across all brands. Even Whirlpool — the American company the tariffs were supposed to protect — jacked up prices. Strategic complementarity, economists call it. Translation: when import competition gets kneecapped, domestic producers cash in too.

graphs of performance analytics on a laptop screen
Photo by Luke Chesser / Unsplash

The Numbers That Politicians Don't Quote

Federal Reserve economists broke down the $831 annual burden: $690 from China tariffs, $75 from steel and aluminum duties, $66 from washing machine and solar panel tariffs. Those aren't abstract statistics. That's rent money.

The geographic breakdowns reveal the political absurdity. West Virginia households — Trump country — faced tariff burdens equivalent to 1.4% of median income. Connecticut families paid just 0.8%. The Tax Foundation's analysis shows why: lower-income families spend more on goods, less on services. Goods get tariffed. Services don't.

Employment effects? Brutal. Steel and aluminum tariffs saved 8,700 jobs in those industries while destroying 12,400 jobs in steel-consuming sectors like automotive and construction. Net jobs lost: 3,700. Each protected steelworker cost consumers $900,000 annually.

The import substitution data tells the real story. Chinese imports in targeted categories fell 21%. American production in those categories? Barely budged. Instead, 73% of lost Chinese imports got replaced by Vietnam, Mexico, and Taiwan. Same products, different flags, higher prices for Americans.

What The Data Actually Reveals

The deeper story here isn't about trade balances or manufacturing jobs. It's about the most regressive tax increase in decades disguised as economic nationalism. Comprehensive Bureau of Labor Statistics analysis shows that tariffs function as a consumption tax hitting exactly the wrong people — families spending 80% of their income on goods rather than services.

Consider apparel and footwear: $198 annually per household in additional costs. Electronics: $165. These aren't luxury purchases that wealthy families can easily substitute. They're necessities hitting paycheck-to-paycheck Americans hardest.

What most coverage misses is the productivity drain. New York Federal Reserve economists project that sustained 15-20% tariffs on Chinese goods could reduce U.S. productivity growth by 0.2 percentage points annually through disrupted innovation networks. That's the kind of compound economic damage that persists long after politicians move on to other talking points.

Expert Reality Check

"The empirical evidence is overwhelming that tariffs are paid by importing firms and passed through to consumers," explains Chad Bown of the Peterson Institute, whose database tracks 18,000 product codes. His research team found consistent pass-through patterns across every industry they examined.

"Tariffs are essentially a consumption tax that falls disproportionately on lower-income Americans who spend more of their income on goods rather than services." — Mary Lovely, Professor of Economics at Syracuse University

Federal Reserve economists Mary Amiti, Stephen Redding, and David Weinstein analyzed monthly price data for thousands of products and reached the same conclusion: "Tariff increases have been almost completely passed through into domestic prices." Foreign exporters absorbed virtually none of the cost through reduced profit margins.

Even industry groups get the math. The National Association of Manufacturers supports targeted anti-dumping duties but opposes broad-based tariffs on inputs. "Steel tariffs might help steelworkers, but they hurt the 6.5 million Americans who work in steel-consuming industries," notes NAM economist Chad Moutray.

The Political Economics Problem

Current trade policy discussions suggest continued tariff usage regardless of which party controls Washington. Both Democrats and Republicans have endorsed "strategic decoupling" from China in semiconductors, renewable energy, and critical technology sectors. Translation: sustained tariff pressure through 2028 and beyond.

Supply chain adaptation requires 18-24 months for major reconfigurations, meaning tariff effects persist long after policy reversals. As our analysis of tariff refund programs showed, administrative complexities create additional hidden costs for businesses navigating trade policy changes.

The question isn't whether tariffs work as intended — comprehensive economic data shows they don't. The question is whether American voters will keep paying $831 annually for policies that deliver the opposite of their promised benefits. Based on current political rhetoric from both parties, that answer appears to be yes.