As many economists have noted, tariffs are not a sneaky way to get free money from foreigners.  They are a tax paid by all of us in higher prices, and that’s exactly what has happened with the Trump tariffs to date.

Since a tariff is a tax, it’s worth thinking about what kind of a tax it is.  We’ll be specific.

For this note we’ll restrict ourselves to the two biggest ones:  the new Chinese tariff of 25% on $250 B of annual inputs and the new Mexican tariffs, scheduled (as Trump says) to reach 25% on all Mexican imports ($346 B annually) by October.  This costs a total of .25 x ($250 B + $346 B) = $149 B annually.

At first glance this doesn’t seem too bad, since aggregate annual income in the US is $13T, and $149 B /$13 T is about 1%.  But that’s the wrong comparison.  The primary impact here is on prices, which means we’re talking about consumption—not income.  This is like a sales tax.   Consumption is much more evenly distributed than income, so lower incomes pay more.

It is a truism that the rich spend less of their income, because they don’t have to.  But in fact we can be specific.  Even if we throw out the highest 5% of incomes, the income distribution in this country is more than 3 times more inequitable than the consumption distribution.  What’s more, the US Bureau of Labor Statistics publishes a breakdown of consumption spending by decile segments of US income.  That is, for each 10% segment of the population—ordered by income—they give a percentage of total consumption spending.  These ten slowly-increasing percent values give a quantitative picture of what it means to say the rich spend less of their income:

4.1, 4.5, 6.1, 7.0, 7.8, 9.0, 10.5, 12.1, 15.1, 23.7

Going back to the $149 B of tariffs, this tells us how those tariff costs are allocated to the different segments of the population:  4.1 % to the lowest tenth of income, 4.5 % to the next tenth, and so forth.  Further, since the total number of households in the country is 141 million, we can even say something about how much households in each segment will actually have to pay.  For the lowest segment this is

$149 B x (0.04) / 14.1 M = $433

The calculation is the same for each segment, using the percent values just listed.  In addition the Labor Statistics figures also include the average income level associated with each population segment.  That lets us directly compare Trump’s much-touted tax cuts with tariff costs.  Putting it all together we get the following chart

What we see is that the Chinese and Mexican tariffs alone overwhelm the Trump tax cuts for 60% of the US population (and come close for another 20%).

Conclusion:  Tariffs are a serious and regressive tax.  Since they are also an expensive and uncertain way to create jobs (at $800-900K each), this is no silver bullet.  And that’s before we even think about retaliation.