The media’s obsession with the ongoing saga of the U.S.-China trade war has given people skewed impressions of how the two economies interact. Most journalists have focused on the minutiae of tariff levels and who is actually paying for them. But this obsession is based on an outdated understanding of global trade.

The real problem is that the United States’ open capital markets keep the U.S. trade balance from adjusting regardless of what level tariffs are set at and who pays for them. The recent phase-one trade deal won’t change that fact, nor will the tariffs that the deal has left in place. If Washington actually wants to move the needle on the trade front, it should pay more attention to the U.S. capital account.

Why Tariff Costs Are a Red Herring

In late November 2019, Liberty Street Economics, a publication sponsored by the New York Federal Reserve, released a paper that has added fuel to one of the most contentious trade-related debates in Washington: who is paying for President Donald Trump’s tariffs on Chinese goods? The paper argues that because prices on imported goods from China have not fallen, U.S. consumers and businesses clearly must be footing the bill:

U.S. businesses and consumers are shielded from the higher tariffs to the extent that Chinese firms lower the dollar prices they charge. U.S. import price data, however, indicate that prices on goods from China have so far not fallen. As a result, U.S. wholesalers, retailers, manufacturers, and consumers are left paying the tax.

The fact that this study is being cited as evidence that U.S. tariffs on Chinese goods harm the U.S. economy only shows how muddled the trade debate has gotten. Tariffs on foreign goods aren’t simply failures if they raise import prices for U.S. consumers. On the contrary, that’s precisely how tariffs are supposed to work. By raising the cost of foreign imports, they effectively transfer income from consumers to businesses, automatically reducing the consumption share of GDP or, to say the same thing another way, raising the aggregate savings share.

One traditional argument in favor of tariffs is that they allow infant industries eventually to move up the productivity scale enough to compete with foreign producers. Ignoring that infant industry argument for tariffs for a moment, the pointed question of whether tariffs benefit or harm the economy that imposes them depends mainly on how that economy fits into the global structure of trade and how that economy adjusts to these higher savings:

  • In surplus-running countries, in which domestic savings exceeds domestic investment, tariffs reduce domestic demand and force the economy to become even more reliant on foreign demand to resolve domestic production.
  • In deficit countries, in which investment is constrained by limited domestic savings (this mainly applies to developing economies), the economy typically adjusts to higher domestic savings by increasing domestic investment.
  • In countries in which investment isn’t constrained by domestic savings (that is, the United States and most other advanced economies), the economy typically adjusts by lowering unemployment, forcing up wages, or reducing household debt. That is why when tariffs supposedly work, as they did during most of the nineteenth century, the higher prices U.S. consumers pay are more than matched by the higher American income they create.

Tariffs may in some cases be harmful to the economy and in other cases they may benefit the economy, but the point is the only way tariffs can boost domestic growth is if they cause the prices of imported goods to rise. If U.S. tariffs on Chinese goods did not cause import prices to rise, either because Chinese businesses lowered their prices or because Beijing devalued its currency, there would be no effect on the U.S. economy, except a small increase in government revenues representing effectively an economically unimportant transfer from Chinese businesses to the U.S. government.

So does the Liberty Street Economics paper that argues that American consumers, not Chinese producers, are paying for the tariffs prove that U.S. tariffs on Chinese goods are good for the U.S. economy? Unfortunately, this isn’t the end of the story. Even if American consumers do pay higher prices for imported goods, U.S. tariffs on Chinese goods will fail to boost growth in the U.S. economy, but not for the reasons most economists think. In reality, the tariffs will fail because the role the U.S. economy plays in stabilizing imbalances in international capital flows in effect prevents tariffs from raising the American savings rate. That’s important because it is the impact of tariffs on the savings rate that determines how they will affect the economy. The special role the United States plays in the global economy means that tariffs do no good.

Why Capital Flows Are the True Culprit

This is because the strategy behind tariffs is based on an obsolete understanding of trade. For much of history, when most countries suffered from scarce savings and a significant amount of slack (that is, underutilized resources and high levels of unemployment and underemployment), trade imbalances were driven mainly by differences in production costs. At that time, more productive producers outsold their more expensive foreign rivals and ran trade surpluses against their deficits. During that period, with most international capital consisting of trade financing, capital inflows adjusted to the trade imbalance, and as tariffs raised the savings rate, it reduced the trade deficit and required less foreign financing.

Today, the causes of a trade surplus and the relationship between the trade and capital accounts have completely changed. Surplus countries do not run surpluses because of production cost advantages. They run surpluses because domestic households are paid so low a share of total income that consumption is too low and savings too high to absorb all that these countries produce. Put differently, the income that surplus countries generate from exports must be recycled not by importing foreign goods but by exporting domestic savings.

That is what has changed the way tariffs work. Because of its deep, flexible, and well-governed capital markets, the U.S. economy is the top destination to which much of the excess savings of the rest of the world flows automatically. The U.S. trade deficit, consequently, cannot adjust as long as the United States has open capital markets and is forced by weak foreign demand and excess foreign savings to run a capital account surplus.

U.S. tariffs on Chinese goods are an overall failure, in other words, but not because the costs are mainly borne by American consumers. Rather, they are a failure because the mechanism that converts high import costs into higher American savings cannot function as long as the United States is forced to absorb the excess savings of the rest of the world. The problem for the U.S. economy is not that China and other surplus countries are more efficient, or that they subsidize domestic manufacturing. It is that their weak domestic demand and excess savings are transmitted into the United States through its open capital account.

That is why U.S. tariffs on Chinese goods will fail: they work mainly by forcing up domestic savings, and it is the way in which the economy adjusts in order to raise savings that ultimately determines whether tariffs will benefit or harm the economy. But flexible financial markets and an open capital account prevent U.S. savings rates from adjusting. That is where the problem truly lies. Whether one thinks it is American consumers who pay for Trump’s tariffs or Chinese businesses is completely irrelevant.

Aside from this blog, I write a monthly newsletter that covers some of the same topics. Those who want to receive complimentary subscriptions to the newsletter should write to me at, stating affiliation. Twitter: @michaelxpettis