In the past two years, President Trump has enacted tariffs on $16 billion worth of Chinese and Canadian goods. Predictably, the Canadians countered with $12.6 billion in retaliatory duties against US exports. The Chinese followed. In addition, in his most sweeping action to date, the president slapped a 25 percent tariff on all foreign steel and a 10 percent tariff on aluminum imports. To this final duty, there are no exceptions. If General Motors wants to import foreign steel to help make cars cheaper for consumers, they must pay a 25 percent tax. If Budweiser wants to drive down production costs with aluminum from abroad, a 15 percent tax. No exceptions. Period.

In sum, we are firing the first shots of what may well become a global trade war; and, the president’s assertions to the contrary, like all conflicts, a trade war won’t be over by Christmas. If the conflict is allowed to escalate, it could be disastrous for the U.S. economy. Tariffs do not cause protracted crashes, but they can be their catalyst. For evidence of this, look no further than the effect of the Smoot-Hawley tariffs in 1930.

After the market crashed in 1929, President Hoover and Congress looked for a quick fix that would help the economy rebound before the 1932 election. Hoover’s 1928 campaign promise to help farmers by raising agricultural tariffs provided additional political cover for what was enacted: a massive tariff bill that raised the average duty rate by 50 percent. The result was a catastrophe: US imports and exports both fell by over 60 percent and GDP collapsed by nearly half in four years. The future President Franklin Roosevelt would run a successful campaign in 1932 in large part by decrying Hoover’s spendthrift policies and promising to repeal Smoot-Hawley and rid the market of its economic nationalism.

This is not to say that President Trump’s tariff policies will be as catastrophic as President Hoover’s. For one thing, the U.S. is a much larger economic actor on the world stage than it was in 1930. The dollar is the reserve currency of the world, and even with only 5 percent of the world’s population, the United States accounts for nearly 22 percent of world trade. In other words, the elasticity of demand for US goods is much steeper than it was in 1930.

Still, the president’s tariff policies could cause significant damage to the U.S. economy. Over 25 percent of US GDP is directly tied to international trade, and the remaining 75 percent goes abroad to purchase inputs for its production processes. This remaining 75 percent is less visible, but it’s where most of the damage wrought by higher tariffs will be concentrated. Higher steel tariffs might make domestic steel producers happy, but they would raise the price of manufacturing inputs for car, appliance, aircraft, railroad, and ship manufacturers, to name only a few—the economy is littered with additional examples. While aluminum tariffs might placate select domestic producers, no one who regularly buys canned soup, beer, cell phones, laptops, or children’s toys is going to appreciate those goods becoming more expensive.

Furthermore, if President Trump seeks to reduce the trade deficit, his actions will likely have the opposite effect. High tariffs actually increase the trade deficit in two major ways (among many others that are beyond the scope of this article). First, US exports will shrink as a result of other countries retaliating with tariffs of their own. Second, the now higher cost of foreign goods will force many consumers to buy from domestic firms, which in turn sends domestic firms looking for additional capital to expand production and meet the new demand. If the domestic savings rate has not changed, they must attract capital from abroad, further increasing the trade deficit.

These effects are not just theoretical. The U.S. trade deficit has actually expanded by 7 percent since 2017, despite the president’s tariffs (or, perhaps, because of them).

What is more, from a strictly economic perspective the trade deficit does not matter a whole lot, and bilateral trade deficits do not matter at all. That is to say, looking at either bilateral trade deficits (say, with China or Canada), or trade deficits generally, misses the real story conveyed when accounting for trade flows.