Investors who see President Trump’s new trade war as more bark than bite have so far been proven right. This could be a problem.

Midmorning on Monday, Mexican President Claudia Sheinbaum said that the U.S. had agreed to a one-month delay before slapping a 25% tariff on its goods. Then, after the stock market had closed, Canada’s Prime Minister Justin Trudeau announced a similar reprieve. Both countries have agreed to reinforce their borders and crack down on fentanyl smuggling.

The S&P 500 ended the day down 0.8%. The Stoxx Europe 600 lost only 0.9%, though it inched lower again early on Tuesday, probably spooked by China’s unveiling of its own retaliatory tariffs.

Even before the detente was announced, however, Monday’s selling hadn’t gotten too dramatic. With the truce in place, shares in many of the companies directly affected by the tariffs ended up only slightly lower than where they had started. These included Ford and General Motors, which make cars in Mexico that they then sell to American consumers. The Canadian dollar and the Mexican peso rebounded and ended up rising against the U.S. dollar.

This served as validation for the optimists on Wall Street, who have insisted on seeing Trump’s tariff threats to nations other than China as mere negotiating tactics to extract concessions on issues such as illegal migration and drug trafficking. 

Such a view is mainstream in financial markets, as suggested by the peso not underperforming peer currencies since the Nov. 5 election, or the Stoxx Europe 600 rising about 5% since the start of the year, despite the threat of duties being levied on goods coming from the European Union. 

Trump’s negotiating style is indeed predicated on an initial display of strength. Last month, Colombia agreed to repatriate migrants in exchange for escaping tariffs and economic sanctions, and something similar may play out this time.

But it could be dangerous to blindly trust that it always will. 

It has now become obvious that Trump’s current protectionist policies are significantly more extreme than the ones he introduced in his first term. Back then, he focused on unfair trade practices and national security, with tariffs primarily levied on China and beyond that largely limited to specific goods and industries such as steel, aluminum and household appliances. They were also gradually introduced, giving businesses time to adapt. Yes, global manufacturing output came under pressure in 2019, but it had gotten a small boost in 2018 as firms stockpiled inventories.

This time, if targeted countries refuse to yield, U.S. effective tariff rates could suddenly become the highest since the late 1930s. On top of hitting domestic growth and spurring inflation, this could derail a network of complex supply chains which, as the pandemic showed, can have unpredictable effects.

Worse, the chances of this happening seem higher when trying to strong-arm long-term allies with robust democratic systems than when dealing with the likes of China, since it is unclear how much public-facing leaders will be allowed to back down.

Take Canada, which despite being knee-deep in a political crisis has responded to the threat of tariffs with newfound unity across the political spectrum, from Trudeau’s center-left party to those on the right formerly seen as close to Trump’s positions, such as Ontario Premier Doug Ford. He has pledged to pull U.S.-made alcohol from shelves in his province, and scrapped a high-speed internet contract he had signed with Elon Musk’s Starlink. 

Regardless of the final policy outcomes, U.S. producers could face a long-lasting backlash from consumers in impacted countries. Furthermore, the experience of the U.K. post-Brexit also shows that the mere uncertainty of on-again-off-again disruptions, and extended negotiations, forces firms into a defensive mode and weighs on business investment. 

This would be done for no long-term economic benefit. Contrary to usual textbook claims, shifting the site of production of complex tradable goods through tariffs can have positive effects, if done selectively while supporting investment by export-oriented firms—as opposed to the pure “import substitution” pursued by Latin America in the 1950s and 1960s. Promoting U.S.-made electric vehicles relative to cheaper Chinese competitors, for example, is one of their present reasonable uses. 

But the new slate of proposed tariffs has no such concrete economic goals in mind: It is just designed to shock. Canada, for one, buys more goods from the U.S. than it sells—including cars—not accounting for energy. Other tariffs proposed by Trump, such as on intermediate goods like semiconductors, could be even more damaging to U.S. industry. 

Finally, consider this: By being too blasé about tariffs, investors have already raised the chances that the U.S. administration will step closer to the edge, confident that markets have signaled that the outcome won’t be too punishing. With the S&P 500 trading at a steep 22 times earnings, the shift from “America first” to “tariffs first” is something to be very weary about.