The Founders knew that Americans, for better or worse, had an insatiable desire for overseas trade. “They are as aquatic as the tortoises and sea-fowl,” observed John Adams, “and the love of commerce, with its conveniences and pleasures, is a habit in them as unalterable as their natures.” As early as 1785 he foresaw that Americans would be compelled to form “connections with Europe, Asia, and Africa,” and he advised that “the sooner we form those connections into a judicious system, the better it will be for us and our children.” Thomas Jefferson would have preferred to cease all commerce with the rest of the world and rely on the simple virtues of the “yeoman farmer,” but he knew this was impossible. “Our people have a decided taste for navigation and commerce … and their servants are in duty bound to calculate all their measures on this datum.” Even that much-caricatured “Jacksonian,” Andrew Jackson himself, as president never fired a shot in anger but negotiated more trade agreements with foreign powers than any of his predecessors.

The American love of trade made using the practice as a weapon against other nations difficult. When Jefferson forgot his own lesson and tried to embargo trade with Great Britain in 1807 in response to the British navy’s abuse of American merchants on the high seas, his efforts backfired, stirring talk of secession in the New England states that conducted most of that trade. It turned out to be easier to get Americans to support a shooting war with Great Britain than a trade war.

Donald Trump is now learning the hard way how vulnerable America is when it comes to trade wars. This is not because the United States doesn’t in theory hold the strongest hand. The American market is the most desired in the world, and any restriction on access to that market should hurt other countries more than it hurts the United States. The ratio of international trade to GDP for the U.S. is roughly 25 percent, compared with more than 60 percent on average for all other nations. In Germany, foreign trade tallies up to 90 percent of GDP. That ought to make the country vulnerable and give the United States leverage. In practice, however, Americans have proved time and again that they have a very low threshold of pain when it comes to trade wars. Jefferson was not wrong to believe that Britain depended heavily on American trade when he launched his embargo in 1807; what he did not anticipate was that his own citizens would cave before Britain did.

[Read: How Republicans learned to love high prices]

The problem is, or at least has been up until now, democracy, and, more specifically, electoral politics in a federal system where narrow, local interests can have broad national political impact. A trade dispute might harm only one sector of the economy, but if that sector happens to coincide with a crucial voting bloc, it can put the United States at a disadvantage in a contest with a nominally weaker power.

A good example of this came during World War I, before the United States had entered the war and Woodrow Wilson was trying to navigate his way through British blockades and German submarine attacks on transatlantic shipping while desperately trying to preserve American neutrality. The United States was far less reliant on international trade then; it was only 11 percent of GDP. But as Wilson learned, even damage to particular sectors of the economy could threaten political upheaval. Although his personal inclinations were pro-British, for instance, London’s threats to blockade cotton as contraband of war infuriated the Democrats’ key southern constituency. Wilson’s secretary of the Treasury, William Gibbs McAdoo, recalled spending “more sleepless nights thinking about cotton” than about anything else during his time in office. The rest of his sleepless nights were spent worrying about finding markets for midwestern grain, much of which had been purchased by Germany and other European nations prior to the war. These specific sectors, because they involved states and regions essential to national political coalitions, had influence on American decision making that exceeded their overall importance to the American economy.  

[Read: Trump’s most inexplicable decision yet]

Trump must believe, as Jefferson did, that the world needs America more than America needs the world, and he may be right—in theory. The problem is that individual voting blocs mean more to him than carrying out a consistent trade war, as he has repeatedly demonstrated during both terms in office. In his first term, the damage done to farmers by his tariffs on imports was sufficiently threatening politically that he had to spend much of the money gained by the tariffs to compensate the farmers for their losses. His vacillations and emendations in his latest rounds of tariffs this year have been similarly motivated by his desire not to alienate Republican voters in particular states—northern-tier states that rely heavily on trade with Canada and automaking states that stand to lose badly from tariffs on auto parts, steel, and aluminum crossing the Mexican and Canadian borders. It is no accident that among the Europeans’ first retaliatory tariffs have been those against Harley-Davidson and American whiskey. Other nations may know their history better than Trump does and have figured out that tariffing sectors of the economy that hit Trump voters can have an impact beyond their dollar value. The United States is a nation split down the middle politically, so marginal voting groups can have a huge effect. This significantly vitiates the American advantage.

It would be one thing if Trump’s supporters were willing to suffer economic hardship in order to show their support for the MAGA way. As Senator Tommy Tuberville of Alabama put it, “There’s going to be some pain with tariffs,” but “no pain, no gain.” The problem for Trump is that, so far, as in the past, even his own voters don’t have much tolerance for pain.