What is the end-game of America’s confrontation with China? Many observers — as well as Chinese President Xi Jinping — probably hope the answer is the 2020 election. But they are likely to be disappointed. Even if President Trump’s tenure ends with the next election cycle, the roots of the conflict are deep, and the actions of the Trump administration will likely have implications that reach far beyond his tenure.

In his presidential campaign, Trump attacked China over and over again as a rival that was “killing us” on trade. Through currency manipulation and other means, the Chinese government arguably created an artificial trade imbalance, in which America persistently bought more Chinese goods and services while the Chinese bought American debt.

As president, Trump has paid some attention to these matters, but less than his campaign would have suggested. He’s been more attentive to the ways, both licit and illicit, in which China has bolstered its domestic industries, and striven to become a direct competitor and even to surpass the United States. These include direct state subsidies for Chinese companies, limits on foreign control of Chinese companies, requirements to transfer technology to Chinese partners as a condition of investment, and rampant disregard for the intellectual property norms that obtain in the U.S.

If that were the extent of the conflict, the Trump administration’s strategic objectives with regard to China might resemble those of his predecessor, Barack Obama. The Obama administration’s approach to China reflected concerns about the direction of Chinese economic and foreign policy, as well as domestic concerns about the state of American manufacturing. The military and diplomatic “pivot to Asia,” which bolstered America’s cooperation with a host of nations in China’s orbit, and the Trans-Pacific Partnership, which aimed to harmonize rules for trade, investment, and intellectual property, were intended to box China into an American-led economic and security order in the Pacific. China, the theory went, would eventually realize that it would be more profitable to join that order than to challenge it, and become a productive partner as well as a powerful competitor.

Obama’s strategy bore, at best, equivocal fruit before the end of his term. While Trump reversed course on the TPP, his escalating tariff war could simply be a more aggressive set of pressure tactics aimed at achieving the same result. Similarly, his so-far-fruitless overtures to Beijing’s North Korean ally could be compared to Obama’s overtures to Myanmar, another Chinese client. But they could be part of a more dramatic shift, not to change the terms of economic integration between the U.S. and China, but to unravel that integration outright.

Consider the administration’s blacklisting of Huawei, China’s leading global high-tech company. America has legitimate security concerns about Huawei’s technology, as well as larger competitive issues with government support of the Chinese champion. But Huawei is central to China’s aims to become a true peer competitor at the top of the value chain, and Beijing reasonably understands the administration’s move not merely as a tactic to change Chinese policy, but as a direct attempt to restrain its rise. Much of the tech world views this as the opening salvo of a technological cold war.

Even if Trump loses his bid for re-election, a shift of this magnitude is unlikely to be reversed quickly, if at all. It’s not just that the winds of economic nationalism are blowing through both parties (Sen. Elizabeth Warren recently made a strong bid to carry that torch in the Democratic camp); even if an advocate of a relatively liberal trade regime, like Joe Biden, wins in 2020, neither the Chinese government nor American multinationals will be able to discount political risk to the U.S.-China relationship as they approach questions of where to invest or how to build their supply chains. And the Chinese will be unlikely to view American foreign and economic policy going forward as anything other than zero-sum, where America views China’s rise as our own defeat. For that reason, a post-Trump attempt to “reset” relations with China could well prove as unsuccessful as the Obama administration’s attempt to reset relations with Russia.

Does that mean conflict between China and America is inevitable? Not necessarily. Even if America adopted a more nationalistic approach to economic policy, and strove to reduce our economic integration with China, we could still engage in trade that did not raise security concerns, and compete peacefully with China for customers — at least in theory.

But in practice, things are unlikely to be so simple or so smooth. China is unlikely to accept limits on its economic rise, and we should expect them to use every form of leverage they have not only to advance their own aims but to disrupt America’s own economic and security relationships. And America is unlikely to accept a Chinese sphere of economic influence in which America’s technology giants are locked out and American investment is unwelcome. Some cautious observers are already anticipating the possibility of direct political conflict, while others are describing our relations as akin to war.

If we want to avoid what is still a metaphor from becoming a reality, we should try to imagine what a relatively stable modus vivendi might look like in a world where economic nationalism continues to be a force. Cold wars don’t have to turn hot, but management of hostility requires at least as robust communication as peaceful alliances do.