Huawei has been in the U.S. government’s crosshairs for over a decade. In 2008, U.S. policymakers convinced the Committee on Foreign Investment in the United States to block the Chinese technology firm’s acquisition of U.S. software company 3‑Com on the grounds that the deal would threaten national security. For many years, I have suspected that the U.S. campaign against Huawei was motivated less by concern over specific security threats than by the desire to respond to China’s aggressive, discriminatory industrial policies in the technology space. If Beijing was going to subsidize indigenous innovation, favor companies that registered intellectual property in China, and encourage Chinese companies to “borrow” U.S. technology in a push to challenge American firms at the technological fore, then the U.S. response would be to inhibit the commercial success of the beneficiaries of those industrial policies. Huawei‘s emergence as a global competitor made it an obvious target.

Although it is certainly plausible that Huawei presents a security threat to the United States, that conclusion has never been demonstrated convincingly in any public forum by anyone with access to the information upon which such a conclusion should be based. There have been closed door hearings in which classified information was discussed and generated, which—if declassified and shared with the public—might convincingly corroborate these threat claims and maybe even justify the administration’s decision to put Huawei on the U.S. Commerce Department, Bureau of Industry and Security’s “Entity List,” a move that could starve Huawei of needed inputs from U.S. companies. But it shouldn’t come as a surprise that policymakers who sit on intelligence committees or who serve in security-oriented federal agencies are probably predisposed to see security threats where others don’t or to discern nefarious intentions where the evidence is benign or even to interpret the absence of evidence as proof of the perpetrators’ craftiness.

Then again, when the standard of proof is the precautionary principle, the evidentiary thresholds aren’t especially rigorous. A threat possibility, however remote, tends to suffice.

Protecting national security is a legitimate function of government. Fulfilling that responsibility sometimes requires that international trade and investment be restricted. Since determinations of threats to national security often are based on classified information, the public has to trust that policymakers have reached the right conclusions and that the prescribed remedies are necessary and appropriate.

It is difficult to trust the Trump administration in this regard, as it has already demonstrated itself an unreliable arbiter of national security threats. President Trump has made a frivolity of the national security rationale for restricting trade. Last year, Trump invoked threats to national security to justify his tariffs on steel and aluminum imports. This year he concluded that U.S. security is threatened by imports of automobiles and auto parts. In those cases, the data and analyses “supporting” the national security threat conclusion were not classified, but publicly available. And you can count on your fingers and toes the number of people convinced that steel, aluminum, and auto imports present such threats.

Based on information that the U.S. public hasn’t seen, the Trump administration has deemed Huawei a national security threat. That may well be the right conclusion, but the U.K., German, and other governments that the administration has been pressuring to purge their networks of Huawei gear, seem unconvinced, and have resisted.

The Trump administration’s latest move to blacklist Huawei escalates already rapidly escalating tensions in the U.S.-China relationship. Putting the company (and 68 affiliates) on the Entity List means that U.S. firms can no longer do business with Huawei without first obtaining a special license, which can only be done after overcoming “a presumption of denial.” Earlier today, Google, Intel, Qualcomm, and other prominent suppliers announced plans to discontinue their current commercial relationships with Huawei. It doesn’t take a creative imagination to foresee worsening troubles ahead for U.S. businesses operating in China and, well, a deepening process of economic disengagement.

The bottom line is that when U.S. economic policy toward China could be successfully sequestered from the geopolitics, the relationship could be managed. Now our economic problems are viewed and magnified through a geopolitical prism and, for many, the calculations suggest that disengagement and decoupling is the optimum. But that, too, will be enormously costly.

To reiterate a conclusion from a recent op-ed:

By banning Huawei gear and putting pressure on third countries to do the same, the United States is effectively saying that a huge swath of 21st century trade—an estimated $12.3 trillion in sales activity across multiple industries involved in developing 5G infrastructure and producing 5G enabled products by 2035, according to the Congressional Research Service—will not be subject to the disciplines of the global trading system. If that doesn’t consign the WTO to insignificance, the ensuing race to carve up the world into spheres of influence based on competing 5G standards will.

In what will look like a replay of the Cold War, Beijing and Washington will compete for the loyalties of the rest of the world by offering carrots and threatening sticks to countries to adopt their respective 5G standards. Dividing the world into these technology blocs will deprive the technology ecosystem of global economies of scale and open the door to bloc-based tariffs and other forms of protectionism, making the world a poorer place. Creation of the open global trading system induced a steady climb in global exports from 4% of GDP in 1947 to 26% of GDP in 2015. Erecting tariffs and non-tariff barriers through that system would undoubtedly cause a decline in global trade and output.