The United States has been on a crusade to block Chinese tech firms out of the development of 5G networks. Its allies, big and small, are reluctant to fall in line as they weigh the potential political and military costs of bucking Washington’s demands against the dollars-and-cents cost of excluding tech companies like Huawei. Ultimately, few countries are likely to adopt a blanket ban on Chinese tech. But it may not matter if the U.S. proves willing and capable of crippling Chinese tech firms unilaterally.
For much of the past half a decade, the U.S. has warned that trouble awaits countries that build their fifth-generation, or 5G, mobile networks with Chinese technology. Fearing that the proliferation of Chinese telecommunications infrastructure would give Beijing unprecedented cyberespionage and network sabotage capabilities, the Trump administration has since tightened the noose, moving gradually to ban Chinese software and equipment – and even foreign tech made or designed in China – from U.S. networks. It wants friends and allies across the globe, on whose telecommunications networks the U.S. military relies, to follow suit. Using Chinese tech was always risky, but the U.S. has threatened to raise the stakes, saying countries that use it could face a future without U.S. military and intelligence cooperation.
This kind of absolutist approach by the U.S. speaks both to just how alarmed it is by China’s creeping telecommunications dominance and how little credence it gives to claims that such threats are manageable. Yet, widespread reluctance to comply with U.S. pressure has raised the question of whether the U.S. is really willing to walk away from the multilateral network of friends and allies it has been cultivating since World War II, with profound potential implications for the global system. But the U.S. won’t have to make this call any time soon. It’s not yet settled whether a blanket ban on Chinese 5G-related tech is really necessary. And U.S. moves to take matters into its own hands and stop Huawei’s rise may well put the whole issue to rest.
Why Other Countries Aren’t Falling in Line
Thus far, the U.S. campaign has found at best mixed success. Only Australia and, to a lesser extent, New Zealand, Japan, Taiwan and Vietnam have come anywhere close to a blanket ban on Chinese telecommunications tech. Elsewhere, responses have ranged generally from “We’re exploring other options, but don’t force us to take an overtly anti-China position” (see: Singapore, South Korea) to “Partial restrictions and careful vetting will be sufficient” (Europe) to “We’ll use as much Huawei tech as we darn well please, so stop nagging us about it” (Malaysian Prime Minister Mahathir Mohammad). Skeptics include the U.K. and Canada – fellow members of the crucial Five Eyes intelligence-sharing network (none of whom, inexplicably, are home to a major Huawei competitor); countries hosting or pursuing major U.S. military bases like Germany, South Korea and Poland; and nominal allies familiar with Chinese aggression like the Philippines. Even the African Union, whose Huawei-wired headquarters reportedly leaked a torrent of data to servers in China every night for five years, recently signed a new cooperation agreement with Huawei.
This reluctance is rooted, above all, in matters of dollars and cents. The physical requirements of 5G make rollouts breathtakingly expensive. It’s not just about upgrading existing cell towers. 5G will operate primarily on high frequency spectrum, which will unleash blistering data processing speeds with exponentially higher traffic capacity, but only at very short range. To ensure network stability and minimize latency, then, it will require a vast and dense network of base stations and antennas, plus millions of miles of new fiber-optic cable. Little of what 5G promises – driverless cars, automation, artificial intelligence, “smart cities,” “the internet of things” and so forth – can be realized without major capital expenditures.
Huawei and ZTE can make the leap to 5G less painful. Just three competitors – Finland’s Nokia, Sweden’s Ericsson and South Korea’s Samsung – are currently capable of delivering a similarly comprehensive suite of network equipment. (The United States’ Cisco and other smaller players will be competitive in narrow segments of 5G systems.) None have Huawei’s ability to achieve economies of scale and its levels of state backing, so it can often undercut its rivals by 20-30 percent. (It’s not a matter of sacrificing quality, either; some Huawei tech is considered the best in the business.) Moreover, the initial phases of 5G rollouts in all but a few countries will be built largely on existing 4G infrastructure – which, in many countries, is already built with Huawei tech. Ripping out all the existing Huawei equipment before upgrading would make the process even more expensive. Vodafone UK, for example, says it would need to replace some 6,000 base stations, costing hundreds of millions of pounds. It would also add costly delays, putting domestic industries behind the curve in developing profitable 5G applications. Germany’s Deutsche Telekom, the largest telecommunications operator in Europe, said a blanket ban on Huawei would set back its 5G roll outs by at least two years.
Poorer and less densely populated countries will benefit the most from Huawei’s cost advantages, of course, but even highly urbanized countries – those best-equipped to develop and reap the economic benefits of 5G applications, and with perhaps the most to lose from delays – aren’t immune. The race to roll out 5G networks is not a winner-take-all contest, despite how it is often portrayed. Still, there are certainly first-mover advantages in the development of new 5G applications, influence over international standards and securing new patents. Even outside the tech world, a firm in any sector – from heavy industry to manufacturing to transportation to healthcare – primed to harness 5G’s power could reap cost and quality advantages over foreign competitors effectively stuck in what might feel like the digital stone age. Add to this the costs associated with potential Chinese economic retaliation and other forms of coercion, and it’s easy to understand why countries insist on exploring protective measures before deciding whether to assume the costs of an all-out ban.
Is a Blanket Ban Really Necessary?
Skeptical governments have relied on four main arguments to explain their reluctance to fully ban Chinese telecommunications firms. Two are falling on deaf ears; two may ultimately gain traction.
The first is that the U.S. has not provided any evidence that Huawei has installed “back doors” into its existing overseas networks or knowingly facilitated state-sponsored cyberespionage. (The U.K.’s Huawei Cyber Security Centre Oversight Board did find defects in Huawei source code and concluded that the firm failed to address security issues in the past, but this doesn’t prove that the company has acted with malicious intent.) Absent evidence, they say, the U.S. is acting primarily on suspicion rooted in its own strategic and trade-related tensions with China – ones that other countries may not share. If the U.S. was really worried about cybersecurity, they say, it wouldn’t have abandoned an Obama-era push to include cybersecurity measures in international 5G technical standards. Nor would the Trump administration be so quick to ease pressure on Huawei and ZTE in the interest of reaching a trade deal with China.
The second argument is that, with proper vetting and oversight, security vulnerabilities in Chinese tech can be detected, obviating the need for a costly ban. To enhance this argument, Huawei has opened up its source code to inspection at security labs it’s established in Brussels, Bonn and the U.K.
To Washington, these two arguments miss the mark. This is, in part, because back doors are largely indistinguishable from common coding errors in network software or firmware, making it nearly impossible to obtain smoking gun evidence of malicious intent. The sheer scale of 5G architecture will also make vetting too slow and expensive, considering the frequency of software and firmware updates involved, to be done thoroughly and regularly. (Modern software testing processes aren’t particularly good at detecting carefully designed back doors, anyway.) Moreover, the full spectrum of potential vulnerabilities with 5G won’t become known for years to come, until its myriad potential applications are developed and until, as expected, tens of billions of “smart devices” are linked into the system. By then, countries may have effectively locked themselves into partnerships with the Chinese. The costs of reversing course would be prohibitive.
To the U.S., then, it’s perfectly rational to want to deprive an adversary of capabilities that might prove dangerous – and to kneecap a company that might act on that government’s behalf. Lack of trust and competing strategic interests have everything to do with it. After all, in the 2000s the U.S. compelled its own tech firms to facilitate government surveillance in the service of national security. It would be naive to expect China to behave any differently, even if you ignored Beijing’s history of coercive activities abroad, the abundance of China-linked cyberattacks, the autocratic nature of the Chinese regime and its national security law requiring firms like Huawei to cooperate.
The other two arguments hint at a possible way for the U.S. and its allies to meet in the middle. One is that, if Chinese tech is limited to the periphery of 5G networks, any damage Beijing could do could be tightly contained. 5G networks consist of a tightly protected “core,” where servers and software execute the most sensitive and crucial functions, and the radio access network equipment (towers, masts, small cells inside buildings and along streets, and so forth) on the “edge” that connect wireless devices to the core.
If China could slip back doors into the core, where encryption keys are stored and authentication functions take place, it could gain unprecedented snooping power and even the ability to shut down key parts of a network altogether. As the dependency of critical infrastructure (including power grids and hospitals) on 5G networks increases, so too would Beijing’s capabilities to conduct crippling sabotage attacks. By comparison, if a Chinese firm slips a backdoor into edge components like, say, the base station or antennas outside your house or the operating system of the phone in your pocket, it could potentially monitor unencrypted data and encrypted metadata or infect user devices with malware, posing a small-scale espionage problem (especially if you happen to be a high-level intelligence target). But it’s doubtful that edge equipment could be used to conduct mass espionage or to bring down large parts of the network.
At this point, governments in France, Germany and the U.K. all plan to ban cheaper Chinese tech from the core but not the edge. Since the edge is where the bulk of new capital investment will be required – and where most Huawei equipment is located in Western 4G networks – this ostensibly makes it possible to harness Chinese cost advantages without incurring Chinese risks. But others, including the U.S. and Australia, say the decentralized nature of 5G networks will erode the distinctions between the core and the edge over time, with edge devices taking on more and more “smart” computing power and sensitive functions. To them, the only sure solution is a blanket ban. Skeptics of this argument say components in an even more decentralized core will still be distinct and protected from edge components.
The final argument is basically that supplier-inserted back doors are just one of a dizzying array of cyber threats facing 5G, and fixating on who makes the equipment addresses the problem too narrowly to justify the cost of a blanket ban. Indeed, this approach could make some cybersecurity challenges harder to address. Any network equipment, whether manufactured in Shenzhen, Silicon Valley or Sarawak, will inevitably be laced with exploitable security flaws, and the biggest threats will still be familiar ones like spearfishing and malware-infected software inadvertently downloaded by users. It’s certainly easier for a malicious actor to hack a system if it built in a backdoor itself. But ultimately, the best way to prevent espionage is widespread adoption of sound end-to-end encryption practices and use of other tools like virtual private networks. And the best protection against network sabotage is system redundancy. This means additional spending on backup network infrastructure from multiple suppliers. Cutting out one of the few major telecommunications suppliers available (and the cheapest one, to boot) would make redundancy harder.
The debate is clearly far from settled. But if the U.S. can be persuaded that the distinction between the edge and core will hold, and if protective measures like end-to-end encryption can be adopted widely enough (no small feat, considering that billions of connected devices will need to be configured to operate on secure channels), the U.S. may be willing to compromise and adopt a more tailored approach to Chinese tech.
Is the U.S. Bluffing?
There’s another, largely unspoken reason countries are resisting U.S. pressure on the issue: They think the U.S. might be bluffing on its threats to sharply curb military and intelligence cooperation. Consider the potential costs. The U.S. currently has troops in dozens of countries. Its warships stop in dozens more. Its critical logistics networks crisscross the globe. Its intelligence-sharing agreements allow it to act nimbly and entrench its partnerships. It would be one thing if there were enough strategically located countries shunning Huawei that the U.S. could keep its global operations humming. But there are not. So, to make good on its threats, the U.S. would have to dramatically scale back its global military footprint and deprive itself of access to vital intelligence flows, potentially putting the global balance of power in flux. It defies imagination to see how the risks of 5G outweigh these costs.
To be sure, the proliferation of Chinese 5G tech could indeed pose extraordinary new challenges to U.S. intelligence and military operations abroad, especially if its arguments about the network security risks prove valid – and particularly when operating in or with countries that allow Chinese access to the network core or fail to adopt prudent network security practices. The battlefield implications could likewise be dramatic; China could realistically shut down military communications, disrupt critical supply lines, collect and exploit signals intelligence, and so forth. The U.S. will need to become ever more judicious about how and where it sets up logistics networks, with whom it shares sensitive information, and how much it can afford to rely on next-generation weapons systems that depend on unhindered connectivity. It will probably need to develop more sophisticated and secure communications systems and consider helping partner governments bear the expense of ensuring network diversity and redundancy.
To an extent, the nature of these challenges isn’t new. The U.S. has long been a global superpower well practiced in handling adversaries keen to steal U.S. secrets, frustrate its best-laid plans and exploit asymmetric capabilities to blunt inherent U.S. advantages. And it’ll certainly be able to exploit these same capabilities itself. (A leaked National Security Agency document from 2014 claimed the agency had penetrated Huawei networks so thoroughly that it didn’t know what to do with all the data it collected.)
The scale and complexity of the new risks are too much for the U.S. to ignore. Yet if it can’t pressure the world to shun Chinese tech or make peace with available security measures, it won’t blow up its alliance network. Washington will instead try to make the whole debate moot by taking matters into its own hands. The U.S. already started this process in earnest in May, when it announced a ban on exports of U.S.-made component parts and software to Huawei, ZTE and other Chinese firms. Last week, to restart negotiations with Beijing on a trade deal, Trump signaled a willingness to relax the ban, though exactly how much remains unclear. But it remains an enormously powerful measure that the U.S. will likely return to eventually. U.S. firms no longer dominate as many sectors of the telecommunications industry as they once did, but they do dominate some of the fundamental building blocks such as semiconductors and mobile chips. This means any foreign firm in Huawei’s supply chain whose products contain these components also has to comply with the ban, lest it be sanctioned by the U.S. (Huawei’s own research and development into semiconductors and microchips is widely believed to be inadequate for its needs.) Whether a ban would kill the company, or just weaken the quality of some of its tech and force it to scale back its product offerings, is impossible to say. U.S. pressure has already damaged Huawei’s reputation and revenue streams. At minimum, even the continued threat of a ban will make some countries think twice about partnering with a company that may not be able to continue to innovate.
There’s also a risk that the move would backfire by eventually accelerating China’s pursuit to develop indigenous components and ushering in an era of Chinese tech parity. Yet the U.S. has enormous incentives to follow through. A ban wouldn’t just hit the Chinese telecommunications sector; it would also hamper China’s broader drive to dominate high-tech industries, its breakneck military modernization, and its sprawling diplomatic ambitions. The U.S.-China strategic rivalry isn’t going away anytime soon, and the U.S. has an opportunity to cement its alliance structure and strike at multiple dimensions of Chinese power with a single blow. In other words, this is one of the few cases in which the U.S. may be better off acting alone.