Thomas Donahue is a Cipher Brief expert and former Senior Director for Cyber Operations on the U.S. National Security Council Staff. His article was originally published by National Defense University Press PRISM.
As a global power, the United States must be able to sustain military forces and project power anywhere in the world, even in the face of resistance from a sophisticated adversary with the ability to infiltrate or disrupt telecommunications and other critical infrastructure within the United States, in space, under the ocean, and in other regions of the world. Policy must consider the worst possible day, not the routine day.
The emergence of the Chinese company Huawei as a leading provider of integrated telecommunications systems is seen as such a security threat that the U.S. Government (USG) has sought to raise barriers to the use of the company’s technology in U.S. infrastructure, and even threatened long-standing intelligence sharing arrangements with the nation’s closest allies who choose to use less expensive Chinese technology.
While arguably helpful in the short term, the USG must confront the basic problem of whether an acceptable alternative must be based in the United States, and whether the U.S. Government should support even a foreign champion with foreign, trade, and industrial policies to promote a viable global competitor to Huawei. As was done in the wake of initial successes by the Soviet space program, the United States should consider whether its strategic role as a global power requires increased government investment in research, industrial capacity, and programs with specific objectives comparable in scale to the U.S. space program of the 1960s.
This article considers how the United States fell into this dilemma, how the government has assisted infrastructure development in the past, and options for how the USG—in partnership with allies and the private sector—might be able to promote competition in the global market place with a view toward enhancing the resilience of national security communications and critical infrastructure.
Opening Gambit
Telecommunications is both a homeland security and a national security issue because the entire economy depends on a shared, interoperable infrastructure, and because many national security communications ride on top of commercial infrastructure both in the United States and abroad. Nonetheless, the United States generally considers free market solutions as more effective and agile, particularly when it comes to consumer products, even when that means foreign companies dominate a technology market. Even when the government is involved in the creation of a technology, privatization is the preferred outcome, as occurred with the internet. The business of integrated telecommunications equipment, however, is not a “consumer” issue; it is about systems used by critical infrastructure. In addition, the business of integrated telecommunications equipment is not a “free marketplace” in the usual consumer sense because the multi-decade timelines for investing in national infrastructures inevitably create cyclic demands as carriers seek to recover their capital expenditure investments and manage financing of these procurements.
The United States, in the wake of a major economic downturn in 2001–02, made “free market” decisions by allowing U.S. industry to succumb to global market forces even as China’s companies, Huawei and ZTE, began to emerge in global markets. Now the United States finds itself struggling to control a national security issue without the usual means to compete, namely its own industry.
Despite U.S. efforts to create a more diverse market of telecommunications service providers, the trend for equipment manufacturing has been toward re-consolidation to achieve economies of scale. Market consolidation during the past 15 years has reduced the global telecommunications equipment integrator market to primarily Finland’s Nokia, Sweden’s Ericsson, China’s Huawei and ZTE, and South Korea’s Samsung (all but Huawei are publicly traded). The United States now mostly depends on Nokia and Ericsson, with a smattering of Huawei deployments in low-density rural areas (the Rural Wireless Association in December 2018 told the FCC that 25 percent of its members use Huawei equipment). Nokia and Ericsson in recent years have been financially shaky, with debt bonds emerging from junk bond status in 2017 only to lowest-grade investment status in 2018 and 2019, and both companies have suffered negative net income most quarters since the beginning of 2017. Nokia suffered a significant downturn in the stock market after announcing that it would not meet its original growth targets in 2020 and would not be paying dividends in order to help fund development costs.
According to industry market research, Huawei’s global market share (including China) for service provider equipment is now greater than that of Ericsson and Nokia combined. As of 2016, Huawei reportedly supplied more than half of the 537 “fourth generation” (4G) mobile networks globally and 59 of the 90 4.5G networks, an intermediate step before 5G. European providers use Nokia and Ericsson equipment but increasingly have turned to Huawei for better prices and “advanced” capabilities that are ready to deploy. The Dutch telecommunications carrier in April 2019 cited a 60 percent price advantage when choosing Huawei. More than half of Huawei’s 5G contracts as of July 2019 were in Europe.
This proliferation of Chinese technology is further enabled by Chinese telecommunications service providers working together with other Chinese industries and cities to develop 5G applications within China and offering to build 5G infrastructure for other countries as part of the broader investment Beijing offers under its Belt and Road Initiative. As part of this overall effort, Huawei seeks to dominate the application of 5G for the “Internet of Things”—used in infrastructure and manufacturing—through technology development and the setting of international standards.
Can’t We Just Keep Huawai Out of the United States?
U.S. concerns with Chinese telecommunications companies—especially the more successful Huawei—are not new. The USG for years has sought to limit the proliferation of telecommunications equipment manufactured by China’s Huawei largely on the grounds that Huawei enables espionage for China’s security services. The U.S. Congress released a report on the threat posed by Chinese telecommunications manufactured equipment in 2012. Reports in 2019 from British and U.S. cybersecurity experts have cited a high degree of risk from a large number of security vulnerabilities well in excess of current industry norms. The African Union in May 2019 renewed its partnerships with Huawei on a range of technologies, from broadband and cloud computing to 5G and artificial intelligence, despite accusations in the media that China used Huawei equipment during a period of five years to steal data from the Union’s headquarters in Addis Ababa (construction of which was funded by Beijing).
President Trump in May 2019 issued an Executive Order on supply chains and ordered the placement of non-U.S. affiliates of Huawei on the Commerce Department’s Entity List. In addition, Section 889 of the Fiscal Year 2019 National Defense Authorization Act began to take effect in August 2019. This bill prohibits federal agencies from using “covered” entities or their subsidiaries and affiliates, including Huawei and ZTE. The bill prohibits federal contracts with companies that use covered entities as of August 2020. The actual impact of these recent policies will not be fully determined until rules and regulations are put in place; however, such impacts may be blunted significantly as part of a broader trade agreement. Other proposed bills call for promoting the development of 5G industry and for ensuring the security of 5G and future mobile telecommunications systems and infrastructure within the United States.
The exclusion policy has the potential to prevent further Chinese deployments within the United States but does not solve the longer-term problem of ensuring trusted communications infrastructure on a global basis for the United States or the proper functioning of other foreign critical infrastructure depended on by the U.S. military and other U.S. interests overseas. Even for domestic infrastructure, the exclusion policy would not protect U.S. interests in the event that the Nordic companies continue to run into financial difficulties despite positive outlooks for growth in 5G sales.
Most discussion of the Huawei issue centers around “espionage”; however, the greater concern is actually availability, given that encryption and authentication technology can be used to protect confidentiality and integrity of communications. Every critical infrastructure system in the United States, including that of the USG, depends on commercial telecommunications infrastructure. On the worst possible day in a conflict with a peer adversary, the United States may not be able to count on the survivability of communication satellites, and satellites do not provide sufficient data rates for the full scope of information-based warfare strategies regardless. Even surviving fiber links would be subject to disruption if the communications must pass through equipment provided by vendors from hostile countries, notably Huawei and ZTE. For example, Huawei completely dominates the telecommunications infrastructure in Iraq that the USG and military presence depend on.
Current policy depends on convincing other countries to use more expensive European equipment with capabilities reportedly lagging Huawei features by a year or more. U.S. policy also potentially has the effect of asking other countries to rip out existing investments in Huawei equipment prior to installing new 5G equipment, which in the case of Europe would cost more than $60 billion, according to some estimates, although others suggest it would cost much less. Within the United States, small rural carriers would face existential financial hardships transitioning away from Huawei. This policy impact could be mitigated in part with equipment that will be available by the end of 2019 to bridge the gap between old Huawei and new western equipment. To counter future transition problems, telecommunications service providers, in part through the Open Radio Access (O‑RAN) Alliance, are promoting the global use of interoperable standards that would allow the service providers to avoid vendor lock-in for future generations.
Some countries are seeking mitigation of the threat by limiting Huawei to the “edge” of their networks and excluding Huawei from the “core”; however, telecommunications companies and U.S. Government officials have noted that, with 5G, the distinction between the edge and the core largely disappears, suggesting this approach would not satisfy U.S. security concerns. The British experience with inspections of Huawei equipment highlights the challenge of guaranteeing security through inspection as another means of mitigation. A nation might seek to adopt a “zero trust” model for 5G; however, industry experience in developing “trusted computing” suggests that hardware as the “root of trust” has a degree of privilege that cannot be controlled against an untrusted supplier. Meanwhile, Huawei is seeking to improve its code review process to mitigate concerns raised by the British reports.
Some argue that U.S. policy should solve the global problem by “killing” Huawei through the cutoff from western supply chains; however, this view ignores the likelihood that Beijing would step in to assist its national champion even more so than it has already. China to date has always fallen short of the information technology state of the art and thus succumbed to the market imperative to use better western components; however, bifurcation of the global market would alter this dynamic decisively. The founder and chief executive of Huawei, Ren Zhengfei, prior to the G20 summit in June 2019 said:
The U.S. is helping us in a great way by giving us these difficulties. If we aren’t allowed to use U.S. components, we are very confident in our ability to use components made in China and other countries.
In response to U.S. policy and concern for their Chinese markets, Ericsson and Nokia reportedly are planning how they might need to bifurcate organizations and supply chains to remain in both western and Chinese markets, with potentially significant costs on top of already weak financial positions.
According to Triolo and Allison of the Eurasia Group, in their paper on “The Geopolitics of 5G”:
The United States and China also are competing to develop innovative technology applications that will run on top of deployed 5G networks. The United States has an advantage in terms of innovation capacity, but China will benefit from its head start building out its domestic 5G ecosystem and as Chinese companies then compete for market share abroad.
Triolo and Allison note that, “The push for a China-free 5G alternative is likely to delay 5G deployment where backup suppliers are forced to invest in new manufacturing capacity and human capital, further cementing China’s first-mover advantage.” They also note that:
A bifurcated 5G ecosystem would increase the risk that the global technology ecosystem would give way to two separate, politically divided technology spheres of influence. Such a split could result in some interoperability issues or lower economies of scale and higher transaction costs.
Potential Damage to the U.S. Semiconductor Industry
Meanwhile, the semiconductor industry and other high-technology industries in the United States have expressed concern about the collateral effects of the Huawei policy resulting from market “uncertainty” and the potential loss of Chinese markets, which in some cases account for a significant fraction of their revenue (for example, Qualcomm, Micron Technology, Qorvo, Broadcom, and Texas Instruments each earn more than 40 percent of their revenues from China; Intel and Nvidia get more than 20 percent from China). Huawei depends particularly on U.S. optical and analog components from companies such as NeoPhotonics (49 percent of revenue from Huawei alone), Lumentum (18 percent), Inphi (14 percent), Qorvo (13 percent), II-VI (9 percent), and Finisar (8 percent).
Because U.S. economic policies have so strongly favored globalization, key capabilities in the high-technology sector during the past 30 years have moved overseas to a significant degree. At first, this primarily involved assembly of electronic components; however, the most complex manufacturing—including for advanced printed circuit boards and semiconductors and their associated supply chains—increasingly are centered outside the United States. U.S. industry has focused on maintaining ownership of intellectual property through the design process but often depends on companies such as the Taiwan Semiconductor Manufacturing Company (TSMC) to make products.
How Did We Get to This Point?
The lack of a U.S. industrial base for integrated telecommunications equipment manufacturing that could compete with Huawei and the European firms for the global deployment of 5G systems is the direct result of a “perfect storm” of regulatory, technology, and economic shifts at the end of the 1990s. As detailed in the extensive study published in 2011 by Lazonick and March, the primary U.S. company, Lucent Technologies, ultimately failed because it made short-term financial decisions without a long-term vision for technology and global market development. In particular:
- The USG breakup of the Bell System in 1984 led by 1996 to the spinoff of Western Electric manufacturing and the world renown Bell Laboratories. The new company, Lucent, was intended to serve as a neutral provider to all of the emerging U.S. service providers. Canada’s Nortel had spun off from the Bell System in 1949 as a result of an earlier anti-trust suit.
- Telecommunications markets benefited from the boom times of the late 1990s and early 2000s but were then crushed in the wake of a general downturn in western economies in 2001–
- Lucent had an incumbent advantage with legacy technologies but failed to make the pivot to internet services and applications, in part because it sought to develop its own protocols rather than use the broadly accepted Internet Protocol.
- Lucent with the help of Bell Labs, along with Canada’s Nortel, made great progress in optical network technology; however, the technology was deployed far faster than anyone was prepared to use it, leading to a collapse in demand after 2000 for Lucent, Nortel, and the submarine cable business. Lucent spun off its optical cable division to Furukawa Electric in 2001. Global demand for optical networks did not recover for more than a decade.
- Lucent contributed to the early expansion of mobile networks with arguably superior technology for 3G networks, such as CDMA, but failed to capture markets in Europe and Asia that used other technology such as GSM.
- Lucent in 2000 spun off its profitable microelectronics division after forcing the division to compete for Lucent business with third-party providers.
- Lucent in 2000 also spun off its “slower growing” enterprise network division, leaving the company without an ability to compete when this market segment grew faster than the general telecommunications market after the 2001–02 economic downturn.
This toxic mix led to consolidation of the integrated equipment manufacturing firms in the West even as Huawei and ZTE emerged in global markets after the worst of the downturn, building on top of rapidly modernizing and expanding infrastructure in China. The UK’s Marconi went through a complicated series of mergers and divestments, with the telecommunications group eventually ending up in 2005 as part of Sweden’s Ericsson. Ericsson in 2010 absorbed what was left of Nortel after that company collapsed into bankruptcy in 2009. Finland’s Nokia absorbed Motorola Solutions in 2011, and then the communications group of Germany’s Siemens in 2013. Meanwhile, in 2006 France’s Alcatel absorbed Lucent; however, even this new firm could not thrive and was absorbed by Nokia in 2016. The only other major player to arise during this period has been South Korea’s Samsung, which so far has little overall market share in telecommunications equipment but seeks to build on its mobile phone and semiconductor reputations and South Korea’s investment in broadband networks.
Government’s Longstanding Role in Promoting Critical Infrastructure
History provides a guide for how the government could help reset the playing field with the goal of creating stronger competition against China’s heavily supported national telecommunications champion. Consensus on the government’s role in infrastructure development did not emerge immediately and will remain a matter of discussion in terms of when such interventions are appropriate. Discussions in the early days of the nation primarily centered on toll roads and canals that were deemed vital to commerce and military mobility for the growing nation. Competition for commercial traffic led to fights among the States until the 1824 Supreme Court decision in Gibbons v. Ogden ruled that the commerce clause of the U.S. Constitution granted power to regulate interstate commerce to the U.S. Congress, overriding any decisions by the States. From that point forward, the U.S. Government acted more decisively.
- The U.S. Army Corps of Engineers took on a direct role in the planning, building, and management of national waterways once the U.S. Congress passed the General Survey Act of 1824.
- The USG provided companies with financing and land rights for the building of the Transcontinental Railroad during the 1860s (the companies eventually paid off the loans).
- During the Depression of the 1930s, the Roosevelt Administration led the effort to build dams and the national power grid to accelerate the spread of electric power.
- The government has been involved in projects that created new infrastructure and involved long periods of investment, including for communications.
- In 1956, the Eisenhower Administration worked with the U.S. Congress to begin building the nation’s Interstate Highway system of almost 50,000 miles, largely funded by the U.S. Government.
- Starting in the late 1960s, the USG-funded research and development for packet-switched networks that by 1995 evolved into the commercial internet. USG interest initially was to create communications that would be resilient in the event of wartime destruction of key telecommunication nodes; however, the interest pivoted to the creation of new commercial services that had been piloted inside government-funded networks.
In other cases, the government sought to create private sector capabilities that required a large investment boost to get started.
- During the 1960s, the USG under President Kennedy funded a large-scale expansion of the nation’s space industry through the Apollo program. In 2019 dollars, NASA spent more than $250 billion between 1960 and 1973 to put a man on the moon—creating a major impetus for communications, electronics, and computer development. At its peak, the NASA budget represented more than 4 percent of the federal budget.
- The international satellite communications company IntelSat, with a constellation of about 50 geostationary satellites, in the early 1960s was created by a multinational government consortium and then fully privatized by 2001. InMarSat, with about a dozen geostationary satellites, has had a similar history, beginning in the late 1970s as an intergovernmental organization and transitioning to privatized operations in 1998.
The government has used a combination of loans, investment, subsidies, taxes, fees, and procurement to save U.S. industry from imminent collapse, to stimulate innovation, or to stimulate the economy as a whole.
- Since 1958 the Defense Advanced Research Projects Agency (DARPA) has funded the development of emerging technologies for military applications, most famously the networking technology that underlies the internet. In the late 1980s DARPA provided matching funds for SEMATECH, an industry consortium that managed grants for semiconductor manufacturing research.
- The U.S. Congress in early 1980 granted Chrysler $1.5 billion in loan guarantees to help the number three automaker recover from bankruptcy. Chrysler paid off the loans in 1982.
- The Department of Defense (DOD) in 2000 used procurement funds to help the Iridium satellite phone business stay afloat and remained a major customer as the infrastructure went through bankruptcy, sale, and then regeneration, because the constellation provided a unique global communications capability.
- DOD in 2005 used $50 million of Title III Defense Production Act funds to restore manufacturing capabilities of high-purity Beryllium metal that had been mothballed five years earlier because of declining demand and liability for health hazards to workers.
- The USG exercised a major role in preserving industry and infrastructure during times of economic difficulties, most dramatically in the wake of the 2008 financial crisis, when the U.S. Treasury and the Federal Reserve intervened on an unprecedented scale to save financial institutions and automakers on the brink of collapse resulting from bad loans. The U.S. Congress authorized the spending of up to $750 billion under the Troubled Assets Recovery Program (TARP), primarily through loans and stock purchases. Assistance to the insurance company AIG alone amounted to $182 billion. The government at one point owned 92 percent of AIG stock but over time sold all of the stock, resulting in a net gain of about $22 billion for taxpayers.
- The 2009 American Recovery and Reinvestment Act made extensive use of tax incentives for individuals and companies, but also set aside procurement funds, all intended as a short-term economic stimulus. Of almost $800 billion, only about $100 billion was specified for infrastructure, and less than a third of that went to building infrastructure (including about $7 billion for broadband internet)—illustrating the challenges for the USG to influence infrastructure either quickly or without a specified objective.
The rise of the defense industrial base after World War II, which helped create “Silicon Valley,” is the best example of a long-term government investment for national security purposes. This capacity was nurtured throughout the Cold War with massive, long-term investments through Defense Department procurements in combination with USG capabilities and facilities. By the year 2000, however, Defense Department policies sought to control costs through commercial-of-the-shelf products and an international supply chain. Policy guidance continues to be “buy, not make.”
Options for the Nation
Given the shortfalls of a “just say no” policy, the United States will need to compete in the telecommunications equipment integration sector, both in terms of products and trade strategy. The U.S. Government typically seeks to use procurement for federal networks and research and development investment as the primary levers for influencing high technology. U.S. industry already leads in component and subsystem technologies (notably in optics); however, that advantage has not overcome the boom and bust cycles of the equipment integration market. Thus, a new element will be required that will involve some combination of direct investment, subsidies, loans, and tax incentives as has been done for other industries, either for national security purposes or to preserve national economic or industrial capabilities. In addition, the USG could include preferred telecommunications equipment manufacturers (no matter where they are from) in U.S. trade, defense, and foreign policy packages that the United States seeks to implement with other nations that are upgrading their telecommunications infrastructure.
Similar ideas have been raised before, including by this author and by James Lewis of the Center for Strategic and International Studies. Lewis cited three options: build networks from insecure components, build a national champion, or subsidize European producers. According to Lewis, the Obama Administration considered funding a national champion using the Defense Production Act, “but it could at most allocate 1 percent of what China spent. The discussion of how to respond to the telecom problem made it as far as a Deputies Committee meeting, but none of the major information technology companies wanted to reenter this field. Though a few medium-size companies could have been candidates for investment, the administration ultimately decided to rely on Google and Silicon Valley to innovate our way out of the problem without the need for the government to spend anything.”
The U.S. Defense Science Board’s June 2019 report on “Defense Applications of 5G Network Technology” notes that “the lack of a U.S. integrator and Radio Access Network vendor industrial base” creates challenges. The report recommends that the Department of Defense “should provide seed funding for western industrial base alternatives of key system components, e.g., Radio Access Networks.”
The scale of investment required—as can be seen from the size of the European companies—would require the U.S. Congress to appropriate additional funds, even if implemented under existing authorities, such as Title III of the Defense Production Act (annual appropriations typically range only in the 10s to 100s of millions of dollars). Ericsson and Nokia each employ about 100,000 or more workers (although not just for telecommunications integrated equipment manufacturing), and each as of 2018 had net equities in the range of $10-20 billion and net assets in the range of $25–45 billion. Nokia spent $16.6 billion acquiring Alcatel–Lucent in 2016.
Maintaining leadership requires huge research investments. Huawei is participating comprehensively in the international standards process and makes large investments in research and development, now increasing to $15–20 billion per year from levels of $13–15 billion in 2017–18. European firms lag significantly. Nokia has increased investment in research and development to about 20 percent of its revenue or roughly $5 billion per year after a significant decline during 2013–15. In addition, the European Investment Bank in August 2018 provided a $583 million five-year loan to Nokia in 2018, and Canada in January 2019 provided Nokia with a $40 million research grant. Ericsson in 2017 increased investments to at least 15 percent of its revenue—a bit more than $4 billion per year—despite concurrent net income losses.
The major U.S. telecommunications service providers with operations in the United States and abroad would need to be included at least in the planning process for such an investment policy given that they would be the ultimate customers for most of the equipment, have expertise on the markets and systems and, most likely, would serve as the final systems integrators and operators during implementation and deployment. Indeed, the service providers could be provided incentives to participate directly in the investment strategy; however, they are also burdened with high levels of debt from capital expenditures. Other operators of critical infrastructure (financial systems, electric power, oil and gas distribution, transportation, etc.) also might benefit by participating in the planning and investments.
Tomorrow, The Cipher Brief will bring you part two of The Worst Possible Day: U.S. Communications and Huawei, by Cipher Brief expert and former Senior Director for Cyber Operations on the U.S. National Security Council Staff. Thomas Donahue.
PRISM is the flagship professional journal published as a public educational service by National Defense University. In its 10th year of publication PRISM has a quarterly print run of 11,000 with recipients in 83 countries. A complimentary subscription is available to Cipher Brief members by request to PRISM@NDU.EDU. Please provide an accurate postal mailing address and the preferred number of copies desired.
Read more national security news, insights and analysis in The Cipher Brief.
If you’re interested in being a part of the conversations that lead to solutions, consider joining the Cyber Initiatives Group, chaired by principals including General Keith Alexander, General Michael Hayden, and FireEye CEO Kevin Mandia. Find out more.
