By Ryan Murphy '23; Photo by Bloomberg
“Over the past 18 months of the COVID-19 pandemic, it has become clear that US supply chains have firmly established themselves as an issue of both economic security and national security. Late night calls in search of masks for our nurses, hand sanitizer for our citizens, and microchips for our automakers, laid bare these vulnerabilities in the commercial sector. That searing experience put new focus on defense supply chains – meaning the international networks that provide the goods and services needed to deliver finished products to the Department of Defense – the defense industrial base, and the ways that our defense supply chains were prepared to respond to supply shocks. It is now incumbent on the US Government, in concert with industry and allied nations, to mitigate critical defense supply chain risks, increase surge capacity, and enhance resilience…”Report of the Defense Critical Supply Chain Task Force, House Armed Services Committee, July 22, 2021
In an international environment confronted by the coronavirus pandemic, breakdown of global value chains, and accentuated national security concerns, Industrial Policy (IP) and Supply Chain Risk Management (SCRM) have emerged as unassuming but critical objectives of the Trump and Biden Administrations. In a July 2021 statement sponsoring the Creating Helpful Incentives to Produce Semiconductors (CHIPS) for America Act, Senator John Cornyn (R–TX) observed that “What [the United States] is doing in industrial policy is unlike anything people with my free-market, conservative background would ordinarily be comfortable with” but whose “driving impetus is what China is doing and the [security of the] supply chain” (Ip 2021). Similarly, Senator Mark Warner noted that “[He] was very impressed with the Chinese model” and that “It’s hard to see how a company in America or any normal, traditional market-based economy can compete against that kind of juggernaut and win” (Ip 2021). Accompanying a strong bipartisan consensus on China relating to issues of national security (conducting Freedom-of-Navigation Operations in the South China Sea and maintaining the de-facto sovereignty of Taiwan) and international norms (condemning alleged human rights violations in Xinjiang Province), analysts cite a dramatic reconfiguration of industrial intervention embodied by executive order (“Buy American Laws” including the federal procurement of electric vehicles and general federal procurement from domestic producers) and legislation (the CHIPS for America Act, United States Innovation and Competitiveness Act, and Infrastructure Investment and Jobs Act) promoted by Congress and the Biden Administration. Overall, the development of IP and SCRM embodies significant continuity of twentieth century industrial intervention in response to national security concerns. Nevertheless, current IP and SCRM initiatives oriented toward four areas identified by the Department of Defense including (i) semiconductors and electronic components, (ii) rare earth elements or REEs, (iii) propellants and supporting chemicals, and (iv) active pharmaceutical ingredients or APIs may reverse increasing transnationalization of the U.S. Defense Industrial Base (DIB) while encouraging private sector growth in breakthrough industries, although such a strategy must address the “disconnected innovation pipeline” of defense-related product deployment, demonstrate clear empirical standards of success, and acknowledge the complexity of globalized manufacturing of the twenty-first century.
Broadly defined, Industrial Policy (IP) refers to a “set of policies and programs explicitly designed to support specific targeted industries and technologies” (Bonvillian 2021, p. 3), while as defined by the National Institute of Standards and Technology (NIST), Supply Chain Risk Management (SCRM) refers to a “systematic process for managing supply chain risk by identifying susceptibilities, vulnerabilities, and threats throughout the supply chain”. The recent shift toward active IP and SCRM bears stark contrast to economic neoliberalism and government nonintervention of the preceding thirty years of Washington Consensus standards oriented toward “deregulation, de-unionization, privatization, and free-trade agreements” (Wade 2014, p. 382). A number of scholars have noted that “Industrial policy remains a dangerous subject in America, because to express sympathy risks being classed as an incompetent or worse” (Wade 2014, p. 381) while “For years, if you wanted to be dismissed in debates over how to help the U.S. economy, you merely had to utter the words ‘industrial policy’” (Bernstein 2021). However, such statements belie the extensive historical role of the federal government in the promotion of defense and eventually “dual use” defense-civilian industries. In his first State of the Union address in 1790, President George Washington noted that the “safety and interest [of Americans] require that they should promote such manufacturers as tend to render them independent of others for essential, particularly military, supplies” (Bernstein 2021). Similar advocacy of subsidies and tariffs for strategic industries, included in the 1791 Report on Manufactures of Alexander Hamilton found expression in the “American System”, whereby U.S. industrial tariffs 1800-1930, averaging 30%, significantly exceeded those of other industrializing economies (Wade 2014, p. 386). Finally IP initiatives included those of agricultural research and development of land-grant universities under the Morrill Acts of 1862 and later, public corporations of the New Deal under the Roosevelt Administration including the Tennessee Valley Authority (TVA) (1933) and the Works Progress Administration (1935) (Wade 2014, p. 388). Thus, although uncoordinated in their implementation, U.S. IP initiatives emerged as strongly associated with an industrializing national economy over the course of the nineteenth and early twentieth centuries.
The expansion of U.S. IP responded first to heightened national security concerns following the Second World War with the advent of protracted geopolitical competition with the Soviet Union and its aligned states. Reacting to the 1957 launch of the Sputnik 1 satellite, the Defense Advanced Research Projects Agency (DARPA) formed in 1958 under the Eisenhower Administration as a research and development branch of the Department of Defense. Critically, DARPA projects followed (i) a “right-left” approach whereby potential breakthrough technologies would be identified (“right”) and only subsequently the requisite technical research conducted (“left”) and (ii) “follow-on” product procurement whereby the Department of Defense participated in all stages of development (testing, prototyping) and provided an initial market for new products (Bonvillian 2021, p. 6). Thus, U.S. IP, as claimed by Wade (2014) emerged first through national security considerations and basic research, whereby “America has had three types of industrial policy: first, World War II, second, the Korean War, and third, the Vietnam War” (388).
However, despite notable successes in aviation, electronics, and computing, DARPA projects which followed the “right-left” approach failed to achieve commercialization in civilian markets. In the case of U.S. semiconductor manufacturing, which lost market share to Japanese manufacturers between 1970-1980, new IP initiatives emerged specifically to counteract the disconnected innovation pipeline resulting from lack of an initial market, including the Bayh-Dole Act of 1980 (granting ownership of federally-funded research to individual universities), and Small Business Technology Transfer (STTR) and Small Business Innovation Research (SBIR) initiatives launched in 1982 under the Small Business Administration formed to provide competitive R&D funding for small and medium-sized enterprises (SMEs). Finally, Manufacturing Extension Partnerships (MEPs) formed in 1988 to provide advanced manufacturing techniques and facilitate information sharing among SMEs. Thus, U.S. IP, as claimed by Wade (2014) has expanded beyond basic research to include new decentralized forms of public-private cooperation and market coordination to form “America’s home-grown version of state capitalism” (p. 396).
The current degree of U.S. dependency on global value chains demonstrates the increased attention to U.S. IP and SCRM initiatives. As noted by Bonvillian (2021), U.S. manufacturing capacity witnessed a dramatic decline between 2000-2010. Output of U.S. manufacturers returned to 2000 levels only in 2018, while significant trade deficits in all goods ($911 billion) and technological goods ($191 billion) have emerged by 2020 (p. 8). Moreover, 92 percent of advanced (7mm) chip manufacturing capability is held by the Taiwan Semiconductor Manufacturing Company (TSMC), which has been threatened by increased activity of Chinese security forces and massive state-led investment in domestic manufacturing capability in its Semiconductor Manufacturing International Corporation (SMIC) (Bonvillian 2021, p. 11). Finally, 80 percent of REE and 87 percent of API processing capability rely on supply chains vulnerable to Chinese interference (Bonvillian 2021, p. 18).
Examination of the beneficial impact of U.S. IP demonstrates several key successes. According to William B. Bonvillian of the Information Technology & Innovation Foundation (ITIF), over a thirty-year period of implementation, MEPs have yielded $18.8 billion in cost savings and $111.3 billion in sales (expanded or maintained) to SMEs based in the United States (Bonvillian 2021, p. 7). Similarly, studies of the Advanced Technology Program (ATP) under the Department of Commerce have claimed a fifty percent reduction in research cycle time for firms who received such funding versus those who applied (Wade 2014, p. 393). However, perhaps the most significant example of successful U.S. IP was the 1987 formation of Sematech, a consortium of 14 semiconductor manufacturing firms, in response to falling U.S. semiconductor market share to Japan. Supported by DARPA, which provided funding of $500 million annually between 1989-1994, Sematech is claimed to have restored American semiconductor manufacturing capacity, such that by 2019, they formed the fifth-largest export totaling $42 billion (Bonvillian 2021, p. 11). Therefore, a positive precedent exists over the past four decades for U.S. IP, whose continuation in areas of semiconductors, RREs, propellants/explosives, and APIs may prove critical to confronting supply chain and technological dependency risk of an ascendent China, while maintaining domestic industry leadership in high-value manufacturing activities.
Examination of the neutral or detrimental impact of U.S. IP however bears some qualifying consideration. According to Scott Lincicome and Huan Zhu of the Cato Institute, DARPA projects, despite having been credited with development of global positioning systems (GPS), light-emitting diodes (LEDs), solar panels, and mRNA technology, have consistently failed to generate “mission-oriented directionality” necessary for project commercialization (Lincicome and Zhu 2021, p. 5). Despite the apparent success of government-funded research in development of Pfizer/BioNTech vaccines for the coronavirus pandemic, the authors note that government contracts leveraged “preexisting manufacturing capacity” alongside multinational research and logistics chains, and was thus not emblematic of U.S. IP defined as being “generated within national borders” (Lincicome and Zhu 2021, p. 4). Moreover, increased investment from federal sources may also generate moral hazard. A frequently-cited example of failed implementation of U.S. IP occurred in the case of Solyndra, a manufacturer of thin film solar cells, which received $535 million in loans from the Department of Energy in 2009, and whose bankruptcy in 2011 generated a federal investigation on the supposed misrepresentation of finances (Bonvillian 2021, p. 9). However, perhaps the difficult task of U.S. IP is epistemic. Attribution of industry “success” to U.S. IP requires “evidence that can ‘hold constant’ all the required variables…[which] does not exist and likely never will” (Lincicome and Zhu 2021, p. 20). Even without such strict causal parameters, Lincicome and Zhu note that “American industrial policy should be considered on its own merits” rather than against state-coordinated Chinese firms. Indeed, while government investment under the Made in China 2025 initiative increased Chinese global share of semiconductor manufacturing from 1 percent to 15 percent between 2000-2020, 75 percent remains owned by foreign firms (Lincicome and Zhu 2021, p. 42). Therefore, a more ambiguous evaluation can be made for U.S. IP, as such initiatives may not necessarily divorce U.S. manufacturing and development from global information and logistics networks, may generate moral hazard, and may not prove necessary or sufficient to maintain domestic industry leadership.
Under the preceding and current administrations, the United States has shifted from a declared norm of regulation to active research and investment in selected industries of the U.S. DIB and high-value manufacturing activities. In their relation to accentuated national concerns of a potentially revisionist China, U.S. IP and SCRM initiatives have gained attention and broad bipartisan support, and may yet achieve objectives beyond defense-oriented manufacturing, including renewable energy sources and conservation, as well as expanded social programs and subsidies for childcare. In historical context, U.S. IP emerged as strongly associated with an industrializing national economy over the course of the nineteenth and early twentieth centuries. Moreover, by the mid-twentieth century, DARPA and associated market “pipeline” programs including SBIR, STTR, and MEPs proved critical in the development of high technology in the civilian sector. Nevertheless, U.S. IP bears several qualifications including the importance of global information and logistics networks to the development of the coronavirus vaccine, tendency toward moral hazard, and its relative ineffectiveness for generating domestic industry leadership. In a critical moment of growing national consensus, however, I propose three areas for further consideration.
- U.S. IP and SCRM initiatives should continue to promote a clear “pipeline” for defense-related product deployment. Existing programs (SBIR, STTR, and MEPs) should be strengthened, with a focus on ameliorating commercialization difficulty posed by the “right-left” development model and lack of initial markets.
- U.S. IP and SCRM initiatives should develop clear empirical standards of success based on alternative scenarios, competitive project submission, and continuous oversight of funding and project development goals.
- U.S. IP and SCRM initiatives should consider the difficulty of a “national industrial policy” in today’s globalized information and logistics networks, and include development of joint regulatory standards and incentives structures for U.S. firms connected to overseas markets and human/physical assets.
Thus, as the United States under the Biden Administration seeks to “win the twenty-first century” through expanded IP and SCRM initiatives, policymakers should consider both the possibility embodied by federal intervention in selected industries, as well as the potential for misdirection and misallocation of resources which may lead the new millenia to be both less secure and less prosperous.