Our $35 trillion national debt costs us 2.3 percent of our GDP to service, and that will rise to 7.2 percent of GDP by 2053, according to the Peterson Institute, levels we associate with Third World nations skirting national bankruptcy. The implications for national security are dire. We know that America’s military power is declining. But nothing on the political horizon will persuade Americans to pay higher taxes to support higher defense outlays, especially after a quarter-century of frustrating forever wars. The dominant meme of our national strategic discourse is lifeboat economics: Do we throw Ukraine under the bus or Taiwan?

　　The question has become moot during the past several months, as the combined West ran out of artillery ammunition, air defense missiles, and other ordinance to support Ukraine against Russia. NATO is scrounging through its inventories, cajoling Greece and Spain to send its limited stocks of Patriot missiles to Kyiv. 155mm shells, the workhorse of the Ukraine war, can’t be had in large numbers anywhere. The West can write checks, but it can’t produce enough weapons.

　　In the meantime, it is now widely acknowledged that Chinas massive investment in anti-ship missiles, modern aircraft, and submarines can overwhelm any naval assets that the United States might place within 1,000 miles of Chinas coast. This has been true for some time, but the continuing buildup of China’s strategic assets makes it harder to ignore.

　　We are well past the point where it makes sense to quibble about whether we should prioritize Europe or Asia. There is only one way to avoid strategic decline. That is to prioritize innovation. That is what we did when the United States fell behind the Soviet Union during the 1970s. We succeeded brilliantly then. We should try to do it again.

　　That also opens a path out of budget constraints. Federal investment in high-tech research and development (RD) has paid for itself many times over in the form of civilian spinoffs. In fact, the only kind of industrial policy that the United States has ever executed successfully proceeded from national security concerns.

　　China’s Advantage

　　Let us review the grim facts about the military balance in Asia. Two years ago I wrote that “China has invested massively in technologies that well may determine the outcome of any future war in the Western Pacific,” including long-range missiles, lasers that target communication satellites, and “AI-driven methods of submarine detection.”

　　In the interim, according to outgoing head of the Pacific Fleet Admiral John Aquilino,“China has built more than 400 aircraft, 20 major warships, and doubled its missile inventory” during his three-year tour. The Pentagon last year estimated that China has 1,000 carrier killer DF-21 intermediate-range missiles, enough to swamp the defenses of any US carrier group. China has tested the satellite-guided weapons system successfully for use against moving targets more than 1,000 miles from its coast.

　　In addition to its inventory of ballistic missiles, China is mass-producing anti-ship cruise missiles, according to the country’s media. One automated factory allegedly can build 1,000 cruise missile motors per day. China can make as many cruise missiles as it wants to.

　　Cruise missiles like the American Tomahawk or Popeye are an older technology. An American destroyer carries interceptors that can destroy them reliably, but a destroyer can only carry 100 interceptors in its hold, while China can fire as many cruise missiles from land as it wants, confirming the US military’s long-held belief that their surface vessels are particularly vulnerable to Chinese missiles.

　　China’s 66 diesel-electric submarines are also an effective threat to the US presence in the Western Pacific. The boats are very quiet. During Pacific exercises in 2023, similar German and Israeli diesel-electric submarines traversed the entire diameter of a US carrier strike group undetected.

　　China also has over 200 J-20 stealth interceptors and over 1,000 fourth-generation fighters. We do not know how good the Chinese planes are—and do not wish to find out. They are armed with the PL-17 air-to-air missile (range of 400 kilometers, twice that of the US AIM-260 Joint Advanced Tactical Missile). China has never fought an air war, and US analysts can only guess at the efficacy of Chinese aircraft, but the little we know suggests that China has a strong air force.

　　With a home-court advantage in logistics, China has a massive preponderance of firepower in its own theater. The last thing the US military wants is a kinetic situation with the PLA Navy. Defense analysts will continue to spin scenarios to ward off a Chinese invasion of Taiwan, imagining that the Chinese would traverse the 90 miles of the Taiwan Strait in landing craft after the fashion of Saving Private Ryan. That is a bait-and-switch: Taiwan has storage capacity for about 11 days of natural gas, and a single Chinese missile aimed at an LNG carrier would shut down the island’s economy (43 percent of its electricity comes from natural gas). The only country likely to attack Taiwan directly is the United States, which might elect to destroy the country’s state-of-the-art semiconductor fabrication plants to prevent them from falling into PRC hands.

　　China, as well as Russia, have deployable hypersonic missiles (maneuverable cruise missiles that travel at Mach 5 or faster). Conventional missile defense is ineffective for the most part against these systems. Former Under Secretary of Defense for Research and Engineering (USD[RE]) Michael Griffin has testified to Congress that the United States does not “have systems which can hold [China and Russia] at risk in a corresponding manner, and we don’t have defenses against [their] systems.”

　　In summary, there is nothing to prioritize in Asia, that is, no tactical or strategic objective that could be achieved by shifting existing resources at the margin from Europe to Asia or vice versa. The United States and its allies have allowed the defense industrial base to decay to the point that catching up with China’s production capacity would take a long time, if it is indeed possible.

　　China’s advantage in industrial automation is formidable. Its large automakers can manufacture electric vehicles at a fraction of American prices (the BYD Seagull sells in China for less than $10,000) and remain profitable. Its flagship telecommunications equipment maker Huawei produces 5G base stations at low cost. In July 2023, I visited a Huawei facility in Shenzhen that turns out 2,400 5G base stations per day, or nearly a quarter of the world’s installed capacity annually. Huawei claims to have 10,000 industrial automation projects underway using dedicated 5G networks. I am aware of fewer than ten in the United States.

　　The economics and logistics of missile war make it virtually impossible for the United States Navy to overtake China’s combination of industrial capacity and home court logistics within the existing framework of military technology.

　　The Alternative: Innovate

　　The United States needs to prioritize defense RD that pushes the envelope of physics: chips that use the quantum properties of matter to multiply computing power, quantum computers that can solve complex problems many times faster, directed-energy weapons that defeat projectiles traveling at Mach 8—and things that have not yet appeared on our thought horizon.

　　We have done this before. Russian surface-to-air missiles and anti-aircraft artillery downed nearly 100 US planes flown by Israeli pilots during the 1973 Arab-Israeli war, in a devastating display of Russian superiority in conventional air war. Only nine years later, at the so-called Beqaa Valley Turkey Shoot, the Israeli Air Force destroyed nearly 100 Soviet planes flown by the Syrian Air Force, using a combination of new technologies: AWAC surveillance, look-down radar, suicide drones, and other innovations. A combination of new computing, communications, and sensing technologies gave America and its allies a decisive advantage in aerial combat. None of these inventions could have been anticipated in 1973. The determination of the United States to innovate its way out of a losing position enabled a set of key discoveries.

　　We need the courage to pursue the unknown unknowns.

　　We do not know a priori which inventions might obviate China’s advantage in missiles. But we surely know where to begin work.

　　Directed-energy weapons may be the most promising field of research. The original Star Wars concept envisioned x-ray lasers based in space powerful enough to destroy ICBMs. The Defense Department currently spends $1 billion a year on directed-energy RD. According to media reports, the Pentagon in April deployed the first laser weapon capable of destroying drones. This appears similar to Israel’s “Iron Beam” laser system and related developments among America’s allies. Drones are easy to destroy, but using million-dollar interceptors to destroy drones that might cost $10,000 to manufacture isn’t viable. The cost per shot of a laser anti-drone weapon is less than $10.

　　Destroying fast-moving targets with lasers is more difficult by orders of magnitude. The laser must stay on target for long enough to inflict damage. That requires extremely fast computing, improved sensing, and enormous power output. There are other forms of directed energy—microwaves, for example—which are now subject to weapons research. In theory, weapons that project force at the speed of light should be able to defeat missiles that travel at several times the speed of sound, but directed-energy defense is only practical against relatively primitive, slow-moving projectiles. Because of the potential of these technologies, the Defense Department should take a Manhattan Project approach to directed-energy defense, that is, with no budget constraints.

　　Drone swarms represent a possible breakthrough in many aspects of warfighting. Manipulating drone swarms requires fast downloading of massive amounts of information, fast computation of alternatives, and fast response to new instructions, all on secure communications with secure geo-positioning. None of these tasks is trivial and their combination is dauntingly complex. Quantum satellite communications may hold the key to practical drone swarms, and China is the leader in this field. New kinds of computer chips might build vastly increased computing power into drones, just as the advent of CMOS chip manufacturing in the 1970s enabled look-down radar in military aircraft. Increases in chip speed and efficiency now depend on etching smaller circuits onto hunks of silicon; the next generation of chips might employ the interaction of molecules at the quantum level.

　　The Only Viable Industrial Policy

　　Industrial policy is now popular on the New Right. The free-enterprise wing of the Republican Party remonstrates that industrial policy produces rent-seeking, cronyism, and corruption—and the critique is correct, with one critical exception. Pouring public resources into existing industries begs the question of why the private sector wont invest in those industries. Is it because tax or regulatory policy inhibits investment? Then change the tax code and leash the regulators instead of handing out government checks. Is it because American labor cant compete against cheaper labor overseas? Then subsidize labor. The moment the federal government proposes to spend a dollar, demons awake and go abroad seeking rent.

　　The only occasions on which we have pursued an industrial policy of sorts conscientiously and in the national interest arose from crises in national security. We became Freedom’s Forge in 1940, as Arthur Herman entitled his book on the World War arms industry. We beat Russia to the moon under JFK. We responded to the loss of Vietnam and other strategic setbacks of the 1970s by reinventing warfare with smart weapons and modern avionics.

　　NASA RD generated a historical rate of return of 43 percent, according to a 1976 study commissioned by the agency. No one has tallied up the benefits of RD funding by the Defense Advanced Research Projects Agency (DARPA) during the late 1970s and early 1980s, but every invention of the Digital Age—the Internet, optical networks, plasma, and LED displays, the GUI interface, the computer mouse, CMOS chip manufacturing, and look-down radar—began with a DARPA grant.

　　Can we afford to do this? The federal development budget (mainly building and testing prototypes) at the peak of the Cold War in the early 1980s was about 0.8 percent of GDP versus about 0.3 percent today.To restore Reagan-era fund levels would cost about 0.5 percent times $29 trillion, or $145 billion a year. We couldn’t spend that much even if we passed a magic wand over Congress and got funded: The great corporate laboratories of forty years ago no longer exist, and we can’t train engineers fast enough. Suffice it to say that the objective seems out of the range of imagination.

　　J. P. Morgan’s observation about the price of his yacht applies: If you have to ask how much it costs, you can’t afford to be a superpower. High-tech RD falls under the broad definition of public improvements in Alexander Hamilton’s sense. It is an investment whose outcome cannot be predicted in advance, but whose long-run benefits have exceeded all possible expectations.

　　One of the architects of the revolution in military technology during the late 1970s and early 1980s was the late Andrew W. Marshall, director of the Pentagons Office of Net Assessment from 1973 to 2015. When I brought a group of newly hired National Security Council staffers to meet Marshall at his apartment in Alexandria in 2017, he told the group: “Never ask the service heads what they want. They will always ask for more of what they have now. For [a technological transformation] to work, it has to be run out of the office of the Secretary of Defense—not the ‘office’ as defined on the organizational table, but out of the SecDef’s actual office. He has to tell everyone at the Pentagon, ‘This is what I want,’ and make sure they do it.’” He died in 2019, at the age of 98.

　　China’s accomplishments are impressive, but the Chinese aren’t ten feet tall. Give them a specific task, for example, to work around American technology restrictions, and they will get it done. A senior Huawei manager told me, “If we have a problem to solve, we’ll put a thousand engineers on it. And if they can’t do it, we’ll put ten thousand engineers on it, until we get it done.” But in RD, the creative 1 percent can accomplish more than the merely competent 99 percent.

　　We owe most of the decisive technological breakthroughs of the Digital Age to stubborn mavericks who hijacked a DARPA-funded research product and did something with it that no one anticipated. Without exception, every major invention of the Digital Age began with DARPA funding, but none of these inventions conformed to the original purpose of the research product. Optical networks, the technology that made the Internet possible, is a case in point. DARPA-funded research to illuminate battlefields at night. An RCA Labs engineer used the funds to perfect the semiconductor laser, which transmits the signals through glass fiber that now carries the worlds data.

　　We need the courage to pursue the unknown unknowns. We can’t know in advance which of the countless competing technologies will transform computation, communications, propulsion, sensing, and industrial automation. The only thing we know with 100 percent certainty is that doing the same things we have been doing for the past thirty years will get us nowhere.