1. Five Lines on a Blank Page
February 2025. A large conference room on the seventh floor of the Ministry of Trade, Industry and Energy, Sejong Government Complex. Beyond the windows, the wide, low skyline particular to this administrative city stretched in every direction. Inside, fluorescent panels and a projector beam overlapped in the room's characteristic brightness. Twelve binders marked with security labels sat on the table, one of them labeled "Structural Analysis of Strategic Assets in Advanced Industries." The head of the Trade Negotiation Bureau stood before the whiteboard. This was a closed strategy session. Twelve working-level directors — semiconductors, batteries, biotech, and foreign affairs — were seated around the table. The subject was "compiling a list of Korea's indispensable assets."
The deputy minister wrote five items on the board. Technology. Supply chain. Institutions. Talent. Security. He left a blank beside each — space for a score out of five. Then, after a brief hesitation, he added a sixth. Cyber. Beside that one, instead of a blank, he wrote a question mark.
Two hours of discussion filled the board. HBM 62 percent. DRAM 70 percent. These numbers had circles drawn next to them. But beneath the circle beside HBM, someone had added a note in small handwriting. "Micron closing in at 21 percent." The speed at which that gap narrows determines how long the circle stays valid. Foundry 8 percent — an X. The three battery companies — a triangle, somewhere between circle and X. Fertility rate 0.72 — underlined in red. After the meeting, one participant said to a colleague in the hallway:
"Half of this country is the world's best, and half of it is the world's worst. Is there another country like this anywhere?"
That remark is where this chapter begins.
Part I showed, inductively, what history teaches. Chapter 1's Hanseatic League — a network of 200 cities monopolized trade between the North Sea and the Baltic, then collapsed under Dutch technological innovation. Chapter 2's Switzerland — 210 years of neutrality, swapping out the contents of its indispensability era by era, from watchmaking to finance. Finland in the same chapter — practicing strategic patience under a friendship treaty with the Soviet Union while building Nokia.
Part II showed, deductively, what seven contemporary nations prove. Chapter 4's TSMC — commanding 90 percent of advanced foundry capacity to become Taiwan's silicon shield. Chapter 7's ASML — locking 100 percent of EUV lithography, the headwaters of the global semiconductor supply chain. Chapters 3 and 6's Singapore — turning institutions themselves into a product, designing an Indispensable Node out of nothing.
When these two lines of analysis converge, five conditions remain. In Chapter 3 we first set them out. Irreplaceable technology. A position in the supply chain that cannot be bypassed. Institutional trust accumulated over decades. A talent system that produces the next generation. The capacity to preserve autonomy between great powers.
Measured against these five conditions, Korea scores as follows: technology (4 points) + supply chain (3.5 points) + institutions (2.5 points) + talent (3.5 points) + security (3 points). The combined total is 16.5 — 66 percent of a perfect score of 25. And the sixth condition, cybersecurity, carries a provisional score of 1.5. Lower than any of the five — but with more levers for improvement than any of them.
Pass or fail? That judgment comes at the end of this chapter. First, let us open each condition one by one.
Before reading the scores, the basis for arriving at them should be stated clearly. Five points represents a monopoly position — a state in which no practical substitute exists for the given condition. Four points represents structural advantage — substitution is possible but costly and time-consuming. Three points represents competitive parity — multiple actors hold comparable capability. Two points represents structural disadvantage — a clearly inferior position relative to other actors. One point represents dependent vulnerability — a state in which national interest is determined by the decisions of external actors.
Three to five indicators were used to assess each condition. Technological irreplaceability was judged by global market share, the existence of substitute technologies, the number of generational gaps versus competitors, and the difficulty of reproducing the process. Supply-chain position was measured by export-import asymmetry coefficients, the possibility of bypass within global supply chains, and the domestic-content rate of key inputs. Institutional flexibility was assessed using the World Bank Government Effectiveness Index, the speed of regulatory transition, MSCI classification, and the volume of FDI inflows. Talent reproduction was evaluated through PISA rankings and R&D intensity, unicorn output rates, and the medium-to-long-term demographic trajectory implied by the total fertility rate. Security-economy balance was grounded in the geopolitical distance between security ally and largest trading partner, the actual scale of damage from economic retaliation, and the progress of arms-export diversification.
This scorecard is not a precise quantitative measurement — it is a snapshot of structural position. On the same indicators, analysts may differ by 0.5 points. That is not evidence of uncertainty; it is an expression of the complexity of reality. What matters here is not the absolute number but the relative position across conditions and what that position means.
2. Cut One Thread and AI Stops
In technological irreplaceability, Korea scores 4 out of 5. Among the seven countries, this is third highest — behind Taiwan and the Netherlands.
Nearly the entire score is explained by HBM. In Chapter 11 we visited the M16 fab at SK hynix's Icheon campus. A process that stacks 12 layers of DRAM dies vertically, packing 1.18 terabytes per second of bandwidth inside a structure 775 micrometers thick — thinner than ten strands of human hair. SK hynix holds 62 percent of the HBM market; combined with Samsung, the figure is 79 percent. A single NVIDIA H200 GPU contains six HBM3E chips. If the cleanroom in Icheon stops, global AI infrastructure stops.
In Chapter 4 we saw how TSMC, commanding 90 percent of advanced foundry capacity, became Taiwan's silicon shield. Korea's dominance in HBM is comparable — but there is a decisive difference. TSMC commands the entire market for foundry as a category. Korea's dominance is concentrated in memory as a domain, and within that, in HBM as a single product line. In DRAM overall, Korea's share is still overwhelming at 70.5 percent; in NAND flash, 52.6 percent. In OLED displays, Samsung and LG divide the world market. But the one domain where it can be said with certainty that "this is impossible without Korea" is HBM.
The weakness in foundry shaves off one point. Samsung 8 percent against TSMC 67 percent. That gap of 59 percentage points is not a simple numerical difference — it is a gulf in customer trust. Samsung became the world's first to introduce a 3-nanometer GAA process in 2022, but recurring yield problems caused Qualcomm and NVIDIA to migrate to TSMC. In Chapter 7 we saw ASML achieve 100 percent monopoly on EUV lithography systems. The core of that monopoly was a barrier to entry built from 30 years of R&D and the complexity of 100,000 components. Samsung Foundry lacks that kind of structural barrier. The technology exists — but it has not reached the level of "cannot do without it."
Even so, 4 points is not a low score. The fact that HBM is a physical bottleneck in the AI era structurally reinforces Korea's technological standing. No matter how sophisticated an AI model becomes, it cannot run without memory. SK hynix's MR-MUF technology — a stacking method that uses non-conductive film instead of mold resin — is process know-how that competitors cannot replicate immediately.
Semiconductors are not the only story. In OLED displays, Samsung and LG divide the global market between them. Premium displays for smartphones and tablets are almost entirely Korean. In automobiles, Hyundai-Kia recorded all-time-high exports of $70.9 billion — roughly ₩94 trillion — in 2023, and the shipbuilding arm of HD Hyundai takes more than half of global orders in LNG carriers. These technologies do not rise to the level of HBM — where cutting one thread stops the world. But they demonstrate the breadth of Korea's technology portfolio. The 4-point score comes from HBM's depth; that depth is supplemented by breadth.
Consider the lesson history teaches. In Chapter 1, the Hansa's monopoly collapsed under Dutch technological innovation because that monopoly was based on structural position — a geographic lock that made trade impossible without passing through Lübeck and Hamburg. Once someone found a way around the structure, it collapsed. Korea's HBM dominance is based not on position but on accumulated process capability. TSV through-silicon vias, microbump bonding, thermal management — these are not a shipping lane that can be circumnavigated. They are the product of decades of repeated trial and error.
Warnings have emerged that China's CXMT is releasing DDR5 and closing the technology gap to one or two generations. In 2024, Korean prosecutors indicted ten former Samsung employees on charges of leaking 10-nanometer DRAM technology to CXMT. The pursuit is real. CXMT's revenue has exceeded roughly $3 billion — roughly ₩4 trillion — triple its 2022 level, and its DDR4 chips are supplied 50 percent more cheaply than Korean products.
But the stacking process for HBM demands a level of complexity categorically different from planar DRAM. Stacking 12 layers is not building 12 rooms — it is building a 12-story structure inside the thickness of a human hair. The collapse pattern of Chapter 1 will not repeat immediately — with the emphasis on "immediately." History offers no guarantee that it will never repeat. Investing more than ₩600 trillion in the Yongin Semiconductor Cluster and extending the technology gap again with HBM4 is the only way to protect this score of 4.
3. The Holes in the Net
In supply-chain asymmetric dependence, Korea scores 3.5 points. Strong, but not complete.
Start with the strong side. The three battery companies — LG Energy Solution, Samsung SDI, and SK On — hold roughly 15 percent of the global EV battery market. They trail CATL's single-company share of 39 percent, but their position as non-Chinese battery suppliers to the United States and Europe is solid. In Chapter 11 we saw LG Energy Solution's plant in Wrocław, Poland — the largest battery factory in Europe — and the Korean EV cluster forming around Hyundai Motor's Metaplant America in Georgia. For Western automakers, Korean battery companies carry structural demand as "the alternative to China." After the IRA introduced the FEOC provision excluding Chinese-sourced critical minerals, that demand was locked in by policy.
Samsung Biologics is a different kind of supply-chain node. The world's largest CDMO — contract development and manufacturing organization — it recorded revenue of ₩4.55 trillion (roughly $3.3 billion) in 2024, a 23 percent increase year-on-year. With the groundbreaking of Plant 5, total production capacity reaches 780,000 liters. In early 2025, it signed the largest contract in its history — $1.4 billion — roughly ₩1.85 trillion — with a European pharmaceutical company. As global big pharma shifts toward outsourced production, a structure is forming in which large-scale biopharmaceutical supply becomes unstable without Samsung Biologics. Celltrion's biosimilars — generics of Remicade, Herceptin, and Avastin — also play a structural role in cutting original drug prices by 40–60 percent in global markets. Korea's global market share in biosimilars is roughly 30 percent, and in this domain too the situation is approaching the point where "the cost structure changes without Korea." As de-China CDMO demand grows among Western pharmaceutical companies on account of geopolitical risk, the factories in Incheon's Songdo are becoming increasingly difficult to bypass.
But there are holes in the net. For the materials, components, and equipment — the so-called "sobu-jang" inputs — that Korean semiconductor manufacturing requires, Korea depends heavily on Japan, the Netherlands, and the United States. Photoresists, blank masks, and the key materials of hydrogen fluoride come from Japan. Japan's 2019 semiconductor materials export restrictions exposed that dependence head-on. Lithography systems come from ASML; etch equipment is dominated by Lam Research and Tokyo Electron. Korea's domestic-content rate for semiconductor equipment stands at just 20 percent. The target is 40 percent by 2035, but when one considers the complexity of ASML that we examined in Chapter 7 — Carl Zeiss mirrors, Trumpf lasers, the precision assembly of tens of thousands of components — full domestic production of core equipment is a task that is unlikely to be achieved within ten years.
In Chapter 8 we saw the paradox of Indonesia controlling 55–60 percent of global nickel production while Chinese companies control 75 percent of the refining. The resource belongs to Indonesia, but the core of the value chain is held by China. Korean semiconductors face a structurally similar paradox. In HBM, Korea commands the world; but in the equipment and materials needed to make that HBM, it is commanded. The direction of asymmetry is favorable to Korea in product, and unfavorable in inputs. That two-directional asymmetry is the meaning of 3.5 points — not full supply-chain dominance, but standing at the center of mutual dependence.
The dependency structure appears in battery raw materials as well. More than 80 percent of lithium refining, 70 percent of cobalt refining, and 91 percent of anode material production are Chinese. In Chapter 11 we saw the IRA's FEOC provision shake the entire supply chain of Korean battery companies. Removing Chinese intermediate materials requires redesigning the supply chain from the ground up. POSCO Future M's construction of a precursor plant in Gwangyang, South Jeolla Province — connected to the Indonesian nickel ladder seen in Chapter 8 — and the launch of a Hyundai Motor-LG Energy Solution joint battery cell plant in Karawang, Indonesia, mark the beginning of that redesign.
Semiconductor exports themselves are breaking all-time highs. $141.9 billion in 2024, $173.4 billion in 2025 — accounting for 28.3 percent of total exports. The HBM supercycle created by AI demand is the pillar of Korea's trade balance. But the fact that this pillar is concentrated in a single product line is a structural risk.
The first-ever trade deficit with China in 2024 is a larger signal of this structural shift. Imports ₩139.9 trillion, exports ₩132.9 trillion — a deficit of roughly $7 billion, approximately ₩9.2 trillion. Korea now buys more from China than it sells to China. In a history of decades-long surplus in Korea's trade with China, this is a structural turning point. The direction of asymmetry is gradually tilting unfavorably.
One point must be faced directly. Is the target of raising equipment domestic-content from "20 percent to 40 percent" realistic? We who dissected ASML's EUV in Chapter 7 know the answer — Carl Zeiss mirrors, Trumpf lasers, and the precision assembly of 100,000 components are not things that can be replicated within ten years. Aiming to domestically produce all sobu-jang inputs is not only unachievable but undesirable — because it means purchasing sovereignty at the cost of efficiency. The realistic approach is bifurcation. Focus domestic production on materials, components, software, and back-end equipment where it is achievable; in the domains where it is not achievable — EUV core optics being the prime example — reduce vulnerability through multiple sourcing and supply assurance agreements. The goal is not "complete self-sufficiency" but "management of critical vulnerabilities." In Chapter 8 we saw Indonesia claim resource sovereignty over nickel while depending on China for the technological sovereignty of refining — possessing the resource is not the same as controlling the value chain. Similarly, measuring supply-chain security by a single domestic-content figure is an illusion.
4. The Slow Clock, the Drying Pipe
In institutional flexibility, Korea scores 2.5 points. The lowest of the five conditions.
What that score means becomes clear in comparison with Chapter 6. Singapore created the Asian Dollar Market in 1968, striking precisely into the gap left by Hong Kong's 15 percent withholding tax on interest income earned by non-residents. Lee Kuan Yew's government decision-making structure — rapid judgment and execution by a small elite — made that possible. World Bank Government Effectiveness: global rank 1. Transition from regulatory sandbox to full regulation is smooth. Korea's Government Effectiveness percentile is 82 — not bad, but inferior to Singapore's 100, the UAE's 90, and the Netherlands' 93.
Labor market rigidity is another factor cutting into the institutional score. Employment protection for regular full-time workers at large companies is among the highest in the OECD, while the dual structure between large corporations and small-to-medium enterprises and irregular workers is entrenched. For startups to recruit top-tier talent, they must compete on salary with the chaebol — without the flexibility to dismiss. That structure blocks the formation of an Israeli-style agile startup ecosystem.
Korea's problem is not the speed of legislation. The AI Basic Act (인공지능 기본법) passed as the world's second AI law, following the EU's. Cumulative regulatory sandbox approvals across ICT and industrial sectors exceed 1,000. Korea was also the first in the world to commercialize 5G. The problem is the gap between legislation and execution. Laws are made quickly, but the speed at which law turns into reality is slow. The rate at which sandbox approvals convert to full regulatory status is low, and companies fall into the dilemma of "it works in the pilot, but not as a real business." It is as though a fast clock and a slow clock hang on the same wall.
The Korea Discount is the financial expression of this institutional rigidity. Samsung Electronics' historical price-to-earnings ratio is 8–12 times — extremely low compared with TSMC's 23 times and ASML's 38 times. Despite equivalent technology assets, it trades at a 50–70 percent discount to TSMC. Chaebol founder-family governance, insufficient protection of minority shareholders, low dividends — these are institutional problems, not technological ones. The failure to secure inclusion in the MSCI developed-market index in 2024 is symbolic. In Chapter 3 we saw Singapore turn institutions themselves into a product — the trust that "Singapore keeps its promises" attracted the regional headquarters of 4,200 multinational companies. Korea's institutions are working in a direction that erodes the value of the technology assets.
In Chapter 9 we saw the UAE — after 2020 — increase FDI by 35 percent through bold institutional innovation: 100 percent foreign ownership, the Golden Visa, and the appointment of a Minister of AI. The caveat of authoritarian speed applies, but the results were clear. Korea's FDI inflows run at roughly $13 billion — roughly ₩17 trillion — one-twelfth of Singapore's $160 billion. Insufficient institutional flexibility is repelling foreign capital and talent. The 2024–2025 impeachment crisis compounded the policy continuity risk. What investors dislike most is not loss — it is uncertainty.
Yet the impeachment was an exceptional event, and Korea's deeper structural constraint lies in the five-year single-term presidential system itself. Each time power changes, the policy horizon resets. A predecessor government's core policies are reduced or discarded by the successor, and the new government re-approaches the same problems under new names. The National Pension and National Health Insurance survived changes of administration only because their beneficiary base was too large to reverse. But regulatory sandbox conversion, visa simplification, and MSCI eligibility reform — the institutional innovations needed now — do not connect directly to the felt experience of millions of voters. The reset that comes every five years blocks the compounding effect of institutional improvement. That is the root of the paradox: legislation is fast, execution is slow.
Talent ecosystem reproducibility scores 3.5 points. The two sides are sharply divided.
Start with the strong side. In PISA 2022, Korea scored 527 in mathematics (ranked 3rd–4th in the world), 515 in reading, and 528 in science. University enrollment is above 76 percent. R&D investment stands at 4.96 percent of GDP — second in the OECD, behind Israel. In absolute terms, roughly $105 billion — roughly ₩139 trillion — the highest by far among the seven countries. Samsung Electronics alone filed more than 6,000 AI patent applications in 2024. At 7,309 patent applications per $10 billion of GDP, Korea ranks first in the world. Korea's education system produces talent that scores at the world's highest levels on standardized tests.
But in Chapter 5 we saw Israel's Unit 8200. A structure in which intelligence veterans found startups, and military service functions as a technology-training incubator. Startups born from Defense Ministry programs grew 72 percent in 2023–2024. As of 2025, Israel has 23 active unicorns; Korea has 13 — a country with one-fifth of Korea's population produces nearly twice as many unicorns. Korea lacks a pipeline from military service to technology to entrepreneurship. Military duty is perceived as a gap in a career, not as a technical launching pad.
Brain drain is another deduction. A significant share of top graduates from KAIST, POSTECH, and Seoul National University head to graduate programs at Stanford, MIT, and Carnegie Mellon. Finding Korean researchers at Google, Meta, and NVIDIA's AI labs is not difficult. The problem is that they do not return. Israel's talent gains experience in Silicon Valley and then returns to Tel Aviv to found startups — the network formed by Unit 8200 provides the pull to come home. Korea lacks that return mechanism. Samsung and SK's internal research centers offer world-class compensation, but that is a story inside the chaebol. Outside the chaebol, the reasons to return have not been created.
The duality of the chaebol system is also reflected in the score. The four major chaebols — Samsung, SK, Hyundai, LG — account for 40.8 percent of GDP. That concentration enables internal technology-training systems and sustains R&D investment at global scale. That is the distinctively Korean structure we saw in Chapter 13. But the ecosystem outside the chaebol is impoverished. The path for a promising startup to grow into a global company independently is weak; dependence on acquisition by or partnership with a large company is high. The kind of global mega-exit — like Israel's Wiz being acquired by Google for $32 billion — has not yet appeared in Korea.
The drying pipe is the heaviest deduction in this score. Total fertility rate 0.72 — the lowest in the world. In the fourth quarter of 2023 it fell to 0.65; Seoul's rate is 0.64. The lowest figures recorded in peacetime in human history. Compared with the seven countries: Israel at 2.9, Indonesia at 2.1 — a gap of three to four times. Singapore is also low at 0.97, but compensates for population decline through immigration — foreigners make up 30 percent of its population. Korea's foreign-resident proportion is just 4.9 percent (2023).
Looking further into the numbers, the situation is more severe. In 2025, Korea enters super-aged society, with more than 20 percent of the population aged 65 or over. Japan took 12 years to transition from aged to super-aged society; Germany took 36. Korea is taking seven years. The old-age dependency ratio in 2060 is projected to exceed 80 elderly per 100 working-age people — the highest level in the OECD. The absolute number of 20-something engineers available to work in semiconductor cleanrooms is shrinking — not an abstract threat but a physical reality.
In Chapter 12 we saw the human meaning of these numbers. The paradox of a PISA-world-3rd-ranked education system producing the world's lowest fertility rate. Private education spending accounts for 3.2 percent of household income; the private education market totals ₩26 trillion — roughly $19 billion — with a participation rate of 78 percent. Educational overheating produces avoidance of childbearing, and avoidance of childbearing produces the physical contraction of the talent pipeline. In historical cases, there is no direct precedent of indispensability collapsing due to population decline — because a population collapse of this magnitude is itself unprecedented. Korea is entering uncharted territory.
But the fact that demographics cannot be fixed by law does not mean the talent pipeline cannot be fixed. Just as Singapore compensates for population decline with a foreign-resident proportion of 30 percent, opening the pipeline's borders is more realistic than reversing the fertility rate. In Chapter 12 we saw that path — an accelerated visa for high-skilled talent, long-term residence pathways in manufacturing and care sectors, settlement-type visas for depopulating regions. Raising the foreign-resident proportion from its current 4.9 percent to 8–10 percent by 2035 is the fastest mechanism for compensating the drying pipe from outside. If this correction operates, the downside risk to the talent score is reduced — it can prevent the 3.5 from sliding to 3.
5. The Corridor Between Two Great Powers
In security-economy balance management, Korea scores 3 points.
Among the seven countries, Korea alone carries the structural impossibility of scoring a perfect 5 on this condition. In a structure where the security ally and the largest trading partner are adversaries of each other, 5 points is theoretically impossible. A score of 3 is close to the realistic ceiling within those constraints.
To understand this score, one must begin with the Prologue. In October 2022, when the United States activated advanced semiconductor export controls, the fate of Samsung's Xi'an NAND factory and SK hynix's Wuxi DRAM factory became uncertain. Roughly 40 percent of Samsung's total NAND production and roughly 40 percent of SK hynix's DRAM production were located in China. Korea's core production assets were sitting in precisely the place the United States had said "do not touch." A 120-day grace period, a one-year extension, and then a shift to an annual licensing system — there is no "permanent exemption." As we saw in Chapter 11, that tension has not been resolved — it has become an annual event.
The Korea-U.S. alliance is the pillar of Korea's security. 28,500 American troops stationed in Korea, the 1953 Mutual Defense Treaty, the nuclear umbrella. Defense spending is 2.6 percent of GDP — roughly $47.6 billion, roughly ₩63 trillion. But the largest trading partner is China — $124.8 billion in exports, 19.5 percent of the total; imports accounting for 22.1 percent. The economic retaliation China imposed when the THAAD system was deployed in 2017 demonstrated the vulnerability of this structure in material terms. Lotte's exit from China, estimated damage to tourism and the Korean Wave of roughly $5 billion — roughly ₩6.6 trillion. Distribution of K-content inside China was blocked. The wounds from that episode, which we saw in Chapter 14, remain.
In Chapter 3 we saw Singapore provide a port to the U.S. Navy while building an industrial park in China. In Chapter 2 we saw Finland maintain democracy internally under the friendship treaty with the Soviet Union. Both countries achieved a balance that delivered value to both sides without being fully absorbed by either. Korea occupies a harder position than both. Singapore is separated from China by sea. Finland's Soviet trade share was 25 percent — but contained within the buffer of a bilateral barter structure. Korea is the only country with both security dependence and economic dependence simultaneously at extreme levels — with the added variable of North Korea's military threat layered on top.
The North Korean variable is a burden unique to Korea among all seven countries. Nuclear weapons sit 40 kilometers north of the demilitarized zone. That is a structural discount factor for foreign investment, and the reason Korea must spend 2.6 percent of GDP on defense — a level that exceeds the NATO standard. Yet paradoxically, this threat has become the foundation of arms export capability. Military technology accumulated over 70 years of continuous war readiness carries competitive value in global markets on a price-to-performance basis. The 2022 mega-contract to export 1,000 K2 tanks, 672 K9 self-propelled howitzers, and 48 FA-50 light combat aircraft to Poland is the yield of that accumulation.
As decoupling accelerates, the corridor narrows. From the American side: demands to restrict HBM exports to China, pressure to join Chip 4, the IRA's FEOC provision. From the Chinese side: the precedent of the THAAD retaliation, the rare-earth export control card, and CXMT's drive to substitute Korean memory. Korea stands in a position where it must exercise leverage toward both the United States and China, while simultaneously receiving leverage from both.
Arms exports are opening new passages in this corridor. In 2022, at roughly $17.3 billion — roughly ₩23 trillion — Korea ranked ninth in the world. Following the Polish mega-contract, export destinations are expanding to Australia, Saudi Arabia, and Romania. The strategic significance of arms exports extends beyond revenue. A country that introduces Korean weapons becomes dependent on Korea for long-term maintenance, components, and follow-on upgrades — a kind of security supply-chain lock-in effect. This is the military version of what Singapore achieved through its trading-hub function in Chapter 3. But the expansion of arms exports simultaneously deepens security ties with the Western camp and raises China's vigilance. This, too, is a double-edged sword.
The sixth condition — cybersecurity — was proposed in Chapter 5 through the Israeli case. The provisional score is 1.5 points. The lowest of the five conditions, and the only one below 2.
This score can be broken down along three axes. First, defensive infrastructure scores at roughly 2 points. The world's highest internet penetration rate at 99.9 percent, the world's first commercial 5G network — on top of that, the National Intelligence Service Cyber Security Center and KISA operate the defensive network. North Korea's Lazarus Group stole more than an estimated $600 million — roughly ₩790 billion — in cryptocurrency-related hacking in 2023 alone, and cyberattacks against Korean government and defense industry targets number in the thousands per year. The defensive experience accumulated from daily exposure to the world's most aggressive threat is battle-tested — but whether that experience has been systematized and converted into an exportable form is a different matter.
Second, industrial conversion scores at 1 point. In Chapter 5 we saw Israel convert cybersecurity into an economic asset through a startup ecosystem staffed by Unit 8200 veterans. Second in the world in cybersecurity exports. Wiz, Check Point, CyberArk — cases where threat became industry. Korea is failing at that conversion. The Galapagos effect we saw in Chapter 5 operates here — the network separation regulation maintained for 21 years and the public certificate system have isolated the domestic security market from global standards. Wiz not being able to enter Korea and Korean security companies not being able to go global are two sides of the same regulation.
Third, the institutional environment scores at 1.5 points. In Chapter 9 we saw the UAE's G42 sever its relationship with Huawei and align with Microsoft — an era in which which camp's infrastructure a nation rides on determines its strategic position. Korea's cyber institutions do not match the speed of that era. Blanket rule-based network separation rather than risk-based approaches, the absence of public-private threat-intelligence sharing, and a military-secrets protection law that blocks technology transfer. The reverse formula proposed in Chapter 5 applies again here — in the cyber domain, institutional redesign must precede technological innovation.
A score of 1.5 is lower than any of the five conditions. But fertility rates cannot be fixed by law, and HBM dominance cannot be built without decades of accumulation. Cybersecurity is different — this is a score that can be raised by changing institutions and designing a talent pipeline. Holding a shield while unable to grip a sword. This is a risk — and an unexplored opportunity.
6. A Country That Belongs Nowhere
16.5 points. 66 percent of a perfect score of 25. How should this number be read?
One approach is to compare Korea with the archetypes of the seven countries. Taiwan, Israel, Singapore, and the Netherlands are Type A — the small-nation technology-and-finance model. Small nations of 5 to 24 million people that have secured global monopolies in specific technology domains and use those monopolies as shields for survival. Indonesia is Type B — the large-nation resource-and-population model. Its leverage is 55–60 percent of global nickel production and 280 million people. The UAE is Type C — the pivot type, buying its future with sovereign capital. A $1.7 trillion sovereign wealth fund purchases technology stakes.
Korea is not Type A, not Type B, not Type C. It does not fit cleanly into any of them. This is what makes Korea difficult to analyze — and simultaneously what widens Korea's strategic space.
HBM 79 percent, DRAM 70.5 percent — this is a technology monopoly on the level of Taiwan and the Netherlands. Korea holds the core asset of Type A. Yet a population of 52 million, manufacturing's share of GDP at roughly 26 percent on a value-added basis (the highest in the OECD), exports of $683.8 billion, and GDP of $1.76 trillion ranked 13th in the world — too large to call a small nation, too small to call a great power. Korea partially holds the industrial scale of Type B.
Samsung's five-year ₩450 trillion — roughly $310 billion — investment plan, SK's AI infrastructure investment, Hanwha's defense-and-space portfolio — these are capital deployments at the level of national strategy, partially carrying the elements of Type C. The Korea Investment Corporation (KIC) manages roughly $200 billion in assets — one-eighth of the UAE's — but the global investment portfolio of the chaebol functions in practice as a quasi-sovereign wealth fund.
K-content exports of $13.1 billion and a Korean Wave fanbase of 225 million represent soft power that is an independent asset unique to Korea among the seven countries. This is a fourth dimension absent from Type A, Type B, and Type C alike. One estimate holds that a $10 billion increase in cultural exports produces a $18 billion chain increase in consumer-goods exports. A structure in which soft power expands the market for hard power — an Indonesian consumer in Jakarta who watches Korean dramas on Netflix and then buys a Hyundai is not a coincidence.
This hybrid character is a double-edged sword. Not specializing in any single category means there is no single lever of the kind that can be said unequivocally — "the world needs us for this one thing" — the way Taiwan's TSMC or the Netherlands' ASML can. Even Korea's strongest leverage, HBM, is narrower in domain than TSMC's foundry monopoly or ASML's EUV monopoly.
But at the same time, holding multiple levers means that if one weakens, another can compensate. Even if HBM dominance is eroded by China's CXMT, all-solid-state batteries after 2027 could create a new technology monopoly — in Chapter 11 we saw Samsung SDI's energy density target of 900 watt-hours per liter. Even if semiconductor exports are constrained by U.S.-China regulation, K-content at $13.1 billion — roughly ₩17.5 trillion — operates as a soft-power buffer — the way the Korean Wave fandom in Southeast Asia and the Middle East formed informal resistance to their own governments' anti-Korea policies during the THAAD crisis. Samsung Biologics' CDMO growth is forming a new supply-chain node in biotech.
History confirms this principle. Switzerland's 210 years of survival was not due to finance alone. As we saw in Chapter 2, when banking secrecy was eroding, there was the pharmaceutical industry — Novartis and Roche — and the 40 international organizations of Geneva. A portfolio of multiple levers provided resilience against single shocks.
Three scenarios lie before Korea. The most probable is managed competition — a 60–70 percent probability. The current trajectory in which the United States and China compete in technology and military affairs while avoiding complete separation in finance and commerce is maintained. In this scenario, Korea sustains a dual technology track — advanced HBM for the American side and legacy memory for the Chinese side — while accelerating export diversification toward ASEAN and the Middle East. ASEAN exports have already reached 17.8 percent of Korea's total, marking the first steps in reducing China-dependence concentration. Exports to Vietnam alone stand at $54.3 billion — nearly twice Japan's $27.4 billion — the center of gravity on the export map is shifting.
The second is abrupt decoupling — a 15–25 percent probability. A scenario in which economic bloc formation accelerates following a Taiwan Strait crisis or direct U.S.-China clash. A rapid decline in Korea's China-directed exports of 19.5 percent of the total would imply a short-term GDP contraction of 5–8 percent. Structurally, however, Korea's standing as a "trustworthy supplier" within the Western supply chain could rise. In Chapter 10 we saw Japan attempting to reclaim semiconductor sovereignty through Rapidus. In an abrupt decoupling scenario, Korean semiconductors stand in a different position from Japan's — possessing something to protect rather than something to reclaim. Even in an extreme scenario where assets at Samsung's Xi'an plant and SK hynix's Wuxi-Dalian plants are frozen or confiscated, the domestic production base of the Yongin cluster and Icheon campus remains. It hurts — but it is not fatal.
The third is selective re-engagement — a 10–20 percent probability. A scenario in which the United States and China resume cooperation in domains such as climate change or pandemic response. In this case, the indispensability premium may weaken. When tension eases, the value of "being needed by both sides" diminishes. If China gains direct access to U.S. markets, the value Korea holds as an intermediary routing around restrictions is reduced. But even in this scenario, the technologically irreplaceable position of HBM and batteries is maintained — geopolitical relaxation does not change the laws of physics.
This scorecard is not a fixed verdict. A change in one condition moves the total, and when the total moves, the doors of scenarios open or close. The condition with the greatest leverage is institutional flexibility — currently at 2.5. If MSCI developed-market index inclusion passes the 2025 review, the high-skilled foreign talent visa is simplified to the level of Singapore's Pass Plus, and the sandbox-to-full-regulation conversion rate rises from 30 percent to 60 percent — these three changes alone would push the institutional score close to 3. The total moves from 16.5 to 17, and the possibility of Korea being incorporated as a core partner in the U.S.-led supply chain under the managed competition scenario opens. A difference of 0.5 points lowers the threshold of scenario B. Conversely, the talent reproducibility score of 3.5 is the most fragile condition — capable of sliding below 3 within the next ten years if the fertility rate falls further or brain drain accelerates — and it is a domain where policy does not take immediate effect. Demographics cannot be fixed by law.
Paradoxically, intensification of U.S.-China competition opens larger strategic space for Korea — because neither side can afford to make Korea an enemy.
In Chapter 2 we saw Switzerland strengthen its indispensability during the Napoleonic Wars and two World Wars — precisely in those vast collisions. There is a dynamic in which as great-power confrontation deepens, the value of the intermediary rises. Korea is now at the center of that dynamic.
But the probabilities of these three scenarios are not fixed. Korea's choices move the probabilities. Institutional innovation and accelerated export diversification strengthen Korea's position in the managed competition scenario. Expanding arms exports and industrializing cybersecurity reduces the impact in the decoupling scenario. Korea can be not a passive recipient of scenarios but an active designer of conditions.
Across all three scenarios, Korea's task is the same. Maintaining the technology monopoly while reforming institutions and rebuilding the talent pipeline. The missing 8.5 points from the 16.5 total came mostly from institutions (2.5 points), security balance (3 points), and talent (3.5 points) — the deductions are concentrated not in technology but in the domains surrounding technology.
This is paradoxically a ground for hope. Technology monopolies are easy to lose and hard to create. Institutions and talent are the reverse — given political will and policy design, they can be improved faster than a technology monopoly can be built. Reforming short-selling rules for MSCI developed-market inclusion, simplifying visas for high-skilled foreign talent, improving the rate of sandbox-to-full-regulation conversion, shareholder return policies to reduce the Korea Discount — these are domains where the score can be raised at lower cost than building a semiconductor fab.
Remember what Lee Kuan Yew proved on the island expelled from Malaysia. Indispensability is not given — it is designed. Singapore built an Indispensable Node in a place without resources, without population, without even the will toward independence. Korea has far more than that — HBM, batteries, CDMO, K-content, the world's third-ranked education system by PISA scores, and second-in-the-OECD R&D investment relative to GDP. What is lacking is not assets — it is design.
To belong nowhere is to be free to go anywhere.
That is the real message of 16.5. Not a crisis because it falls short of perfect. An opportunity because the domains where the score can be raised are visible. Protecting the 4 points in technology while lifting the 2.5 in institutions to 3.5 and the 3 in security balance to 4. That is the equation Korea's next ten years must solve.
In Chapter 10 we traced Japan's trajectory. A country that held 51 percent of the global semiconductor market in 1988 had fallen below 10 percent by 2019. It took another 30 years to recover, and with Rapidus as a national project, it has barely taken its first step. What is lost is lost quickly; what is reclaimed comes back slowly. Korea is now in a position to reinforce before the loss occurs. The 30 years of disappearance Japan endured need not be repeated by Korea — provided Korea moves now.
Yet the number 0.72 places a deadline on this equation. Korea's population, currently 51.73 million, will shrink to 45.6 million by 2050, and according to the median projection of Bayesian statistical models, to below 30 million by 2100 — 58 percent of today's figure. The median age rises from 45.2 in 2024 to 54.6 by 2040 — Korea will be the oldest country in the world.
All indispensable nations in history shared one thing: a human reproduction system that produced, educated, and motivated the next generation. Venice's 500 years, Switzerland's 210, Singapore's 60 — at the base of each was an intergenerational human continuity. The Hanseatic League's 260 years, the Dutch Golden Age's 120 — ultimately, at the base of it all were people. When that continuity is broken, the scorecard itself becomes meaningless.
Return to the whiteboard in the conference room. Five lines of conditions and the sixth — cyber. Circles and X-marks and triangles and red underlines.
The numbers on that board show simultaneously what Korea has and what it does not have. What it has must be defended. What it does not have must be built. Time is not unlimited.
The next chapter asks the precise length of that deadline.