TW-F-04: The Cross-Strait Military Balance, Asymmetric Defence, and the Overall Defense Concept (2017–2026)

1. Key Takeaways

• The 2017 appointment of Admiral Lee Hsi-min (李喜明) as Chief of the General Staff (參謀總長, Canmou Zongzhang) of the Republic of China Armed Forces (中華民國國軍, Zhonghua Minguo Guojun) and the corresponding promulgation of the Overall Defense Concept (整體防衛構想, Zhengti Fangwei Goouxiang, "ODC") constitute the most-substantive post-democratisation revision of Taiwan's defence doctrine and the foundational starting point for the post-2017 asymmetric-defence trajectory. Admiral Lee, appointed Chief of the General Staff under Defence Minister Feng Shih-kuan (馮世寬) and then Yen Teh-fa (嚴德發) in the Tsai Ing-wen administration's first term, articulated the ODC as a three-stage operational concept built around: (a) force preservation (戰力保存) — dispersing and concealing forces during the PLA's initial precision-strike phase; (b) littoral combat (濱海決勝) — engaging the PLA invasion-fleet at the Taiwan Strait littoral zone with large numbers of small, mobile, lethal anti-ship platforms; (c) coastal annihilation (灘岸殲敵) — destroying PLA forces that reach Taiwan's coast through dispersed coastal-defence and territorial-defence elements. The ODC's signature shift from the legacy doctrine was the rejection of expensive prestige-platform procurement (heavy frigates, additional fighter squadrons of the most advanced types, conventional submarines optimised for blue-water operations) in favour of large numbers of mobile, survivable, lethal asymmetric platforms — sea-mines, coastal-defence cruise missiles, man-portable air defence (MANPADS), shoulder-fired anti-tank weapons, small fast-attack craft, and air-defence-suppression munitions. The ODC framing has been the canonical reference point for US-Taiwan defence-procurement conversations from 2017 onwards, even where actual ROC MND procurement decisions have diverged from the strict ODC prescription.

• Admiral Lee Hsi-min's June 2019 retirement and the subsequent partial post-Lee reversion toward traditional-platform procurement under his successors as Chief of the General Staff and as MND Ministers (Yen Teh-fa, Chiu Kuo-cheng 邱國正, and from May 2024 Wellington Koo 顧立雄) produced a hybrid Taiwan-defence-procurement posture that mixes asymmetric and traditional elements rather than fully implementing the ODC's pure-asymmetric prescription. The post-Lee MND has continued to procure asymmetric platforms — Stinger MANPADS (post-2019 packages), HIMARS (October 2020 package, 11 launchers plus rockets), the AGM-84L Block II Harpoon Coastal Defence System (December 2022 package, 400 missiles plus 100 launchers, US$2.37 billion), Volcano anti-tank-mine systems, and the NCSIST-indigenous Hsiung Feng II/III anti-ship cruise missiles — but also continued procurement of traditional prestige platforms: the F-16V Block 70/72 (August 2019 package, 66 aircraft, US$8 billion), the M1A2T Abrams main battle tank (108 tanks, US$2 billion), and the indigenous Hai Kun (海鯤) submarine programme. The post-Lee posture, alongside Lee's continued post-retirement advocacy through Brookings lectures, War on the Rocks essays, and his 2022 book Storm Coming, has been the subject of sustained debate among Taiwan-defence analysts and US strategic-community commentators including Michael Hunzeker, Alexander Lanoszka, Eric Lee, and Tanner Greer over the degree to which Taiwan has actually internalised the ODC's prescriptive force-design implications.

• The People's Liberation Army (PLA, 中国人民解放军) modernisation trajectory across the 2017–2026 period has substantively widened the cross-Strait military balance against Taiwan, with the PLA acquiring force-projection, joint-operations, precision-strike, and amphibious-assault capabilities that the ROC Armed Forces cannot match in symmetric force-on-force terms. The PLA Navy (PLAN) reached approximately 370+ surface combatants by 2024 [TBD-VERIFY exact 2024 PLAN battle-force composition — US DoD-published estimates ranged from 340 to 400 across the 2022–2024 China Military Power Reports] including three aircraft carriers (the Liaoning, the Shandong, and the Fujian, the latter undergoing sea trials from May 2024), 50+ destroyers, and the Type-075 amphibious-assault helicopter-dock vessels; the PLA Rocket Force (PLARF) has deployed the DF-17 hypersonic glide vehicle, the DF-26 intermediate-range ballistic missile, and the DF-21D anti-ship ballistic missile across thousands of launcher-equivalents; the PLA Air Force (PLAAF) operates approximately 200+ Chengdu J-20 fifth-generation fighters and large numbers of Shenyang J-16 and Chengdu J-10C fourth-generation-plus fighters [TBD-VERIFY exact J-20 inventory — published 2024 estimates range from approximately 200 to 280]. The 2015–2016 PLA reorganisation establishing the five Theatre Commands (with the Eastern Theatre Command, 东部战区, having Taiwan as its primary contingency focus) and the post-2017 Joint Operations Command Centre architecture have produced a PLA force-employment capacity for Taiwan-contingency operations qualitatively distinct from the pre-2015 PLA.

• The PLA's four-iteration Joint Sword (联合利剑, Lianhe Lijian) exercise series — Joint Sword August 2022 post-Pelosi-visit response; Joint Sword April 2023 post-Tsai-McCarthy-transit-meeting response; Joint Sword-2024A May 2024 post-Lai-inaugural response; Joint Sword-2024B October 2024 post-Lai-National-Day-address response — and the 2025 Strait Thunder (海峡雷霆) successor series have operationalised PLA grey-zone-and-coercion-pressure architecture and established the new post-2022 cross-Strait military-pressure baseline. The August 2022 exercise — held 4–10 August 2022 in response to the 2 August 2022 House Speaker Nancy Pelosi visit to Taipei (treated in TW-C-06) — crossed multiple thresholds previously observed: ballistic-missile overflight of Taiwan proper (5 August 2022 DF-15 trajectories transiting over Taipei airspace at high altitude); naval-and-air-force operations encircling Taiwan in six designated exercise zones around the island; the de facto suspension of the Taiwan Strait median line. The April 2023 exercise (8–10 April 2023) was a smaller-scale response to the 5 April 2023 Tsai-McCarthy transit meeting in California. The May 2024 Joint Sword-2024A (23–24 May 2024) was the post-Lai-inaugural response, naming the exercise Joint Sword-2024A with the explicit implication of a successor 2024B iteration; the October 2024 Joint Sword-2024B (14 October 2024) responded to the 10 October 2024 Lai National Day address. The 2025 Strait Thunder-2025A (1 April 2025 [TBD-VERIFY exact dates]) and subsequent iterations sustained the post-2022 pressure pattern.

• The Davidson Window — the canonical formulation from US Indo-Pacific Command Commander Admiral Philip Davidson's 9 March 2021 Senate Armed Services Committee testimony stating that "the threat is manifest during this decade, in fact, in the next six years" — has framed US strategic-planning conversations on a Taiwan contingency through the 2021–2027 window, even where subsequent US Defense Department, intelligence-community, and academic analyses have substantively disputed the 2027 timeline. The Davidson testimony's "next six years" formulation produced the 2027-deadline reference framework that has shaped US war-planning, force-posture decisions (the 2023 EDCA-expansion agreement with the Philippines, the 2024 trilateral Japan-Korea-US summit defence-coordination architecture, the 2024 AUKUS Pillar Two technology-cooperation tracks), and Taiwan-defence procurement-prioritisation conversations. CIA Director William Burns in February 2023 affirmed that Xi Jinping had directed the PLA to be "ready by 2027" to conduct a successful invasion of Taiwan; subsequent analytical work by Bonny Lin, Ryan Hass, Richard Bush, and others has emphasised that "ready" does not entail "will invade" and that the strategic calculus around an actual invasion decision involves political, economic, and operational factors well beyond the binary readiness assessment. The Davidson Window debate has not been resolved at the analytical level; its operative effect on US-Taiwan policy has been substantial regardless of its analytical merits.

• The ROC defence-budget trajectory from NT$327 billion (FY2017) through NT$606 billion (FY2024) and NT$683 billion (FY2025) to the proposed NT$949 billion (FY2026) represents a near-tripling in nominal terms across the nine-year Tsai-Lai sequence and the most-aggressive sustained defence-spending expansion in the post-democratisation era. The FY2017 defence budget of approximately NT$327 billion (~2.0 per cent of GDP) reflected the legacy late-Ma-era and early-Tsai-era spending baseline; the FY2024 budget of approximately NT$606 billion (~2.5 per cent of GDP) reflected the Tsai second-term expansion incorporating special-budget supplementary allocations for the F-16V Block 70 programme and the indigenous-submarine programme; the FY2025 budget of approximately NT$683 billion (~2.5–2.6 per cent of GDP) was the first Lai-era budget cycle [TBD-VERIFY: the budget passage was affected by the January 2025 KMT-TPP freeze-and-cut sequence treated in TW-D-04, with disputed published reduction figures ranging across NT$197.5 billion, NT$207.5 billion, and NT$216 billion aggregating frozen-versus-cut items]; the proposed NT$949 billion (~3.3 per cent of GDP) FY2026 budget submitted by the Executive Yuan in late 2025 represents the response to the Trump-2 "five per cent of GDP" demand and the post-2025 strategic-pressure environment. The FY2026 budget submission's actual LY passage trajectory through May 2026 will be shaped by the post-recall LY arithmetic (treated in TW-D-05).

• The US Foreign Military Sales (FMS) pipeline to Taiwan across 2017–2025 totalled approximately US$26 billion across more than thirty separate notification packages [TBD-VERIFY exact aggregate figure — published Congressional Research Service and Defense Security Cooperation Agency totals vary depending on whether DCS notifications are included] and constitutes the structural backbone of Taiwan's post-2017 defence-modernisation programme. Key FMS packages included: the August 2019 F-16V Block 70/72 package (66 aircraft, US$8 billion, delivery scheduled through 2026); the December 2020 MQ-9B SeaGuardian package (4 aircraft); the October 2020 HIMARS, SLAM-ER, and MS-110 reconnaissance-pod combined package (11 HIMARS launchers, 64 ATACMS, 135 SLAM-ER); the December 2022 AGM-84L Block II Harpoon Coastal Defence System package (400 missiles, 100 launchers, US$2.37 billion); the multiple Stinger MANPADS packages 2019–2024; the M1A2T Abrams package (108 tanks, US$2 billion); the 2023 NASAMS package and Volcano anti-tank-mine package; multiple ammunition-and-spare-parts packages including the 2023 announcement of accelerated munitions stockpiling under Presidential Drawdown Authority [TBD-VERIFY: the Presidential Drawdown Authority usage for Taiwan-bound transfers from 2023 onwards constituted a procedurally novel mechanism distinct from FMS, drawing weapons from US military stockpiles for direct transfer]; post-2024 announced packages including the post-March 2025 Wellington Koo–US-DoD packages [TBD-VERIFY exact post-March 2025 announced packages].

• The indigenous defence industry architecture — anchored by the National Chung-Shan Institute of Science and Technology (NCSIST, 國家中山科學研究院, Zhongshan Kexue Yanjiuyuan), CSBC Corporation Taiwan (台灣國際造船) for the indigenous-submarine programme, the Aerospace Industrial Development Corporation (AIDC, 漢翔航空工業) for the Indigenous Defense Fighter (IDF) and Brave Eagle (勇鷹) advanced jet trainer programmes, and a constellation of smaller indigenous suppliers — has been expanded across the Tsai-Lai sequence and constitutes the resilience-and-self-sufficiency complement to the US-FMS-dependent pipeline. NCSIST's key 2017–2026 programmes include: the Hsiung Feng II (雄風二型) subsonic sea-skimming anti-ship cruise missile in continued production; the Hsiung Feng III (雄風三型) supersonic ramjet anti-ship missile reaching mass deployment from approximately 2018 onwards (the "carrier killer" framing); the Hsiung Feng IIE (雄風二E) land-attack cruise missile; the Yun Feng (雲峰) extended-range surface-to-surface missile programme [TBD-VERIFY exact 2024–2026 deployment status of the Yun Feng programme — open-source reporting indicates initial production from 2023 onwards but precise deployment counts remain classified]; the Tien Kung III (天弓三) and Tien Kung IV (天弓四) surface-to-air missile systems; the Sky Bow / Sky Sword (天弓 / 天劍) air-defence and air-to-air-missile architecture; and the post-2022 expanded military-grade unmanned-aerial-vehicle (UAV) and loitering-munition production. The CSBC indigenous-submarine Hai Kun (海鯤, Narwhal) was unveiled on 28 September 2023 with sea-trial commencement in 2024–2025 and projected first-of-class delivery to the ROC Navy by 2027 [TBD-VERIFY exact delivery schedule].

• The 2022 announcement and 2024 implementation of the extension of compulsory male military service from four months (the 2013–2018 transitional and 2018-onwards baseline) to twelve months for the post-2005-born cohort represents the most-substantial post-democratisation conscription policy reversal and reflects the post-2022 strategic-environment hardening. The post-2013 ROC conscription regime had progressively shortened the male compulsory-service term from the historical eighteen-to-twenty-four-month standard through staged reductions to the four-month "military training service" (軍事訓練役) standard implemented from 2018 onwards, alongside the transition to an all-volunteer professional force as the primary war-fighting establishment. The December 2022 Tsai Ing-wen announcement extending the compulsory-service term to twelve months for males born from 1 January 2005 onwards — first cohort entering service in 2024 — represented the formal reversal. The implementation has covered: a substantially restructured training curriculum incorporating Stinger MANPADS familiarisation, anti-tank weapon familiarisation (Javelin and shoulder-fired anti-armour), combat-medic training, and urban-warfare basics; an increased monthly stipend from approximately NT$6,500 (four-month service) to approximately NT$20,320 (twelve-month service); and a planned post-conscript-service reservist-cycle integration. The MND-reported 2024 first-cohort intake [TBD-VERIFY exact 2024 first-cohort numbers] is being tracked through 2025–2026.

• The 26 June 2024 Lai Ching-te launch of the Whole-of-Society Defence Resilience Committee (全社會防衛韌性委員會, Quan Shehui Fangwei Renxing Weiyuanhui) at the Office of the President — the institutional centrepiece of the Lai administration's defence posture — operationalised the post-2022 civil-defence-and-resilience-architecture turn and integrated civil society, military, and inter-ministerial coordination on a quarterly-meeting cadence. The Committee — chaired by President Lai with Premier Cho Jung-tai (卓榮泰), MND Minister Wellington Koo, National Security Council Secretary-General Joseph Wu, Mainland Affairs Council Minister Chiu Tai-san (邱垂正), and selected civil-society partner-organisation representatives — has covered five strategic pillars: (a) civil-defence training and volunteer-reservist mobilisation incorporating Kuma Academy (黑熊學院, Heixiong Xueyuan, founded by legislator Puma Shen 沈伯洋 and Lin Bin-jian) and Forward Alliance (壯闊台灣聯盟, founded by Enoch Wu 吳怡農); (b) strategic-stockpile and supply-chain resilience including petroleum, grain, and pharmaceutical reserves; (c) energy and critical-infrastructure resilience; (d) financial-and-economic resilience; (e) information-warfare and cognitive-warfare resilience. The Committee's institutional integration of civil-society civil-defence-training providers represents the formal recognition of the post-2014-Sunflower-era civil-society defence-mobilisation institutional architecture as a national-resilience asset.

• The post-2022 PLA grey-zone-pressure architecture — incorporating ADIZ incursions, naval-vessel encirclement, subsea-cable cuts, cyber operations, and information-warfare campaigns — constitutes the operational baseline of cross-Strait military-pressure short of kinetic conflict and the most-sustained day-to-day military challenge facing the ROC Armed Forces. The ROC MND-published PLA-aircraft ADIZ daily tracker recorded approximately 1,737 individual PLA aircraft intrusions into Taiwan's ADIZ in calendar-year 2022, approximately 1,711 in 2023, approximately 3,070 in 2024 [TBD-VERIFY exact 2024 ADIZ-incursion total — published MND-tracker aggregations range from approximately 2,900 to 3,100 depending on counting methodology around dual-zone incursions], and a sustained 2025 pace [TBD-VERIFY 2025 running totals]. The subsea-cable cut record — the February 2023 Matsu-Taiwan cable severance (attributed to a PRC-flagged fishing vessel anchor-drag, with sustained Matsu connectivity disruption for approximately 50 days), the January–February 2025 Pratas-and-Taishi cable incidents (the post-25 February 2025 Hongtai 58 vessel detention near the Penghu Channel), and subsequent post-March 2025 incidents — has produced the dedicated subsea-cable-protection programme and the post-2024 satellite-internet redundancy architecture incorporating Eutelsat OneWeb and other non-Starlink providers given the Starlink-Taiwan availability question. The information-warfare and cognitive-warfare pressure — covering disinformation operations targeting Taiwan electoral and policy processes, PRC-affinity media and entertainment-industry penetration, and the 2024 Bluebird Movement and 2025 Great Recall mobilisation (treated in TW-D-04) — operates within the broader cross-Strait grey-zone architecture.

• The post-January 2025 Trump-2 administration "five per cent of GDP" defence-spending demand and the associated friction over Taiwan's strategic positioning — including the Trump February–March 2025 statements that "Taiwan stole our chip business" and that Taiwan should pay for US-provided defence — represents the most-substantial post-1979 friction in the US-Taiwan strategic-architecture and has restructured the Lai administration's defence-budget and policy calculus. The Trump-2 demands articulated across the post-January 2025 period have included: a numerical demand for Taiwan defence spending to reach 5 per cent of GDP (against the FY2025 ~2.6 per cent baseline and the proposed FY2026 ~3.3 per cent submission); a framing of Taiwan's semiconductor-industry centrality as a problem (the "stole our chip business" framing) rather than a strategic-alignment asset (the "silicon shield" framing operationalised under Tsai-Biden); ambiguous statements on the US defence commitment to Taiwan under Article 3 of the Taiwan Relations Act and the operative Six Assurances framework; and the linkage of defence-and-tariff tracks in the Trump-2 negotiating architecture (treated in TW-D-03). The Lai administration's response — including the proposed NT$949 billion FY2026 budget, the post-March 2025 Ten Measures operationalisation, the post-March 2025 TSMC US-investment announcements (treated in TW-D-03), and the sustained Wellington Koo–US-DoD engagement track — has aimed at preserving the operational defence-cooperation pipeline while managing the political friction.

2. Background: The ROC Armed Forces, the Post-1996 Doctrine Evolution, and the 2017 Inception of the Overall Defense Concept

The Republic of China Armed Forces (中華民國國軍) that confront the People's Liberation Army across the Taiwan Strait in the 2017–2026 period are the institutional descendant of the National Revolutionary Army (國民革命軍) of the 1920s Northern Expedition era and the Nationalist forces that fought the Second Sino-Japanese War and the Chinese Civil War. The 1949 retreat to Taiwan brought approximately 600,000 mainland-China-origin military personnel together with several million civilian refugees, and the subsequent 1950s build-up under the US Military Assistance Advisory Group (MAAG) and the 1954 Mutual Defense Treaty produced the post-1950s ROC Armed Forces as a US-aligned force structure oriented to deter a PLA cross-Strait amphibious assault. The 1958 Second Taiwan Strait Crisis (the Quemoy / Kinmen-Matsu artillery duel beginning 23 August 1958) was the post-1949 high-water mark of large-scale PLA-ROC kinetic engagement and concluded with the US-supplied AIM-9 Sidewinder-equipped ROC Air Force F-86 Sabres establishing air superiority over the Strait. The 1971 United Nations General Assembly Resolution 2758 transferring the China seat to the People's Republic and the 1979 US-PRC normalisation with the corresponding Taiwan Relations Act (TRA) and Six Assurances framework reshaped the diplomatic context within which the ROC Armed Forces operate, but the post-1979 US-arms-supply pipeline under the TRA preserved the structural basis of US-Taiwan defence cooperation.

The 1996 Third Taiwan Strait Crisis — when the PLA conducted ballistic-missile launches into waters near Taiwan in March 1996 ahead of the 23 March 1996 presidential election, and the US Clinton administration responded by dispatching the USS Nimitz and USS Independence carrier strike groups to the region — was the post-1979 inflection point for the ROC defence-doctrine evolution. The 1996 crisis exposed the vulnerability of Taiwan's pre-1996 deterrence posture to PLA missile coercion and established the rationale for the subsequent decade-and-a-half of ROC defence-modernisation procurement, including the 1992 F-16A/B sale (delivered 1997–2001), the Mirage 2000-5 acquisition from France, the Kidd-class destroyer acquisition from the United States (commissioned in ROC service 2005–2006), and the post-2001 Bush-administration approval of the diesel-electric submarine sale — which the Chen Shui-bian administration was unable to consummate due to a combination of US-side and ROC-LY-side budget-and-procurement frictions. The 2002 Chen administration Quadrennial Defense Review introduced the formal doctrinal shift from territorial-defence to "offshore engagement" (境外決戰, Jingwai Juezhan), positioning the ROC Armed Forces to engage PLA forces in the maritime approaches rather than awaiting them at the coast — a doctrinal framing that subsequent military analysts have characterised as ambitious-to-fanciful given the cross-Strait force-balance trajectory.

The 2008–2016 Ma Ying-jeou era saw the legacy doctrinal framework largely sustained alongside the Ma administration's cross-Strait political-economic-engagement track (the 2008–2010 ECFA negotiation-and-signing sequence, the 2015 Ma-Xi Singapore meeting; treated at TW-B-02, TW-B-03, TW-B-07). The Ma-era defence budget remained at approximately 2.0 per cent of GDP, with significant special-budget supplementary allocations for specific platform procurement (the post-2008 P-3C Orion maritime-patrol aircraft, the post-2010 F-16A/B Block 20 mid-life upgrade programme that became the foundation for the subsequent F-16V Block 70/72 build-out). The 2013–2018 transition from compulsory conscription to an all-volunteer force, initiated under the late-Ma administration and continued under the early-Tsai administration, structurally reshaped the ROC Armed Forces' personnel architecture by reducing the conscript-personnel intake from the historical eighteen-to-twenty-four-month standard through staged reductions to the four-month "military training service" (軍事訓練役) standard for males born from 1 January 1994 onwards. The transition's premise was that a smaller, professional, high-quality volunteer force would be operationally more effective than a larger conscript-based force; the practical 2018-onwards outcome was an ROC Armed Forces with a target active-duty strength of approximately 215,000 personnel (compared to the pre-2013 strength of approximately 275,000) and a reservist force whose readiness and training quality became a sustained policy concern through the late-2010s and early-2020s.

The 2016 Tsai Ing-wen administration's first-term defence-policy review concluded in the 2017 National Defense Report (released August 2017) and the corresponding 2017 Quadrennial Defense Review, both of which signalled the doctrinal pivot toward asymmetric defence under Admiral Lee Hsi-min's incoming Chief of the General Staff appointment. Admiral Lee — a career naval officer with prior commands including Commander of the ROC Navy (海軍司令, January 2015 – April 2017) — was appointed Chief of the General Staff in April 2017 by President Tsai under Defence Minister Feng Shih-kuan, succeeding Admiral Chiu Kuo-cheng (邱國正, the later 2021–2024 MND Minister). Lee's tenure as Chief of the General Staff from April 2017 to June 2019 was the formative period of the Overall Defense Concept (ODC) articulation. The ODC's intellectual lineage drew on the strategic thinking of the late 1990s and early 2000s "porcupine" or "hedgehog" framings developed by William Murray (US Naval War College), James Holmes, Toshi Yoshihara, and Andrew Krepinevich (Center for Strategic and Budgetary Assessments) — analysts who had argued through the late-Bush and early-Obama-era period that Taiwan should structure its defence around large numbers of mobile, survivable, lethal asymmetric platforms rather than fewer-and-more-expensive prestige platforms that would be vulnerable to PLA precision-strike preemption. Lee's contribution was to translate the external analytical framing into an ROC MND-internal operational concept with three formal stages: force preservation, littoral combat, and coastal annihilation.

The ODC's force-preservation stage envisaged ROC Armed Forces dispersing from peacetime garrison locations to dispersed, concealed, and hardened positions across Taiwan's mountainous interior, urban areas, and offshore islands during the initial PLA precision-strike phase (PLARF ballistic-and-cruise-missile salvoes targeting ROC airfields, command-and-control nodes, and major military installations). The doctrinal premise was that the ROC Air Force F-16, Mirage 2000-5, and IDF Ching-kuo fighter fleet — based at a handful of major airbases (Hualien, Taitung, Chiayi, Hsinchu, Taoyuan, Tainan, Pingtung) — could not survive a sustained PLARF saturation strike unless dispersed to highway-strip alternative basing, hardened-shelter facilities, and rapid-redeployment cycles; the ROC Navy surface-combatant fleet could not survive at port and required dispersal-and-concealment among Taiwan's many small fishing harbours; the ROC Army's mechanised and armoured units required dispersal-and-camouflage rather than concentrated garrison positioning. The littoral-combat stage envisaged engaging the PLA amphibious-assault fleet at the Taiwan Strait littoral zone with large numbers of small fast-attack craft (the Kuang Hua VI / 光華六號 missile boats; the Tuo Chiang-class 沱江級 stealth corvettes from the 2014 first-of-class delivery); Hsiung Feng II / III anti-ship cruise missiles fired from land-based mobile launchers; sea-mines deployed from minelayers and aircraft; and dispersed coastal-defence batteries. The coastal-annihilation stage envisaged dispersed coastal-defence and territorial-defence forces — including reservists, Volunteer Reserve units, and civil-defence elements — engaging PLA forces that reached Taiwan's coast.

3. The Lee Hsi-min ODC and Its Post-2019 Trajectory

The ODC's signature procurement-policy implication was the rejection of expensive prestige-platform procurement in favour of large numbers of asymmetric platforms. This implication produced sustained internal-MND debate across 2017–2019 and beyond, because the ROC Armed Forces' established procurement-pipeline included multiple expensive prestige-platform programmes that the ODC framework called into question. The F-16V Block 70/72 programme — formally announced as a US sale to Taiwan on 20 August 2019 (66 aircraft at approximately US$8 billion, with deliveries scheduled to commence in 2023 and complete by 2026) — represented exactly the kind of high-end fighter-aircraft acquisition that the ODC framework would deprioritise relative to dispersed anti-aircraft capability (Stinger MANPADS, NASAMS, Patriot PAC-3). The indigenous-submarine Hai Kun programme — formally launched in 2017 under the CSBC Corporation Taiwan in partnership with overseas suppliers — represented a substantial multi-decade investment in a small (target eight-submarine) fleet that would have limited operational survivability and impact in a Strait-contingency scenario. The M1A2T Abrams main battle tank programme — formally announced as a US sale on 8 July 2019 (108 tanks at approximately US$2 billion) — represented an investment in a heavy-armour capability that under ODC reasoning would be less useful than equivalent investment in dispersed anti-armour weapons.

Admiral Lee's response to this implementation tension was to argue that the prestige-platform pipeline could not be cancelled retroactively given the political-and-bureaucratic-commitment costs but that the marginal-procurement dollar should be redirected toward asymmetric platforms — and that the ROC MND should formally adopt the ODC as the doctrinal framework that would shape subsequent procurement decisions. In practice, the Lee-era 2017–2019 procurement decisions reflected the hybrid posture: the F-16V Block 70/72, M1A2T, and Hai Kun programmes proceeded; the AGM-84 Harpoon Coastal Defence System, Stinger MANPADS, HIMARS, MS-110 reconnaissance, and SLAM-ER programmes were initiated or accelerated; the NCSIST Hsiung Feng II/III production was sustained at higher rates. The Lee June 2019 retirement (succeeded as Chief of the General Staff by Admiral Huang Shu-kuang 黃曙光 [TBD-VERIFY exact successor as Chief of the General Staff at June 2019 and subsequent rotation], with Yen Teh-fa continuing as Defence Minister through May 2020 and succeeded by Chiu Kuo-cheng from February 2021) saw the post-Lee MND continue the hybrid procurement posture without formal repudiation of the ODC framework but without consistent prioritisation of asymmetric platforms over prestige platforms in budget-line terms.

The post-Lee period also saw the Admiral's continued external advocacy for the ODC framework through Brookings Institution lectures (notably the September 2022 lecture), War on the Rocks essays, the 2020 Prospects & Perspectives paper co-authored with Eric Lee, and the 2022 book Storm Coming: My Vision for Taiwan's Asymmetric Defence (the English-language working title; the Chinese-language original was published in Taiwan in 2022 under a different title [TBD-VERIFY exact Chinese-language title — the Mandarin title rendering varies across published sources]). Lee's post-retirement advocacy has been read by US strategic-community commentators as both an authoritative source on the ODC's intellectual content and a critique of the post-Lee MND's incomplete implementation. The sustained debate on the asymmetric-versus-traditional-platform trade-off has continued through the post-2020 period, with US-side analysts including Michael Hunzeker (George Mason University Schar School), Alexander Lanoszka (University of Waterloo), Eric Lee (Project 2049 Institute), Tanner Greer (Scholar's Stage), Mark Cancian (CSIS), and Bradley Bowman (FDD) writing across War on the Rocks, The Diplomat, Foreign Policy, and CSIS commentary on the prescriptive force-design implications.

The Wellington Koo MND-Minister appointment from May 2024 — the first civilian MND head in the post-democratisation era apart from interim arrangements [TBD-VERIFY: prior civilian MND heads in the ROC post-democratisation era include Cai Ming-hsien 蔡明憲 in the late-Chen administration period and others over discontinuous tenures] — was widely interpreted as a deliberate Lai-administration signal of intent to deepen the asymmetric-defence-and-civil-society-integration architecture under a civilian-led MND. Koo's prior tenure as Chairman of the Financial Supervisory Commission (FSC) and as Secretary-General of the National Security Council under Tsai (2022–2024) had positioned him within the senior national-security architecture without traditional uniformed-service career-progression credentials; his May 2024 MND-Minister appointment was paired with the appointment of Po Horng-huei (柏鴻輝) and other deputy MND heads with traditional military-career credentials, producing a civilian-political-military leadership architecture intended to operationalise the Whole-of-Society Defence Resilience Committee vision.

4. PLA Modernisation, the Joint Sword Exercise Series, and the Davidson Window Debate

The People's Liberation Army's 2015–2016 reorganisation under the Central Military Commission (中央军委) Chairman Xi Jinping (习近平) transformed the PLA from the pre-2015 seven-Military-Region territorial-defence structure into the five-Theatre-Command joint-operations structure, with the Eastern Theatre Command (东部战区, Dongbu Zhanqu) — headquartered in Nanjing with operational area covering Fujian, Jiangsu, Zhejiang, Anhui, and Shanghai — having Taiwan as its primary contingency focus. The reorganisation's impact has been the development of PLA joint-operations capacity through standing joint headquarters, integrated air-sea-ground-rocket-cyber-and-space operational planning, and theatre-level command-and-control architecture for a Taiwan-contingency campaign. The PLA Navy (PLAN) build-up across the 2017–2026 period has produced an approximate 370+ surface-combatant fleet [TBD-VERIFY 2024–2025 battle-force composition — US DoD-published 2024 China Military Power Report aggregates the surface battle force at approximately 370 vessels including 50+ destroyers, 50+ frigates, and 70+ corvettes, with substantial year-on-year additions through new construction], three aircraft carriers (the Liaoning, the Shandong, and the Fujian undergoing sea trials), and the Type-075 amphibious-assault helicopter-dock vessels (three commissioned by 2024 plus additional hulls under construction). The PLA Rocket Force (PLARF) has deployed the DF-17 hypersonic glide vehicle (operational from approximately 2019), the DF-26 intermediate-range ballistic missile (operational from approximately 2016 with sustained deployment expansion across 2017–2024), and the DF-21D anti-ship ballistic missile (operational from approximately 2010 with the modernised DF-21D variants entering service through the post-2017 period); the PLARF's total operational launcher inventory of intermediate-range-and-shorter ballistic missiles is variously estimated [TBD-VERIFY exact 2024 PLARF launcher-inventory] as the most- PLARF deployment relative to a Taiwan contingency.

The PLA Air Force (PLAAF) operates approximately 200+ Chengdu J-20 fifth-generation fighters [TBD-VERIFY exact J-20 inventory at end-2024 — published 2024 estimates range from approximately 200 to 280, with the wider range reflecting different sourcing across IISS Military Balance, US DoD China Military Power Report, and open-source order-of-battle trackers], with mass deployment to multiple PLAAF brigades from approximately 2019 onwards. The PLAAF also operates large numbers of fourth-generation-plus Shenyang J-16 and Chengdu J-10C fighters that constitute the bulk of the deployed PLAAF combat-aircraft fleet. The PLA Army (PLA-A) Eastern Theatre Command formations include the 71st, 72nd, 73rd Group Armies, with the 73rd Group Army (headquartered in Xiamen, Fujian) postured for the Taiwan-contingency amphibious-assault role alongside the Type-075-equipped PLAN Marine Corps elements. The PLA's qualitative modernisation across cyber, electronic-warfare, space, and counter-space domains under the Strategic Support Force (战略支援部队, Zhanlue Zhiyuan Budui) from 2015 through its 2024 reorganisation into the Information Support Force, Aerospace Force, and Cyberspace Force has expanded PLA non-kinetic-operations capacity.

The August 2022 Joint Sword exercise (4–10 August 2022) responding to House Speaker Nancy Pelosi's 2 August 2022 visit to Taipei (treated in detail at TW-C-06) marked the operational baseline shift in PLA cross-Strait military-pressure activity. The exercise — designated by Beijing as a "joint military operation" rather than a formal exercise — crossed multiple thresholds: (a) ballistic-missile overflight of Taiwan proper, with the 5 August 2022 DF-15 trajectories transiting over Taipei airspace at high altitude (approximately 100+ kilometres) — although technically transiting through space rather than airspace given the trajectory altitudes, the symbolic-political effect was substantial; (b) naval-and-air-force operations encircling Taiwan in six designated exercise zones around the island that included zones within Taiwan's claimed territorial waters; (c) the de facto suspension of the Taiwan Strait median line as a tacit-restraint reference; (d) the sustained post-exercise PLA pattern of routine median-line crossings, ADIZ incursions, and naval-vessel presence around Taiwan, establishing the post-August 2022 "new normal" of cross-Strait military pressure that has been sustained through the 2023–2026 period.

The April 2023 Joint Sword exercise (8–10 April 2023) was a smaller-scale response to the 5 April 2023 Tsai-McCarthy transit meeting at the Ronald Reagan Presidential Library in California, during which President Tsai met with House Speaker Kevin McCarthy. The exercise focused on PLA Navy and PLA Air Force operations around Taiwan with PLA Eastern Theatre Command-published exercise zones to the north, south, and east of Taiwan, and substantial ADIZ-incursion volume during the exercise window. The May 2024 Joint Sword-2024A exercise (23–24 May 2024) followed the 20 May 2024 Lai Ching-te inaugural address (TW-D-01, TW-D-02), with the PLA Eastern Theatre Command-published designation of "Joint Sword-2024A" explicitly signalling a successor 2024B iteration. The October 2024 Joint Sword-2024B exercise (14 October 2024) responded to the 10 October 2024 Lai National Day address, with the exercise architecture emphasising naval encirclement and joint air-sea operations around Taiwan and offshore islands.

The 2025 Strait Thunder-2025A exercise (April 2025 [TBD-VERIFY exact dates — published reporting indicates the exercise was conducted in early-to-mid-April 2025]) and subsequent iterations sustained the post-2022 pressure pattern under a new exercise-name framework, with the PLA Eastern Theatre Command continuing to use post-2024 exercise iterations as cross-Strait pressure tools coordinated with PRC political messaging on Lai-administration cross-Strait actions. The transition from the Joint Sword to the Strait Thunder naming framework has been interpreted by Taiwan-side analysts as a deliberate PLA-and-Beijing signal that the exercise series is institutionalised rather than ad-hoc, with sustained quarterly-to-semi-annual frequency through the foreseeable post-2025 horizon.

The Davidson Window debate has framed US strategic-community conversations on the Taiwan-contingency timeline since Admiral Philip Davidson's 9 March 2021 Senate Armed Services Committee testimony stating that "the threat is manifest during this decade, in fact, in the next six years". The "next six years" formulation produced the 2027-deadline reference framework that has shaped US war-planning, force-posture decisions, and Taiwan-defence procurement-prioritisation conversations through the post-2021 period. CIA Director William Burns in February 2023 affirmed that Xi Jinping had directed the PLA to be "ready by 2027" to conduct a successful invasion of Taiwan — a formulation that distinguishes "ready to invade if directed" from "will invade in 2027". The CSIS First Battle of the Next War wargame report (Cancian, Cancian, and Heginbotham, January 2023) — running multiple iterations of a 2026-set Taiwan-invasion scenario — produced a baseline finding that the United States, Japan, and Taiwan could defeat a PLA invasion in most iterations but with substantial losses including multiple US aircraft-carrier sinkings, hundreds of US aircraft losses, and tens-of-thousands of US service-member casualties, alongside very substantial Taiwan-and-Japan losses. The CNAS wargame series (Wasser, Metrick, Pettyjohn) has produced similar findings emphasising the high-cost-of-victory framing.

Subsequent analytical work by Bonny Lin (CSIS), Ryan Hass (Brookings), Richard Bush (Brookings, Difficult Choices 2021), and others has emphasised that the Davidson Window's "next six years" formulation does not entail an actual PLA invasion decision in 2027, that the strategic calculus around an actual invasion decision involves political, economic, and operational factors well beyond the binary readiness assessment, and that the political-economic costs of a Taiwan invasion to the PRC under any plausible scenario substantially exceed any plausible benefits. The Davidson Window debate has not been resolved at the analytical level — the gap between "PLA capability readiness" and "PLA invasion decision" cannot be closed through public-domain analysis alone — but its operative effect on US-Taiwan policy through 2021–2026 has been substantial regardless of its analytical merits.

5. Defence-Budget Trajectory 2017–2026: From NT$327 Billion to the Proposed NT$949 Billion

The ROC defence-budget trajectory across 2017–2026 represents the most-aggressive sustained defence-spending expansion in the post-democratisation era and reflects the cumulative response to the post-2017 cross-Strait military-balance evolution. The FY2017 defence budget of approximately NT$327 billion (~2.0 per cent of GDP) was approved by the Legislative Yuan in the late-2016 budget cycle as the Tsai administration's first full-year defence allocation; the FY2018 budget remained near the 2 per cent of GDP baseline. The FY2019 budget reflected the initial Tsai-era expansion approaching NT$346 billion. The FY2020 budget incorporated the post-2019 announced F-16V Block 70/72 special-budget allocation (approximately NT$250 billion total special-budget over the FY2020–FY2026 delivery window), bringing the aggregate defence-and-defence-related spending substantially above the headline figure. The FY2021 budget of approximately NT$453 billion reflected continued expansion incorporating the special-budget allocations and the post-2020 supplementary procurement packages. The FY2022 budget of approximately NT$471 billion was supplemented by a special-budget allocation for the Hsiung Feng II/III, Tien Kung III, Wan Chien stand-off-attack missile, and other NCSIST-produced weapons programmes (the "Sea-Air War Force Improvement Special Budget", 海空戰力提升計畫採購特別預算, totalling approximately NT$240 billion over the FY2022–FY2026 window).

The FY2023 budget of approximately NT$586 billion (~2.4 per cent of GDP) reflected the post-2022-Pelosi-visit expansion and the integration of the special-budget supplementary allocations into the operative defence-spending baseline. The FY2024 budget of approximately NT$606 billion (~2.5 per cent of GDP) was the Tsai administration's final full-year defence allocation and incorporated the cumulative special-budget allocations alongside continued procurement-pipeline expansion. The FY2025 budget of approximately NT$683 billion (~2.5–2.6 per cent of GDP) — the first Lai-era budget cycle — was the budget cycle affected by the January 2025 KMT-TPP freeze-and-cut sequence treated in TW-D-04; the disputed published reduction figures ranged across NT$197.5 billion, NT$207.5 billion, and NT$216 billion aggregating frozen-versus-cut items, with the defence-budget-specific component of the cuts variously estimated [TBD-VERIFY exact defence-specific cut and freeze totals — the KMT-TPP-framing emphasised that the defence-spending headline was preserved while specific line-items (NCSIST programmes, NSC operations, satellite-procurement) were targeted, while the DPP-framing emphasised the operational disablement of cross-Strait deterrence infrastructure].

The proposed NT$949 billion FY2026 defence budget submitted by the Executive Yuan in late 2025 — approximately NT$266 billion above the FY2025 baseline and approximately 3.3 per cent of GDP — represents the Lai administration's response to the Trump-2 "five per cent of GDP" demand and the post-2025 strategic-pressure environment. The proposed budget incorporates: a substantial expansion of the regular-budget defence allocation; the continuing special-budget allocations for the F-16V, Hsiung Feng / Tien Kung / Yun Feng / Hai Kun programmes; a new special-budget allocation framework for post-2025 expanded asymmetric-platform procurement and civil-defence-resilience programmes; the integration of the Whole-of-Society Defence Resilience Committee operational budget across MND, MAC, NSC, and inter-ministerial coordination accounts. The proposed FY2026 budget's actual LY passage trajectory through May 2026 will be shaped by the post-recall LY arithmetic (treated in TW-D-05); as of the document version date the post-recall LY arithmetic is partially established with the May–August 2025 Great Recall outcomes integrated into the post-September 2025 LY configuration, though the November 2026 Nine-in-One election cycle and the inter-election dynamics across late 2025 to mid-2026 add uncertainty to the operative LY-passage probability.

The defence-spending-to-GDP ratio trajectory benchmarked across SIPRI Military Expenditure Database and IISS Military Balance sources shows the ROC defence-spending-to-GDP-ratio rising from approximately 2.0 per cent in 2017 to approximately 2.6 per cent in 2024 — a substantial expansion in real terms given the underlying GDP growth across the period, but still well below the Trump-2 demanded 5 per cent and below the post-2014 NATO 2 per cent floor that has become the operative benchmarking standard in transatlantic defence discussions. The proposed FY2026 budget's 3.3 per cent ratio would represent the highest defence-spending-to-GDP ratio in the post-democratisation ROC, exceeding the late-1980s Chiang Ching-kuo-era post-democratisation high-water marks. The KMT-Blue-camp framing has positioned the proposed FY2026 increase as fiscally irresponsible given the ROC's other fiscal pressures (pension reform, social-welfare expansion, infrastructure investment); the DPP-Green-camp framing has positioned the increase as the necessary minimum-credible response to the post-2022 cross-Strait threat environment and the Trump-2 strategic-context.

6. US Foreign Military Sales Pipeline: F-16V, M1A2T, HIMARS, Harpoon, Stinger, MQ-9B

The US Foreign Military Sales (FMS) pipeline to Taiwan across 2017–2025 totalled approximately US$26 billion across more than thirty separate notification packages [TBD-VERIFY exact aggregate figure — published Congressional Research Service, Defense Security Cooperation Agency, and US-Taiwan Business Council totals vary depending on inclusion of Direct Commercial Sales notifications and FMS modifications]. The pipeline constitutes the structural backbone of Taiwan's post-2017 defence-modernisation programme and reflects the post-2017 Trump-1, post-2021 Biden, and post-2025 Trump-2 US-administration sequence of Taiwan-related FMS approvals.

The August 2019 F-16V Block 70/72 package — formally announced by the Defense Security Cooperation Agency on 20 August 2019 — covered 66 aircraft at approximately US$8 billion, with deliveries scheduled to commence in 2023 [TBD-VERIFY exact delivery commencement — initial deliveries were delayed beyond the originally scheduled 2023 commencement due to Lockheed Martin supply-chain and software-integration issues, with first F-16V Block 70 arrivals in Taiwan reported through 2024–2025] and complete by approximately 2026–2027. The package included the AN/APG-83 Active Electronically Scanned Array (AESA) radar, the AN/AAQ-33 Sniper targeting pod, conformal fuel tanks, and advanced electronic-warfare suite. Combined with the post-2017 F-16A/B Block 20 mid-life upgrade programme (approximately 142 aircraft upgraded to the F-16V configuration through the AIDC-Lockheed Martin joint programme at the AIDC Taichung facility), the post-2026-completion ROC Air Force F-16V Block 70/72 fleet will number approximately 200+ aircraft and represent the backbone of the ROC's post-2026 fighter-aircraft force structure.

The October 2020 combined package — formally announced 21 October 2020 — covered 11 M142 HIMARS launchers plus 64 ATACMS missiles, 135 SLAM-ER air-to-ground stand-off-attack missiles, and the MS-110 reconnaissance pod for the F-16V fleet, at approximately US$1.8 billion aggregate. The HIMARS component delivered Taiwan a long-range precision-strike capability against PLA amphibious-assault and force-concentration targets across the Taiwan Strait, with the ATACMS range of approximately 300 kilometres allowing strikes against PLA staging areas on the Fujian coast. The December 2020 MQ-9B SeaGuardian package covered 4 aircraft providing extended-endurance maritime-surveillance capacity. The December 2022 AGM-84L Block II Harpoon Coastal Defence System package — formally announced 5 December 2022 — covered 400 missiles plus 100 launchers and 25 radar-and-control vehicles at approximately US$2.37 billion, providing Taiwan a substantial coastal-defence anti-ship capability matching the ODC framework's prescription for large numbers of mobile anti-ship platforms.

The multiple Stinger MANPADS packages 2019–2024 cumulatively delivered to Taiwan several thousand FIM-92 Stinger missiles [TBD-VERIFY exact cumulative Stinger transfers — published Congressional Research Service and Defense Security Cooperation Agency aggregations indicate cumulative transfers of approximately 2,500 Stingers across multiple packages, with delivery schedules running through 2026 due to US production-line constraints]. The Stinger transfers have been complicated by post-2022 Ukraine-war-driven US Stinger inventory pressures and Lockheed Martin production-line ramp-up requirements; the 2023 Presidential Drawdown Authority transfers to Taiwan (the first PDA usage for Taiwan-bound transfers, announced 29 July 2023 at approximately US$345 million) addressed the near-term delivery-acceleration question by drawing weapons from US military stockpiles for direct transfer rather than through the longer-cycle FMS production-and-delivery pipeline.

The M1A2T Abrams package — formally announced 8 July 2019 — covered 108 tanks at approximately US$2 billion, with deliveries commencing in late 2024 and continuing through 2026. The package's ODC-framework-implications were debated within Taiwan-defence analytical circles given the asymmetric-defence framework's emphasis on dispersed anti-armour capability over heavy-armour platforms; the MND defence of the M1A2T package has emphasised the asymmetric-resilience role of heavy armour in coastal-annihilation-stage operations and the deterrence-signalling value of US-supplied advanced main battle tanks. The 2023 NASAMS package (approximately US$1.16 billion announced 25 May 2023 for Norwegian Advanced Surface-to-Air Missile System launchers plus AIM-120 AMRAAM munitions for ground-launched air-defence) and the 2023 Volcano anti-tank-mine package (approximately US$180 million for vehicle-mounted Volcano anti-armour mine-laying systems) added further asymmetric-defence-aligned procurement.

Post-2024 announced packages have included multiple smaller and medium-sized packages covering sustainment, ammunition replenishment, and specific platform-modernisation items. The post-March 2025 Wellington Koo–US-DoD engagement track has aimed at sustaining the post-Biden FMS-pipeline cadence under the Trump-2 administration, with the post-March 2025 announced packages [TBD-VERIFY exact post-March 2025 announced packages — published 2025 reporting indicates several Trump-2-administration-approved packages including extended-range precision munitions and continued sustainment packages, with the specific notification dates and dollar values requiring verification against DSCA publications]. The cumulative FMS-pipeline architecture has produced a Taiwan-defence-procurement posture deeply integrated with US-supplier capacity and US-administration political-approval cycles, with the corresponding strategic-vulnerability that a sustained US-administration policy shift away from Taiwan arms-supply would have substantial operational consequences for the ROC Armed Forces' modernisation trajectory.

7. Indigenous Defence Industry: NCSIST, CSBC Submarines, and the Yun Feng Programme

The indigenous defence industry architecture — anchored by the National Chung-Shan Institute of Science and Technology (NCSIST, 國家中山科學研究院), CSBC Corporation Taiwan (台灣國際造船) for the indigenous-submarine programme, the Aerospace Industrial Development Corporation (AIDC, 漢翔航空工業) for the IDF Ching-kuo and Brave Eagle (勇鷹) advanced jet trainer programmes, and a constellation of smaller indigenous suppliers — provides the resilience-and-self-sufficiency complement to the US-FMS-dependent pipeline. NCSIST, headquartered in Longtan District, Taoyuan with major research-and-production facilities at multiple locations across Taiwan, is the ROC's primary indigenous defence-research-and-development organisation with origins in the early-1970s Chiang Ching-kuo-era nuclear-and-missile development programmes and the subsequent decades of expansion under MND oversight.

NCSIST's key 2017–2026 missile programmes include: the Hsiung Feng II (雄風二型) subsonic sea-skimming anti-ship cruise missile, in continued production at expanded rates from 2017 onwards, with cumulative production exceeding 1,000 missiles [TBD-VERIFY exact cumulative Hsiung Feng II production — published estimates vary and the specific production totals are classified]; the Hsiung Feng III (雄風三型) supersonic ramjet anti-ship missile, reaching mass deployment from approximately 2018 onwards with the "carrier-killer" framing reflecting its design optimisation for PLAN aircraft-carrier targeting at extended range; the Hsiung Feng IIE (雄風二E) land-attack cruise missile, with reported range of approximately 600–1,200 kilometres providing Taiwan a counterstrike capability against PLA mainland targets [TBD-VERIFY exact Hsiung Feng IIE range — published estimates vary widely from approximately 600 to 1,500 kilometres]; the Yun Feng (雲峰) extended-range surface-to-surface missile programme, reportedly with range up to 2,000 kilometres allowing strikes against PLA targets across Eastern and Southern Theatre Command areas [TBD-VERIFY exact 2024–2026 Yun Feng deployment status — open-source reporting indicates initial production from 2023 onwards but precise deployment counts remain classified].

NCSIST's air-defence missile programmes include the Tien Kung III (天弓三, Sky Bow III) surface-to-air missile, comparable to the US Patriot PAC-3 in role and approximate performance, with substantial deployments across Taiwan's air-defence network; the Tien Kung IV (天弓四) under development with extended range and improved seeker performance [TBD-VERIFY exact Tien Kung IV deployment status]. The Sky Sword II (天劍二, Tien Chien II) air-to-air missile arms the ROC Air Force IDF Ching-kuo and F-CK-1 fleet. NCSIST has also expanded its post-2022 military-grade unmanned-aerial-vehicle (UAV) and loitering-munition production substantially, with multiple platform types entering service across the 2023–2026 period including the Teng Yun (騰雲) extended-endurance UAV, the Chien Hsiang (劍翔) anti-radiation loitering munition, and smaller tactical-UAV systems.

The CSBC Corporation Taiwan indigenous-submarine programme — formally launched in 2016–2017 under the Tsai administration's strategic decision to pursue indigenous submarine construction given the absence of sustained US-side willingness to supply post-2001-approved diesel-electric submarines, and developed in partnership with overseas suppliers (the so-called "red team" of US, UK, European, and other Asia-Pacific suppliers under reported coordination with the Trump-1 and subsequent administration permissions) — produced the first-of-class Hai Kun (海鯤, Narwhal) submarine unveiled on 28 September 2023 at the CSBC Kaohsiung facility. The Hai Kun is reported to have entered sea-trial commencement in 2024–2025 with projected first-of-class delivery to the ROC Navy by 2027 [TBD-VERIFY exact delivery schedule — the published programme timeline anticipates first-of-class operational delivery in 2027 with the eight-submarine target fleet completing through approximately 2038, though programme-progression uncertainty is substantial given the complexity of indigenous submarine construction]. The programme's ODC-framework implications have been debated within Taiwan-defence analytical circles; the MND defence of the programme has emphasised the asymmetric-deterrence role of indigenous submarines (the difficulty PLA anti-submarine warfare faces in tracking small-numbers of well-operated diesel-electric submarines in the Strait littoral zone) and the strategic-signalling value of indigenous-construction capacity.

The AIDC Brave Eagle (勇鷹, Yong Ying) advanced jet trainer programme — replacing the legacy AT-3 trainer and supplementing the F-5 fleet retirement — has produced an indigenous twin-engine advanced jet trainer with initial operational capability from approximately 2020 onwards. The Brave Eagle programme is intended to deliver approximately 66 aircraft by 2026, with the AIDC Taichung facility serving as the production site. The IDF Ching-kuo F-CK-1 fleet (approximately 130 aircraft delivered through the original 1990s production run) remains operational through the post-2026 period with sustained upgrade programmes. The AIDC's continued co-production partnerships with US suppliers for the F-16V mid-life upgrade and the F-16V Block 70/72 sustainment programmes integrate the indigenous defence-industry capacity with the US-FMS-pipeline.

8. Reservist Reform and the 2024 Conscription Extension to Twelve Months

The 2013–2018 transition from compulsory conscription to an all-volunteer force structurally reshaped the ROC Armed Forces' personnel architecture but produced sustained policy concerns about reservist readiness and training quality. The transition's premise was that a smaller, professional, high-quality volunteer force would be operationally more effective than a larger conscript-based force; the practical 2018-onwards outcome was an ROC Armed Forces with a target active-duty strength of approximately 215,000 personnel (compared to the pre-2013 strength of approximately 275,000) and a reservist force whose readiness was widely perceived as inadequate. The four-month "military training service" (軍事訓練役) standard implemented from 2018 onwards for males born from 1 January 1994 onwards produced a training cycle widely characterised by both critics and trainees themselves as insufficient: the four-month period covered approximately five weeks of basic military training plus eleven weeks of specialty training, with limited live-fire training, limited tactical-unit-training exposure, and limited integration with the operational reservist mobilisation cycle.

The post-2022 strategic-environment shift — the August 2022 Pelosi-visit PLA response, the 2022 Russian invasion of Ukraine and its demonstrations of mass-conscript-army war-fighting in the post-2014 European security environment, and the sustained 2022–2024 Davidson-Window pressure — produced the political-and-bureaucratic preconditions for conscription-policy reversal. The December 2022 Tsai Ing-wen announcement extending the compulsory-service term to twelve months for males born from 1 January 2005 onwards — first cohort entering service in 2024 — represented the formal reversal. The implementation has covered: a substantially restructured training curriculum incorporating Stinger MANPADS familiarisation, anti-tank weapon familiarisation (Javelin and shoulder-fired anti-armour systems including the Kestrel domestically-produced anti-tank weapon), combat-medic training, urban-warfare basics, and improved live-fire training cycles; an increased monthly stipend from approximately NT$6,500 (four-month service) to approximately NT$20,320 (twelve-month service); and a planned post-conscript-service reservist-cycle integration ensuring that twelve-month-service-completers cycle through annual reservist training with live-fire-and-unit-tactics components.

The MND-reported 2024 first-cohort intake [TBD-VERIFY exact 2024 first-cohort numbers — published MND-reporting indicates the 2024 first-cohort included approximately 9,000 males born in the first quarter of 2005, with the cohort scaling up through subsequent quarters as the policy comes into full effect] is being tracked through 2025–2026. The post-2024 implementation has faced multiple challenges: barracks-and-training-facility capacity constraints requiring substantial post-2022 infrastructure investment; instructor-and-training-cadre capacity constraints requiring the recall of retired non-commissioned officers and the expansion of professional-soldier training-cadre allocations; the equipment-supply chain for individual-soldier weapons and protective gear including the post-2022 expanded helmet-and-body-armour procurement.

The reservist reform programme alongside the conscription extension has aimed at improving the readiness of the approximately 2 million-strong nominal reservist pool. The post-2022 reservist mobilisation cycle reforms have included: the introduction of more-substantial 14-day reservist refresher training (against the legacy 5-7 day cycle); the introduction of regional reservist-mobilisation exercises integrated with active-duty unit operations; the expansion of the Volunteer Reserve (志願役後備) programme. The civil-society civil-defence training organisations — Kuma Academy (黑熊學院, founded 2021 by legislator Puma Shen 沈伯洋 and Lin Bin-jian); Forward Alliance (壯闊台灣聯盟, founded by Enoch Wu 吳怡農); and a constellation of regional civil-defence-training operations — have operated alongside the formal MND-administered reservist training, providing civil-defence basics, first-aid, communications, and search-and-rescue training to civilian volunteers. The post-June 2024 Whole-of-Society Defence Resilience Committee architecture formally integrated the civil-society civil-defence-training organisations as recognised national-resilience partners.

9. Civil Defence and the Whole-of-Society Defence Resilience Committee

The 26 June 2024 Lai Ching-te launch of the Whole-of-Society Defence Resilience Committee (全社會防衛韌性委員會, Quan Shehui Fangwei Renxing Weiyuanhui) at the Office of the President was the institutional centrepiece of the Lai administration's defence posture and operationalised the post-2022 civil-defence-and-resilience-architecture turn that had begun under the late-Tsai administration. The Committee — chaired by President Lai with Premier Cho Jung-tai, MND Minister Wellington Koo, National Security Council Secretary-General Joseph Wu, Mainland Affairs Council Minister Chiu Tai-san, and selected civil-society partner-organisation representatives — held its inaugural meeting on 26 June 2024 with subsequent quarterly meetings through 2024–2025 and into 2026. The Committee's five-pillar architecture covered: (a) civil-defence training and volunteer-reservist mobilisation; (b) strategic-stockpile and supply-chain resilience; (c) energy and critical-infrastructure resilience; (d) financial-and-economic resilience; (e) information-warfare and cognitive-warfare resilience.

The civil-defence-training pillar operationalised the integration of Kuma Academy, Forward Alliance, and other civil-society civil-defence-training organisations into the national-resilience architecture. Kuma Academy — founded in 2021 by legislator Puma Shen and Lin Bin-jian with funding from Robert Tsao (曹興誠), the former UMC chairman who returned to Taiwan in 2022 to invest substantially in civil-defence-training infrastructure — provides civilian volunteer training in civil defence basics, first aid, communications, and tactical awareness. Forward Alliance, founded by Enoch Wu Yi-nung (former MND consultant and 2022 Taipei legislative candidate), provides similar civilian volunteer training with a focus on operational-readiness drills. The integration of these organisations into the Whole-of-Society Defence Resilience Committee architecture — recognised through inter-ministerial memoranda of understanding and budget-allocation arrangements — represents the formal recognition of post-2014-Sunflower-era civil-society defence-mobilisation institutional architecture as a national-resilience asset.

The strategic-stockpile and supply-chain resilience pillar covered: strategic-petroleum-reserve expansion (post-2022 expansion of the Chinese Petroleum Corporation strategic reserves, with reported reserves of approximately 90+ days of consumption [TBD-VERIFY exact post-2024 strategic-petroleum-reserve days-of-consumption — published reporting varies across approximately 90 to 145 days depending on calculation methodology]); grain reserve expansion through the Council of Agriculture; pharmaceutical-and-medical-supply reserve expansion through the Ministry of Health and Welfare. The energy-and-critical-infrastructure-resilience pillar addressed the post-Hualien-2024-earthquake (3 April 2024) grid-reinforcement programme covering Taipower's transmission-and-distribution network, post-2023 LNG-terminal expansion (the controversial third LNG terminal at Datan and the post-2023 fourth LNG terminal at Keelung), and the satellite-internet redundancy programme through Eutelsat OneWeb and other non-Starlink providers given the Starlink-Taiwan availability question (with Starlink not operationally available in Taiwan through 2024–2025 due to Elon Musk's PRC-aligned business interests and the corresponding ROC NCC licensing-and-political concerns).

The financial-and-economic-resilience pillar covered post-2024 measures including: the Financial Supervisory Commission's stress-testing of the banking sector for cross-Strait conflict scenarios; the Central Bank's foreign-exchange-reserve diversification under continued NT dollar-USD exchange-rate management; the post-2024 anti-money-laundering-and-counter-terrorism-financing enhancements addressing PRC-aligned financial-infiltration concerns. The information-warfare-and-cognitive-warfare resilience pillar covered: the post-2020 Disinformation Acts framework operationalisation; the post-2022 expanded social-media-platform engagement with Meta, Google, X (Twitter), TikTok, and others on content-moderation cooperation; the post-2024 PRC-affinity-programme due-diligence requirements in primary, secondary, and tertiary educational institutions.

The Committee's quarterly meeting cadence through 2024–2025 — with publicly documented meetings on approximately 26 June 2024 (inaugural), 26 September 2024, 26 December 2024, 27 March 2025, 26 June 2025 (one-year-anniversary), and continuing quarterly meetings through 2026 — has produced inter-ministerial coordination outputs across the five pillars. The KMT-Blue-camp critique of the Committee has emphasised perceived politicisation of national-defence-resilience architecture and the integration of DPP-aligned civil-society organisations into government-resourced programmes; the DPP-Green-camp framing has emphasised the operational necessity of integrated civil-military resilience architecture given the post-2022 cross-Strait threat environment.

10. Information Warfare, Grey-Zone Pressure, and the Subsea-Cable Architecture

The post-2022 PLA grey-zone-pressure architecture — incorporating ADIZ incursions, naval-vessel encirclement, subsea-cable cuts, cyber operations, and information-warfare campaigns — constitutes the operational baseline of cross-Strait military-pressure short of kinetic conflict. The ROC MND-published PLA-aircraft ADIZ daily tracker, maintained from approximately September 2020 onwards via the MND X (Twitter) account, recorded approximately 1,737 individual PLA aircraft intrusions into Taiwan's ADIZ in calendar-year 2022 [TBD-VERIFY exact 2022 ADIZ-incursion total — published aggregations vary from approximately 1,700 to 1,800 depending on counting methodology], approximately 1,711 in 2023, approximately 3,070 in 2024 [TBD-VERIFY exact 2024 ADIZ-incursion total — published MND-tracker aggregations range from approximately 2,900 to 3,100 depending on counting methodology around dual-zone incursions and revisit-versus-unique counting], and a sustained 2025 pace [TBD-VERIFY 2025 running totals — through end-2025 the running total exceeded approximately 3,000 incursions with the post-Joint-Sword-2024B and Strait-Thunder-2025A pattern continuing]. The median-line crossings, naval-vessel presence around Taiwan, and Taiwan-encircling exercise zones have established the post-August 2022 baseline.

The subsea-cable cut record has been particularly consequential for civil-and-strategic-resilience. The February 2023 Matsu-Taiwan cable severance — when two PRC-flagged fishing vessels separately severed the two submarine cables connecting Matsu to Taiwan within several days of each other, with the resulting connectivity disruption lasting approximately 50 days through cable-repair operations — was attributed to anchor-drag operations whose deliberateness-versus-accident character was contested. The Matsu islanders' sustained connectivity disruption produced the first systematic Taiwan-side civil-and-military response to subsea-cable vulnerability. The January–February 2025 Pratas-and-Taishi cable incidents — with the post-25 February 2025 Hongtai 58 vessel detention near the Penghu Channel under the post-March 2025 Foreign Hostile Force designation operational framework — extended the post-2023 pattern with the addition of formal-legal countermeasures against vessels implicated in cable severance. The post-March 2025 incidents [TBD-VERIFY specific 2025 incident dates and attribution status] sustained the pressure pattern.

The Taiwan subsea-cable protection programme — operationalised through the Ministry of Transportation and Communications, the National Communications Commission, the National Security Council, and Chunghwa Telecom — has covered: increased subsea-cable redundancy through new cable construction (the post-2024 announced new cable routes connecting Taiwan to Japan, the Philippines, and the United States with intermediate landing points); the satellite-internet redundancy architecture through Eutelsat OneWeb (with reported Taiwan-government contracts from 2024 onwards) and other non-Starlink providers; the post-2025 reinforced-cable construction with armoured outer-sheathing reducing vulnerability to anchor-drag operations; the post-March 2025 increased coast-guard and MND-naval monitoring of vessels operating near subsea-cable landing zones. The satellite-internet redundancy architecture's structural limitation — Starlink's non-availability in Taiwan through 2024–2025 due to Elon Musk's PRC-aligned business interests and the Trump-2 administration's post-January 2025 reorientation — has constrained the bandwidth-and-resilience capacity of the satellite-redundancy backstop, with Eutelsat OneWeb's bandwidth substantially below Starlink's at comparable cost.

The information-warfare and cognitive-warfare pressure architecture has covered disinformation operations targeting Taiwan electoral and policy processes, PRC-affinity media and entertainment-industry penetration, and social-media-platform information-environment manipulation. The Taiwan FactCheck Center (台灣事實查核中心), Cofacts (真的假的), MyGoPen, and other civil-society fact-checking organisations have tracked sustained PRC-aligned disinformation operations across 2017–2026 including: 2018 and 2020 election-cycle disinformation around the Kaohsiung mayoralty, Han Kuo-yu, and the 2020 Tsai re-election; 2022 mid-term-elections-cycle disinformation; 2024 election-cycle disinformation; the post-2024 sustained disinformation campaign around the Bluebird Movement, the Great Recall, and the post-March 2025 Ten Measures designation. The post-March 2025 Foreign Hostile Force designation operationalised the formal-legal framework for countering PRC-aligned information operations under the Anti-Infiltration Act framework, with corresponding enforcement actions against PRC-aligned media operations, entertainment-industry penetration cases, and educational-institution affinity-programme cases.

The cyber-operations pressure — covering PRC-state-affiliated and PRC-state-tolerated cyber operations against ROC government, military, critical-infrastructure, and private-sector targets — has been tracked by the National Information Security Center, the National Security Bureau, and Taiwan Computer Emergency Response Team Coordination Center. Reported attack volumes against ROC government networks have been on the order of millions of attempted-intrusions per day, with the post-2024 sustained pressure pattern reflecting both PRC-state and PRC-aligned criminal-actor operations. The post-2024 cyber-resilience architecture has covered: the Ministry of Digital Affairs (MODA, 數位發展部, established August 2022 under Minister Audrey Tang 唐鳳, succeeded after April 2024 by Huang Yen-nun 黃彥男) operationalisation of cyber-resilience programmes; the post-2024 cyber-defence-cooperation expansion with US Cybersecurity and Infrastructure Security Agency (CISA), US Cyber Command, and partner-nation cyber agencies; the post-2025 expanded private-sector cyber-defence cooperation with TSMC, Foxconn, and other major Taiwan companies.

11. Trump-2 "Five Per Cent of GDP" Demands and the US-Taiwan Friction Track

The post-January 2025 Trump-2 administration's Taiwan-related demands have produced the most-substantial post-1979 friction in the US-Taiwan strategic-architecture and have restructured the Lai administration's defence-budget and policy calculus. The Trump-2 demands articulated across the post-January 2025 period have included: a numerical demand for Taiwan defence spending to reach 5 per cent of GDP, articulated most clearly in pre-and-post-inaugural statements and reiterated through multiple media engagements across the post-January 2025 period; a framing of Taiwan's semiconductor-industry centrality as a problem rather than as a strategic-alignment asset, articulated through the February 2025 Trump statements that "Taiwan stole our chip business" and through the corresponding Section 232 semiconductor-tariff track (treated in detail at TW-D-03); ambiguous statements on the US defence commitment to Taiwan under the Taiwan Relations Act and the Six Assurances framework, with multiple post-inauguration interviews where Trump declined to clarify whether the United States would defend Taiwan in the event of a PRC attack; the linkage of defence-and-tariff tracks in the Trump-2 negotiating architecture.

The Lai administration's response has aimed at preserving the operational defence-cooperation pipeline while managing the political friction. The principal response measures have included: the proposed NT$949 billion FY2026 defence budget representing approximately 3.3 per cent of GDP — a substantial increase from the FY2025 baseline of approximately 2.6 per cent of GDP, though still well below the demanded 5 per cent threshold; the post-March 2025 TSMC US-investment announcement of additional US$100 billion bringing total post-2020 TSMC US investment to approximately US$165 billion (treated in detail at TW-D-03); the post-March 2025 Ten Measures operationalisation including the Foreign Hostile Force designation signalling Taiwan's seriousness about cross-Strait deterrence-and-resilience; the sustained Wellington Koo–US-DoD engagement track including multiple visits to Washington and the operational continuation of FMS package approvals through the post-March 2025 period.

The KMT-Blue-camp framing of the Trump-2 friction has emphasised the structural risk of over-dependence on a transactional US-administration whose Taiwan-policy commitment is unreliable, with KMT-aligned commentators including former MOFA officials, retired military officers, and policy analysts advocating for greater cross-Strait dialogue restoration and for a Taiwan-defence posture less dependent on US-FMS pipeline continuity. The DPP-Green-camp framing has emphasised the necessity of preserving US-Taiwan defence cooperation while increasing indigenous-defence-resilience capacity and diversifying international defence-cooperation partnerships including expanded Japan-Taiwan, Australia-Taiwan, and European-Taiwan defence-and-security engagement. The post-January 2025 Lai administration sustained outreach to Japan (under Prime Minister Sanae Takaichi from May 2025), to the Philippines (under President Ferdinand Marcos Jr.'s continuing administration), and to selected European partners — discussed in detail at TW-D-02 and at JP-F-04 — has aimed at constructing a diversified international-cooperation architecture less vulnerable to Trump-2-administration policy reversal.

The structural Trump-2 friction question is whether the post-January 2025 US-Taiwan-policy trajectory will sustain the post-1979 Taiwan-Relations-Act-and-Six-Assurances framework or whether the Trump-2 administration will retreat from the operational defence-cooperation pipeline. The post-March 2025 evidence through the document version date suggests the operational pipeline has been preserved with continued FMS package approvals, continued AIT-MOFA engagement, and continued AIT-MND engagement, but with substantially increased Trump-2-administration extraction of Taiwan-side concessions on semiconductor investment, defence-spending, and trade-balance pressure. The medium-term trajectory through 2026–2028 will be substantially shaped by the post-2026-LY-election arithmetic, the 2028 ROC presidential election cycle, and the corresponding US-presidential-election cycle.

12. Forward View: Lai's Defence Agenda 2026–2028

The Lai administration's defence agenda through 2026–2028 — the remainder of Lai's first four-year presidential term concluding May 2028 — will be substantially shaped by the proposed NT$949 billion FY2026 budget's actual LY passage trajectory, the November 2026 Nine-in-One local-election cycle and the resulting partisan-political environment, the trajectory of cross-Strait PLA-pressure architecture, and the Trump-2 administration's evolving Taiwan-policy posture. Several discernible trajectory components can be tracked.

First, the post-2026 LY defence-budget passage will determine whether the proposed 3.3-per-cent-of-GDP FY2026 budget can be sustained in subsequent fiscal years. The post-Great-Recall LY arithmetic and the November 2026 Nine-in-One election cycle (treated in TW-D-05) will shape the post-2026 LY's willingness to sustain defence-spending expansion. The KMT-TPP working coalition's post-recall arithmetic — partially established with the May–August 2025 Great Recall outcomes integrated into the post-September 2025 LY configuration — will determine the operational majority for budget-cycle decisions through the FY2027 and FY2028 budget submissions. The Trump-2 administration's continued 5-per-cent-of-GDP pressure (and potentially escalating demands), combined with the post-2025 evolving Section 232 tariff track, will continue to shape the political-economic context for defence-spending decisions.

Second, the indigenous-submarine Hai Kun programme's first-of-class delivery to the ROC Navy is projected for 2027 [TBD-VERIFY exact delivery schedule], with the eight-submarine target fleet completing through approximately 2038. The 2027 delivery — coinciding with the Davidson-Window-debate 2027 reference date — will be a substantial strategic-signalling moment regardless of the operational-capability question. The continued NCSIST missile-and-UAV production expansion through 2026–2028 will further substantiate Taiwan's indigenous asymmetric-deterrence capacity.

Third, the F-16V Block 70/72 fleet completion through 2026–2027 will produce a post-2027 ROC Air Force F-16V fleet of approximately 200+ aircraft, with the AIDC-Lockheed Martin sustainment partnership ensuring continued post-delivery operational support. The post-2026 fighter-aircraft modernisation question — whether Taiwan will pursue F-35 acquisition (variously suggested but never formally announced through the document version date), an additional F-16V batch, or expanded indigenous-and-partnership fighter development — will be a major post-2026 procurement-pipeline decision.

Fourth, the Whole-of-Society Defence Resilience Committee architecture's institutional consolidation through 2026–2028 will determine whether the civil-defence-and-resilience integration becomes a durable feature of the Taiwan national-security architecture or a Lai-administration-specific arrangement subject to subsequent-administration reversal. The post-2026 KMT chair (succeeding Eric Chu's October 2025 successor) and the 2028 KMT presidential-candidate selection (treated in TW-D-05) will shape the post-2028 partisan-political environment for the Committee's institutional sustainability.

Fifth, the cross-Strait PLA-pressure-architecture trajectory through 2026–2028 — the continuing Strait Thunder successor exercise series, the continuing ADIZ-incursion baseline, the continuing subsea-cable-pressure pattern, and any new PLA-capability deployments — will continue to shape the strategic-environment baseline within which Taiwan's defence posture operates. The post-2024 PLA capability-development trajectory — the Type-076 amphibious-assault vessel (reported sea trials from 2025); the Fujian aircraft carrier's expected operational commissioning by 2026 or 2027; the PLAAF J-20 fleet's continued expansion; the PLARF DF-27 hypersonic intermediate-range missile reported development — will further widen the cross-Strait military-balance through the medium-term horizon.

Sixth, the post-2025 US-Taiwan defence-cooperation track under the Trump-2 administration will be shaped by the operational continuity of FMS package approvals, the AIT-MOFA-and-AIT-MND engagement track, and the broader Trump-2 administration's evolving Indo-Pacific-strategy framework. The post-March 2025 evidence through the document version date suggests substantial operational continuity, but the medium-term political risk remains substantial. The post-2026 US-presidential-election cycle's effect on US-Taiwan policy will further shape the post-2028 trajectory.

13. Conclusion: The Asymmetric-Defence Hypothesis in Practice

The 2017–2026 Taiwan defence trajectory has operationalised the asymmetric-defence framework articulated by Admiral Lee Hsi-min in the Overall Defense Concept, though the implementation has been a hybrid rather than a pure-asymmetric force-design that retains substantial traditional-platform procurement alongside the asymmetric-platform expansion. The defence-budget trajectory from NT$327 billion (FY2017) to the proposed NT$949 billion (FY2026) reflects a near-tripling in nominal terms and the most-aggressive sustained defence-spending expansion in the post-democratisation era. The US Foreign Military Sales pipeline of approximately US$26 billion across 2017–2025 — anchored by the F-16V Block 70/72 package, the M1A2T Abrams package, the HIMARS-and-SLAM-ER package, the AGM-84L Block II Harpoon Coastal Defence System package, the multiple Stinger packages, the MQ-9B SeaGuardian package, and the NASAMS-and-Volcano packages — constitutes the structural backbone of Taiwan's modernisation programme.

The indigenous defence industry architecture — anchored by NCSIST, CSBC, and AIDC — provides the resilience-and-self-sufficiency complement, with the Hsiung Feng II/III, Tien Kung III, Yun Feng, indigenous-submarine Hai Kun, and Brave Eagle programmes operationalising the indigenous-deterrence-capacity expansion. The 2022 announcement and 2024 implementation of the conscription extension to twelve months for the post-2005-born cohort represents the most-substantial post-democratisation conscription-policy reversal and reflects the post-2022 strategic-environment hardening. The 26 June 2024 Lai launch of the Whole-of-Society Defence Resilience Committee operationalised the post-2022 civil-defence-and-resilience-architecture turn and integrated civil society, military, and inter-ministerial coordination on a quarterly-meeting cadence.

The PLA's 2017–2026 modernisation trajectory has widened the cross-Strait military balance against Taiwan, with the PLA acquiring force-projection, joint-operations, precision-strike, and amphibious-assault capabilities that the ROC Armed Forces cannot match in symmetric force-on-force terms. The Joint Sword exercise series of 2022–2024 and the 2025 Strait Thunder successor series have operationalised PLA grey-zone-and-coercion-pressure architecture and established the post-2022 baseline. The Davidson Window debate has framed US strategic-planning conversations through the 2021–2027 window, even where subsequent analytical work has disputed the 2027 timeline.

The post-January 2025 Trump-2 administration's "five per cent of GDP" demands and the associated US-Taiwan friction have restructured the Lai administration's defence-budget and policy calculus, with the proposed NT$949 billion FY2026 budget representing approximately 3.3 per cent of GDP as the Lai-side response. The medium-term trajectory through 2026–2028 will be substantially shaped by the post-2026-LY-election arithmetic, the indigenous-submarine programme's 2027 first-of-class delivery, the F-16V Block 70/72 fleet completion through 2026–2027, the Whole-of-Society Defence Resilience Committee architecture's institutional consolidation, the continuing cross-Strait PLA-pressure-architecture trajectory, and the evolving Trump-2 administration's Taiwan-policy posture.

The deeper question — whether Taiwan's post-2017 asymmetric-defence-and-resilience architecture can sustainably deter PLA aggression across the medium-term horizon, and whether the Lai-administration's substantially-increased defence-spending-and-resilience-investment can offset the cross-Strait military-balance trajectory's adverse direction — cannot be definitively answered from the 2026 vantage. What can be said is that the post-2017 Tsai-Lai sequence has produced the most-substantial Taiwan defence-modernisation effort in the post-democratisation era, that the asymmetric-defence framework articulated by Admiral Lee Hsi-min has shaped the procurement-and-doctrine trajectory if imperfectly implemented, and that the post-2022 civil-defence-and-resilience architecture's integration of civil society, military, and inter-ministerial coordination represents a qualitatively new feature of the Taiwan national-security architecture. The cross-Strait military-balance trajectory's medium-term direction will continue to favour the PLA in symmetric force-on-force terms; the Taiwan-side hypothesis that asymmetric-defence-and-resilience can offset that direction sufficiently to maintain cross-Strait deterrence remains the operative national-security-strategy premise.