New Southbound 2.0: You Can Diversify, But You Can't Replace — Taiwan's Market Breakout and Its Ceiling

Any discussion of New Southbound has to start by acknowledging one fact — Taiwan's economic center of gravity really is moving away from China. This is not rhetoric; it is statistics.

The share of Taiwan's exports going to mainland China (including Hong Kong) has fallen from a 2020 peak of 43.9% to 31.7% in 2024, slid further to 28.3% in the first quarter of 2025, and reached just 26.6% for full-year 2025 — falling below the share going to the U.S. (30.9%) for the first time in 26 years ^[1][7][14]. In just five years, a huge chunk has been cut away. (A note on the figures: 28.3% is a first-quarter-2025 number and 26.6% is a full-year number; both come from the Ministry of Finance but cover different periods.) This trend is even more striking on the investment side: the share of Taiwan's outbound investment flowing to mainland China has collapsed from 83.8% in 2010 to 11.4% in 2023 ^[1][3]. Fifteen years ago, Taiwanese firms going abroad had almost only one destination; today China doesn't even get the leftovers.

The reality of this structural shift cannot be denied. Southeast Asia has now formally overtaken China as Taiwanese firms' top choice for overseas investment, with Vietnam attracting the most attention ^[10]. At the 2024 Ketagalan Forum, President Lai Ching-te proudly announced that in the first half of that year, Taiwan's exports to the 18 New Southbound countries reached US$50.2 billion, the highest for the same period on record ^[5]. From Tsai Ing-wen launching the New Southbound Policy in 2016, targeting 18 countries across Southeast Asia, South Asia, Australia, and New Zealand ^[2], to the Lai administration's upgrade to "2.0," this southbound path has genuinely produced real results.

Here's the question. China's share dropped more than ten percentage points — did that "missing" share really all go to New Southbound?

The numbers tell a more complicated story. Over the same five years, Taiwan's export share to the United States surged from 14.7% in 2020 to 30.9% in 2025 ^[1] — the U.S. has nearly caught up with, or even overtaken, China as Taiwan's largest export market. The engine behind this is the demand tsunami for AI servers and advanced chips.

This leads to a piece of sober arithmetic: when your "numerator" of exports to the U.S. explodes, the overall "denominator" of exports also gets inflated, and China's share is naturally diluted — even if the absolute dollar amount going to China hasn't changed much. The decline in China's share is the joint product of "actual relocation" and a "denominator effect," not New Southbound single-handedly eating away at the China market.

This is exactly where the Eastern Eye and the Western Eye clashed most fiercely in our panel session. The Eastern Eye said bluntly: "The biggest credit goes to AI exports to the U.S. inflating the denominator, not New Southbound actually seizing market share." The Western Eye pushed back: "Don't reduce this to just a denominator effect — investment in China collapsing from 83.8% to 11.4%, and Taiwanese firms' top overseas choice shifting from China to Vietnam, is real, hard structural change." The economics expert's final calibration: both sides are half right — this is a dual effect; crediting New Southbound entirely is dishonest, and crediting AI entirely is unfair too (the precise breakdown of attribution shares; medium confidence).

Acknowledging this is not meant to belittle New Southbound, but to put it back in its true place: it is one leg of this larger restructuring, not the whole of it.

If you're looking for evidence that New Southbound is "actually delivering something," India and Vietnam are the two hardest data points. But interestingly, they tell two different stories.

India is more like a new market that is opening up. Bilateral Taiwan-India trade hit a record US$10.6 billion in 2024, then climbed further to US$12.5 billion in 2025, up 17% year on year ^[4][8]. In terms of composition, Taiwan's exports to India totaled US$9.2 billion (integrated circuits, plastics, electrical machinery, steel), while India's exports to Taiwan totaled US$3.3 billion (mineral fuels, aluminum, steel, organic chemicals) ^[8]. Even more telling is the on-the-ground reality: the number of Taiwanese firms in India is now more than triple the 2017 level, exceeding 250 firms ^[8]. India's massive domestic demand and the pull of "Make in India" make this closer to "opening a new market" than simply relocating factories.

Vietnam, on the other hand, is more like a China+1 manufacturing depth zone. Cumulative Taiwanese investment in Vietnam has reached roughly US$39.3 billion, making Taiwan the fourth-largest source of foreign investment in Vietnam ^[10]. Foxconn, Pegatron, Compal, Wistron, Chimei — these electronics contract-manufacturing giants relocated capacity from mainland China in bulk to Hanoi and Hai Phong in northern Vietnam after the U.S.-China trade war and the pandemic ^[11]. Vietnam offers low wages and CPTPP/RCEP tariff preferences, making it a natural landing spot for redirected orders.

But hidden in Vietnam's story is a trap that has to be named. Many Taiwanese firms that have "gone south" have only relocated the "last mile" of assembly — the machinery, engineers, and upstream materials still come from China. Products carry a "Made in Vietnam" label while the underlying supply chain is still Chinese at its core. This is what we have repeatedly discussed in our "The Three-Layer Dividend of De-Risking from China" series as the "China+0.5" phenomenon. Its implication is pointed: moving assembly out of China does not mean moving dependence on China out of China. Cumulative Taiwanese investment in Vietnam is large, but Vietnam's largest foreign investors are actually Singapore, Hong Kong, and China; Taiwan ranks fourth, and its influence has its limits ^[10].

Where New Southbound 2.0 has truly upgraded is not just in the "hard numbers" of trade and investment, but in person-to-person "soft ties." The Lai administration has framed the policy as "six corridors" — the government leads three: digital technology and semiconductors, public health, and disaster resilience; civil society drives three social-connection corridors: think tanks, youth, and NGOs ^[6]; and the National Science and Technology Council's research cooperation focuses on frontier biomedicine, semiconductors, and ICT ^[6].

The most concrete connection is happening in the labor market. In March 2026, the number of foreign migrant workers in Taiwan reached 873,000, a record high, and together with those who have gone missing from their contracts, the total is approaching one million — over 5% of Taiwan's resident population ^[12]. Of these, Indonesian nationals number 317,000 (37.1%), now slightly ahead of Vietnamese nationals at 292,000 (34.2%) — the labor forces of these two major New Southbound countries are, in very concrete terms, propping up Taiwan's electronics factories and long-term-care system ^[12].

This is a double-edged sword. On one hand, the movement of migrant workers, overseas students, and researchers is the connection Beijing finds hardest to cut off within New Southbound — it is rooted in civil society, not dependent on official agreements. On the other hand, Taiwan's heavy reliance on this labor also means New Southbound is not just "Taiwan positioning itself with others," but "Taiwan and these countries deeply depending on each other." This mutual dependence is precisely where New Southbound can quietly grow resilience by going around the diplomatic ceiling.

Having covered the achievements, the limits must also be addressed — and they are limits that cannot be avoided. New Southbound's biggest pain point is not that it hasn't done enough trade; it's that it has run into an institutional and diplomatic ceiling.

The most glaring example is CPTPP. Taiwan's application to join has now failed three times. At the ninth CPTPP Commission meeting in Melbourne in November 2025, four applicants (Uruguay, the UAE, the Philippines, and Indonesia) were found to meet the "Auckland Principles," and Uruguay's accession process was launched — Taiwan wasn't even mentioned ^[13]. Deputy Foreign Minister Wu Chih-chung said bluntly that China is "substantively blocking" Taiwan's accession; more awkwardly, Vietnam's leadership, while visiting China, publicly voiced support for "China's" accession ^[13]. Japan and the UK support Taiwan, but lack the ability to break through.

IPEF (the Indo-Pacific Economic Framework) is no help either — it does not itself include market access or tariff reduction, and Taiwan isn't even a member. Taiwan is thus caught in an awkward position: it is doing friend-shoring, yet it is excluded from the most important multilateral trade club, lacking any institutional anchor. Will the dividends be siphoned off by "institutionally recognized" players like Vietnam, India, and Mexico? This is what worries the Western Eye most.

An even deeper layer of the ceiling is Beijing's diplomatic pressure. Through its enormous economic leverage and its continued consolidation of the "One China" position, China keeps obstructing Taiwan from signing FTAs or investment agreements with regional partners, and pressures ASEAN leaders not to engage with Taiwan too visibly ^[9]. New Southbound carries an inherent political ceiling: as long as China is willing to apply pressure, official exchanges between Taiwan and Southbound countries will shrink back. That said, scholars also point to a margin of room — Beijing's economic leverage has not fully translated into ASEAN support for its values or diplomatic positions, which leaves Taiwan strategic space for hedging and maneuver ^[9].

Laying the achievements and the ceiling side by side, the neutral synthesizer's judgment from our panel session is: New Southbound 2.0 is a diversification strategy that is real, but has an upper limit.

It has genuinely and significantly "de-concentrated" Taiwan's trade and investment structure — that is a high-confidence fact. But it has not, and is unlikely to, let Taiwan "escape" the China market: China (including Hong Kong) remains one of Taiwan's largest export markets to this day, and China+1 relocation is often accompanied by continued upstream dependence on China. Writing New Southbound off as either a "cure-all for escaping China" or dismissing it as "pure political rhetoric" both depart from the facts.

Its real value lies not in the unachievable goal of "swapping out China," but in something more pragmatic: raising Taiwan's indispensability and freedom of action. Turning diversification into resilience, and turning Southbound into depth among "trustworthy partners" — as Taiwan's markets, supply chains, and talent connections spread across more partners who don't put all their eggs in one basket, the risk of Taiwan being "choked" by any single power drops a notch. This is the standard by which New Southbound 2.0 should actually be measured.

The state: stop managing expectations with the unachievable slogan of "escaping China," and tell society honestly — New Southbound is diversification, not replacement. Strategically, replace multilateral gridlock with bilateral breakthroughs: since CPTPP is being substantively blocked by China, put the effort into bilateral tracks such as the Taiwan-U.S. 21st Century Trade Initiative and deeper Taiwan-India economic and trade ties; at the same time, use technology, talent, and migrant labor — soft ties that go around the diplomatic ceiling — to make up for the lack of an institutional anchor. New Southbound's success should not be measured by "how much China's share fell," but by "how much Taiwan's freedom of action has increased."

Industrial intermediaries: help member firms see clearly the different playbooks for India and Vietnam — India for building out the domestic market and local roots (it is a new market, not just a new factory), Vietnam for China+1 manufacturing depth (but watch out for the China+0.5 trap: changing the assembly plant while the upstream stays tied to China). Trade associations and institutes should act as guides on specifications and certification, "writing" Taiwanese products into the specs of Southbound countries and de-risked supply chains, turning diversification into positioning rather than a mere cost-driven escape.

SMEs: with limited resources, the biggest mistake is treating Southbound as a "relocate the whole factory" gamble. The pragmatic approach is to treat Southbound as a second market — test the waters first with niche products and local partners, diversifying away from over-reliance on any single market, whether China or the U.S. Avoid the compliance landmines and political risk of pure transshipment origin-washing, and head toward niches "that others cannot easily replicate and that are needed locally." For a small firm, the point of going south is not "betting on a new China," but "opening one more window so you're not locked out by a single closed door."

Five years ago, Taiwan had staked nearly forty-four dollars out of every hundred on the single card table of China; today that figure is under thirty. This is a remarkable pivot, and it deserves credit. But after the pivot, Taiwan faces a more complicated world — a world where you can diversify your markets but cannot replace China, where you can escape the factory but cannot escape the supply chain, where you can make the trade work but still cannot get into CPTPP.

New Southbound 2.0 is not a one-way ticket "out of China" — it is an insurance policy against being "choked" by anyone. Buying the right insurance won't make you rich, but it will give you an extra way out when the storm hits. For an island sitting at the very center of a geopolitical storm, having one more way out is itself a strategic victory. The real test is not how many orders Taiwan can move south, but whether — as Beijing tries to close this southbound window inch by inch — Taiwan has already welded it wide enough, and strong enough, in advance.