Vietnam and Thailand are not competing for the same manufacturers.
Industry Insights
Manufacturers frequently treating Vietnam and Thailand as alternatives are starting from the wrong premise. These are not two versions of the same bet. They serve structurally different roles in the global value chain, and have been diverging toward those roles for over a decade.
Companies that recognize this early and position accordingly will hold a compounding structural advantage. Companies that don’t will spend the next decade managing a decision they can’t fully unwind.
This piece will cover three things most market analyses get wrong about this decision, one framework for getting it right, two markets clearly positioned, and one signal that will tell you in the next 18 months whether your thesis holds.
The three mistakes most manufacturers make
Most comparative analyses of Vietnam and Thailand fail before they conclude, because their structure is off.
Mistake one: treating all criteria as equally weighted.
Labor costs, tax incentives, logistics infrastructure, and FTA coverage are laid out in a table, side by side, as if each carries the same weight in a 15-year facility decision. They don’t. The sequence in which you apply these criteria is not a detail. It is the decision. Apply them in the wrong order, and the right data will lead you to the wrong answer.
Mistake two: optimizing for today’s snapshot.
Both markets are moving.
Vietnam’s labor cost advantage is narrowing at 8–10% annually. Thailand’s ecosystem depth is compounding, each layer of capability pulling the next into place.
Carbon costs that are negligible today will be material by 2030. A location decision built on current numbers is being made for a market that will not exist at full capacity. The analysis needs to run forward, not be anchored to the present.
Mistake three: treating the decision as either/or.
The twin-sourcing model, Thailand upstream, Vietnam final assembly, is already the dominant structure among sophisticated complex manufacturers. It did not emerge by accident.
EV players, including Changan and Hyundai, evaluated Vietnam seriously before committing to Thailand. Not because Vietnam was cheaper or offered worse incentives, but because Thailand’s upstream automotive supply chain eliminated a supply disruption risk that Vietnam could not mitigate at any price. Meanwhile, Apple and Foxconn chose Vietnam for final assembly, not despite rising wages, but because the FTA stack creates a market access advantage that compounds regardless of labor cost trajectory.
These are not opposing decisions. They are the same decision, applied correctly to two structurally different roles.
The framework for manufacturers: three filters, applied in sequence
A 15-year facility commitment is not a spreadsheet problem. It is a sequencing problem. The criteria that feel most concrete: labor costs, tax rates, and incentive packages, are the ones that should be evaluated last. Here is the right order.
Filter one: geopolitical stability and market access
| Vietnam | Thailand | |
| Domestic political continuity | CPV one-party system. FDI and industrial policy have been stable for 30+ years. | History of coups and judicial reshuffles. BOI and EEC industrial policy have survived all government transitions. |
| External geopolitical exposure | High. US exports ~$153.2B in 2025, ~30% of total. April 2025 tariff shock exposed this directly. | Lower. Balanced trade relationships across the US, EU, China, and ASEAN. BRICS allied partner from January 2025. |
| FTA coverage | 17 active FTAs. The only ASEAN economy with EVFTA, CPTPP, and RCEP simultaneously. | RCEP and ASEAN FTAs only. Not a CPTPP member. No EVFTA equivalent. EU FTA unsigned. |
Source: Compiled by Arches’ team
This is the foundation filter. If a market cannot guarantee policy continuity and global market access across your investment horizon, no other criterion matters. Evaluate this first, before anything else.
Vietnam passes on domestic continuity. Over 30 years of consistent FDI policy. The strongest FTA stack in ASEAN. However, it fails partially on external exposure. That concentration is a structural vulnerability that no incentive package resolves.
Thailand passes on the external balance. Neutral geopolitical positioning, diversified trade relationships, and no equivalent single-market dependency. It fails partially on internal continuity; coups, judicial reshuffles, and PM dismissals have been a recurring feature of the past two decades. Industrial policy has survived every transition. The operating environment has not always.
The directional implication is clear: if more than 35% of your finished goods are destined for EU or Pacific markets, Vietnam’s FTA stack is a structural anchor that no incentive package in Thailand can offset. If your supply chain is more regionally contained, Thailand’s external balance becomes a stronger foundation.
Filter two: ecosystem depth and supply integration
| Vietnam | Thailand | |
| Ecosystem position | Downstream only. Final assembly and back-end semiconductor packaging. PCB sector nascent. | Upstream present. PCB, HDD, IC testing, and EMS are all locally available. Produces ~80% of the world’s HDDs. |
| Component import dependence | $102B in component imports in 2024. ~36% of total national imports. | Net electronics trade balance: -$13B. Significantly less import-dependent. |
| Local supplier integration | 14–15% of local firms are integrated into FDI supply chains. Of ~340 Vietnamese firms in Samsung’s supply chain, only 35–39 are Tier-1. | 16 of the global top 100 PCB suppliers are operating locally. Upstream material suppliers are now co-locating. |
| FDI composition | Assembly-focused. Samsung, Foxconn, LG. | Upstream capital entering. PCB clusters, SiC wafer fab, EV component manufacturing bases. |
Source: Compiled by Arches’ team
This is the ceiling filter. The question it answers is not whether a market can support your operation at entry. It is whether the market can support your operation at the complexity level you need, five years from now, at full capacity.
Vietnam operates on an import-to-assemble-to-export model. Component imports were strong in 2024. Only 14–15% of local firms are integrated into FDI supply chains. The indispensable link, the local supplier ecosystem that makes a manufacturing hub self-reinforcing, has not been built. For assembly operations, this is sufficient. For complex upstream manufacturing, the ecosystem is not there yet. Building it requires simultaneous upgrading across an entire supply chain, and that has never happened quickly, anywhere.
Thailand presents a different picture. PCB, HDD, IC testing, and EMS capabilities are all locally available. Upstream material suppliers are co-locating around anchor facilities. The pattern is the tell; each layer of capability forces the next layer into position. This is what an indispensable link looks like while it is forming.
The directional implication: if your operation requires Tier 2 and Tier 3 supplier proximity to function, Thailand is currently the only market in Southeast Asia where that condition is met at scale. Vietnam is building toward it. But building toward it and having it are not the same condition at a 15-year horizon.
Filter three: workforce technical depth
| Vietnam | Thailand | |
| STEM pipeline | Government target: 50,000 semiconductor engineers by 2030. An aspiration, not a current capability. | ~80,000 STEM and robotics graduates annually. |
| Institutional depth | 28% of the labor force holds formal training certificates. | 4 government-funded semiconductor labs, including KMITL. 20+ years of precision manufacturing depth in HDD and PCB. |
| Talent development reality | Must be built from the ground. Cannot be imported at scale. | Same constraint, but the build started two decades earlier. |
Source: Compiled by Arches’ team
This is the compounding filter, and in the long run, the decisive one. Capital can be raised. Technology can be transferred through licensing agreements, joint ventures, and direct investment. Talent cannot be imported at scale and cannot be built overnight.
Thailand’s foundation is built and actively compounding: large STEM and robotics annual graduates, available semiconductor facilities, and over 20 years of precision manufacturing depth accumulated through HDD and PCB production.
Vietnam’s ambition at the advanced engineering level is genuine. The government’s target by 2030 reflects real political intent. It does not reflect the current reality.
The historical proof that this filter is decisive: ITRI received RCA’s technology transfer in 1976. TSMC was not spun out until 1987. Eleven years separated the transfer from the industry. What filled that gap was not additional capital or more licensing agreements. It was engineers who could understand, adapt, and improve what was handed to them. Absorptive capacity is what converts a technology transfer into a functioning industry. Ambition alone does not.
Thailand is building absorptive capacity. Vietnam is still in the ambition phase.
The directional implication: for any operation requiring advanced process engineering, semiconductors, precision EV components, and high-end PCBs, workforce depth is the filter that will determine whether your facility can scale. Right now, only one market passes this filter cleanly.
What the filters reveal for manufacturers
Apply the sequence, and the picture becomes unambiguous.
For complex upstream manufacturers such as EV, PCB, and semiconductors, Thailand clears all three filters more cleanly today. The decision to locate upstream operations in Thailand is supported at every level of the framework. Vietnam’s ambition is real. Its execution at this level of complexity is unproven.
For assembly-focused manufacturers: Vietnam clears filter one strongly. Filter two is sufficient for the role. Filter three is less critical at assembly complexity. The FTA stack is a structural anchor that compounds regardless of wage trajectory. The sequence supports Vietnam unambiguously for this role.
For manufacturers who need both: Stop treating this as an either/or. The twin-sourcing model is the logical conclusion of applying the sequence honestly. Thailand for upstream. Vietnam for final assembly and Western market access. Each market is doing exactly what the filters say it is built for.
What could make this wrong
For Vietnam to win the upstream race by 2035, three things would need to be simultaneously true. Viettel’s 32nm fab would need to deliver trial production on schedule by the end of 2027, and generate genuine Tier 2 and Tier 3 supplier co-location, not just an isolated anchor plant. Core material suppliers, leadframe producers, and substrate manufacturers would need to begin establishing local facilities. And the administrative transformation of 2025–2026 would need to produce consistent policy execution, not just a consolidated vision. All three. Simultaneously. That is a high bar. It is not impossible. It has not happened yet.
For Thailand’s ceiling not to matter, technology transfer from Japan, Taiwan, and China would need to accelerate beyond its current pace, and government coordination on high-end materials would need to reach the scale of the EEC cluster model applied to semiconductors. Neither condition is currently in place.
The honest read: Vietnam’s upside-down scenario is real, but it requires simultaneous systemic upgrading that has no recent historical precedent in Southeast Asia. Thailand’s ceiling is real, but it sits higher than most analyses acknowledge, and the path to it is better defined.
The 18-month signal that will tell you if the thesis holds
Both markets are moving, and the framework reflects today, not full capacity. For Vietnam, watch Viettel’s 32nm fab: on schedule by the end of 2027 is a genuine ecosystem signal; a slip means the anchor isn’t pulling. For Thailand, watch whether semiconductor coordination moves beyond PCB into wafer-level processes; progress pushes the ceiling higher, absence confirms it.
If you want the macro forces behind this decision, green energy access, carbon costs, and supply chain realignment through 2050, I contributed research to a RECCESSARY piece that covers exactly that.
The position, restated
Vietnam and Thailand are not competing for the same manufacturers. They never were.
The companies that figure this out and apply the right framework in the right sequence will make a decision in 2026 that looks obvious in 2035. The ones still comparing labor costs and tax incentives will still be deciding.
If you are ready to apply this framework to your specific situation, what’s your experience with it so far? Find me on LinkedIn.
About Arches Expert Solution
Looking for deeper market intelligence?
At Arches, we help corporate and consulting teams go beyond surface-level research. Our Expert Solution combines structured market frameworks with targeted expert validation, giving teams the clarity they need before committing to a new market.
Learn more about Expert Solution →
To discuss how Expert Solution is typically applied in real market entry decisions, meet Hoang Le, our Expert Solution | Consulting Manager, who works closely with corporate and consulting teams on early-stage market assessments across the region.
Follow Arches on LinkedIn to get the latest updates from us.
Subscribe to Arches’ monthly newsletter for grounded perspectives from experts across Asia and beyond, where global decisions meet local insights.
