The headline story around Malaysia is unmistakable: the country is riding the AI and chip cycle to become Asia's growth standout. The National Semiconductor Strategy has attracted RM63 billion in investment. IntelINTC-- is beginning operations at its $7 billion advanced packaging facility in Penang. Malaysia handles 13% of global chip assembly and test. The electrical and electronics sector accounts for 40% of national exports.
That story is incomplete. The RM63 billion figure is loaded almost entirely with foreign direct investment - RM58 billion from multinationals versus RM5 billion from domestic sources. Malaysia's 13% share of global assembly volume captures only about 6% of semiconductor value-add. Not a single Malaysian company today is capable of advanced packaging, the highest-value segment of the business. The country's OSAT players - Inari, Pentamaster, Unisem, Globetronics - are watching their core wire-bond and flip-chip business commoditize and migrate to cheaper competitors in Vietnam and China.
This is not a demand story. It is a supply-side bifurcation: multinational capital building advanced work on Malaysian soil, while local companies fight for margin in the low-value tail of the value chain. The question is whether the gap closes or widens.
Quick Backtesting Tool
The Multinational Investment That Carries the Headline
Within roughly a year, the government had secured RM63 billion in commitments - RM58 billion from foreign investors and RM5 billion domestically. The foreign projects include Intel's advanced packaging complex in Penang, Infineon's world's largest 200mm silicon carbide power fab, NXP semiconductor operations, and Carsem's advanced packaging expansion.
Intel's Penang facility is the anchor. Intel's $7 billion advanced packaging facility in Penang is expected to commence operations in 2026.
The implication is structural. Malaysia is not generating indigenous semiconductor capacity. It is attracting capital that would have deployed elsewhere. The multinationals bring process knowledge, customer relationships, and technology access that do not transfer to local operators through proximity alone. Intel's packaging plant will run Intel's packaging technology for Intel's customers. The economic benefits for Malaysia are employment, tax receipts, and cluster effects - real, but not equivalent to building a domestic champion.
The Value-Add Gap: 13% Volume, 6% Value
The data tells a more precise story than the headline suggests.
Table 1 below summarizes the mismatch:
| Metric | Value |
|---|---|
| Global semiconductor assembly/test/packaging share | ~13% |
| Semiconductor value-add captured by Malaysia | ~6% |
| E&E share of total exports | ~40% |
| E&E contribution to GDP | 6-7% |
| Bursa-listed semiconductor companies share of total market cap | ~1.7% |
| NSS investment from domestic sources (of RM63B total) | RM5B (~8%) |
Source: Bank Negara Malaysia economic reports, The Edge-HSBC E&E Symposium 2026, Apex Academy sector research.
Malaysia's 13% share of global assembly volume is widely cited. The less-discussed figure is the 6% value-add capture. In semiconductor terms, value-add measures how much profit, margin, and intellectual property a country retains within its borders versus how much flows to foreign parent companies. The gap between 13% volume and 6% value exists because the work Malaysia does - traditional assembly, testing, marking, and packaging for multinational subsidiaries - sits at the lowest-value end of the industry's smile-shaped curve. Design and IP capture roughly half of total value-add. Fabrication takes around a quarter. Assembly and test, Malaysia's traditional stronghold, account for the remainder.
The 40% export share and 6-7% GDP contribution are another way of seeing the same dynamic. E&E is voluminous but low-margin. It moves a lot of dollars through customs without retaining much economic surplus.
Local OSAT Companies Under Structural Pressure
Malaysia's listed OSAT companies - Inari Amertron, Malaysian Pacific Industries, Unisem, and Globetronics - have a combined market capitalization of roughly RM27 billion. Their core business is traditional packaging: wire-bond and flip-chip processes that have dominated the industry for three decades but are increasingly commoditized.
The competitive threat is twofold. On the cost side, basic OSAT work is migrating to Vietnam and China, where labor costs are lower and local governments offer aggressive incentives. On the technology side, the highest-growth and highest-margin segment - advanced packaging, which integrates multiple chips (processors, memory, AI accelerators) into a single unit and carries ASP premiums of 3-10x over traditional packaging - is dominated by foundries (TSMC, Intel, Samsung) and Chinese OSAT leaders. None of the top 10 global OSAT companies by revenue is Malaysian. The combined 2024 revenue of the world's top 10 OSATs reached $41.6 billion; no Malaysian company appears on the list.
Chinese OSATs JCET and HT-Tech grew 19% and 26% respectively in 2024, fueled by domestic localization policy and government investment funds. They are moving up the packaging value curve faster than Malaysia's local players.
The Malaysian OSAT sector is not broken. It is structurally squeezed between commoditization below and advanced packaging above, with neither cost advantage nor process capability to differentiate itself.
The Advanced Packaging Gambit
The government's response has been the Malaysia Advanced Packaging Consortium, launched in May 2026. The consortium brings together five local companies - SkyeChip (IC design), FusionAP (packaging process R&D), Inari Amertron (OSAT production infrastructure), Pentamaster, and NSW Automation (testing and equipment) - with RM92 million in government funding matched by RM93 million from industry. Its initial project is an HBM4 pilot line, with a stated target of capturing 7% of the global advanced packaging market by 2035.
The ambition is real. The challenge is the starting point. Industry veteran Tan Eng Tong, a Silicon Valley veteran who has played a leading role in building MAPC, put it plainly at the SEMICON Southeast Asia conference: "Malaysia is at zero. Nothing." On the measure of local advanced packaging capability - process knowledge, IP, proven customer relationships - the country has none. The 13% global assembly share belongs to multinational operations on Malaysian soil, not to Malaysian companies.
The advanced packaging market was worth approximately $52 billion in 2025 and is forecast to nearly double by 2031. MAPC's 7% target by 2035 would represent roughly $5 billion in annual revenue. For comparison, ASE Technology, the global OSAT leader from Taiwan, generated $18.5 billion in 2024 revenue. MAPC would need to reach roughly a quarter of ASE's size in a market where TSMC's CoWoS capacity remains sold out through 2026 despite a near-quadrupling of capacity since 2023.
The consortium model - IP sharing among members, milestone schedules, split roles - is credible as a framework. It mirrors the approach that China used to build its advanced packaging ecosystem by incentivizing multinationals to develop local supply chains. But MAPC's independently estimated pilot line cost of RM400 million, a Series-A/pre-IPO funding gap of $10-20 million for domestic IC design firms, and a talent retention crisis (an estimated 1.1 million Malaysian passport holders work in Singapore at higher wages) represent concrete obstacles, not aspirational ones.
The Talent Constraint Is a Pay Problem, Not a Supply Problem
At the The Edge-HSBC E&E Symposium in July 2026, panelists were consistent on one point: Malaysia's constraint is compensation and retention, not raw talent supply. Local IC design and OSAT wages are benchmarked against Taiwanese and Singaporean offers, and Malaysian operators cannot match them. The government's RM1.2 billion talent pipeline over five years, targeting 60,000 engineers, addresses supply. It does not address the fact that engineers who complete that pipeline will be paid substantially more if they work in Singapore, which is a 2.5-hour drive from Penang.
Concrete measures raised included linking property-sector housing benefits to semiconductor cluster retention, extending work permits for foreign STEM graduates, and making ESOS (employee share option schemes) and share grants tax-deductible in Malaysia - they currently are not, unlike in Western markets.
Talent retention is not a background condition. It is the bottleneck that determines whether MAPC's pilot line produces commercial results or remains a government-funded demonstration project.
Investor Takeaway
Malaysia's semiconductor boom is real, but its benefits are distributed unevenly. Multinational operators expanding in Penang and Selangor are the primary beneficiaries of AI-driven packaging demand. Local OSAT companies face structural pressure from commoditization below and technological leapfrogging above. The MAPC consortium is the most credible vehicle for narrowing the gap, but it starts from zero process capability and faces a talent retention crisis that policy has yet to solve.
The key issue is not whether semiconductor demand remains healthy. It is whether Malaysia's local operators can capture value beyond the assembly-and-test segment before the global OSAT market consolidates further around Taiwanese, Chinese, and foundry-led advanced packaging. The forward condition to watch is whether MAPC's HBM4 pilot line transitions from demonstration to commercial production within its stated 24-month window, and whether the government's proposed tax and retention reforms materialize in Budget 2027. If the pilot succeeds and the talent reforms land, the gap between Malaysia's 13% volume share and 6% value-add can close. If not, the country remains a well-paid factory floor for multinational operations.
