The silicon challenge: Can Malaysia move up the chip value chain?

An agreement with Arm Holdings, the chip design unit of SoftBank, signed in March 2025 gives Malaysia access to the company’s semiconductor intellectual property.

This comes at a time when the country has established strong capabilities in the back-end segment. Semiconductors dominate the RM711bn electrical and electronics sector, which remains Malaysia’s largest export category. By 2030, the semiconductor industry is expected to reach a global market size of about US$1tn (RM4tn).

There is also a growing need to expand into front-end activities such as wafer fabrication.

This initiative aligns with the new industrial master plan for 2030 and the national semiconductor strategy, positioning Malaysia as a premier hub for semiconductor innovation, while cultivating an ecosystem to support the artificial intelligence sector.

AI is expected to contribute between RM13bn and RM20bn annually to Malaysia’s gross domestic product (GDP) by 2030, potentially adding between 0.8 and 1.2 percentage points to GDP growth each year within the digital economy.

The digital economy already contributes over 23% to Malaysia’s GDP, amounting to more than RM450bn in economic value as of 2025.

These developments follow the signing of two memorandums of understanding (MOUs): one between Mimos Bhd and Brazil’s Eldorado Institute, and another between the Malaysia Semiconductor Industry Association, the Brazilian Association of the Electrical and Electronic Industry and the Brazilian Semiconductor Industry Association.

When Prime Minister Anwar Ibrahim visited Brazil in November 2024, he consolidated Malaysia’s semiconductor initiatives with a bilateral meeting with President Luiz Inácio Lula da Silva on 18 November 2024.

The discussions around Brazil–Malaysia cooperation began in 2023, following the Brazil-Asean and Semiconductors: Unveiling Global South Synergies event, organised by the Brazilian Embassy in Kuala Lumpur in October 2023. This event marked the beginning of dialogue between stakeholders from both Global South nations.

Brazil and Malaysia’s collaboration in the semiconductor industry will involve the development of joint initiatives in integrated circuit design.

This comes at a time when Malaysia has been trying to break free from the dependency syndrome of a semi-peripheral nation.

While the country’s IT position is decidedly positive, certain important tasks still need to be executed carefully to ensure a successful path ahead.

The energy imperative

At the heart of Malaysia’s semiconductor industrial push is energy – and it is a vulnerability that policymakers cannot afford to underestimate. Advanced packaging and high-performance computing facilities serving AI workloads are energy-intensive.

Malaysia’s existing generation mix – historically reliant on fossil fuels from abroad, despite being an oil and gas producer and exporter – has been under stress from rising demand, weather variability and infrastructural bottlenecks, including the ongoing West Asia crisis.

The stakes are real. Grid instability poses a direct production risk: voltage fluctuations and outages can cost semiconductor fabrication plants millions in yield losses.

Khazanah’s capital injections into power infrastructure signal an awareness of this constraint.

But resilience is more than capacity expansion. It also encompasses redundancy, distribution automation, microgrids and rapid restoration capabilities.

Taiwan and South Korea did not arrive at their semiconductor dominance by accident. Both have integrated, sophisticated grid management systems that allow fabrication plant clusters to maintain near-continuous operations –essential for advanced packaging where thermal precision and process stability are critical.

Malaysia’s ongoing grid upgrades must therefore prioritise not just generation but smart grid technologies and renewables integration to ensure energy cost competitiveness over time.

Skills and the talent gap

Malaysia’s semiconductor labour force has traditionally excelled in back-end assembly, testing and packaging – a foundation built over five decades.

But five decades of doing something well does not make it sufficient for what comes next. Advanced packaging demands higher cognitive skills, precision engineering and cross-disciplinary capabilities.

Technologies such as 2.5D/3D integrated circuit integration, fan-out wafer-level packaging, and system-in-package solutions require engineers trained in microelectromechanical systems, materials science and thermal management, alongside production teams skilled in ultra-precise process control.

The national education and vocational pathways are misaligned with these demands. While Malaysia’s technical universities produce competent graduates in electronics and electrical engineering, industry-specific competencies for advanced packaging remain underdeveloped.

This is not a gap that can be papered over with short-term fixes. It requires curriculum redesign, industry–academy partnerships and direct placement programmes with global leaders in advanced packaging.

Building an innovation ecosystem

Advanced packaging straddles the divide between hardware and innovative design for manufacturability.

Simply importing equipment is not enough. Local firms must be able to absorb, adapt and build upon these technologies. That requires an ecosystem, not just a purchase order.

Here, Malaysia’s research architecture works against it. The country has yet to build a robust ecosystem of research and development centres, intellectual property creation and internationally collaborative research that feeds into domestic firms’ competitive capabilities.

Singapore’s success in packaging has hinged on institutions such as the the Institute of Microelectronics, a research institute under the Agency for Science, Technology and Research. These bodies operate with explicit mandates to co-create intellectual property with industry partners and spin out firms capable of competing globally.

Malaysia has no comparable institutional engine. The research ecosystem remains fragmented, with agencies duplicating efforts and limited pathways for research outputs to scale commercially.

Policy coherence

Khazanah’s funding for energy and semiconductors is necessary. But money alone does not build an industry.

Without co-ordinated industrial policy that aligns fiscal incentives with long-term capability building, the risk is real. Tax incentives and grants can attract investment.

But without performance-linked conditions, such as local R&D commitment, export orientation and workforce development, capital inflows risk perpetuating precisely the low-value activities Malaysia is trying to move beyond.

Countries that have successfully upgraded their semiconductor industries have employed mission-oriented policy frameworks with measurable milestones and accountability.

Malaysia’s policy architecture, by contrast, remains a patchwork. Multiple agencies have overlapping roles, diluting strategic focus and increasing transaction costs for firms.

A consolidated semiconductor strategy under a single coordinating body – with clear key performance indicators (KPIs) tied to capability progression from assembly, testing and packaging to advanced packaging – would streamline policy implementation and sharpen Malaysia’s strategic competitive edge.

From vision to viability

Malaysia’s aspiration to move up the semiconductor value chain towards advanced packaging is technically feasible. But it remains politically and institutionally underpowered.

While Khazanah’s investments in energy infrastructure acknowledge a crucial enabler, broader success depends on addressing four areas:

Without these structural reforms, Malaysia risks stagnating at mid-value stages of semiconductor production.

And without the Brazil–Malaysia collaboration and the Arm Holdings agreement, the country risks missing the economic and geopolitical dividends of fully participating in the semiconductor and emerging AI investment cycle.

...