Foreigners in Malaysia — who gets to stay, and who doesn’t
Source: Foreigners in Malaysia — who gets to stay and who doesn’t? 📹 — CNA (CNA Correspondent), 30 Sep 2026, 20:56. Host: Annelise Grainger; guest: Melissa Goh, CNA senior correspondent in Malaysia. Sent via Discord, 30 Sep 2026, as source material on the inequality/dependency theme.
The condensed read
One country, two completely different questions about “foreigners”, asked in the same 21 minutes:
- The capital everyone wanted, at a scale nobody planned for. Mainland-Chinese and Korean businesses have been in Malaysia for decades; what changed in the last two years is scale and scope, driven by the US–China trade and tariff war making business inside China brutally competitive. Klusters of Kuala Lumpur now read as Chinese or Korean — Nanjing Street’s Mandarin signage and mainland clientele, Ampang’s Koreatown — and rents in expat areas (Mont Kiara, Hartamas) have risen sharply. William Ong of the Malaysian SME Association states the ask precisely: “we are not against legitimate foreign operator in the Malaysia. What we want is really that they adhere to the local rules” — and that local players get included in the supply chain. The friction case is the “Alibaba business”: foreign operators using local proxies to trade without licences, tax, premises or staff costs, undercutting locals on price. Malaysia has revoked thousands of licences and is tightening registration, certification (electrical safety, halal) and online trading.
- The population nobody recruited at all. Around 215,000 refugees and asylum seekers are in Malaysia per UNHCR — ~193,000 registered from Myanmar (end-Feb 2026), more than 90% of them Rohingya. They have been hosted for decades, many born there, and they have no right to work, no access to public schooling, no public healthcare. Aman Hussein’s testimony: no permit, so day work only — construction, cleaning, moving — “sometimes 80 ringgit, sometimes 60, sometimes 50”, and often not paid at all. UNHCR third-country resettlement has effectively halted. The government’s DPP documentation programme was pitched as a route to matching refugees with jobs in manufacturing, construction and plantations; reporting now suggests its purpose has shifted toward repatriation, and Myanmar’s military government has agreed to take back 5,000. On 29 September 2026, Malaysia deported around 1,500 Myanmar nationals, including children, over UN and human-rights objections (Reuters).
The numbers that matter
- 2.13 million legally registered foreign workers (Malaysia, mid-Oct 2025) — down 13% year-on-year, against a government target of cutting dependence on migrant labour, and a video claim that Malaysia imports workers from Bangladesh and Nepal while barring 200,000 refugees from the labour market.
- ~215,000 refugees and asylum seekers; ~193,000 registered from Myanmar (UNHCR).
- 5,000 Myanmar nationals agreed for return; ~1,500 deported on 29 Sep 2026.
- Singapore: 6.21 million population at June 2026, up 1.6%, with non-residents contributing 70,300 of the 97,300 increase — i.e. the region’s richest state is itself built on migrant labour, while remaining not a signatory to the 1951 Refugee Convention or its 1967 Protocol, and with no formal asylum system.
- ASEAN, 21 July 2026: foreign ministers agreed to develop benchmarks and a longer-term special envoy to revive the stalled Five-Point Consensus on Myanmar.
Fact-check
| Claim | Verdict |
|---|---|
| ~200,000+ refugees and asylum seekers in Malaysia, >90% from Myanmar | Holds — UNHCR figures reported at ~215,000 total and ~193,000 registered from Myanmar (Feb 2026). |
| Refugees cannot work legally, and their children cannot attend public school | Holds — Malaysia is not a party to the 1951 Refugee Convention; UNHCR cardholders have no work right and no public education access. |
| Myanmar agreed to accept 5,000 returnees; repatriation starting “later this month” | Holds, and has already begun — the Home Ministry screened Rohingya at Bukit Kayu Hitam from July 2026; ~1,500 were deported on 29 Sep 2026 (Reuters). |
| ”Easily a million” migrant workers | Understated — 2.13 million legally registered (Oct 2025), before undocumented labour. |
| Malaysia’s target to cut migrant dependence to 10% by 2030 | Not verified in this pass — the machinery exists (13th Malaysia Plan 2026–2030 labour reforms; a multi-tier levy on foreign-labour-heavy employers), but I did not find the 10% figure stated in an official document. Treat as the video’s figure. |
| No official count of mainland-Chinese residents | Holds — neither the Chinese embassy nor Malaysian authorities publish one; UNHCR resettlement has been the only route out for many refugees. |
My assessment (Alf)
The requester’s framing is the right lens, and the video supplies the correction it needs. Two points.
First, “regional core” fits Malaysia better than the phrase suggests — but for a reason that cuts against the comfortable version. Malaysia is a middle-income core with a periphery inside it: it imports over two million workers and protects its SMEs from foreign competition by price and licence, while 200,000 people who live there, work informally and were often born there have no legal right to the job they are already doing. That is dependency theory’s structure in miniature — a rich middle, an unprotected poor — and it is why this video belongs under the inequality thread rather than only under migration.
Second, the honest test of “doing more for migrants”. The strongest argument in the video is Charles Santiago’s, and it is not sentimental: refugees spend their wages locally and cannot remit — there is no country to send money to — so employing them is more domestically stimulative than importing another worker who sends the money home. That is a rare case where compassion and self-interest point the same way, and it fails for one reason only: public consent. The government’s own communication failure, in an election window, is what the reporter names as the binding constraint. Which is the uncomfortable finding — the constraint is not fiscal, legal or economic capacity. It is politics, in a country that has the capacity.
Third, on Singapore. It is the region’s richest state and, per its own reporting, still not a party to the refugee convention with no asylum framework — so the “could do more” claim is factually available. But the sequencing matters: a state that cannot process an asylum claim has no mechanism to grant the work right, which is why the honest ask to Singapore is procedural before it is generous (join the framework, or build an equivalent, then decide the numbers). Demanding “more compassion” from a state with no legal channel produces nothing but a headline.
And the part that belongs to the second half of the frame — the failing source state. Repatriating 1,500 people to a country whose military is the defendant in a genocide case is not a migration policy; it is an outsourcing of the problem to the place it cannot be solved. That is the argument for regional cores working on Myanmar rather than on their own borders: ASEAN’s July 2026 move to benchmarks and a long-term envoy is the only mechanism in the region aimed at the source, and it exists because the Five-Point Consensus has failed for five years. See the Envisioning note for what that could look like.
Related
- Global Inequality and Dependency — the core–periphery frame this belongs to (
#crisis/global-dependency). - Migration and Displacement — the direct crisis thread (
#crisis/migration-displacement). - Regional Cores and the Duty of Care — Asia’s Rich States and the Migrant Question — the third step: what the vision looks like if taken seriously.
- SDGv2 Draft — the multilateral version of the same obligation.