The only economy that matters
Six Asian apparel exporters and the claim that worker protections are what keep India out of low-cost manufacturing.
Comparison of statutory employer social contributions, minimum wages, apparel export growth and market access across China, Vietnam, Bangladesh, Cambodia, Indonesia and India.
The political economy is the only economy that matters. A familiar argument runs the other way: that provident fund, insurance and the rest of the protection stack are the reason India cannot do low-cost manufacturing, and that the people asking for them are the problem.
That is a testable claim. Below are the six largest Asian apparel exporters, what each one legally makes an employer pay on top of wages, what its wage floor is, how its exports have moved since 2010, and what tariff its garments face at the border. If protections were the constraint, the countries with the heaviest ones should be losing.
Statutory employer social contributions, as a share of gross wage
Pension, health, injury, unemployment, housing and union levies that are the employer's cost by law. 2026 rates. Ranges where they vary by city or risk class.
China: pension 16%, medical ~10%, unemployment 0.5%, injury 0.2–1.9%, maternity 0.5–1% (merged into medical in many cities), housing fund 5–12%; all-in employer add-on commonly cited at 30–40%. Vietnam: social insurance 17.5%, health 3%, unemployment 1%, trade union levy 2%. India: EPF 12% (incl. EPS), EDLI 0.5%, admin 0.5%, ESI 3.25%; EPF applies at 20+ workers, ESI at 10+ and wages ≤₹21,000. Indonesia: BPJS Kesehatan 4%, JHT 3.7%, JP 2%, JKM 0.3%, JKK 0.24–1.74% by risk class. Cambodia: NSSF injury 0.8%, health 2.6%, pension 2% (4% from 2027). Bangladesh: no payroll social insurance; export factories pay 0.03% of export value to a central fund; the 2026 Labour Act amendment mandates a matched provident fund (7–8% of basic pay) at 100+ workers — hollow bar shows that mandate at ~4.5% of gross, not yet in force at the factory floor.
Apparel exports, indexed to 2010 = 100
HS 61+62, reported exports in current USD. Bangladesh uses the mirror series (partners' reported imports) because it stops reporting after 2016.
UN Comtrade, preview API, accessed August 2026. Vietnam 2024 not yet reported; hollow dot extends 2023 by the 9.3% growth WTO records for 2024. Bangladesh mirror values are CIF and run above its own export figures; the index is unaffected by the level gap, the level chart uses WTO instead.
Apparel exports, 2024, USD billion
WTO World Trade Statistics 2025. One source for all six so levels are comparable.
Bar colour carries the statutory burden from section 1: full ink above 20%, mid-tone 10–20%, faint below 10%. The two that stalled since 2010 sit in the middle band, not the top.
Monthly wage floor in the main garment region, with the statutory add-on stacked on top
USD per month, 2026. Solid segment is the legal minimum wage; hatched segment is the employer's statutory contribution on that wage.
Wage floors: China, Guangdong class B (Dongguan, Foshan) RMB 2,300; Vietnam Region I VND 5.31m; Cambodia garment sector USD 210; India, Karnataka garment scheduled wage ≈₹12,800; Indonesia, Central Java provincial minimum IDR 2.33m; Bangladesh garment sector BDT 12,500. Converted at RMB 7.25, INR 87.5, IDR 16,600, BDT 121 per USD; rounded. Add-on uses the midpoint of each range in section 1 (Bangladesh at zero, since the 2026 mandate is not yet operative). Not a labour-cost survey: actual garment wages run above the floor everywhere, most of all in China.
What the same T-shirt pays at the border
Duty on apparel by origin. The US column is the country-specific tariff added on top of the ordinary MFN rate, which is about 16.5% on cotton knits for everyone.
| Origin | EU duty | US add-on above MFN | What is changing |
|---|
EU: MFN on knitted cotton apparel is 12%; Everything But Arms gives least-developed countries zero. US: the IEEPA "reciprocal" tariffs (Bangladesh 37%, Vietnam 46%, Cambodia 49%, India up to 50%) were struck down in February 2026, replaced by a flat 10% under Section 122, then by country-specific Section 301 rates from 24 July 2026. China additionally carries the 2018–19 Section 301 lists (7.5–25% on apparel).
Reading one: protections are not the variable
Sort the six by statutory burden and the export story does not sort with it. The two heaviest regimes, China at roughly 37% and Vietnam at 23.5%, are the largest and the fastest-growing exporters respectively. The two lightest, Bangladesh and Cambodia, also tripled. The two that stalled since 2010, India and Indonesia, sit in the middle of the range.
Vietnam raised its minimum wage 7.2% in January 2026 and does so nearly every year; its social insurance law was widened in 2025 to catch anyone on a one-month contract. Its apparel exports grew 9.3% in 2024, the fastest of the top exporters. If protections were the constraint, that is the country that should be losing orders to India.
Stacked all-in, a legal-minimum worker in India costs about $170 a month with every statutory contribution paid. That is the second-cheapest floor in the set, below Cambodia, Vietnam and China. Whatever is keeping orders away from Tiruppur, it is not the PF line.
Reading two: the unprotected path exists, and it is not available to India
Bangladesh did build the world's second-largest apparel export industry on a wage floor of $103, no payroll social insurance, and a minimum wage reviewed once every five years. Cambodia's rise came with a 5.4% burden. The cheap-and-unprotected route is real.
But both walked it with zero-duty access to the EU under Everything But Arms, a preference reserved for least-developed countries and one India has never had. Indian apparel has paid 12% into the EU throughout; Bangladesh's own graduation from LDC status in November 2026 is the reason its industry is nervous. The unprotected model is a tariff story wearing a wage story's clothes.
What actually moves India's position is the EU agreement concluded in January 2026, which takes that 12% to zero on entry into force, and the collapse of the US reciprocal tariffs. Both are trade policy. Neither requires a garment worker to give up her pension.
The data does not force a verdict on how much protection a poor country should legislate. It does settle the narrower claim. Among the six countries that make the world's clothes, the heaviest protection regimes are not the losers, the two stalled exporters are not the most protected, and India's all-in statutory labour floor is already among the lowest. Anyone arguing that the way to Vietnam's order book runs through denying workers a provident fund has not looked at what Vietnam pays.