The Textile Trap

Mauritius did not fail to industrialise in textiles. It industrialised on terms that never fully belonged to it. For decades, the garment industry stood at the centre of the island's development story. Sugar receded, factories rose, and the Export Processing Zone became the emblem of post-colonial reinvention. By 1990, the EPZ employed approximately 89,900 workers, with garments accounting for roughly 90 percent of EPZ employment and 83 percent of EPZ exports. Even then, the warning signs were structurally visible: garment manufacturing was a low-profit, highly mobile, labour-cost-driven industry, exposed to rising wages, shifting quotas, external competition and the sustained threat of production relocation to cheaper jurisdictions.

That early warning has proven correct in ways more damaging than anticipated. The export-manufacturing machine has not disappeared entirely, but it has thinned, hardened and shed most of the sovereign promise once attached to it. By the fourth quarter of 2025, total Export Oriented Enterprise employment had fallen to 28,224. Of those workers, 13,486 were foreign nationals, meaning migrants made up 47.8 percent of the entire EOE labour force. What was once presented as an industrial answer to Mauritian unemployment now survives to a remarkable and uncomfortable extent by importing labour from poorer countries. The transition from local employment engine to migrant labour platform is not a side note. It is the central structural fact about the current state of the sector.

The core problem is not that textiles declined. Many industries do, and structural change is not the same as failure. The deeper problem is that Mauritius built an export sector whose structural logic was always externally disciplined. The industry imported much of what it needed to produce, exported into markets shaped by trade preferences, relied on foreign buyers to sustain volumes and competed in a segment of world manufacturing where wages, speed and compliance mattered more than deep technological ownership. It created jobs, activity and exports. But it never fully became sovereign industry in the sense that matters: an industry Mauritius controls rather than one it participates in.

The historical record matters because it strips away hindsight bias. Mauritius' garment ascent was not accidental and it was not passive. The state actively constructed it. The early EPZ model combined tax holidays, import-duty exemptions on inputs and equipment, subsidised credit, labour-market flexibility, free repatriation of dividends and a sustained investment-promotion drive that made the island genuinely attractive for foreign garment manufacturers looking for a new production base in the late 1960s and 1970s. That was how a small island with no significant cotton base and no heavy industrial tradition entered global apparel manufacturing so quickly and successfully.

But the structure carried its own warning from the beginning. Garments were a fragmented, footloose and low-margin business. Raw materials could be sourced from world commodity markets. Entry barriers were modest. Cost competition was intense and geographically mobile. That meant production would tend, over time, toward the cheapest viable locations unless Mauritius moved decisively up the value chain toward design, branding, technical textiles or specialised manufacturing. The island succeeded not because garments were a stable industrial frontier, but because it briefly aligned incentives, labour cost advantage, trade access and timing in a way that was always contingent rather than permanent.

The scale of the employment retreat is stark. EPZ employment peaked at roughly 91,800 around 2001 when the textile miracle still appeared intact. By the fourth quarter of 2025, total EOE employment had fallen to 28,224. Even allowing for definitional shifts between the EPZ and EOE frameworks over time, the direction is unmistakable and the magnitude is not marginal. The labour base has been dramatically hollowed out over two decades, falling to roughly 30 percent of its peak level.

This is not only a labour story. Manufacturing's share of GDP has fallen from 15.4 percent in 2014 to 12.8 percent in 2025. The EOE sector's share of gross value added is now just 3.8 percent. Textiles are no longer the commanding productive frontier of the Mauritian economy. They have become a thinning enclave whose political and historical memory considerably exceeds its present economic weight, and whose survival increasingly depends on the structural conditions that the sector's defenders tend not to discuss openly.

The trade data expose the structural reality even more directly than the employment figures. In 2024, Mauritius exported approximately USD 478 million in textiles and garments while importing approximately USD 397 million in textile goods and inputs. China alone supplied around USD 121 million of those imports, with India also a major source. The net trade surplus in textiles is real but modest relative to the gross flows, and the import intensity of the export activity is substantial. This does not look like sovereign textile manufacturing. It looks like an import-to-export assembly platform: Asian materials enter at one end, garments are cut, sewn and processed in Mauritian factories, and the finished product is shipped outward to foreign buyers.

That does not mean there is no value added. There is, and the workers inside those factories are doing real work under real conditions. But the question is what kind of value added, and where in the chain Mauritius sits. The island does not control the upstream raw-material supply. It does not own the technology frontier. It does not sit at the most commanding or profitable end of global demand. It occupies an intermediate processing position: necessary enough to function, but not central enough to set terms. The distinction matters analytically because the benefits of controlling a value chain and the benefits of participating in it are fundamentally different in scale and in sovereignty.

The labour story is perhaps the most analytically significant dimension of the diagnosis. By the fourth quarter of 2025, foreign workers represented nearly half of all EOE employment. In the garment sector specifically, the foreign share is estimated above 60 percent, with large contingents from Bangladesh, Madagascar and India. The industry that once helped absorb Mauritian unemployment and provide industrial pathways for a generation of workers now depends substantially on imported labour to remain operationally viable.

That shift is not an administrative footnote. It is structural evidence of a sector that can no longer sustain itself on domestic labour at prevailing wage levels and working conditions. If an industry cannot recruit or retain enough domestic workers without structural reliance on importing poorer workers from abroad, the question is no longer simply whether the sector is still producing. The question is what kind of industrial logic it is reproducing and who is bearing the human cost of sustaining it.

Mauritius has not merely lost local labour to other sectors with better terms. It has, in effect, had to internationalise labour vulnerability to preserve what remains of the garment base. That creates a moral dimension as well as an economic one. A sector built in the name of national development, national employment and Mauritian industrial capability now survives substantially by drawing in workers from poorer countries under conditions of structural dependence on employers for housing, immigration status and income continuity. That is not the profile of a self-confident industrial republic. It is the profile of a narrowing margin being defended through labour substitution.

The old textile model faces a simple and unresolvable geopolitical fact: Mauritius is no longer cost-competitive against the dominant and fast-rising producers of global apparel. Bangladesh, Vietnam, India and, in some categories, Madagascar operate with lower labour costs or stronger scale advantages than Mauritius can match. Ethiopia has entered the global garment supply chain with wage levels far below anything an island economy with Mauritius' cost structure can approximate. Once the era of quota-protected preferential advantage weakened, Mauritius found itself exposed more directly to the fundamental arithmetic of commodity garments: labour cost matters decisively, volume matters, speed matters, and producers who cannot move up the value ladder face sustained competitive pressure from those who remain cheaper at the same tier.

The sector has attempted to defend itself through quality positioning, reliability, niche market access and regional trade preferences. That has kept it alive, and there are genuine examples of successful upgrading in specific product categories. But keeping a sector alive is not the same as restoring its strategic centrality. The basic problem is that Mauritius moved out of the low-wage phase of garment production without fully securing the high-value phase. Costs rose, margins compressed, and sovereignty over the supply chain did not follow. That is the textile trap in its simplest economic formulation.

The country's export geography reinforces the structural dependence argument. South Africa is now the leading textile export destination at roughly USD 169 million, followed by Madagascar, the United Kingdom, the United States and France. That concentration matters analytically. It means the sector does not sell broadly into an autonomous global market under its own commercial power. It remains tied to a relatively narrow and specific geography of market access, much of it dependent on preferential trade arrangements that Mauritius benefits from but does not control.

AGOA for the United States, the EPA for the European Union, and regional access to southern African markets under SADC and COMESA frameworks have been critically important to the sector's survival and orientation. But preferences are not sovereignty. They are tolerated openings inside frameworks negotiated and governed by larger powers, subject to political conditions, compliance requirements and the right of those larger powers to revise the terms. A country that depends structurally on preferential market access remains externally disciplined even when its exports look strong in annual trade statistics. The garment sector has long been a beneficiary of global rule structures rather than a commander of its own industrial trajectory.

The most analytically significant recent event came not from price competition but from regulatory enforcement. In November 2025, U.S. Customs and Border Protection issued a Withhold Release Order on garments and apparel from Firemount Group Ltd, one of Mauritius' major denim manufacturers. The WRO cited multiple indicators of forced labour as defined under U.S. law, including abuse of vulnerability, debt bondage, deception and intimidation or threats. All goods from Firemount were presumed to have been produced using forced labour and detained at U.S. ports of entry until the manufacturer could demonstrate compliance.

That matters for reasons that go beyond the specific company and the specific order. First, it directly damaged Mauritian garment exports to the United States at a moment when AGOA access was already under review pressure across the continent. Second, and more fundamentally, it struck at the one ethical narrative through which Mauritius has consistently distinguished itself from cheaper rivals: the claim to be a rules-compliant, well-governed, reliable manufacturing partner with better labour standards than competitors operating at lower cost. Once a major U.S. enforcement action targets a Mauritian manufacturer for forced labour indicators involving migrant workers, that claim requires substantive defence rather than rhetorical assertion.

The Firemount episode should be read as diagnostic rather than anecdotal. It is what happens when a sector that can no longer compete cleanly on cost, that has lost its domestic labour base and replaced it with structurally vulnerable migrant workers, and that operates under intense margin pressure from buyers and competitors, encounters the consequence of those structural conditions in regulatory form. The WRO did not create the problem. It revealed it.

There is a further contradiction that deserves direct attention. Mauritius produces garments substantially for foreign buyers while continuing to import large volumes of textile goods and clothing for its own domestic consumption. The island exports apparel, yet Mauritian consumers rely heavily on imported garments, often from cheaper Asian producers, for their everyday clothing needs. Local garments are either too expensive, too export-oriented in product profile, or too disconnected from domestic mass market preferences to serve the population that lives on the island year-round.

This is not merely ironic. It reveals something fundamental about what kind of industry has been built. If locally produced garments cannot satisfy domestic demand at prices that Mauritians can afford, then the sector is not deeply integrated into national economic life. It is an export enclave with local employment and local factories, but without the domestic commercial integration that would make it a genuinely embedded part of the Mauritian economy. The republic manufactures for others while clothing itself from elsewhere. It bears the imported inputs, the assembly labour, the factory overhead, the compliance burden and the export vulnerability, while the domestic market continues to depend on supply chains that do not involve Mauritian producers.

The old story of Mauritian textiles was one of triumph over monocrop dependence. In one important sense, that story has real truth to it. Garments helped transform the economy, absorb labour at scale, diversify exports and build entrepreneurial and manufacturing capability that did not previously exist. The EPZ model genuinely changed what Mauritius was economically capable of producing. That contribution should not be dismissed or minimised.

But success under one set of global conditions is not the same as sovereignty under another. The same features that made garments a good entry point into manufacturing for a small post-colonial island, namely mobile production organisation, accessible technology, quota-protected market access and labour-cost advantage, also made the sector inherently fragile once any of those conditions shifted. A country that enters manufacturing through the low end of the value chain on the basis of temporary cost and access advantages needs to move decisively up-chain before those advantages erode. Mauritius did not move far enough or fast enough, and the sector's current condition is the consequence.

Today the sector remains: smaller than its history, more import-intensive than its narrative acknowledges, more dependent on migrant labour than its founders intended, more exposed to trade preference revision than its advocates admit, and more vulnerable to the kind of regulatory intervention that Firemount experienced than the official story of Mauritian governance competence would predict. Mauritius did not build a textile nation. It built a garment-export corridor that has been narrowing for two decades.

The hollowing out of Mauritian textiles was not simply the result of impersonal global forces. It was also the product of specific and identifiable domestic policy choices made in the period between approximately 1995 and 2000, when the window for a different outcome was still open. This was the decade in which rising Mauritian wages created the pressure that should have forced an industrial upgrade. In Taiwan and South Korea, comparable wage pressure in the 1980s and early 1990s had been used as exactly that: a forcing mechanism. Governments in those economies made the upgrade conditional rather than optional. Firms were required to automate, invest in technology and move up the value chain, or accept that they would be outcompeted and exit the sector. The discipline was deliberate, costly in the short term and transformative over the longer period.

Mauritius made three different choices in that window, each individually defensible in the short term and each collectively damaging over the longer period. The first was labour importation rather than technology investment. When local wages rose and factory owners sought relief, the government of the late 1990s opened access to imported migrant labour rather than requiring firms to automate and upgrade. That decision was understandable from the perspective of maintaining employment levels and avoiding factory closures in the near term. But its structural consequence was severe: by removing the wage pressure that would otherwise have compelled private sector investment in technology and higher-value production methods, the State eliminated the primary economic incentive for industrial upgrading. It subsidised stagnation and called it industrial policy. The foreign labour it imported to sustain the low-cost assembly model became the structural foundation of the migrant dependency the sector now depends upon to survive. The second choice compounded the first. Rather than investing the institutional effort required to build industrial depth, engineering capability or advanced manufacturing, the same period saw heavy policy commitment to the offshore financial sector and luxury tourism. Both were commercially rational from the perspective of foreign exchange and tax revenue. But neither transferred technology, built backward linkages into domestic manufacturing, or positioned Mauritius to own any commanding share of a productive value chain. The sectors that were strengthened in this window create lawyers, accountants and hotel managers rather than mechanical engineers and industrial designers. The easy revenue of regulatory arbitrage was chosen over the harder work of sovereign production. The third choice compounded both of the preceding ones. Free secondary education introduced in 1976 had been a genuine social achievement that created a literate, bilingual workforce well suited to the textile boom of the 1980s. By 1995, however, the world had entered the early phase of the digital economy and the workforce Mauritius needed had fundamentally changed in character. A decisive pivot toward science, technology, engineering and mathematics, combined with serious investment in technical institutes and research infrastructure, was required. That pivot did not happen with the urgency the moment demanded. The education system continued producing generalists and administrators suited to the service economy being built, rather than the technicians, engineers and coders required for a technologically competitive manufacturing sector.

Each of the three choices was individually defensible at the moment of decision. Keeping factories open, generating foreign exchange through financial services and providing broad literacy through secondary education were all reasonable short-term objectives. The problem is that each choice, when combined with the others, produced a structural outcome that was far more damaging than any one of them would have been in isolation. Together they locked Mauritius into a position in which the private sector had no incentive to upgrade its technology, the economy was orientated toward service intermediation rather than industrial command, and the workforce was systematically underequipped for the technical economy the world was entering. When quota-protected preferential access narrowed and cost competition intensified, all three of these structural weaknesses became simultaneously visible. Mauritius found itself with an export sector it could not upgrade, a service economy it could not quickly industrialise and a workforce it could not easily redirect. The textile trap was not only about garments. It was about the compounding cost of three decisions that, in the critical window of the mid-to-late 1990s, chose continuity over transformation.