Writing a thesis on living wages as a legally enforceable human right taught me something: the further you dig into the law, the more you realise that law alone cannot explain whether people actually get protected. You can have the rights, you can have the right conventions ratified, you can have the right institutions on paper, and still, we have millions of workers with no meaningful floor beneath them. That gap between recognition and reality is what pulled me beyond what I had researched: what does a safety net actually need to look like to hold?
The scale of the problem:
According to ILOSTAT (2025), approximately two billion workers, around 58% of the global workforce, are currently employed informally, outside the reach of the law, social insurance, or protections. These workers are often not in transition to something better. Increasingly, they are the permanent condition of how global supply chains are organised.
The garment worker on a seasonal contract in Bangladesh, the agricultural labourer in Kenya paid by the harvest, the platform driver in Jakarta who bears all the risk and owns none of the platform – this is the model we allowed to be built.
Social protection systems were not designed for them. The twentieth-century welfare state was built around a set of assumptions that made sense at the time: stable, full-time, formal employment; a single dominant employer who could be made responsible for contributions; a household with a male breadwinner and a dependent spouse; and a nation-state with the administrative capacity to collect and redistribute.
Those assumptions are breaking down, and the safety nets built on top of them are breaking with them. Keynes built his framework for a world of full employment and stable demand, but I am positive he would have had something to say about the globalisation of supply chains nowadays. On his place, Guy Standing named it the rise of “the precariat“, a class defined not by income level but by the structural absence of labour security, voice, and protection (Standing, 2011).
Protections that do not follow:
Let’s take a closer look to Ethiopia. The country’s industrial park strategy, centred on Hawassa Industrial Park, was designed to replicate East Asia’s export-led development model by attracting garment manufacturers with cheap land, logistics infrastructure, and a young, low-cost labour force.
At peak capacity, the park employed around 28,000 workers across 20 firms producing for major Western brands (LSE, 2024). These were, technically, formal jobs. They came with contracts and employment relationships. What they did not come with was a minimum wage. Ethiopia has no private-sector minimum wage, a structural choice the government marketed to investors as affordable labour. Entry-level garment operators earned around $32 to $50 per month, approximately one quarter of the minimum wage in Bangladesh, and well below any credible living wage benchmark (Capital Newspaper, 2025; Sahley et al., 2024).
The fragility of this model became visible with two successive shocks. COVID-19 caused a 40% drop in Hawassa’s production volumes, and nearly half of workers surveyed in mid-2020 had been placed on leave or dismissed without meaningful recourse (World Development, 2023). Then, in 2022, the United States suspended Ethiopia’s trade preferences in response to human rights concerns related to the Tigray conflict, prompting major buyers to exit.
More than 11,000 workers lost their jobs; the majority, women who had migrated from rural regions with limited options for return (IndustriALL, 2024; LSE, 2024).
The ILO’s Siraye programme, evaluated in 2024 by researchers from SOAS and King’s College London, drove measurable gains in unionisation, occupational safety, and social dialogue inside the parks. But its own evaluators noted explicitly that the absence of a minimum wage remained a fundamental vulnerability for the sector’s long-term sustainability, a floor so basic that global framework agreements between brands and unions now treat it as a threshold condition for responsible sourcing (Better Work, 2024).
Ethiopia illustrates a pattern that is easy to miss when formal job creation is the primary metric of development success: formal employment without a protection floor is not the safety net we think it is.
The challenge is similar in Indonesia. Where Ethiopia’s problem is the absence of any floor for workers in formal manufacturing, Indonesia’s problem is the near-total failure of existing social insurance to reach informal and platform workers. It is estimated that approximately 84.2 million Indonesians, around 60% of the active labour force, work informally, including an estimated 41.6 million gig workers (Kantar, 2024).
Asia’s gig workers strive to match global gains on labour rights https://sightmagazine.com.au/news/asia-s-gig-workers-strive-to-match-global-gains-on-labour-rights
The country has a functioning employment social security system, that offers accident insurance, death benefits, and old-age savings. The problem is coverage. Fewer than 8% of informal workers are currently enrolled (IJMRA, 2025).
Social insurance was designed to be collected through payroll, deducted at source from a predictable monthly salary. The income of platform workers is fragmented, volatile, and platform-mediated. Ride-delivery platforms such as Gojek and Grab have explored direct-debit contribution models, deducting modest amounts from driver e-wallets, but as of the most recent available data, pension participation among platform workers remains negligible (Green Network Asia, 2024)
A government-platform partnership was adding roughly 7,000 new BPJS participants per month as of recent reporting, less than 0.3% of Gojek’s driver base alone (IJMRA, 2025). The World Bank identified over a decade ago that there is no contribution-collection infrastructure suited to volatile, fragmented income (World Bank, 2023). The draft “Dependent Worker” classification of 2023 in Indonesia could have brought over a million freelancers into the system with flexible contributions, but as of 2024, the government has preferred non-binding guidelines over binding reclassification law (Green Network Asia, 2024).
These cases are part of a global pattern that the ILO’s World Social Protection Report documents; most workers in supply chain economies operate in a space where neither the formal employment model nor the existing social insurance architecture was designed to reach them (ILO, 2024).
What’s coming:
The regulatory reclassification approach, pursued through Spain’s 2021 Rider Law and the EU Platform Work Directive, extends formal employment status, and therefore existing protections, to platform workers by shifting the burden of proof around their employment classification. Another approach, championed by Guy Standing and the Basic Income Earth Network, argues that the fragmentation of income has become so structural that only a universal, unconditional income floor, funded through dividends on natural resources, data, and intellectual property, can match the reality of how most people now work (Standing, 2023).
However, we need to note that these approaches carry different assumptions about what the underlying problem is. Reclassification assumes the problem is corporate evasion of existing law. Basic income assumes it is the wage-labour model itself. They are not mutually exclusive, and the most credible policy trajectories combine elements of both.
This is where my thesis left me with more questions than it answered. Social protection is not welfare for when supply chains fail; it should be a precondition for supply chains. When workers in a garment park have no minimum wage and no social insurance, their vulnerability to order volatility is total, and the reputational and operational risks that transfers to brands and sourcing companies are real and growing, particularly as due diligence legislation in Europe begins to require companies to account for labour conditions across their value chains.
The insight that runs through the most recent ILO, World Bank, and academic literature seems simple, yet demanding. The unit of social protection must shift from the job to the person (ILO, 2024; World Bank, 2023; Standing, 2023). Every policy, every programme, every sourcing standard that treats employment status as protection is working with a twentieth-century scenario in a twenty-first-century landscape.
That is the question my thesis left open, and the one I keep returning to. Whether the architecture around labour rights can hold in a world where most work no longer fits in the model it was built for.
References:
Better Work. (2024). Q&A: Evaluating impact in Ethiopia’s garment sector. https://betterwork.org/qa-inside-the-siraye-programme-evaluating-impact-in-ethiopias-garment-sector/
Capital Newspaper. (2025, May 6). Textile and garment sector faces policy gaps, urgent calls for minimum wage reform. https://capitalethiopia.com/2025/05/06/textile-and-garment-sector-faces-policy-gaps-urgent-calls-for-minimum-wage-reform/
Green Network Asia. (2024). Social protection for Indonesia’s gig economy workers. https://greennetwork.asia/gna-knowledge-hub/social-protection-for-indonesias-gig-economy-workers/
IJMRA. (2025). The gap in social security participation of informal workers in Indonesia. https://ijmra.in/v8i6/Doc/1.pdf
International Labour Organization. (2024). World social protection report 2024–26: Universal social protection for climate action and a just transition. https://www.ilo.org/sites/default/files/2024-09/WSPR_2024_EN_WEB_1.pdf
ILOSTAT. (2025). ILO modelled estimates database. https://ilostat.ilo.org/methods/concepts-and-definitions/ilo-modelled-estimates/
IndustriALL Global Union. (2024). Job losses wreck livelihoods in Ethiopia’s garment industry. https://www.industriall-union.org/job-losses-wreck-livelihoods-in-ethiopias-garment-industry/
Kantar. (2024). Uncovering the world of gig workers in Indonesia. https://www.kantar.com/Campaigns/APAC/Uncovering-the-World-of-Gig-workers-in-Indonesia
LSE. (2024, January 30). Ethiopia’s industrial parks leave workers to the fate of global shocks. https://blogs.lse.ac.uk/africaatlse/2024/01/30/ethiopias-industrial-parks-leave-workers-to-the-fate-of-global-shocks/
Sahley, C., Tadesse, T., & Dereje, M. (2024). Industrial promises, employment precarity, and disrupted production in the shadow of global pandemics. Discover Sustainability, 5, Article 468. https://link.springer.com/article/10.1007/s43621-024-00468-z
Standing, G. (2011). The precariat: The new dangerous class. https://library.oapen.org/bitstream/handle/20.500.12657/58691/9781849664547.pdf
Standing, G. (2023). The politics of time: Gaining control in the age of uncertainty. Penguin. https://www.penguin.co.uk/books/320800/the-politics-of-time-by-standing-guy/9780241475928
World Bank. (2023). Social protection for online gig workers (Chapter 6, Working Paper). https://documents1.worldbank.org/curated/en/099071923115632890/pdf/P1773020bb7f7c0c50b5f202234fbea3e8e.pdf
World Development. (2023). The impact of firm downsizing on workers: Evidence from Ethiopia’s ready-made garment industry. World Development, 171, 106368. https://www.sciencedirect.com/science/article/abs/pii/S0305750X23002309
Leave a Reply