Global Campus of Human Rights | Education & Research

FULL MENU

What the label doesn’t say: a Global Campus alumnus maps child labour in Madagascar’s vanilla and centella trade

Vanilla and Centella asiatica are unglamorous words for glamorous industries. The first flavours ice cream and perfume; the second, a herb prized for its skin-soothing extract, turns up in the serums and creams that line pharmacy shelves the world over. Few buyers of either product think about who picks the pods or the leaves. GC Alumnus Rado Harintsoa Rakotosamimanana has spent the past several months making sure that companies, at least, have no excuse not to.

Rado, a human rights lawyer and graduate of the Global Campus’s LLM in Human Rights and Democratisation in Africa (Centre for Human Rights, University of Pretoria), has produced a research report and a companion toolkit on child labour in Madagascar’s vanilla and centella supply chains, developed under the Global Campus Alumni’s 2025–26 Cross-Regional Alumni-Led Activities. The two outputs are practical siblings: one diagnoses the problem with as much rigour as the evidence allows; the other hands auditors, cooperatives and buyers the forms to do something about it.

 

 

The diagnosis resists sounding more alarming than the facts warrant. Vanilla, grown in Madagascar‘s north-eastern SAVA region, is unusually well documented: an ILO/USDOL baseline survey found that nearly a third of sampled children had worked in the preceding year, mostly in agriculture generally rather than vanilla specifically, and that two-thirds of working children were in conditions classified as hazardous—heavy loads, unsupervised machetes, scalding water used to cure beans. Centella is a case study in the limits of sourcing-side evidence: no household survey exists, only picker and buyer reports pointing to the low per-kilogram pay that pulls children into collection elsewhere. The report refuses to blur that distinction.

Being able to say that vanilla has a rigorous commodity-specific survey and centella does not—and that this asymmetry should shape where practitioners direct resources—is more useful than a document that treats all commodities as equally documented when they are not,” Rado told us.

 

That discipline extends to his own fieldwork. An interview with a labour controller from a regional child-labour committee in Moramanga, and a situation note from two centella-collection sites, are presented not as proof of a national pattern but as single, attributable testimony—useful precisely because it is honest about its limits. It is also, in places, startling: only one child-labour case from the region has ever reached a tribunal, and retail businesses have reportedly presented child workers as “dependants” to sidestep labour inspection.

 

The toolkit turns this into something a field officer can actually use: a risk-screening template, minimum supplier expectations, an age-verification and safe-recruitment checklist, and a referral pathway for suspected cases—each insistent that no child be punished, and no family abandoned, once a case is found.

 

 

The full research report and toolkit are available from the Global Campus of Human Rights Repository: https://doi.org/10.25330/3214