Serving migrant labour interests in India through data stewardship

By The Data Economy Lab Team

November 9th, 2020

Image sourced by

This article is based on a paper by Aditi Ramesh, with research assistance from Preethi Sunderrajan, and has been summarized here by Aapti Institute.

With the bulk of its economic activity centered around metros, India sees millions migrate from its villages to metropolises to earn a living, often in the form of daily wages for manual labour. Part of the unorganised sector, they often work without job security and for exploitative wages. The nationwide lockdown in the wake of the COVID-19 pandemic in March this year compounded these systemic issues into an existential threat for this vulnerable group. In this  paper, Aapti Institute discusses how the figure of a data steward can direct the value extracted from migrant labourers’ data towards the welfare of the group in a rights-respecting manner to aid in resolving potential crises such as these.

This approach envisages a data governance framework where a data steward enables data sharing in an accountable and privacy preserving fashion, bringing together data from silos of civil society, NGOs, government surveys, and private players. This can then be used in humanitarian response in response to a crisis, guiding the deployment of resources for relief. This framework would ideally ensure that the data steward is free of commercial interests, has fiscal independence, and is transparent and accountable to the public. In addition to adhering to data protection principles such as purpose limitation and data minimisation, a data steward in such a framework could also serve the community by acting as an intermediary on their behalf to bargain for collective data rights, ensuring that the power of self-determination remains firmly with data subjects.

We have seen varying approximations of such a model in action internationally. One is the Humanitarian Data Exchange (HDX), owned by the United Nations Office for the Coordination of Humanitarian Affairs. On the HDX platform, a group of 250 organisations can share data, while a further refined and select set of organisations has the privilege of uploading data directly onto the platform itself. The data as well as the organisations they come from, are vetted and approved by the HDX. Data must be adequately aggregated and anonymised and must meet HDX’s standards of being “humanitarian” data that pertains to a crisis, its context, and the organisations working on said crises. The data can then be used by policy makers, relief organisations, researchers etc. Another example is the International Organisation for Migration’s (IOM) Global Migration Data Portal (MDP). Here, the IOM acts as a quasi-data steward in collating data from its 173 member states on 70 migration indicators, checking data for inconsistencies and maintaining confidentiality.

The Indian context, needless to say, would require a specifically tailored response suited to its everyday functional exigencies. For starters, both the quality and quantity of data on Indian migrants is lacking. In the specific case of the pandemic-induced migrant crisis, the federal ministry of labour and employment itself admitted the absence of data “maintained by the government on the number of migrant deaths that occurred due to restrictions imposed by the Covid-19 lockdown.” Private sector and civil society responses to the crisis at the time, like Facebook’s Data for Good or India Observatory’s CoAST India, used limited means and ad hoc data to deliver insights on movement of migrant population, map relief centers, and highlight the need for transport.

At a more general level, existing databases are insufficient for a data-informed response to a crisis. To start with, official estimates on the number of migrant workers in the country range as widely from 2.6 million to 80 million persons. Siloed databases, and varying ways of defining and registering workers have led to such discrepancies. Only 25% of India’s unorganised workers are registered under the Unorganised Workers’ Social Security Act, 2008. The Socio-Economic and Caste Census of 2011 concerns itself with limited sectors of work such as domestic help, construction, sanitation etc, leaving the picture incomplete.

Much of the digital data gathering on migrants in the COVID-19 crisis depended on location data on phones — a commodity one cannot assume the population group to possess across the board. Moreover, there are concerns regarding privacy violations, lack of accountability, and commercial bias with data collected in this manner. Techno-solutionist approaches in this case, have often failed to account for those who do not have access to technologies.

A rights-respecting, transparent and regulated data sharing mechanism between Indian labour unions, NGOs, civil society and government coupled with the standardisation and accountability measures as observed in the international models of the HDX and IOM could help pave the way for an Indian data stewardship model. This would represent migrant labourers’ interests and serve their needs, whilst also bringing them to the table to collectively bargain for their interests.