During the second wave of COVID-19, tens of thousands of migrant workers were abandoned and left to their own devices. Through empty bus depots and highways, many attempted to reach their home state on foot. The workers were from Bihar and Jharkhand who had left their homes and travelled to Surat, Delhi, and Ahmedabad in search of work. These scenes exploded on social media, allowing the country to see, for the first time, the large invisible workforce that keeps the factories running—workers who were present but had never been noticed. Even more surprising was that the state itself had insufficient information on where these workers lived, where they worked, or who they worked for. .
This is not a random knowledge gap, but a systemic problem. India’s internal migrant workforce, which is often seasonal and crosses state boundaries, has never been accounted for by any official system that could track their movements, match their entitlements, or document occupational injuries. The e-Shram portal was launched in 2021 as a national registration portal for unorganised workers, as per Press Information Bureau (2025), and it has registered over 30 crore workers. However, it does not track migration in a real sense: such as where these registered workers went, what they did there, whether they returned, or the difficulties encountered during migration.
A registry is not a replacement for a migration-aware database. The country needs more than a list of names; it needs a system that can dynamically link work, movement and entitlements across state lines, sectors and time.
The Scale of What is Being Missed
The census of 2011 recorded about 450 million internal migrants in the country, the number of which has further increased significantly, as recent estimates suggest (Rajan & Nizam, 2025). The Economic Survey 2016-17 estimated that about 9 million people migrate every year for work between states (Government of India, 2017). These estimates, however, are contested because there is hardly any system to collect reliable data on circular migration. There are also inconsistencies in recording workers who are leaving construction, agriculture, brick kilns or garment manufacturing in October and returning in April, either at the point of origin or destination.
This is an important gap as the current administrative model is not geared to deal with seasonal and circular migration. A worker registered under e-Shram in Sitamarhi, Bihar, can work for eight months in the factories of Surat, return to his village during the monsoon and then go to another city for work. The registration is recorded by the portal, not the journey. The state only sees the name, not the move.
Portability as a Policy Failure
India’s social protection systems, from the Public Distribution System to Ayushman Bharat, the Building and Other Construction Workers welfare boards to the Employees’ State Insurance Corporation (ESIC), are largely designed for fixed locations and stable jobs. Ration cards in most states are still linked to one’s home district, though there has been some progress under the One Nation One Ration Card scheme. Healthcare access presents a similar challenge for migrant workers. Although Ayushman Bharat has expanded health insurance coverage, access remains uneven and medical expenses continue to push households into debt. Data shows that 92 per cent of rural hospitalisations still involve out-of-pocket expenditure. (The Print, 2024).
Workplace injury compensation is largely connected to the place of employment, which for migrant workers is often informal, undocumented and fluctuating. And many times, there is no compensation for injuries. The weak enforcement of labour protections is evident more broadly as well: the International Labour Organization’s India Employment Report 2024 found that nearly 70 per cent of casual construction workers do not receive the prescribed minimum wage (International Labour Organization, 2024).
Studies of migrant welfare have consistently shown that access to ration, healthcare, insurance, and compensation falls off precisely when it is needed the most—in the event of injury, illness, accidents, or death, and when the worker has moved away from their support networks, but has not gained comparable access at their new location (John et al., 2020).
The gap between data and portability is two sides of the same coin. The current system does not track where an individual is working, their industry, or their welfare coverage. This lack of granular information makes it impossible for national or state governments to deliver social protection that moves with the worker. Welfare programs are still tethered to a geographic location even as workers are not, leaving a dangerous protection gap.
When Workers Die Without a Record
Migrant workers are highly represented in hazardous sectors like brick kilns, road infrastructure, construction, small manufacturing, logistics, etc. Workplace injury and death rates in these sectors are significantly underreported for all workers, but the issue is even worse for migrants (Work Fair and Free Foundation, 2022). This is evident in workplace safety data.
If a worker is injured at a construction site in Maharashtra, a hospital may document a trauma case. However, if that worker is then transferred to another facility, goes home for treatment, or dies on the way, the chain of administrative records breaks. The injury may show up in hospital records without any classification linked to their job. It may appear in police records without any connection to the employer. The labour department in the destination state may have no record at all. The source state where the worker’s family lives and where any compensation claim would need to start likely knows nothing about the incident.
The Lancet Journal pointed out that occupational health data of migrant workers in South Asia are fragmented across agencies that do not exchange data (Abubakar et al., 2018). The Migration Story, referencing a Safe in India report on migrant workers in the Indian auto industry, found that processing claims of injured migrant workers can drag on for years, sometimes leading to low settlement payouts or none at all, especially since fundamental pieces of information (like the identity of the employer, the workplace, and date of injury) may not be recorded anywhere (The Migration Story, 2025).
This statistical invisibility is more than just bureaucratic laziness—it leads to a failure of accountability. If data does not exist, companies will not be accountable, welfare councils cannot offer compensation, and the state will fail to highlight the highest-risk sectors and regions. The forgotten workers are even more forgotten, and the absence of data can be a conscious policy decision.
What a Migration-Aware Database Would Actually Require
How can, then, the e-Shram platform better serve the country’s migrant workforce? The current platform follows a static one-size-fits-all approach where once an informal worker registers, they are issued a unique ID that is then linked to a set of social security schemes. This works well for those in stable informal jobs, but is ineffective for those in the migratory, or circular, informal economy. Migrant workers move; their work patterns are fluid, so a migratory-sensitive database would have to be structured differently.
At a minimum, it would need to view migration as a changing factor, not a one-time label. Each worker’s record should be updatable to include their source district, destination district, current sector, type of occupation, employer or contractor, and duration of stay. This is not an impossible technical challenge; India has successfully run real-time payment systems on a large scale, be it the Unified Payments Interface or the Aadhaar, and there is no reason why it cannot be applied to labour mobility.
The database also needs to be portable: a record linked to a worker would have to seamlessly link to Public Distribution Scheme entitlements, health coverage, construction worker welfare boards, ESIC, maternity benefits, and accident insurance, regardless of the worker’s current location. It would need ways of updating records—an employer, a hospital, a local government authority or the worker could update a worker’s place of residence and employment status. It also needs fields to log accidents and deaths, so that a workplace accident can be appropriately cross-referenced to a worker’s record, even if the accident happened in another jurisdiction, and mapped to a compensation scheme.
Privacy should also be built into the design. The design should be oriented toward safeguarding welfare and labour rights, not towards monitoring migrant movement. Migrant workers already navigate an environment of distrust. Local governments in destination countries have been known to have used administrative means to restrict movement or enforce exclusions. Migrant workers will not be prepared to participate in a registration scheme if they fear that it is being used to track their whereabouts rather than safeguard their rights. Such designs should feature role-based access to sensitive location data, limitations on the number of persons with access to such data, plain and transparent consent mechanisms, data retention, and other policies. If it does not, it risks being both ineffectual and abusive.
The Federal Coordination Problem
In India, the Centre and States both play a role in governing labour. This has resulted in fractured policymaking around internal migration for a while now. Source states look after the families, destination states look after the migrant worker, and no one has a bird’s eye view of the situation. For years, destination states and source states, like Jharkhand and Odisha, have established bilateral labour protection mechanisms, allowing states to engage with each other and provide protections for their migrant workers, yet operate without systematic data to track where workers are going, what they are doing, or what happens to them.
While a national migrant database won’t fix this rift, it would create a common ground: a shared database, which would enable source states to track the welfare of their employees, and destination states to find coverage gaps. This calls not for a central hub from the national government but for a standard for state-to-state data sharing. This coordination can and should be handled by the Ministry of Labour. States must report into a common system, and aggregate data should be publicly available for planning.
It’s not as if a precedent does not exist. MG-NREGA (now rechristened as Viksit Bharat-Guarantee for Rozgar and Ajeevika Mission) already proves that a nationally orchestrated policy with state-level execution can account for each worker’s entitlements through time. Health entitlements have been made portable across state borders with Ayushman Bharat, albeit with its own shortcomings. A new worker’s database equipped for migration does not need to be built in silos. What is needed is an expansion of the existing systems, making them dynamic and publicly funded as infrastructure, not as a one-time registration exercise.
So, if technology is not the bottleneck, then what is? If India can manage to monitor each financial transaction and each biometric identity of its citizens, it is fully capable of tracking worker entitlements and their movement with the same accuracy. The crux of the issue boils down to a choice of priority – Are migrant workers seen as resources to be managed by the economy, or as citizens with rights that the state must protect, no matter where they work?
As long as there’s a gap between policy and practical reality, migration policy will remain haphazard. We may see a new portal rolled out, new agreements struck up, new relief schemes rolled out in times of crisis—but the fact of the matter is, many workers will still not be able to get assistance when they need it most. Whether it is compensation after an accident, a welfare claim, or the registration of a death, the mechanism often breaks down at the point of delivery. In the absence of concrete and factual evidence about both the destinations and the reasons for migration, policies targeted to migrants often fail to reach the beneficiaries they are intended for.
