India’s Gig Worker Safety Net Is Still a Blueprint Who Pays Next
A delivery rider can spend the morning logged into one app, the afternoon on another, and the evening trying to recover from a sudden deactivation. The law now recognises that worker. The benefit system meant to protect that worker is still waiting for its starting gun.
In November 2025, India’s Code on Social Security came into force and gave gig and platform workers a formal place in national labour law for the first time. That was a major shift. Until then, platform work was often treated as a private contract between an app and an individual, even when lakhs of people depended on it as their main source of income.
The scale is no longer small or experimental. NITI Aayog counted about 77 lakh gig and platform workers in 2020-21 and projected the number could reach roughly 2.35 crore by 2029-30. That is larger than the population of many Indian states. It includes food delivery riders, cab drivers, quick-commerce pickers, beauty workers, home-service workers, logistics partners and many others who keep urban consumption running.
Almost a year after the Code came into force, the core question has changed. The debate is no longer whether gig workers deserve rights on paper. Public policy has crossed that line. The sharper question is who pays for a minimum floor of protection, when payments start, and whether the system will cover the risks workers face every day.
The law has moved faster than the benefits
The Code on Social Security matters because it creates legal categories for gig workers and platform workers. That is not symbolic. Once a category exists in law, governments can build schemes around it, collect contributions and set rules for registration.
Under the central rules that arrived in May 2026, aggregators must register workers. In time, they are expected to pay into a social security fund. The proposed contribution is 1 to 2 per cent of annual turnover, capped at 5 per cent of the amount paid or payable to gig and platform workers.
That formula tries to balance two aims:
Policy aim | What the rule tries to do |
Create a funding pool | Make platforms contribute to a social security fund instead of leaving protection to ad hoc insurance or goodwill |
Limit the burden | Cap the contribution so it does not exceed 5 per cent of worker payouts |
Link liability to scale | Use annual turnover so larger platforms contribute more |
Keep the model flexible | Leave the exact rate and timing for notification |
The gap lies in that last line. Reporting through late October found that neither the exact contribution rate nor the date from which payments begin had been notified. No scheme for life cover, health benefits or pensions had launched.
That means the legal frame exists, but the worker still cannot point to a clear benefit and say, “This covers me if I am injured,” or “This protects my family if I die on the job,” or “This creates long-term savings for old age.”
This matters because gig work carries an unusual mix of risks. Many workers use their own vehicles. They pay for fuel, repairs, mobile data, rain gear and loan instalments. They face traffic, heat, poor air quality, accident risk and income swings. If the app reduces incentives or changes the delivery radius, the worker’s earnings can shift quickly.
A data-driven view has to separate legal recognition from actual protection.
Legal recognition means the system knows these workers exist. Actual protection means money flows, schemes operate, claims get processed, and workers know what they can access.
Right now, India has made the first move. The second remains incomplete.
The National Social Security Board can recommend, but it cannot compel
On 28 October, the government constituted the National Social Security Board. Its composition reflects the political challenge of platform work. The Board includes five seats for platforms such as Eternal, Swiggy, Zepto, Urban Company and Rapido, and five for worker groups.
That balance matters. Platforms control access to work, pricing systems, incentives, ratings and deactivation tools. Worker groups bring evidence from the road, including accident cases, payment disputes and sudden account blocks. A board without either side would lack legitimacy.
Still, the Board’s legal power is limited. It can recommend schemes. Its recommendations do not bind anyone.
That distinction is central. A recommending body can identify problems, consult stakeholders and suggest designs. It can help the government decide how to spend a social security fund. It can advise on eligibility, portability and benefit types.
It cannot, by itself, force platforms to pay before the notified date. It cannot launch a pension scheme without government action. It cannot order a company to reverse a deactivation. It cannot create a binding minimum payout.
The result is a board with visibility, but not the final switch.
For gig workers India, Code on Social Security debates now sit at the centre of a larger question about platform worker rights, Swiggy Zomato riders and the direction of gig economy law. The issue is not only the wording of the Code. It is whether the rule-making process turns into money, claims and enforceable rights.
The Board’s mandate also leaves out some of the hardest issues in platform work.
It does not directly cover:
Minimum pay or per-kilometre compensation
Fuel cost sharing
Deactivation rules
Appeal systems
Waiting time
Algorithmic allocation of orders
Penalties linked to ratings or cancellations
Those exclusions create a split between social security and working conditions. The state may eventually provide a welfare scheme, while the platform may still decide pay structures and access to work with limited outside review.
That split may prove unstable. For many riders, the immediate crisis is not a pension twenty years from now. It is whether today’s fuel cost leaves enough after loan repayments and food expenses. It is whether an account can be blocked without a fair hearing. It is whether a bad week of orders pushes a household into debt.
Eligibility rules may leave out the most unstable workers
Even when benefits arrive, not every worker will automatically qualify. The rules require a worker to complete 90 days with one app, or 120 days across several apps, in the previous year.
On paper, this creates a basic link between active platform work and benefit eligibility. A public fund needs criteria. Without them, platforms and governments could face inflated claims or unclear liabilities.
On the ground, the rule raises practical concerns.
Gig workers often move between apps because earnings vary by time, location and incentive cycle. A rider may use one food delivery app in the evening, a quick-commerce app in the morning and a bike-taxi app during gaps. Some workers log in only during festivals, college breaks or periods of job loss. Some are pushed off an app through deactivation before they can complete the threshold.
The people with the most unstable work histories may be the hardest to cover.
Consider three common patterns:
Work pattern | Possible issue under the eligibility rule |
Worker depends on one app but gets deactivated after two months | May miss the 90-day threshold despite relying on platform income |
Worker splits work across two or three apps | Needs records across platforms to prove 120 days |
Worker works seasonally or during income shocks | May fail to qualify even if the risk faced during work was real |
The design challenge is not simple. A fund cannot cover every person who ever opened an app. Yet a rigid threshold could exclude workers who need support because their platform work is irregular, unstable or interrupted by the platform itself.
Portability will be crucial. If a rider works across several apps, the benefit system must track work across those apps in a reliable way. A worker should not lose social security because no single platform sees the full picture.
Registration also has to be worker-friendly. If the process depends on unclear app records, mismatched names, old phone numbers or missed data uploads, the safety net could become hard to access. Many welfare schemes fail not because the idea is weak, but because the last-mile paperwork blocks the people it was meant to help.
The eligibility threshold also interacts with deactivation. If an app can remove a worker before the qualifying period, and there is no fair appeal system, the worker may lose both income and future benefits. That creates a strong case for linking social security rules with basic process rights.
The biggest daily conflicts sit outside the current safety net
The platform work debate often gets reduced to one question, employee or contractor. India’s current approach has taken a different route. It recognises gig and platform workers without fully converting them into regular employees.
That may allow a wider safety net. It also leaves key workplace questions unresolved.
A Telangana union told TechCrunch it handled 25 to 30 deactivation cases a week. That number matters because deactivation is the platform economy’s version of dismissal. For a worker who depends on one app, losing access can mean losing the week’s income instantly.
The problem is not only that deactivations happen. Platforms have legitimate reasons to act against fraud, safety violations or misconduct. The problem is process.
A fair system would need basics such as:
A clear reason for deactivation
Evidence that the worker can review
A time-bound appeal
Human review where livelihoods are affected
Protection from automatic or mistaken account blocks
Records that can be used if the worker challenges the decision
Without those steps, deactivation becomes an income shock with no labour-law style remedy.
Pay is another live issue. In May, riders staged a five-hour national shutdown over fuel costs and demanded a minimum payout of ₹20 per kilometre. That demand reflects a simple cost structure. When fuel prices rise, the worker’s cost rises immediately. If base pay or incentives do not adjust, the platform may still show order growth while the rider’s net income falls.
Gross earnings can mislead. A rider may receive a payout, but net income depends on:
Fuel
Vehicle maintenance
Mobile data
Insurance
Loan instalments
Time spent waiting for orders
Time spent reaching pickup points
Penalties, cancellations or unpaid returns
A small change in per-order pay can have a large effect after costs. That is why many workers focus on per-kilometre payment rather than only daily gross earnings. Distance is a direct cost driver.
The current national social security framework does not settle this. It aims to build a welfare fund, not a wage code for platform work. That leaves a gap between risk protection and income protection.
Both matter. Accident cover helps after harm occurs. Health cover may reduce financial stress during illness. Pension support could help with long-term security. But if everyday earnings remain too low after expenses, workers will still face pressure long before any benefit claim is filed.
States are testing models while the Centre moves slowly
The central framework sets the national direction, but states have started moving faster.
Karnataka’s law includes a welfare fee on payouts. In July, the High Court did not stay it. Telangana passed its own bill in 2026. These state-level moves matter because they may create working models before the national system becomes fully operational.
A welfare fee on payouts takes a different path from a turnover-linked central contribution. It connects the funding source directly to platform transactions or worker payments. That may make collections easier to track, depending on how the rules are written and enforced.
State action has three advantages.
First, states are closer to local labour markets. Bengaluru, Hyderabad, Delhi-NCR, Mumbai, Pune, Chennai and Kolkata have different platform densities, traffic conditions and worker organisations. A state can respond to specific patterns faster than a national scheme.
Second, states can experiment. One state may test a welfare board model. Another may focus on fees, accident cover or dispute systems. Over time, the stronger designs can influence national rules.
Third, state-level laws can create pressure. If platforms face different welfare obligations across states, they may prefer a clear national standard. That gives the Centre a reason to notify rates and schemes sooner.
There is also a risk. Too many different state rules can create confusion. A worker who moves between cities may not know which benefits apply. A platform may comply better in one state than another. Smaller platforms may struggle with multiple reporting systems.
The best outcome would combine national portability with state-level responsiveness. A worker’s ID, contribution record and benefit eligibility should travel across states and platforms. At the same time, states should retain room to address local costs and disputes.
This is where India’s existing welfare experience offers a lesson. A scheme can look strong on paper and still fail if registration is patchy, claims are slow or workers cannot understand the process. Gig worker welfare will need real-time data from platforms, simple worker access and clear accountability when records are wrong.
The funding question is the policy fight
The hardest question is who pays.
Platforms argue, directly or indirectly, that high mandatory contributions can raise costs, reduce flexibility and affect prices. Some may say consumers will pay more. Some may say workers could receive fewer incentives if compliance costs rise. These claims should be tested with data, not accepted as fact.
Workers and unions argue that platforms built large businesses by shifting costs and risks downwards. The app creates the marketplace, sets prices, controls access and measures performance. The worker provides the vehicle, fuel, time and physical labour. If the business depends on that labour, the business should contribute to the safety floor.
The proposed central formula tries to avoid an all-or-nothing answer. It does not impose full employee-style social security. It asks aggregators to pay a limited share into a fund. The cap at 5 per cent of worker payouts is meant to prevent open-ended liability.
Still, the rate matters. At 1 per cent of turnover, the fund will collect one amount. At 2 per cent, it will collect twice that. The start date matters even more. A rate with no notified start date raises no money.
The design also needs clarity on what counts as turnover, how payouts are calculated, how refunds or discounts are treated, and how multi-service platforms report income from different business lines. The public does not need every technical detail in a headline, but regulators do. If definitions are loose, contribution disputes can delay the fund.
There is another point. Social security is often cheapest when it starts early. Small regular contributions can support pooled benefits. Delayed protection shifts costs to workers, families, hospitals, charities and state relief systems after crises occur.
The driver with a bike on EMI does not have the same shock absorber as a billion-dollar app. A week without income may mean a missed instalment. An accident may mean debt. A blocked account may mean borrowing at high interest. That imbalance is why a statutory floor exists in the first place.
What a real safety net would need to prove
A working safety net should be judged by outcomes, not announcements.
The first test is registration. How many workers are registered, across which platforms, and how often is the data updated? If NITI Aayog’s projection points to crores of workers by 2029-30, the system cannot rely on slow manual entry or fragmented app records.
The second test is funding. Has the government notified the contribution rate? Has it notified the payment start date? Are platforms paying into the fund? Are collections visible at an aggregate level?
The third test is benefit design. Workers need to know what they are covered for. Life cover, health cover and pensions are different benefits with different rules. A vague “welfare” promise is not enough.
The fourth test is claims. A benefit matters only if workers can file, track and receive it without months of confusion. Claim rejection reasons should be clear. Appeals should exist. Assistance should be available in local languages.
The fifth test is portability. A rider should not lose protection because they moved from one app to another, or from one city to another. Platform work is already fragmented. The welfare system should not make it worse.
The sixth test is grievance handling. Even if the National Social Security Board does not cover deactivation and pay, the wider policy system cannot ignore them. A worker who loses access to work also loses the path to eligibility. A worker whose net pay falls below costs may never reach the point where long-term benefits matter.
A credible framework would connect at least three layers:
Layer | What it should cover |
Social security | Life cover, health support, pensions and other welfare benefits |
Income floor | Transparent pay rules, cost recognition and fair treatment of waiting time or distance |
Process rights | Deactivation appeals, clear notices and access to records |
The current national framework has started with the first layer. States and worker groups are pushing harder on the second and third.
What workers, platforms and governments should watch next
A year after recognition, the next phase depends on specific notifications and scheme design. Broad promises will not settle the issue.
The key signals to watch are clear.
The notified contribution rate
The law provides a 1 to 2 per cent range. The chosen rate will show how ambitious the first national funding pool is.
The payment start date
No date means no contributions. This is the most practical marker of progress.
The first benefit scheme
Life cover, health support and pensions solve different problems. The first scheme will reveal the government’s priority.
The registration mechanism
If registration is automatic, accurate and portable, coverage can grow fast. If it is confusing, many workers may remain outside.
State-level enforcement
Karnataka and Telangana could shape the next stage. Their laws may show whether welfare fees and boards can work in practice.
Dispute systems
Even if they sit outside the current Board’s functions, deactivation and pay disputes will keep driving worker mobilisation.
Young urban consumers also have a stake. Cheap delivery is not free. Some cost is paid through investor money, some through platform margins, some through restaurant or merchant commissions, and some through worker earnings. A serious safety net makes those costs more visible.
That visibility may create discomfort. It may raise questions about delivery fees, platform charges or business models. But hidden risk does not disappear. It travels to the person with the least bargaining power.
The blueprint now needs a date, a rate and a floor
India has crossed an important legal threshold. Gig and platform workers now exist in national social security law. That matters for a workforce counted in lakhs today and projected in crores by the end of the decade.
The next step is harder. The Centre must notify the contribution rate, set the payment start date and launch actual schemes. The National Social Security Board can help design the path, but recommendations alone will not insure a rider or pay a claim. States are already testing faster routes, and their experiments may shape the national model.
The policy choice is now concrete. Either the safety net remains a blueprint while workers carry fuel costs, accident risk and deactivation shocks on their own, or India builds a funded floor that matches the scale of the platform economy.
The app may be backed by deep capital. The worker may be paying off a bike on EMI. A fair system has to start from that imbalance. Recognition was the first line in the law. Payment is the line that will decide whether the law reaches the road.
