entitled, but unenrolled

entitled, but unenrolled

Problem statement

Governments run thousands of social-protection programs — pensions, scholarships, maternity benefits, construction-worker funds, food and cash transfers — that poor and marginalized citizens are legally entitled to but never receive. The barrier is not eligibility but administrative burden: people don't know which of thousands of schemes they qualify for, the registration and claims processes are complex and document-heavy, and the state often holds the data to identify eligible citizens yet never reaches out to them. The result is that an estimated quarter of legislated benefits in India never reach intended recipients, even as the government allocates roughly $150 billion a year across 5,000+ programs — money budgeted for the poor that simply doesn't arrive.

Why this matters

This is one of the largest and least visible equity failures in development: the safety net exists on paper but is gated behind paperwork that the least-resourced people are least equipped to clear. The concrete cost shows up in programs like India's Building and Other Construction Workers welfare fund, where only ~40% of the corpus has been spent and fewer than half of estimated workers are even registered — leaving billions earmarked for some of the most precarious laborers unspent. Administrative burden falls hardest on exactly the people the programs target: those with low literacy, no reliable documents, no internet access, and no time to queue at offices. When entitlements go unclaimed, poverty that policy was designed to relieve persists, and public trust in the state erodes.

What’s been tried and why it hasn’t worked

The dominant fix has been digital self-service portals and eligibility checkers. They help the digitally literate but systematically miss the intended population, because "the most vulnerable are least likely to get online to check their eligibility or pay for a service" — so a tool meant to close the gap can widen it. Document-and-verify enrollment (proving identity, residence, income, occupation) imposes exactly the friction that excludes informal workers without formal paperwork. Awareness campaigns raise demand but leave the onerous application process intact. The deeper constraint is informational and coordinative: eligibility data sits in siloed government databases, the burden of proving eligibility is placed on the citizen rather than the state, and no single actor owns the job of proactively matching people to the benefits they qualify for. Promising "data-driven outreach" approaches (using records the state already holds to identify and notify eligible citizens, as Indus Action does) cut administrative cost dramatically but require cross-department data sharing and a redesign of who bears the burden of enrollment.

What would unlock progress

The reframing that unlocks progress is shifting from "apply and prove you qualify" to "the state identifies you and enrolls you" — using data the government already holds. Tools that match existing administrative records (ration, census, labor, land) against scheme eligibility rules, that deliver assisted (not self-service) enrollment through trusted local intermediaries, and that measure take-up rather than just spending would each move the needle. A single interoperable "entitlements interface" that lets a citizen see every benefit they qualify for, and a human helper to complete the claim, is the systems-level direction the field is converging on.

Entry points for student teams

A team could build a rules-based eligibility-matching engine for a defined population (e.g., construction workers in one district) that takes the kinds of records a local government already holds and outputs a ranked list of schemes each person likely qualifies for, with the documents needed. A service-design team could prototype an assisted-enrollment workflow run through a trusted intermediary (a community worker or kiosk) and measure completed enrollments, not just awareness. A data/policy team could quantify administrative burden for a single benefit — counting the steps, documents, and trips required — and design a reduced-friction version. Relevant disciplines: information systems, public policy, human-centered/service design, data science.