When Digital ID Becomes a Wall: The Hidden Costs of Biometric Governance

Digital identity systems are not neutral infrastructure. They are gateways, and like any gate, they decide who gets through and who stays shut out. A digital ID is supposed to be a key—a way to prove who you are so you can open a bank account, collect benefits, cross a border, or cast a ballot. But for millions of people, the key doesn’t fit. The system wasn’t built for their hands. When designers ignore the messy realities of life on the margins—no fixed address, no birth certificate, no smartphone, no stable fingerprints—the gate slams shut. This isn’t a bug. It’s a predictable result of choices that prize administrative tidiness over human complexity.

Governments and global institutions have poured billions into digital identity, promising to cut fraud, speed up services, and pull the unbanked into the formal economy. The pitch is polished. The reality on the ground is rougher. For refugees, unhoused people, domestic violence survivors, people with disabilities, and those living in deep poverty, a digital ID system often becomes just another locked door—one that replaces a human being who might have listened with a screen that simply says no.

When Identity Becomes a Gatekeeping Mechanism

Strip a digital ID system down and you’ll find a set of rules about what counts as a valid identity. Those rules get baked into the technology: biometric scanners, database schemas, authentication protocols. Every rule is a potential tripwire. Someone staying in a shelter can’t provide a fixed address. A domestic violence survivor had her documents destroyed by her abuser. A transgender person’s appearance doesn’t match the photo on file. An elderly farmer’s fingerprints have been worn smooth by decades of work. The system doesn’t see these as situations that need accommodation. It sees them as failures to reject.

The World Bank’s Identification for Development (ID4D) initiative reckons nearly 850 million people worldwide lack any official ID. The proposed fix? Enroll them. But enrollment is an obstacle course. You need documents many people never had. You need to travel to a center that might be miles away, navigate a bureaucracy in a language you may not speak well, and take time off work you can’t afford to lose. The very people the system says it wants to help are the ones least able to jump through its hoops.

Biometric Barriers and the Myth of Universality

Biometrics get sold as the great equalizer—a way to identify someone even without papers. But bodies aren’t universal. Fingerprints become unreadable from manual labor, aging, or certain medical conditions. Iris scans fail for people with cataracts or eye injuries. Facial recognition systems stumble on darker skin, women, and older faces—a disparity well documented by the National Institute of Standards and Technology. When a system hangs everything on biometrics, it quietly locks out anyone whose body doesn’t match the template.

Look at India’s Aadhaar system, the largest biometric ID program on the planet. Civil society groups and journalists have documented case after case of manual laborers who couldn’t authenticate because their fingerprints had faded. Denied at the scanner, they were turned away from food rations under the Public Distribution System—a program meant to keep them from going hungry. The biometric lock became a starvation lock. The assumption that every person has stable, machine-readable fingerprints wasn’t a small oversight. It was a fundamental failure to understand the lives of the people the system was supposed to serve.

Documentation Requirements as a Poverty Tax

To get a digital ID, you usually need foundational documents: a birth certificate, proof of address, or an existing national ID card. This creates a cruel loop. You need an ID to get an ID. For anyone who has never had formal documentation—people born at home in rural areas, those displaced by conflict or climate disaster, those whose births were never registered—this requirement is a brick wall. Getting a birth certificate after the fact can be staggeringly expensive. It means traveling to a distant administrative center, paying legal fees, and losing income from missed work. It’s a poverty tax, levied on those least able to pay.

In Kenya, the Huduma Namba digital ID system ran into legal trouble precisely because of these exclusionary dynamics. The Nubian community, a historically marginalized ethnic group, argued that the registration process would disenfranchise them by demanding documents many didn’t have. The High Court of Kenya ultimately ruled that the government couldn’t make Huduma Namba mandatory for accessing services, recognizing that the system’s design would deepen existing inequalities. That ruling was a rare moment of institutional accountability, but it only came after years of advocacy and litigation by the communities on the receiving end.

Administrative Violence and the Burden of Correction

When a digital ID system fails someone, the burden of fixing the error lands squarely on the excluded person. A misspelled name, a mismatched gender marker, a corrupted biometric file—these aren’t simple clerical errors you can resolve with a quick phone call. They demand in-person visits to understaffed offices, long waits, and often the same documents the person couldn’t produce in the first place. Scholars call this administrative violence: the harm that happens when bureaucratic systems are designed to be indifferent to human suffering. The system doesn’t reach out to help. It waits for the excluded person to fight their way back in, knowing full well that many will simply give up.

For people with disabilities, the barriers multiply. A registration center might be physically inaccessible. A digital interface might not work with screen readers. A deaf person can’t complete a voice-based authentication step. These aren’t edge cases. The World Health Organization estimates that over 1 billion people live with some form of disability. Designing a system that ignores their needs isn’t an oversight. It’s a decision to shut out a significant portion of the population from civic participation.

Data Privacy and the Weaponization of Identity

Exclusion isn’t only about being locked out. It’s also about the dangers of being locked in. When a digital ID system collects and centralizes sensitive personal data, it creates new vectors for surveillance, discrimination, and violence. A centralized database of biometric and demographic information can be used to track political dissidents, target ethnic minorities, or expose individuals to harm. For domestic violence survivors, a digital ID that links their identity to a fixed address or a family unit can make it easier for an abuser to find them. For LGBTQ+ people in countries where their identity is criminalized, a digital ID that records their sex assigned at birth can become a tool of state persecution.

The United Nations High Commissioner for Refugees has raised alarms about the risks of digital ID systems for displaced populations. When a refugee’s biometric data is shared across borders without their informed consent, it can expose them to danger from the very regimes they fled. The promise of frictionless access to services curdles into a threat when the systems aren’t designed with the safety of the most vulnerable as the primary consideration.

Digital Literacy and the Connectivity Divide

Even when someone has the right documents and compatible biometrics, the digital nature of the system can still shut them out. Digital ID systems assume a level of literacy—both textual and digital—that many people simply don’t have. Navigating an online portal, making sense of privacy policies, and managing authentication credentials demand skills that are unevenly distributed. Older adults, people with limited formal education, and those who speak minority languages get hit hardest.

Connectivity is another layer of exclusion. In many rural and low-income areas, internet access is spotty or unaffordable. A digital ID system that requires online authentication for every transaction becomes useless when the network drops. In Nigeria, where the National Identification Number (NIN) is increasingly required for mobile phone registration, bank accounts, and government services, poor network infrastructure has left millions unable to complete enrollment or verification. The result isn’t just inconvenience. It’s disconnection from the formal economy and civic life.

The Smartphone Assumption

Many digital ID systems are built around smartphone apps, assuming every adult has a personal, secure, up-to-date mobile device. This assumption erases the reality of shared devices, especially among women and older adults in low-income households. When a phone is shared, the security model of one-person-one-device falls apart. Biometric data stored on a shared device can be misused. Authentication codes sent via SMS can be intercepted by an abusive family member. The system’s security features, designed to protect the individual, can instead become tools of control and surveillance within a household.

Accountability Gaps in Public-Private Partnerships

Digital ID systems are increasingly built and run through partnerships between governments and private technology companies. These arrangements often create accountability gaps. When a system fails—when someone is denied food, medical care, or the right to vote because of an ID error—who is responsible? The government agency that mandated the system? The private contractor that built the software? The local official who misentered the data? The answer is often no one. The excluded person is left to navigate a maze of unresponsive institutions, each pointing at the other.

In the United States, the push for digital ID verification in unemployment insurance systems during the COVID-19 pandemic showed exactly these dangers. States contracted with identity verification vendors to screen applicants for fraud. But the systems frequently flagged legitimate applicants, particularly those with thin credit files, unstable housing, or non-traditional work histories—disproportionately Black and Latino claimants. The Government Accountability Office found that these systems lacked adequate oversight and that claimants had little recourse when wrongly denied. The digital gatekeepers were accountable to no one.

Designing for Inclusion: Principles and Practices

Exclusion isn’t inevitable. It’s a product of design choices, and those choices can be made differently. An inclusive digital ID system starts with the people most likely to be excluded and builds outward from their needs. That means involving marginalized communities in the design process, not as an afterthought but as co-creators. It means offering multiple pathways for enrollment and authentication, so no single point of failure can lock someone out. It means building offline fallback mechanisms and human-mediated exceptions that are accessible, well-resourced, and free from stigma.

Inclusion also requires legal and institutional accountability. Independent oversight bodies must have the power to investigate exclusion, order remedies, and compel system changes. Affected individuals need clear, accessible complaint mechanisms with meaningful timelines for resolution. And governments must be transparent about exclusion rates, broken down by demographic categories, so the public can hold them accountable for the systems’ real-world impacts.

Data Minimization and Purpose Limitation

One of the most effective ways to reduce the harms of digital ID is to collect and store as little data as possible. A system designed to verify eligibility for a specific service doesn’t need to become a universal identity platform. Data minimization limits the damage when systems are breached or misused. Purpose limitation prevents function creep—the gradual expansion of a system’s use beyond its original intent. A digital ID created for vaccine distribution should not quietly become a requirement for opening a bank account or accessing education. These boundaries must be encoded in law and enforced by independent regulators.

FAQ

Why do digital ID systems exclude people who lack formal documentation?

Most digital ID systems require foundational documents—such as a birth certificate or proof of address—to enroll. People who are unhoused, displaced, or born in areas without formal birth registration often cannot provide these documents. The system then denies them an ID, which in turn prevents them from accessing the very services that could help them obtain the foundational documents. This creates a self-reinforcing cycle of exclusion.

How do biometric systems fail certain populations?

Biometric systems assume that physical characteristics like fingerprints, irises, and facial geometry are stable and universally readable. In reality, manual labor can wear down fingerprints, aging can alter facial features, and certain disabilities can make biometric capture impossible. Additionally, research has shown that facial recognition algorithms perform less accurately on darker-skinned individuals, women, and older people, leading to higher error rates for these groups.

What can governments do to make digital ID systems more inclusive?

Governments can adopt several measures: design systems with input from marginalized communities, provide multiple authentication methods (not just biometrics), ensure offline fallback options, eliminate fees for enrollment and correction, establish independent oversight bodies, and publicly report exclusion rates by demographic category. Legal frameworks should also limit how ID data can be used and shared, with strong penalties for misuse.

How does digital ID affect people experiencing domestic violence?

Survivors of domestic violence often need to keep their location and personal information confidential. Digital ID systems that link identity to a fixed address, share data across agencies, or allow abusers to access family-linked accounts can expose survivors to danger. Systems must include privacy safeguards that allow individuals to control who can access their information and to decouple their identity from an abuser without losing access to essential services.

Moving Toward Accountable Infrastructure

The conversation about digital ID can’t be reduced to a simple binary of inclusion versus exclusion. We have to face the reality that these systems, by their nature, create winners and losers. The question isn’t whether digital ID can be made perfectly inclusive—no complex system can—but whether the inevitable exclusions are recognized, minimized, and remedied. And whether the communities most affected have genuine power to shape the systems that govern their lives.

Accountability requires that we name the excluded. Not as statistics in a monitoring report, but as people with rights and agency. It requires that we measure success not by the number of people enrolled, but by the number of people who can reliably, safely, and equitably access the services they need. And it requires that we listen when communities tell us that a system is harming them, even—especially—when that message is inconvenient for the institutions that built it.

The architecture of digital ID is not fixed. It can be redesigned, regulated, and in some cases, refused. The choice is not between digital ID and chaos. It is between systems that concentrate power and systems that distribute it. Between infrastructure that treats people as problems to be managed and infrastructure that treats people as citizens with rights. That choice is political, and it belongs to all of us.