Before They Can Be Protected, They Have to Be Found: Birth Registration, Invisible Children and the State That Doesn't Know They Exist

Worldwide, 150 million children under five remain unregistered at birth, invisible to the governments meant to count them (UNICEF, The Right Start in Life, 2024). Sub-Saharan Africa carries the heaviest share of that figure, with a regional registration rate of just 51%. Against that backdrop, South Africa is not where the obvious story would place it. Southern Africa, the sub-region South Africa anchors, posts a registration rate of 88%, the strongest performance on the continent. The country most people would assume is failing its children is, by the numbers that matter most globally, one of the better-performing systems in Africa.

That fact should change how this post is read. This is not a story about a country that never learned to register its children. It is a story about what happens to the residual few it still loses, once a system has become reliable enough to handle nearly everyone else.

A Quiet Policy Success, and Where It Stopped

In 1996, an estimated 25% of South African births were registered within the calendar year of birth. By 2008, that figure had climbed to 76%, and by 2011, 90% of children were registered before their fifth birthday (Nannan, Dorrington & Bradshaw, Bulletin of the World Health Organization, 2019). The authors trace this improvement directly to policy: the introduction of the Child Support Grant, which required a birth certificate as a condition of payment, alongside a deliberate "basket of interventions" the post-apartheid government rolled out from the mid-1990s, eliminating healthcare user fees for pregnant women and children under five, building roughly 1,300 new clinics, and shifting births from home to institutional settings, which rose from 78% to 95% over the same period.

That history matters because it explains why registration was so low to begin with, and the explanation is not apathy. Under apartheid, identity documents were instruments of control, used to restrict where Black South Africans could live, work, and move; registering a child was, for many parents, registering them into a system built to surveil and confine them. Declining to register was, for some families, a deliberate act of protection rather than negligence. One account collected by the public history outlet Reasons to Be Cheerful in 2020 describes a grandmother, referred to as Gogo, who gave birth at home in the 1960s with no government contact at all; when her children later needed identity documents, the only record of their birth dates existed in her Catholic parish's baptism register. Her son, by contrast, registered all three of his own children without incident, by which point the system had changed enough to make registration routine rather than risky.

That generational shift, from a grandmother with no documentary trace of her children's births to a son who registered his own children as a matter of course, is the success story underneath the modern statistics. What it does not resolve is what happens to families for whom that shift was incomplete; where a mother still lacks her own identity document, decades after Gogo's generation, and that absence becomes the single biggest reason her own child's birth goes unregistered today, according to research from the Children's Institute at the University of Cape Town. The cycle did not end. It moved one generation forward and got smaller, but it did not close.

What Registration Means, and What the State Assumes It Means

Before examining the present-day backlog itself, it is worth pausing on an older and still underused piece of South African research that explains why some families have always been slower to register than policy assumes they should be.

In 1998, researchers Rachel Jewkes and Katharine Wood published a study in Social Science & Medicine based on 55 interviews with rural Xhosa women and local officials, conducted at a time when only 18% of births were being registered within their first year. What they found complicates the standard administrative explanation of apathy or ignorance. The women they interviewed generally understood birth registration not as the law understands it, an instant, document-conferred legal personhood beginning at birth, but instrumentally: a step taken to obtain something else, an identity document, a grant, school entry, rather than a meaningful event in its own right. Personhood itself was described as a gradual process unfolding over a child's early years, not a status created in a single legal moment by a piece of paper.

This is a genuinely useful finding, because it reframes what non-registration actually represents. The state's entire system assumes a population that shares its own legal-positivist concept of personhood, in which a person exists, officially, from the moment a form is filed. Jewkes and Wood's research suggests that for some of the families the system has historically struggled to reach, that assumption was never quite true; registration was not refused so much as deprioritised, because the document was understood as useful rather than constitutive. A policy built entirely around the document will always slightly misread a population that does not experience the document as the thing that makes the child real.

This does not explain today's backlog on its own; modern non-registration is driven far more by logistics and bureaucracy than by belief. But it offers an instructive caution: a system can solve its administrative problems, as South Africa largely has, and still fail to fully close the gap with the people it serves, because some of that gap was never purely administrative to begin with.

The Backlog, in the State's Own Words

By the state's own account to Parliament, a backlog of late registration of birth applications had reached just under 258,000 by December 2022, a figure then-Minister of Home Affairs Dr Aaron Motsoaledi disclosed in May 2023, covering only the 2018–2022 period and therefore almost certainly an undercount of the true total. By 2025, current Minister Leon Schreiber's department put the figure for undecided applications at 33,386 for that specific measure, while separate reporting placed the broader backlog at roughly 258,000 people, some of whom had been waiting up to seven years for a decision.

The Department's own position on the scale of the problem has not been consistent. In an October 2025 answering affidavit responding to litigation brought by the Children's Institute at the University of Cape Town and the Legal Resources Centre, Director-General Livhuwani Makhode argued there was no systemic backlog at all that year, a claim the Children's Institute directly disputed in its November 2025 reply. Paula Proudlock, senior researcher at the Children's Institute, offered a more concrete account of what the dispute actually concealed: once Home Affairs became aware the litigation was proceeding, it began processing the files of the families named in the case specifically, and had resolved all but one by September 2025. Her conclusion, drawn from watching that process unfold, was blunt: the files had not even been opened until the court case forced the issue, because no register exists that tracks whether they are being decided at all.

This is the structural heart of the matter. It is not only that a quarter of a million children are waiting. It is that nothing in the system was checking on them while they waited, until nineteen of them were named in a court order.

A Decade in Limbo

The clearest illustration of what this can mean for a single life is Tebogo Khoza's case. Born in South Africa to undocumented parents and orphaned at six, Khoza spent ten years, from 2013 to 2023, trying to have his birth registered and his citizenship recognised. The Pretoria High Court's ruling in February 2023 did not merely order Home Affairs to register him within thirty days; it found that the Minister's failure to give effect to the relevant provision of the Citizenship Act, despite three prior court orders instructing him to do so, amounted to contempt of court. A government department was found in contempt for failing, across a decade and three separate judicial instructions, to recognise that a man born in the country existed in law.

Khoza's case was extreme in its duration, but not in its mechanism. The same gap that kept him stateless for ten years is the gap currently holding an estimated quarter of a million children: a late registration process with no functioning tracking system, no enforced timeline, and, evidently, no consequence for inaction short of a court finding the state in contempt.

What the Gap Actually Costs

The consequences documented in the Children's Institute and Legal Resources Centre litigation are concrete rather than abstract: children denied access to public healthcare and private medical aid; refused enrolment in early learning programmes and school; adoptions that cannot be finalised while a child's birth remains unregistered; and access to the Maintenance Court blocked, so that a child cannot claim financial support from an absent parent even when that parent is known and locatable.

The Child Support Grant carries a particular cruelty. Families can sometimes secure a grant for a child without a birth certificate, only to have it suspended once SASSA requires documentation the family still does not have, withdrawing income precisely when a household has begun to depend on it. And the 2024 Mid-Year Population Estimates already show a national decline in registered live births, with researcher Katharine Hall's calculations, drawing on that series alongside Thembisa demographic modelling and SASSA's own grant statistics, suggesting a corresponding decline in households accessing grants they would otherwise qualify for; a trend, not a blip, and one expected to worsen rather than self-correct.

What Kind of Adults They Become

The research on what happens to a child who grows up without legal recognition is thin in South Africa specifically, but where it exists, it is consistent with a much larger global literature on statelessness, and the pattern it describes is not subtle.

Warria and Chikadzi (2022), writing in the African Human Mobility Review, document the case of a South African-based child, referred to as Andrew, who in 2007 was prevented from writing his matric exams because he had no identity documentation; a single missed exam sitting, in a system where matric is the gateway credential to almost every formal job and tertiary institution in the country, with consequences that do not end when the exam period does. The authors argue that statelessness should be understood through a trauma lens: repeated, cumulative exposure to exclusion from healthcare, education, and basic services produces the same kind of chronic stress response associated with other forms of sustained adversity, and that this stress does not resolve automatically even once documentation is eventually secured. The global literature on statelessness reaches a similar conclusion from a different angle: a 2022 quantitative and qualitative study of stateless ethnic minorities in Northern Thailand found stateless persons reporting measurably poorer mental health than persons with at least some recognised legal status, with respondents describing the condition in terms of inferiority and an absence of rights, even where their material circumstances were otherwise similar to documented neighbours. UNHCR's own synthesis of the evidence describes the psychological toll of statelessness in terms of dashed hope and a pervasive sense of belonging nowhere, a description that maps closely onto what Jewkes and Wood's (1998) earlier South African research found when it asked rural families what registration meant to them: not a bureaucratic inconvenience, but a marker of whether the state regarded you as the kind of person whose existence required confirming.

The more immediate, measurable adult outcome is economic and administrative exclusion, and South Africa's own data on identity documentation, not birth registration specifically, shows exactly where that exclusion concentrates. World Bank Global Findex data places South African adult ID ownership above 90%, among the highest rates in Sub-Saharan Africa. But across the region, when adults without an ID are asked why, the single most commonly cited reason is that they lack the underlying supporting documentation, principally a birth certificate, needed to apply for one in the first place. An adult who falls into that remaining gap in South Africa cannot easily open a bank account, since FICA regulations require a barcoded ID or passport; cannot easily access formal employment, since employers are legally expected to verify a worker's age and identity; and, as former Home Affairs Minister Naledi Pandor noted in addressing a related identity-document problem, risks exclusion from social grants, housing, and further education and training, the same institutions a birth certificate was originally meant to unlock in childhood.

This is where the cycle closes on itself, and it is the single most important fact this post can offer a reader trying to understand what becomes of these children. The Children's Institute's research identifies a mother's own lack of identity documentation as the most common cause of her child's late registration today. The World Bank's regional data identifies a missing birth certificate as the most common reason an adult lacks an ID in the first place. An unregistered child becomes, on the available evidence, an adult statistically more likely to be an undocumented adult; and an undocumented adult is, by the Children's Institute's own finding, the parent most likely to produce another unregistered child. The backlog is not simply a stock of cases waiting to be cleared. Left unaddressed, it is a mechanism that manufactures its own renewal, one generation at a time.

What Needs to Change

None of what follows requires new legal authority; it requires the political will to use the authority the state already has. First, the late registration process needs an actual case-management register, the absence of which is the single fact every account of this backlog, from Proudlock's research to the Khoza judgment, keeps independently confirming. A file that no one is required to check on will sit, regardless of how many are filed. Second, the burden of proof in late registration interviews needs to shift from treating delay as suspected fraud to treating it as the predictable outcome of distance, cost, and missing parental documentation, the causes the Children's Institute's own research consistently identifies. Third, mobile registration units, already piloted in some districts, need funding proportionate to a quarter of a million outstanding cases, not a fraction of it. Fourth, and most directly responsive to the intergenerational pattern documented above, late registration assistance should be actively linked to identity-document support for the parent, not only the child, since processing a child's birth certificate while leaving an undocumented mother undocumented resolves one generation's case while leaving the next generation's most likely cause untouched. South Africa proved with the Child Support Grant that linking registration to something families urgently need accelerates compliance dramatically. The same lesson, applied to the backlog itself rather than only to future births, is sitting unused in the country's own policy history.

A Final Word

A state does not have to fail at counting its children to still lose some of them. South Africa's achievement, raising registration completeness from a quarter to nine in ten within fifteen years, is real, and it should be named as the success it is. But a system that has solved its easy problem can mistake that solving for being finished. The children still waiting, some for seven years, are not waiting because the state cannot register children. They are waiting because no part of that now-functioning system was ever built to notice when a file simply stops moving, and because the one mechanism most likely to break the cycle, fixing a parent's documentation alongside a child's, has never been built into the process at all. Competence at scale and attentiveness to the individual case are not the same achievement. A country that has only ever built the first should not be surprised when it keeps losing track of the second, generation after generation, in numbers it has never quite managed to count.