The Digital Convergence Initiative of USP2030 has sought the endorsement by USP2030 members and partners of certain technical standards for linking various national databases to management information systems for social protection (SP-MIS). USP2030 has endorsed the linking of civil registration and vital statistics databases (CRVS) with SP-MIS. USP2030 has also endorsed linking integrated beneficiary registries (IBR), which consolidates data on the benefits received by programme beneficiaries across several programmes, with SP-MIS. USP2030 endorsement was also sought for linking social registries to SP-MIS.
The Global Coalition for Social Protection Floors (GCSPF) does not endorse these standards, as they fail to meet numerous criteria, including - but not limited to - respect for human rights, universality of social protection, and failure to strengthen national systems of civil registration and vital statistics.
On the basis of available evidence, the GCSPF furthermore holds that governments and multilateral agencies should reconsider their position on the creation, promotion and use of social registries for delivering social protection benefits.
The GCSPF promotes universality of the right to social security and universal access to all benefits collectively comprising social protection. The GCSPF draws attention to the fundamental fact that universal systems of social protection are affordable, and there is substantial evidence in support of this affirmation.
If, nevertheless, countries wish to use proxy means tests for targeting individual schemes, they should be fully informed of the challenges with this methodology and, importantly, of the global evidence on their failures in every situation in which they have been used.
Importantly, also, social registries are not used in high-income countries, and the GCSPF believes that it is highly problematic that they are used in low- and middle-income countries, suggesting a form of global discrimination that derives alone from the misguided frustration of addressing short-term objectives in the infrastructural context of the digital divides.
Human rights are universal, not selective according to national income or individual means.
The GCSPF has elaborated the reasons for withholding endorsement of the DCI’s standards in the paragraphs that follow.
1. What is a Social Registry?
A social registry is a database that includes a range of socio-economic data on households, including income, assets, and housing conditions, along with some characteristics of household members. They are designed to select the recipients for poverty-targeted, household-based social assistance schemes. They make it possible to rank households from poorest to richest in order to determine cut-offs and to select recipients. Eligibility is often determined through mechanisms such as proxy-means tests (PMTs) to estimate a household’s poverty level which we argue are highly problematic. For example, in Guatemala, where only 12% of the population has access to any form of social protection, the government is currently rolling out a social registry census. The collected data will be processed through a PMT algorithm to determine household eligibility for a cash transfer program.1
2. Social Registries and Single Registries2
Social registries were first developed as a cost-cutting measure to reuse PMT databases that were developed for an individual scheme. They enable the high costs of developing the databases to be shared over several schemes, which reduces the cost per scheme. Because of conglomeration, however, the impact of errors is multiplied: for example, exclusion errors are multiplied as incorrectly excluded persons will be excluded from several schemes.
In contrast, a single registry integrates data from various government sources and social security programmes, acting as a central information hub. It links individual programmes and can link external databases when needed, with appropriate firewall safeguards otherwise, such as tax, civil registration, or disability databases to streamline registration and programme management.
3. Social Registries are not fit for purpose for Rights-based Social Security Systems
Social registries are designed for the targeting of “the poorest” in household-based social assistance schemes using methodologies such as PMTs. They increasingly serve as a backdoor to poverty targeting, whereas poverty targeting has been widely found to undermine the right to social security, which is a universal right. Poverty targeting flouts the right to security especially in settings where the entire social security system relies on poverty targeting. In these cases, the system introduces risks to individuals and households, rather than providing protection. Risks that originate in social registries include:
3.1 Failure to respect universality of the right to social security and social protection
Social registries are designed to limit access to social security and social protection and are therefore incompatible with universal access to benefits.
Furthermore, where a national or regional scheme is conceived to be universal, social registries will not be fit for purpose because they usually do not include the entire population of a country and the information held by them rapidly degrades in quality; they cannot be used for implementing benefits in universal schemes such as universal old age pensions or universal child benefits.
3.2 High levels of exclusion: programmes reliant on social registries have high exclusion rates (44–90%). By applying the same targeting methodology to different programmes, social registries systematically exclude many of the poorest members of society from different programmes of a scheme. In effect, they almost guarantee that the majority of the poorest members of society are erroneously excluded from schemes, defeating the essential purposes of a social protection scheme. Furthermore, their inaccuracy and arbitrariness not only obstruct access to benefits but cause widespread distress and dissatisfaction that can undermine trust in the state.
3.3 Unsuitability to determine individual benefits: Even in the case of means-tested schemes that target individuals, social registries cannot be used because they hold information on households. The social registry cannot show whether an individual has passed or failed the means test since they do not hold the required information on the individual.
3.4 Unreliability during periods of shocks or crises: Social registries hold static data and so cannot be used to respond to shocks, because they cannot show how the circumstances of households have changed during a shock or crisis.
Furthermore, they do not hold the details needed to transfer payments to individuals who are not already recipients of schemes and, therefore, it is not possible to make payments rapidly, whereas swift payments may be critical. It is a notable fact that social registries failed to service the response to the Covid-19 crisis.
3.5 Social scoring: Data from social registries may be used to develop social scoring, a system based on assessing individual behaviour and characteristics to determine access to social protection or public services. Such practices can lead to discriminatory allocations and violate human rights.
Human Rights Watch has called for a ban on social scoring due to its potential for abuse.
3.6 Basic data inaccuracy: Economic and social circumstances frequently change, yet social registries often assume static household conditions. Data is rarely updated for periods of years, outdating them rapidly and rendering the data unreliable. For example, if a registry is not updated within 5 years, there will be no information on the number of children 0-4 years in households, a critical omission.
Even when social registries are on-demand, the information is still held for a number of years and rapidly outdated. Households have limited or no incentive to update their data on an on-going basis.
3.7 Risks to privacy and confidentiality: The collection and storage of extensive personal data raise significant risks to the right to privacy. The conglomeration of data by removing any or all firewalls or other safeguards raises the risk of flouting privacy.
For all purposes related to access to health data, serious failures to protect and respect medical confidentiality can arise. In this regard, the Recommendation of the Council on Health Data Governance adopted by the OECD Council on 13 December 2016 outlines the safeguards that must be in place when health-related data are not being used for the benefit of individual treatment, but for other purposes, as is the case in their use in registries to identify potential beneficiaries.
The Council on Health Data outlined the safeguards recognising that “governments have a role in fostering compliance with privacy laws and policies[...] that personal health data being sensitive in nature and subject to ethical standards and the principle of medical confidentiality, require a particularly high level of protection and that technological developments can both enable the privacy protective use of personal health data and also introduce new risks to privacy and data security (italics added)”. Accordingly, their Recommendation furthermore provides for Monitoring and evaluation mechanisms that “should:
I. Assess whether the uses of personal health data have met the intended health-related public interest purposes and brought the benefits expected from such uses and whether any negative consequences of such uses have occurred, including failures to comply with national requirements for the protection of privacy, personal health data and data security; data breaches and data misuses; and feed the results of such assessment into a process of continuous improvement, including through:
a. Periodic review of developments in personal health data availability, the needs of health research and related activities, and public policy needs; and
b. Periodic assessment and updating of policies and practices to manage privacy, protection of personal health data and security risks relating to personal health data governance.
II. Encourage those processing personal health data to periodically review and assess the capabilities, reliability and vulnerabilities of the technologies they use.”
In the same vein, the commitment to human rights in the digital space was clearly outlined in the recent Global Digital Compact, Annex 1 to the Pact for the Future, adopted by Heads of State and Government at the United Nations in September 2024. It was the specific focus of Objective 3 of the Compact: Foster an inclusive, open, safe and secure digital space that respects, protects and promotes human rights. Moreover, it implied the requirement that national safeguards should be aligned with international law.
This commitment is, in particular, to “ensure that the development and implementation of national legislation relevant to digital technologies is compliant with obligations under international law, including international human rights law”. This applies to the creation of purpose-driven national databases containing personal information on individuals or households as much as to the dissemination of internet access and the applications of artificial intelligence.
No initiative in the area of registries of households and/or individual citizens and residents of a country may by-pass such obligations.
3.8 Security concerns: In the context of social registries, interoperability and wide access to social registries introduces security risks, and data breaches or misuse can similarly lead to serious violations of human rights through identification, discrimination and stigmatization.
3.9 High cost: Establishing and maintaining social registries require significant financial and administrative resources, whereas they provide flawed and often harmful results. Therefore, they are not fit for purpose and cannot deliver value for money. The funds they use could be better invested in building universal social security systems or in developing useful registries, such as vital statistics registries that hold information on births, marriages and deaths, for example, or national ID databases. Rather than contributing to closing the digital divide, social registries divert resources away from that objective; the alternative path of action is elaborated in the following point.
4. Failure to contribute to the development of national statistical capacity
Social registries ignore the inadequacies of national statistical systems, by attempting to circumvent their shortcomings with a parallel course of action. Consequently, they become a cost burden to governments and an encumbrance for the national statistical system. Rather than investing large sums in developing complex databases, countries should invest their resources in strengthening their national civil registration and vital statistics systems and ensuring universal access to identification documentation. Overseas development assistance in the forms of grants or loans should be targeted for this purpose wherever necessary, because of the fundamental importance of these systems. It should be ensured that each state has an autonomous and sustainable national system of registration and of vital statistics.
Civil registration of all residents is necessary for implementation of any universal social security system and should be the basis to which other additional, carefully selected data can be linked as and when necessary: such as, for example, data on people with disabilities for universal disability benefits, or geographic data for support in the event of climate-related disasters.
Technical cooperation in matters of national civil registration of births and deaths and the development of vital statistics systems, in particular training of government statisticians, civil registration officials and other stakeholders, is the responsibility of the United Nations Statistics Division. The specific mandate of the United Nations Statistics Division to ensure international principles and recommendations for compiling, processing and disseminating vital statistics has been in place since the UN’s creation and was reinforced under the International Programme for Improvement of Civil Registration and Vital Statistics Systems, adopted by the United Nations Statistical Commission in 1991. This Programme, originally funded partly by UNFPA, has been seriously underfunded and continues to be disregarded in favour of shortcuts to speedier means to determine national statistics. Yet it is only the national statistical systems that we can expect to be sustainable. A ramp-up in resources to national systems would alter the cost equation as a fundamental investment in national statistical capacity with substantial immediate, mid-term and longer-term benefits.
Furthermore, because of the particular need for agglomerated data on causes of death of the global health community for determining health policy, and in response to the digital demands of implementing the SDGs, the World Health Organization has initiated the WHO civil registration and vital statistics strategic implementation plan 2021-2025, bringing on board numerous partners from the UN system, academia, civil society and the private sector.
Again, this WHO initiative, adequately funded and implemented, would go far in establishing or strengthening - and rendering sustainable - national capacity that is owned and operated by national authorities, among whom would be the administrators of national systems of social protection.
In any time frame, the development of sustainable civil registration and vital statistics systems represents the better investment of resources to close the digital divide between countries, as well as the better basis for the administration of a national system of social protection that respects human rights and is founded on universality.
Sources & Further Reading on Social Registries:
Note submitted to USP2030 by the Global Coalition for Social Protection Floors (GCSPF)
December 2024
Notes:
2 There is no consensus on the definition or wording of different registries, and in practice names and features of registries vary in different countries.
