This following is a detailed summary of the panel discussions and keynote address that took place during Harvard FXB Center’s 21st Century National Identification Conference in November 2015.
PANEL 1: Foundations of National Identification Systems: Political, legal and civil frameworks
The first panel made connections between national identification systems and political, legal and civil registration frameworks. As moderator, Jacqueline Bhabha place the panel into a global context where millions are fleeing persecution and hostility, and where questions of legal identity, access to a nationality, and possession of documentary proof of identity could not be more urgent.
Alan Gelb analyzed the increase in nationwide identification programs since 2000, and suggested that the most salient question was how identification systems could contribute to inclusive development, not whether such systems were necessary. He identified the recent drivers of identification systems: security concerns after 9/11, the growth of e-services, an increase in multi-party elections, migration, and the growth of the digital biometrics industry, as well as increased support from donors. He also identified several challenges to the identification systems. For example, the interest in identification has been sector specific, a development which has slowed the creation of more general databases such as civil registration and vital statistics (CRVS) systems. Gelb also described the challenges posed by fragmented identification systems, by gender- and nationality-related discrimination, and by data misuse. He presented three questions that could be used to assess an identification system: first, whether the identity establishment process was robust; second, whether there was high coverage; and, third, whether the system could be used for multiple applications. Finally, he described Sustainable Development Goal 16.9, which affirms the right to identity and is directly relevant to at least 10 other SDGs. Interested parties can strengthen identity systems by fostering cooperation between between donors and development partners, and by incorporating country-specific information into systems.
Mia Harbitz placed birth registration at the center of national identification and stressed the need to ‘get it right from the start’ by meaningfully considering how identification systems can contribute to CRVS systems. Harbitz highlighted how advances in identification technology in the 2000s had not been not matched by a reconsideration of policies relating to legal identity and personal identification. National identification schemes were not improving civil registration, and often competed with other ID projects for funding. Harbitz discussed the consolidation of legal and judicial processes for civil registration and identification. In addition, Harbitz also emphasized that unique personal number from birth can be used to facilitate interoperability between databases, and that it is important for government officials to standardize civil registration procedures so as to avoid duplications. Finally, she stressed the need to communicate with citizens, users, ministries, and the private sector to ensure there was a coordinated approach to identification.
Tarik Malik reviewed the development of the National Database and Registration Authority (NADRA), Pakistan’s identification system, to show how national identification systems can be used to strengthen empowerment, service delivery, and state capacity. Pakistan’s Identification system has its origins in mid-1970s, but use of biometric technologies for citizens registration was undertaken in 2007 after old registration body was reformed and renamed as NADRA in 2000. Biometric technology was deployed in a context where no census had been conducted in decades and was used to rollout social inclusion programs, improve voter registration, identify tax evaders, strengthen branchless banking, and coordinate service delivery after natural disasters or internal displacements. However, Malik emphasized that government must resist capture of a semi-autonomous institution and introduce legislation to protect the privacy of its citizens.
Bronwen Manby showed how new identification systems could be a danger point for creating statelessness and stressed that technological solutions must be connected to underlying legal frameworks. Focusing on Africa, she considered the overlaps and distinctions between lack of documentation and statelessness, noting that efforts to increase levels of documentation of citizens often did not sufficiently consider the question of access to citizenship. Among those most at risk of statelessness are the migrants and their descendants, cross-border populations and nomadic pastoralists, and children affected by discrimination in law (for example, based on gender discrimination in transmission of nationality) or fact (for example, birth out of wedlock). Birth registration is only 50% on average across Africa, and many adults are also undocumented. National ID cards are, however, rapidly becoming more common, especially as new biometric systems are being rolled out: the West Africa region will introduce a common form biometric ID during 2016. The risk is that these new systems, often sold as an almost magic solution to problems of identification, either do not address underlying discrimination, or even make it worse. Using examples from Cote d’Ivoire, Nigeria and Mauritania, Manby showed how identification systems can both define and deny legal identity and nationality.
PANEL 2: Global perspectives on ID numbers and ID cards
The second panel examined the uses of ID cards using examples from India, Iceland, the UK, and from the World Bank’s work on identification systems. Robert Palacios presented data and examples from the World Bank ID4D initiative to show that the vast majority of African countries have a national ID. He then described the role of initiative in providing technical support to countries to develop and improve identification systems. He suggested that contextual factors need to be taken into account when designing or improving identification systems. Particularly important factors include transportation infrastructure, internet penetration, birth/death registration, the competing role of electoral bodies in ID provision, and the presence of personal data protection legislation. He argued that birth and death registration systems could not be relied on meet the identification needs of the current population. However, a central goal to improve identification systems needs to be to improve the flow of birth and death registration while building other systems, so these systems can converge. Palacios discussed the challenge of a separate identification system for voting and highlighted the need to connect national and voter identification to avoid costly duplication. He concluded by stressing the need for minimal, appropriate standards for personal data protection, especially in a context where identification systems were being introduced in the absence of data protection acts and laws.
Ajay Bhushan Pandey described the design and implementation of India’s Aadhaar card, a biometric identification system which generates a unique 12-digit number for all residents, including children. Today, Aadhaar enrollment is greater than 930 million (more than 94% of adults in India) and has surpassed the coverage of other identification systems in India. Total expenditure for Aadhaar over the last five years was close to one billion dollars, and approximately one dollar per card. Aadhaar was launched as a response to the fact that 60% of residents in India had no identification and were being denied services, government spending was approximately $60 billion on direct subsidies and payments while only 5% of the habitations had banks. Aadhaar was designed to be highly scalable, to have robust data protections, to ensure vendor neutrality, to incorporate biometrics, and to include marginalized populations. The system is based on an open architecture and is the Aadhaar authentication service allows service providers (banks, hospitals etc.) to identify users. Aadhaar is being used to open bank accounts, send payments directly to bank accounts, and by a range of national and state-level social welfare schemes that provide cooking gas, pensions, and cash transfers. Pandey also discussed the legal context of Aadhaar: twelve petitions have been filed in the Supreme Court against Aadhaar based on concerns about legal status, concerns with hacking, inconvenience to vulnerable populations which will lead to denial of service, fear of privacy breaches, and potential of data aggregation and profiling. Of these, one petition was filed by Aadhaar administrators, and asked the Supreme court to keep biometric information from being shared with the Central Bureau of Investigation (CBI).
Ian Watson gave a presentation on Iceland’s national identification number, the kennitala. It is ten digits long and includes an individual’s date of birth. In practice, citizens use the number constantly–when paying bills, submitting tax returns, visiting the doctor, or attending university. A publicly accessible database list’s every Icelander’s kennitala. The number is never used as an authenticator. Watson described the kennitala as a kind of public good and emphasized that people use the kennitala like their name. Referring to people by number has, he argued, been ‘desacralized’ in Iceland. He concluded by pointing out that Iceland has no identity card, and suggesting that open national ID systems may be more secure than systems laden with supposedly protective security features.
Edgar Whitley examined the history of the UK National Identity Card Scheme and the current ‘Verify’ service in the UK. He described the government’s unsuccessful attempt in 2005 to institute a biometric identification system, which he argued could also be called a surveillance system. He also described a report entitled ‘The Identity Project’ (which he worked on) which criticized these efforts. Today’s ‘Verify’ service authenticates access to online government services. Unlike the previous attempt to build an identification system, the Verify service has a federated structure, only uses private sector identity providers, and is focussed on ensuring privacy. Identity providers include: Experian, Verizon, Digiidentity, Post Office, and other agencies – Barclays, GB Group, Morpho, PayPal, Royal Mail – that are being onboarded. A set of Identity Assurance Principles have been developed which are focussed on what the citizen needs (not what the state needs) and an dispute resolution service has been instituted to assist citizens without access to the Verify system. Edgar asked whether it was desirable to keep data in one centralized system and offered an example from Austria to show how it is possible to generate different numbers for each government agency, so it is impossible to look up information across government agencies. If there was a need to merge data from different agencies, he argued that this act should only be done through a judicial process.
PANEL 3: Government data – How it is formulated, combined and used
The third panel analyzed how government data is formulated, combined, and used. Deborah Rose examined the structure of government databases and the characteristics of effective ID systems. She argued that ID systems should be available to all residents, especially vulnerable populations. Moreover, care should also be taken to ensure that numeric identifiers embed a minimal amount of personal information. Government officials have multiple policy options, and should not only think carefully about how to structure national ID databases, but also about whether national identity cards or numbers are most appropriate locally.
Ryan Seals presented research using government data from the Danish Civil Registry, which, since 1968, has collected basic information on all individuals living in Denmark. Seals’ research centers on the risk factors for amyotrophic lateral sclerosis (ALS)—a neurodegenerative disease that is difficult to study in part due to its rarity. Using the Civil Registry database and other government databases, Seals was able to link the health and economic data of those affected by ALS and similar but unaffected residents, allowing the disease to be studied with high-quality and uniquely rich data. Merging together this data was only possible because of national ID numbers being present across different data sources.
Finally, Jo Anne Barnhart discussed her experiences at the Social Security Administration, arguing that unique identifiers facilitate inter-governmental communication. Because unique identifiers allow officials to merge together data across different sources, it also facilitates the evaluation of government programs. However, government data systems can become technologically obsolete over time, reducing the usefulness of such data and requiring expensive updates. In addition, she suggested that cost benefit analyses should be done before expanding existing data collection programs.
PANEL 4: Uses of Local and National Identification Systems: Financial inclusion, benefits and services, health insurance and development
The panel on the uses of local and national identification systems drew on examples from the World Bank’s efforts to support the development of identification systems in low-income countries, the role of identification systems to improve financial inclusion, and the introduction of ID cards in New Haven and New York to allow undocumented immigrants access to public spaces and advance equity.
Mariana Dahan described the launch of the Identification for Development ID4D in July 2014, a cross-sectoral group at the World Bank to lead the ID4D agenda, and discussed their engagement with countries on assessments of the current ID infrastructures, design of solutions, followed by implementation. Dahan emphasized the importance of a coordinated approach across donors working on identification systems.
Tilman Ehrbeck offered an industry user’s perspective on UID and highlighted the role of mobile phones in improving access to bank accounts, and the many innovations that leverage the e-money infrastructure in a world where 40% of people are without a bank account and 60% without informal jobs. He discussed the role of technology in reducing the total lifetime channel costs per transaction, and the benefit of reducing the cost of Know Your Customer (KYC) authentication; in India, for example, the cost for paper KYC was $40 and was reduced to $11 with eKYC. Ehrbeck concluded by describing how ID cards were used as a foundation for inclusion and could serve at the foundation for a broader technology ‘stack’ that enables a range of services.
John DeStefano discussed the rationale behind the New Haven identification card, which was launched in 2007 in a city where approximately 20% of the community were immigrants, half of whom were undocumented. The resident ID card was funded by a private foundation and introduced by the Mayor’s office for all residents of New Haven. He described how the card was a way to address the underreporting of crime by undocumented immigrants, wage theft, threats of reporting immigrants, the challenges undocumented immigrants had with opening bank accounts, and to allow people to live and work in New Haven.
Finally, Nisha Agarwal presented the design and implementation of a city-wide ID card for all New Yorkers. Identification was a priority in Mayor de Blasio’s platform who argued it was essential for people to have IDs, access city services, be allowed to enter buildings. In short, the ID card was a way to create a more inclusive city. The IDNYC card was developed through consultation with stakeholders from across NYC. It was designed to look a government ID card, gender was optional (Male, Female, Blank), and language preference and emergency contact could be listed on the back of the card. NYPD was also consulted to ensure they would accept the card when it was presented. To register, any document to prove residency and identity were requested and visa status or citizenship were not essential. A range of benefits were linked to the card: several banks in New York accepted the card to open and access an account, free or discounted membership to 33 of the city’s cultural institutions was included, and the card allowed access to all government buildings, libraries, and to a health prescription benefits program. 630,000 cardholders enrolled in the first ten months. Agarwal concluded by highlighting that data associated with the ID cards will be protected to the maximum extent protected by the law, and law enforcement would need a judicial warrant to access the data for individuals; to ensure these executive orders were issued on the use of data and privacy protections. The card would continue to be used to make the city more inclusive and to improve access to public spaces and services for immigrants.
KEYNOTE: Nandan Nilekani, Former chairman of the Unique Identification Authority of India (UIDAI)
For the keynote, Nandan Nilekani recounted the history of the Aadhaar system in India from his vantage point as former Chairman of the Unique Identification Authority of India (UIDAI). He first discussed the motivations for the Aadhaar system. This system was formed in the late 2000s, partly in response to low birth registration in India, which created a situation where many residents lacked the identification necessary to open a bank account or to participate in India’s economic growth. In addition, Indian social programs previously experienced 10-30 percent of leakage in their transfers, as identity fraud was rampant.
Five years in, the program has achieved considerable success. Because it reduced identity fraud, the introduction of the Aadhaar system dramatically reduced false entitlement claims, saving some programs as much as $2 billion per year. Also, the Aadhaar program has maintained a high level of data security, partly because the program the program uses open source software and was designed with privacy considerations in mind. Nilekani also discussed how the Aadhaar number has taken on numerous uses in finance, online shopping, and more. Because of the system’s multiple uses, it has enjoyed popular support: Nilekani emphasized that, 935 million people have Aadhaar numbers today (even though the numbers are voluntary), and its reach is more widespread than almost any other program in India.
PANEL 5: National identification: Protecting privacy and ensuring data security
This panel brought together experts on privacy and data security, and examined the opportunities and risks present in the use of national ID systems. David Lyon offered a conceptual analysis of ID systems, suggesting that they are fundamentally ambiguous, and can either enhance or can compromise civil liberties. ID cards can facilitate the surveillance capacities of the state, but this does not make their use inappropriate in all contexts. Identification systems should be overseen in a democratic fashion to ensure the accountability of the system supervisors.
Robert Gellman said that differences between national constitutions, laws, traditions, and needs affect the design of a national ID system and the necessary tradeoffs. Three concerns about any ID system are mission creep, centralization of transactional information, and ID theft. Helpful privacy ideas for any ID system include Fair Information Practices, privacy impact assessments, an independent Chief Privacy Officer for the system, and a federated system with different IDs for different purposes rather than a single central ID system. Political and popular opposition to national IDs in the US and some other countries can be a significant factor.
Latanya Sweeney examined identification security in the US context by discussing security risks of the system of Social Security Numbers (SSNs). She noted that the major forms of identity fraud in the US stem from insecurities in the SSN system: that tax fraud alone accounts for $26 billion, and that one-half of identity theft involves fraudulent credit cards. She argued that ongoing reliance on insecure identity systems increases the risk of cyberattacks, and that government officials in the US should seek out both institutional and technological mechanisms to enhance the security of identification systems.
PANEL 6: Inclusion and Exclusion: Ethical and human rights implications of identification systems in context of statelessness<h/3>
The final panel examined the history of using identification cards to discriminate against minority populations, and the use of identification systems in the context of the Syrian and other refugee crises. Panelists examined the ethical and human rights implications of identification systems and emphasized the need to understand how identification can be either a right or a threat, especially for minority populations.
Andrew Hopkins presented UNHCRs efforts to enroll and provide services to refugee populations. He described the transition from paper, wristbands and ration cards to biometrics, and the introduction of standards to address the practices of data collection, identity management, and data sharing. UNHCR used ID cards to improve the distribution of food and other services, improve mobility for refugees, and aid in understanding movement in and out of refugee camps. Hopkins also highlighted that biometric identification had served to increase donor confidence in the delivery of services and contributed to a more accurate aggregation of global statistics. He recognized the need for further work in determining if, and how, data would be shared with host governments. Edwin Black detailed the revelations of his book IBM and the Holocaust: The Strategic Alliance between Nazi Germany and America’s Most Powerful Corporation. He examined IBMs role in co-planning six phases of the holocaust using punch cards and the Hollerith Machine, and the role of corporate power in enabling states to persecute minorities.
Margaret Matache discussed the implications of identification and registration for Roma communities in Europe in a presentation entitled ‘Roma: a stolen social, political and legal identity’. She emphasized the importance of offering registration, while concurrently examining the threats of deportation and discrimination posed by governments to marginal populations. She described how 70,000-80,000 of the Roma population remain stateless and continue to face discrimination when they attempt to register. She examined how fingerprinting and biometric proposals for Roma in France (2009/10) and Italy (2008) were designed not to assist Roma, but to monitor re-entry after deportation, and described how the EU Commission decided that this was a violation of the freedom of movement law.
The conference concluded with a discussion on the legacies of history and how they inform our understanding of national ID systems and their future role in governance. Joseph Atick examined the rollout of ID systems in Mexico, Indonesia, and Brazil, discussing the challenges these countries faced in this process, and lessons learned from their experience. He emphasized how ID systems must be transparent to win public buy-in, how systems should avoid getting locked into proprietary software, and how leaders should promote arrangements that avoid competition between government agencies. Mia Harbitz provided a conceptual tour of the identification issue, emphasizing the legal, administrative, and political dimensions of ID programs. She also discussed the importance of country-specific contexts, and how ID systems must accommodate themselves to local culture and customs. Alan Gelb emphasized the timely nature of the conference, and drew out a theme present throughout: ID technologies are neutral, and can facilitate development or can be used for surveillance purposes. Gelb also discussed lingering tensions around whether countries should adopt a centralized, universal ID system, or should pursue more federated institutions. Deborah Rose concluded by noting that this is the first academic conference on National ID systems; that this important role should continue; and that the more inclusive a national ID number system is, the more useful a tool it can be for development.
Launched in September 2013, the National ID Number (NIN) Project works to to examine the scientific, technical, social and political aspects of national ID numbers. Under the leadership of Jacqueline Bhabha, Director of Research at the FXB Center for Health and Human Rights, along with Conference Co-Chair and Visiting Scholar Deborah Rose, the program aims to build an interdisciplinary body of knowledge on national identification numbers, examining the necessity, benefits, challenges, feasibility and implications of the growing adoption of such as system by countries worldwide.
The NIN project is consistent with pursuit of a central Harvard FXB Center goal: The advancement of scholarly inquiry at the boundary between normative and scientific research on the one hand and pragmatic policy related implementation and progress on the other. Scientific progress, innovation and a robust adherence to international norms of human rights and social justice are essential elements to achieving prosperity and reducing inequality in our century.