Select Committee on Home Affairs Fourth Report


4  BENEFITS AND WEAKNESSES OF THE GOVERNMENT'S SCHEME

65. In this section we first examine the possible contribution the proposed identity card scheme might make to each of the Government's stated aims, taking into account as appropriate existing means for attaining these goals. We then consider possible areas in which the Government needs to take additional action if a scheme is to be fully effective, before considering aspects of the Government's proposals such as the National Identity Register; the identity cards and readers; biometrics; and procurement and costings.

The stated aims

66. The Government argued that the scheme will help in the following areas:

67. We note that the relative prominence given to these has varied. In July 2002 the original consultation document[45] asked for views on whether cards would help with: providing better services; identity fraud; tackling illegal immigration and illegal working; a convenient travel document; proof of age; reducing crime; electoral registration and voting; and emergency medical information. In this list there was no mention of terrorism. The card was called an entitlement card and there was clearly a greater emphasis on the link between the card and the use of public services than in later proposals.

68. In November 2003 the Home Secretary argued that biometrics enabled the Government to deal with "the growing threats to the security and prosperity of Britain, from identity theft, fraud, and illegal migration". He added "the security services have indicated to me that they would value improved methods of verifying identity and counteracting the use of multiple identities. It is obvious that terrorist networks would target the countries that had made the least progress in developing the capacity to provide this protection".[46] The paper setting out the Government's plans,[47] which the Home Secretary was introducing, argued for identity cards on the grounds of fighting illegal immigration and illegal working; disrupting terrorist use of false and multiple identities; combating identity fraud and money laundering; and providing a convenient way for those entitled to services to access them and preventing unauthorised access to services, including health tourism.

69. Explaining the change from an entitlement card, the Home Secretary told us:

    "Post-11 September 2001 I was asked on a number of occasions, starting on the end of the week of 11 September, whether I believed that we should have ID cards as a consequence of the attack on the World Trade Center, and I said on record several times, and I still believe it, that whilst there could be a contribution towards countering terrorism this was not the primary purpose, and although it would be part of any such scheme it should not be seen as the sole focus. I went on to say that it was probably sensible, if we were going to move towards such a programme, to describe it as being part of entitlement—entitlement to services and benefits—which we had built up by the contributions we made and the mutuality that has stood us in good stead and is part of the National Insurance concept of the post-Second World War settlement. I then took that to the appropriate Cabinet committee the following January, that is January 2002. When we launched the consultation proper in the July it soon became clear that people did not like the term "entitlement" card. They thought that it should be an ID card, that it should be explicit rather than implicit, that it should give a clearer picture that it encompassed tackling terrorism and organised crime, and they believed that it would be more honest and transparent of the Government to do so, so in a nutshell we agreed after listening to the results of the consultation that that is what we should describe it as."[48]

70. It is reasonable for the Government to have refined the aims of its scheme after a consultation exercise and development of proposals for its implementation. It has now set out its reasons for introducing identity cards, in its most recent document, Legislation on Identity Cards: A Consultation, which accompanied the publication of the draft Bill.

71. However, many elements of the design of an identity card scheme, from the national register, to the design of the card and to its operational use, depend greatly on the precise purpose for which it is designed. Although some core functions are consistent and clear, the changing aims of the scheme do not give total confidence that the Government has arrived at a complete set of clear and settled aims for the card. The Government has not yet clarified how it intends to deal with some elements of the original proposals for entitlement cards, such as which services should be linked to the card and whether there should be unique personal numbers across public services. We consider these issues further below, but it is clear that they are central to the functioning of the scheme.

72. The draft Bill might have been expected to clarify the Government's aims but we do not believe it has done so. It is essential that the Government explain its intentions on issues raised in this report before the Bill is published.

ILLEGAL WORKING AND IMMIGRATION ABUSE

73. Since January 2002, when an asylum application is made, the applicant is screened, and his or her personal details are recorded and put on an Application Registration Card (ARC) which is issued to the applicant. The ARC provides a secure means of recording asylum seeker's details and biometric identity. Asylum seekers now use the ARC to access the services provided for them. Up to the end of March 2004 200,000 cards have been issued and the Immigration Service say they have seen very few cases of suspect cards. Of the ones they have seen, only two were in fact forgeries, and only one of these was of a quality which might have deceived untrained personnel.[49] Both the Law Society and the Commission for Racial Equality thought that the ARC had been a success in that it provided a secure form of identification for asylum seekers, who might have no other means of showing who they were.[50]

74. The consultation paper accompanying the draft Bill put tackling illegal immigration at the top of the aims of the Government's scheme:

    "The identity cards scheme is intended primarily as a United Kingdom wide measure to help deter and control illegal immigration by helping to establish the nationality and immigration status of UK residents […]."[51]

75. The Home Office memorandum to the Committee argued that an identity card system would enable employers to establish the status of prospective employees more quickly and more simply.[52] It also argued that the creation of the National Identity Register would, by verifying entitlements to public services, make it more difficult for those here illegally to gain access to them, and thus would reduce the "pull factor"the perception that once in the country people can work and obtain benefits and public services with impunity.

76. The Government argued that an identity card will make it easier to establish that a foreign national has the right to work. The Commission for Racial Equality recognised "a potential benefit to ID cards scheme in that they could enable immigrants and other groups to authenticate their identities (for accessing the labour market and public services where they are entitled)".[53] The Law Society, however, argued that the real problem was the small number of employers who did not at the moment carry out the appropriate checks, who would continue to employ people illegally regardless of whether or not there was a national identity card.[54]

77. In our recent report on Asylum Applications, we said:

    "We believe that a significant factor in the problem of illegal working is the deliberate decision by some employers to break the law. We recommend that the Government should target such employers, who are not only easier to identify than those they employ but arguably more culpable. We refer below to the Government's commitment to use the Proceeds of Crime Act as a weapon against people traffickers. We recommend that the Act should also be used to seize profits made from the employment of illegal labour. The Home Office should be pro-active within Government in seeking to ensure that other departments take action against illegal working—for instance, by means of a concerted attempt to prosecute employers of illegal labour for other related breaches of employment legislation..."[55]

78. The Minister of State for Citizenship and Immigration told us that as of 1 May this year the Government required employers to use "a far more secure form of identification" and that employers now knew what the position was and what was expected of them. He added that employers would be given help and assistance to ensure that they could enforce regulations and that "we will police that properly and investigate it properly between now and the time when we can start to see the roll out of ID cards offering employers a more secure and simpler form of identity".[56] The Government also maintained that identity cards would reduce the "pull factor" for illegal immigration by making both illegal working and unauthorised access to benefits more difficult.

79. Identity cards could make it easier for those seeking work to demonstrate their right to do so, and, by the same token, make it easier for the police to show that a company employing illegal labour had done so knowingly.

80. We believe that identity cards can make a significant contribution to tackling illegal working. However this will need to be as part of wider enforcement measures, including action against culpable employers. We repeat our recommendations that the Government should target employers who deliberately break the law and that the Proceeds of Crime Act should also be used to seize profits made from the employment of illegal labour. We welcome the steps the Government has taken so far, but to be fully effective there must be properly resourced enforcement of existing regulations.

81. The Government must clarify what action will be expected from the employer, including whether presentation of card by a job applicant is enough or whether an employer would have to check the biometrics or the authenticity of the card. If so, the Government needs to be clear how often this will be required and what access to biometric readers or the National Identity Register will be available to employers or other agencies.

82. It is clear that a non-EU national with false EU documents would be able to enter the UK and to work. The Home Secretary argued that such illegal immigrants would be caught by the system: after the three-month period during which EU nationals would not be required to register, they would have to register.[57] We are concerned that the three-month period for EU nationals, or those claiming to be such, might constitute a significant loophole: it is difficult to see what would stop someone moving from job to job on false papers. The Government must bring forward proposals to deal with this loophole, as well as making a substantial commitment to robust enforcement of laws against illegal working.

83. It is also clear that the integrity of the UK system will be dependent on the integrity of the passport, asylum and visa regimes in other EU countries. In our visit to Germany we were told of a pilot scheme involving biometrics to prevent fraudulent asylum and visa applications. The Minister of State has set out the UK's involvement in similar schemes. As part of the development of the identity card scheme, the Government should report regularly to Parliament on progress being made across the EU to tackle any weaknesses in other EU countries, and, in particular, those countries currently judged to be the least secure.

84. We noted in our Asylum Applications Report referred to above that, according to research commissioned by the Home Office, "expectations relating to welfare benefits and housing did not play a major role in shaping the decision to seek asylum in the UK within the response group".[58] Such benefits may well, of course, be more significant to illegal immigrants if they can access services without adequate proof of identity. Identity cards will need to be used effectively to control access to public services if they are to reduce this "pull factor". We conclude that identity cards, by reducing the "pull factor" from work, and public services, could make a contribution to preventing illegal immigration, but only if the scheme is properly enforced and complemented by action on access to public services.

ORGANISED CRIME AND TERRORISM

85. The Home Office said that a card scheme would disrupt the use of false and multiple identities by terrorist organisations. The Home Secretary told us that "in excess of a third of those who are engaged in supporting terrorism use multiple identities in order to be able to evade detection and to evade us being able to disrupt their activities, and in tracing those who have undertaken terrorism, even the limited identification that is possible from traditional ID cards has been helpful".[59] He accepted that the scheme would not prevent atrocities such as those in Madrid and Istanbul, but argued that a new database, created from scratch with the use of biometrics, would help. Nicola Roche, Director, Children, Identity Cards and Coroners, in the Home Office, also told us that their estimate was that about £390 million a year was laundered through the use of multiple identities.[60]

86. Evidence from the police endorsed the Home Secretary's view. For example, the Association of Chief Police Officers argued that it would put "an additional hurdle in the path of those who aid and support terrorism by providing funding, false identities and 'safe locations'".[61] The Minister of State for Citizenship and Immigration told us that from his experience in Northern Ireland stopping the use of false identities had helped with police work against terrorism.[62]

87. ACPO also argued that an identity card scheme would be beneficial in areas such as organised crime, people trafficking, the sex trade and money laundering.[63] The Metropolitan Police took the view that while the introduction of an identity card would not of itself lead to a reduction in crime or an increase in detection rates, a society built around an individual's true identity and their ability to prove it would significantly reduce the opportunity for crime in a number of areas.[64] Jan Berry, Chairman of the Police Federation, told us:

    "I do not think that we should underestimate how much time is spent by police officers, checking identities. […] we could be using that time to far better effect."[65]

88. Police witnesses argued that the identity card scheme would only demonstrate its full benefits when it was in the compulsory stage. But while ACPO and the Police Federation argued in favour of making it compulsory to carry a card, the Metropolitan Police stated: "We agree that the carrying of an identity card should never be compulsory nor do we seek powers for individuals to produce their card on the spot".[66] The Metropolitan Police did argue in favour of their having access to the Register under certain circumstances in order to enable them to establish an individual's identity through his or her biometrics.

89. The Minister of State for Citizenship and Immigration has set out how the introduction of identity cards would affect police powers to establish identity:

    "..there will be no new power for the police to stop someone and demand to see their card. Existing police powers to require drivers to produce their driving licence (which could be designated as an ID card) on demand or within seven days at a police station will remain. Added to this, if someone has been arrested for a recordable offence, existing powers will allow the police to take reasonable steps to identify them. This currently includes powers to check biometric information. There are also classes of criminal offences which are non-arrestable and are enforced by sending a summons. In these situations, the police have to be certain of a person's name and address. If a person refuses to identify themselves in these circumstances or the police are not satisfied with the information given, they have a power of arrest after which biometric checks can be made. If it were not possible to identify an arrested person otherwise, for example checking police records, a check could then be made on the person's biometric against the National Identity Register."[67]

90. He also set out what police powers to access the Register would be:

    "There is an exception to the general bar on disclosing information from the register without consent where disclosure to the police is in the interests of national security and for the prevention and investigation of crime, as set out in Clause 20(3) of the draft Identity Cards Bill. Such disclosure may be authorised only where the Secretary of State is satisfied that it was not reasonably practicable for the police to have obtained the information by other means. For example, if fingerprint information is recorded on the register, the police would first have to search their own fingerprint records before resorting to the register. In addition, Clause 24(2) enables the Secretary of State to impose other requirements that must be satisfied before such a disclosure is made."[68]

91. Liberty said that all of those involved in the 9/11 atrocities had either legitimate identification papers or very compelling forgeries and argued that sophisticated terrorist networks would be able either to forge the cards themselves or produce false papers enabling an individual to register under a false identity. During the course of our inquiry, Privacy International released a study on identity cards and terrorism, which argued that "of the 25 countries that have been most adversely affected by terrorism since 1986, 80% have national identity cards, one third of which incorporate biometrics".[69] When we put this argument to the Minister of State for Citizenship and Immigration, he replied that other countries did not have the biometric database that was being proposed as part of this scheme, and that this, rather than the card, was the defence against the use of multiple identities.[70]

92. The Law Society doubted that identity cards would be effective in reducing crime, arguing that the biggest problem for the police lay not in identifying individuals, but rather in linking an individual to a crime. Liberty shared these doubts and argued that the Government had not produced any evidence, for example from experiences in the rest of Europe, that identity cards were cost-effective; they believed that funding for an identity card scheme could be better spent on police recruitment, retention and training.

93. Critics of the Government's proposals argue that it has not made the case that identity cards will help fight crime and terrorism. We think that it is perfectly reasonable for the Government to rest its case on the support of those responsible for fighting crime and terrorism.

94. We understand that the contribution to fighting terrorism would be the ability to disrupt the use of multiple identity, identity fraud and related activities like money-laundering, and illegal migration by terrorists and their networks. While, of course, not all terrorists make use of activities based on false identities, and some will have legitimate national or international identity documents, we believe that effective action on identity would be a real and important contribution to restricting the ease with which terrorists can operate.

95. We note, however, the real benefits of an identity card in fighting serious crime and terrorism are only likely to be achieved with a compulsory scheme covering all citizens and residents. It will also be dependent on the effective use of the scheme to check identities, an issue we discuss in the next sections.

IDENTITY FRAUD

96. The Government estimates that identity fraud costs the country above £1.3 billion a year, about a tenth of the total annual economic cost of all fraud.[71] The Home Office argued that the scheme will combat both identity fraud and identity theft, since starting the registration process from scratch would mean that incorrect data would not be imported. Biometrics would reduce identity theft, since attempts to change an identity would show up automatically. Similarly, lost and stolen cards would be put on a stop list, taking a card out of circulation and making it impossible to use fraudulently.

97. The Association of Chief Police Officers stated that:

    "The proposed scheme has the potential to reduce identity fraud. ACPO acknowledges that the proposed scheme provides further barriers for those seeking to fraudulently create a false identity. This is clearly beneficial in the prevention and detection of crime. ACPO is currently working with partners in both the public and private sectors to develop policies, practices and procedures to combat identity fraud."[72]

98. The Foundation for Information Policy Research were sceptical about the effects of identity cards on identity theft, arguing that it was more a regulatory issue centred on enforcing the correction of inaccurate records; they also said that "fraud patterns do not appear to vary across Europe according to the existence or absence of ID cards".[73] Professor Anderson, the Foundation's Chair, told us that in his experience the main determinant of levels of fraud was not the card technology but the diligence and frequency of online checks on whether a transaction was valid or not. The Law Society also pointed to international comparisons, arguing that no evidence had been produced to demonstrate how similar schemes across the world had made an impact on identity or improving the prevention and detection of crime.

99. We believe there is a danger that in many day-to-day situations the presentation alone of an identity card will be assumed to prove the identity of the holder without the card itself or the biometrics being checked, thus making possession of a stolen or forged identity card an easier way to carry out identity fraud than is currently the case. The availability of readers of cards and biometrics, including to the private sector, is therefore a crucial factor.

100. We think it would be likely that identity cards would help combat identity fraud, but only as part of a wider package of measures. The Government should be clearer both about how and when it expects the card and biometrics to be checked and about what levels of security are appropriate in different circumstances.

ENTITLEMENT TO PUBLIC SERVICES

101. The Home Office argued that an identity cards scheme would allow public services to check whether an individual's entitlement, for example to benefits, had expired or had never existed. "An identity card would give everyone a recognised Government-confirmed proof of identity, would mean that an individual did not need to provide many different cards or pieces of paper, and would combat impersonation and identity fraud."[74]

102. The Parliamentary Under-Secretary at the Department of Work and Pensions, Mr Chris Pond MP, told us that his Department estimated that of the estimated £2 billion total annual benefit fraud, £50 million came from people not being who they said they were when making a claim. He argued that as the Department clamped down on other forms of fraud, there would be more attempts at this type of fraud. He also believed that identity cards might help tackle an additional £50 million of other types of benefit fraud annually.[75]

103. The British Medical Association strongly supported the Government's intention to reduce the inappropriate use of health services, and recognised that the NHS had a legal duty to check the eligibility of individuals wishing to use free health services. Dr John Chisholm, Chairman of the BMA's General Practitioners Committee, expected the checking to continue to be done by the primary care trust, rather than by doctors and their staff.[76] The BMA was, however, concerned that any potential system should not cost the NHS more, in terms of additional bureaucracy, than it saved.[77]

104. Entitlement to services and entitlement to an identity card will not always coincide. For example, the Rt Hon John Hutton MP, Minister for Health, told us that: "It is the case of course that you can have entitlement to NHS care, but not an ID card and it is perfectly possible that you could have an ID card, but not be entitled to NHS care".[78] The Rt Hon Charles Clarke MP, Secretary of State for Education and Skills, noted that "Education legislation does not impose any conditions relating to the nationality or immigration status of children seeking entry to a maintained school. So questions of proving entitlement to schooling are irrelevant".[79]

105. It is not obvious to us why the Government want to charge for NHS treatment, but to provide free education to children of illegal immigrants (and whom the Government wishes to remove from the country by using the identity card scheme).

106. We do not believe that entitlements are currently checked as rigorously as they could befor example, the Secretary of State for Education and Skills conceded that this was the case for post-16 education.[80] The development of identity cards should not be used as a reason to avoid tougher measures now on identity fraud in relation to public services. Indeed, if no culture of checking identity is developed now, it is difficult to believe that identity will be checked rigorously once identity cards have been introduced. If checks are not carried out thoroughly now, they may still not be after identity cards have been introduced. Providers of public services, such as schools and hospitals, need to check identity regularly now.

107. Identity cards would make it easier to establish entitlement to public services. But the Government should take action now to ensure that measures to check identity are developed across public services prior to the introduction of the new card.

108. The Government should also review entitlements to public services across the board with the aim of rationalising and standardising them, since there does not appear to be a consistent set of principles underlining access to government services.

109. Some public services fall to the devolved administrations. The Scottish First Minister, Mr Jack McConnell MSP, has said that the position of the Scottish Executive is that "any proposals for voluntary, compulsory or any other form of identity card system in the United Kingdom that might be used for any matter that comes under the United Kingdom Government's remit should not and will not be compulsory for use in relation to devolved services in Scotland."[81] Andy Kerr MSP, the Scottish Executive Minister for Finance and Public Services, explained that Scottish Ministers had decided to proceed with a voluntary Scottish card[82] because, in their view, the focus of the Home Office proposals had narrowed to primarily the reserved issues of nationality, immigration and employment rights, while the authentication requirements of the UK identity card would be significantly higher than for the Scottish card and the proposed time table for implementation of the identity card was considerably longer than for the Scottish card.[83]

110. The Welsh First Minister, Rhodri Morgan AM, has said that "The current Assembly Government has no intention of cutting back on or denying services to people who do not have identity cards".[84] He explained to us "We will consider the implications for Wales of the Home Secretary's proposals for identity cards as these proposals are firmed up. But, as the Home Secretary has announced, decisions on the use of these cards to access services for which the Welsh Assembly Government is responsible will be a matter for this administration. I used the opportunity of an Assembly Question to provide reassurance to people in Wales that their access to public services will not, in the foreseeable future, be dependent on the possession of an identity card".[85]

111. The situation in Northern Ireland was set out by the Secretary of State:

    "The Identity Card scheme involving the registration of individuals and the issue of identity cards will operate on a UK-wide basis. If, as we hope, we can restore the devolved institutions at an early date, then it would be up to the devolved administration in Northern Ireland to decide what use they make of the identity cards."[86]

112. The existence within the United Kingdom of up to four different systems for checking entitlement to public services will be a possible cause of confusion, particularly where cross-border services are provided. The UK Government should liaise closely with the devolved administrations on these issues, both to avoid confusion and to learn from the experiences of the devolved administrations' own entitlement cards.

EASIER ACCESS TO PUBLIC SERVICES

113. The Home Office said that the card will make it easier to access not only public services, but also those provided by the private sector:

The Parliamentary Under-Secretary at the Department of Work and Pensions argued that identity cards would, for example, make life easier for those claiming benefits, by making the process of establishing identity both more secure and more convenient for the customer.[88]

114. He added that the Register might also make it easier for his Department to contact groups who were not receiving the benefits to which they are entitled, such as Council Tax Benefit and Pension Credit.[89]

115. For his part, the Minister for Health emphasised that it would be important to keep a balance between checking cards and allowing access to services, suggesting that appropriate points to confirm entitlement to free NHS care would be when first registering with a GP or at the first in a series of hospital appointments.

116. The Finance and Leasing Association argued that identity cards would also help with access to private sector services: "an identity card could also assist the financially excluded to access mainstream financial services. Currently due to the information sources available, the main one being the electoral roll, individuals may find it difficult to access these services".[90] Martin Hall, Director General of the Association, explained that "financially excluded people, who typically do not have a passport, do not have a driving licence, perhaps have a meter for electricity and gas" did not have the documents they need to establish an identity under the Know Your Customer money laundering rules. [91] Identity cards would therefore, in his view, help such people establish who they were. Councillor Gerald Vernon-Jackson of the Local Government Association told us: "If you are going to do itif you are going to do an identity cardlet us do it properly and have one that actually works for a range of things, otherwise it becomes a charter for wallet makers that we have to carry five or six of these things around, and that just seems to be inefficient and really does not do government a great service".[92]

117. The Government claims that identity cards will make access to public services easier, but it has not set out in detail how this will be achieved. Witnesses indicated two ways how cards might in principle help. First, a single card could provide a substitute for multiple forms of identity where proof of identity was required, including public services, such as benefits and local authority services, and private services, such as proof of age in bars or meeting the requirements of money-laundering legislation. Second, an identity card could make it easier to access other types of services. For example, it could potentially enable an individual to access their NHS records or ensure that the electoral register was up to date with changes of address.

118. The Government's current proposals would improve access to public services to the extent to which this depends on identification. It is important to ensure that the convenience to the state of having a comprehensive system of identifying individuals and accessing data about them is accompanied by an increase in convenience to the individual. The benefits must not be entirely, or even predominantly, to the state.

119. The Government has not developed coherent proposals for using the identity card in other ways to improve access to a wider range of services and information or to promote greater coherence across public services. As a result, citizens are still likely to be required to carry a wide range of cards and documents to use many local and national, public and private services. We believe that this is a missed opportunity.

120. Many of the benefits noted above are not dependent on a card being compulsory. The Secretary of State for Education and Skills, the Minister for Health and the Parliamentary Under-Secretary at the Department of Work and Pensions agreed that the benefits of even a voluntary identity card, such as the better management of services and the elimination of fraud, would lead to efficiencies and benefits both to the citizen and the State across a wide range of different services.

121. The Home Office argued, however, that some of the expected benefits of identity cards, in particular those relating to countering organised crime and terrorism, would only accrue if a scheme were compulsory. They believe that this would encourage service providers, public and private, to use the scheme consistently, and that only a compulsory scheme would be non-discriminatory, since everyone would have a card.

The Government's proposalssome key issues

122. There are some common issues which arise from our discussion of the government's aims.

123. Identity cards would assist with each of them, although for many compulsion would be necessary. However, in each case, cards will have to be properly checked, and their utility will depend on verification and security being at the most appropriate level. The system will have to be sufficiently reliable, and the proportion of cards incorrectly rejected sufficiently low, for checks to be accepted as a normal part of the provision of services.

124. The Home Office told us that there would be a number of different ways of checking identity through a card. They gave the following examples:


125. We note that at the moment there is very little clarity about the level and nature of checks that will be required and carried out, even though this is fundamental to the whole scheme. We recommend that the Government should provide estimates of the proportion of checks that would be biometric and therefore highest security.

126. It is not clear that Government departments have identified how the operation of their services, or entitlement to them, need to be changed to make best use of an identity card system.

127. In most cases, identity cards will only be fully effective if complementary enforcement action can be taken.

128. Finally, more could be done to check identities today and there is a danger that action will be delayed pending the introduction of an identity card.

Public support

129. We note that public support for the principle of identity cards has so far been high. A poll carried out for the Daily Telegraph in September 2003[93] found 78% in favour of the introduction of national identity cards (15% opposed, 7% don't know); 81% were in favour of compulsory cards (dividing roughly half and half on whether it should be compulsory to carry one at all times). 92% would welcome or not mind the introduction of identity cards or reluctantly go along with them, while 7% would strongly object and absolutely refuse to acquire one. Around 80% believed that cards would help cut down on 'health tourism' and benefit fraud and make it easier to catch bogus asylum seekers and others attempting to avoid deportation (13% disagreed in each case); 60% also thought that it would be easier for the police to catch criminals (26% disagreed). A poll carried out for the IT consultancy Detica gave similar results in April 2004.[94] 50% were strongly and 30% moderately in favour of identity cards (5% were moderately and 6% strongly opposed), while 73% said they were unconcerned that identity cards would affect their civil liberties, and 83% were happy to carry a national identity card at all times.

130. However, a poll commissioned the following month by Privacy International found only 61% in favour of identity cards, with the number saying they were strongly opposed at 12%.[95] It is also the case that those questioned were less than enthusiastic about paying for them (see paragraph 134), and were distinctly sceptical about the Government's ability to introduce them smoothly58% of those questioned by Detica had little or no confidence in thisor to protect information on the databasein the same poll 48% were not very or not at all confident in the Government's ability to store personal information securely.

131. The Home Secretary thought that demand for identity cards would be high from the opening of the scheme:

    "I happen to believe that once we have got this up and running, as with the pilot for biometrics that we announced a week ago, people will queue up for it and we will have to deal with the flow and the flood of people wanting it much earlier, wanting to renew their passport and get an ID card very fast."[96]

Intellect, the trade association representing IT, telecoms and electronics companies in the UK, was less sanguine and argued that "an ID Card will have to deliver rapid and compelling benefits to citizens to shift negative perceptions and establish the foundations for long term success".[97]

132. Other witnesses expressed the concern that even a voluntary scheme would become effectively compulsory if a card was required to access large numbers of services; the Law Society was concerned that "in practice, a voluntary scheme could become in effect compulsory as more and more organisations and service providers required production of the card to prove identity".[98] The Information Commissioner raised the linked issue of inappropriate demands to produce a card:

    "I would be unhappy if a local authority gymnasium required you to produce your identity card in order to make use of that local authority gym. I think that is going well beyond what should be acceptable. The local video shop, when you are renting a video: for them to insist upon the production of an identity card for you to rent a video would in my view be disproportionate and unacceptable and I would want to see appropriate restrictions stopping that sort of activity."[99]

133. It may be that citizens will choose to use identity cards voluntarily on an extensive basis. However, until identity cards are compulsory there should be realistic alternatives to their use in every case. There should also be effective restrictions on inappropriate demands for them.

134. We also note that public support for the principle of identity cards does not extend to the reality of having to pay for them. In the Daily Telegraph poll of September 2003, cited in paragraph 129 above, only 1% believed that £40 was a reasonable charge for an entitlement card, while 86% thought the card should be provided free. Similarly, in the Detica poll also cited in paragraph 129, 40% did not want to pay anything, while a further 32% were only prepared to pay up to £25.

The 'voluntary' stage

135. The Home Office made a distinction between the second phase, when it will be compulsory to register, and the opening phase of the scheme, which they describe as 'voluntary'.[100] The Law Society were opposed to a compulsory system, though not necessarily to a voluntary scheme, but argued that even before the Government moved to the compulsory stage of the proposed scheme, cards would not in reality be voluntary, since anyone who needed a new passport or driving licence, or who moved house, and thus had to update their driving licence, would have no option but to apply for a card and to be put on the National Identity Register. They characterise the word 'voluntary' as 'very disingenuous', since anyone renewing a passport or driving licence would have no choice but to register and pay for an identity card version of those documents: they suggested that it should be possible to choose a non-identity card version until the compulsory stage was reached.[101] The Information Commissioner similarly talked of "an illusion of choice".[102] The Home Office said that the Government's proposal would be 'more convenient to the public' and increase take-up; they also argued that it would not be in the public interest to continue to issue less secure passports and driving licences.[103]

136. For most people, to travel abroad and to drive are fundamentals. It cannot be argued that these would be given up voluntarily. To describe the first phase of the Government's proposals as 'voluntary' stretches the English language to breaking point.

137. The Government expected to cover 80% of the economically active population within five years of the beginning of the first phase.[104] The Director of the Home Office's Identity Cards Programme told us that they would expect to issue between 10 and 17 million cards a year.[105] This may be optimistic: according to the Home Office, the highest annual figure for new and renewed passports and driving licences over the past five years is less than 12.5 million.[106]

138. Given the Government's decision to base identity cards on passports and driving licences, we believe the incremental approach to introduction is justified. We set out our concerns about the implications of this choice in paragraphs 19-20 above.

Vulnerable groups

139. A wide range of witnesses expressed concern to us over the effects of any identity card system on vulnerable groups, such as the elderly or those suffering from mental illness. The British Medical Association, for example, said "Socially excluded groups are among the least likely to access public services effectively, and the need to register for and hold an ID card in order to access vital services could operate as an additional disincentive."[107] The Metropolitan Police linked this concern to the issue of how the card would affect minority ethnic groups:

    "The scheme could become compulsory prematurely for those disadvantaged members of society, because they would have to have an ID Card in order to access Social Security Benefits, etc. It should also be noted that many of the visible ethnic minorities are over-represented in this socio-economically deprived group. We have severe reservations that the scheme could add to tensions at a time when the police service is investing greatly in gaining confidence across all communities."[108]

Trevor Phillips, Chair of the Commission for Racial Equality, also noted the concerns of minority ethnic groups over the operation of an identity card, pointing to "a pretty profound lack of evidence about the potential impact of the scheme on different ethnic groups and communities."[109] The Director of Liberty was concerned that identity cards might exacerbate existing tensions with minority groups;[110] Mr Philips pointed out that the perceptions and expectations of such groups regarding identity cards were as important as the reality in affecting community relations.[111]

140. Press for Change (PFC), "the largest representative organisation for transsexual people in the UK", made the point that even before the move to full compulsion, anyone applying for a new passport or driving licence would have to apply to be put on the National Identity Register and would be given an identity card version of the relevant document. They argued that "Protection for trans people and other vulnerable minorities must therefore be assured before any implementation of the NIR or of enhanced versions of existing documents, rather than delayed until consideration of universal compulsion"[112].

141. The effect of the identity card scheme on minorities, such as the elderly, the socially excluded and ethnic groups, is of the utmost importance. The Government should ensure that the scheme imposes no new disadvantages on these groups, and do so before it is implemented. In paragraph 256 we make a practical recommendation on how legislation should approach such groups.

The National Identity Register

142. There are a number of basic questions about the design of the scheme. These include whether its aim is to enable an individual to prove his or her identity or to enable him or her to access services. There are also different degrees of possible control over information on the database: an individual might wish to allow some, but not others, to have access to their address or their past or current names. The answers to these fundamentally affect the design of the scheme, how it would be used in practice and how it would affect individuals in their daily life.

143. There is a wide range of options for the design of the identity cards scheme. Andy Jebson, of Cubic Transportation Systems, argued that the possibility of a pure identity card that would never be anything else had to be balanced against the richness of data that might be used for other purposes later.[113] John Harrison, a Director of Edentity, argued that an individual had many different identities, each a function of a relationship with a different organisation or body, and that the question was therefore how an identity card, evidence of an individual's relationship with the Home Office, should be used to identify that individual in other relationships, such as with a health provider.[114] There were also differences of opinion, as noted below, on the proper architecture of the database, the relationship between the card and the database, the type of card and the information held on the card itself. Witnesses were, however, unanimous on the importance of the Government taking the over-arching decisions at the beginning of the process and sticking to them; as Mr Harrison put it:

    "You take, you should take, the big decisions about the infrastructure, the basic shape of the thing, very early on. You create the outline, and then, as time goes by, you can fill in the detail. What is very, very expensive and almost catastrophic is to go a number of years down the path and then change the overall outline."[115]

144. We do not ourselves have the expertise to make judgements on the technical issues involved in setting up a national identity card system, but we have been struck by witnesses' insistence on the importance of the Government getting the structure right from the beginning and sticking to its decisions. We are concerned that the Government's approach has not taken into account the need to ensure adequate technical debate and public scrutiny of the design of the system.

ARCHITECTURE OF THE DATABASE

145. There are technical issues of considerable importance arising from choices made about the nature of the database. Professor Martyn Thomas, representing the UK Computing Research Committee, as noted in paragraph 62, argued that to create a single national database was risky; Professor Ross Anderson, the Chair of the Foundation for Information Policy Research, also warned against trying to cover too many functions at once:

John Harrison, of Edentity, was cautious about the ways in which a single database would be used, arguing that occasions on which identity needed to be proved to the Home Office were relatively few, while identity might have to be proved more frequently for other purposes, such as in the work place or a hospital, and that it would be "nonsensical" to create an infrastructure that threw all of those back at one central database.[117]

146. Andy Jebson, of Cubic Transportation Systems, argued from his experience that a single database should not be ruled out, since it was possible to contain a database in a very secure facility in one single site and thus to limit the access to the data over communications links. Furthermore, existing systems could be completely backed up in a separate, highly secure site. He acknowledged that security could never be 100% certain, but argued that even using currently available technology it was possible to reach such a high level of security that risk was significantly minimised.[118]

147. The structure of the database, and how to set it up and manage it, are among the most important choices the Government has to make. We are greatly concerned that the Government's procurement process appears to be taking these key decisions without any external reference or technical assessment, or broader public debate. We recommend the Government publishes details of consultations with any external bodies and also any technical assessments that have been undertaken.

ACCESS TO THE DATABASE

148. Richard Thomas, the Information Commissioner, told us:

149. The Director of Liberty said "the tradition in this country has always been that […] the state holds a piece of information about you in one place for one purpose."[120] She was concerned that this was being undermined by a culture of data-sharing that disregarded the purposes for which the data had been gathered. The Law Society expressed similar fears. The Director of Liberty also argued that experience showed that compliance with existing data protection laws was not sufficient and that they should therefore be broadened.[121] The Law Society's Head of Law Reform, Vicki Chapman, noted that widespread access to the database would have consequences for the security of the system.[122]

150. The Home Office expected the 'vast majority' of cases involving other Government departments would involve checking whether information provided by an individual matched that on the National Identity Register, rather than requesting additional information.[123]

151. A balance needs to be struck between, on the one hand, protecting individuals from unnecessary access by public and private bodies to information held on them and, on the other, ensuring that users of the database have the information they need for the scheme to fulfil its purposes. Above all, it is important that the public should know who may be able to see information about them, and what that information is. We consider the Government's proposals for access to the Register by the police, security services and officials in paragraphs 266-72.

'FUNCTION CREEP'

152. Concerns were also expressed over "function creep". The Information Commissioner drew a historical parallel:

153. On the other hand, the Finance and Leasing Association took the view that "the database should be available for all legitimate users of information to access either directly or indirectly to facilitate the uses of the card which an individual may make".[125] Their Director-General stated that no lender would be seeking direct access to that central database and stipulated that the purposes for which that data could be used would have to be very strictly defined, such as when there were prima facie grounds for suspecting fraud, for instance, or for the purpose of credit-checking, but not for marketing purposes.

154. It is almost certain that a successful identity card scheme will bring forward new proposals for its use. However it is essential that there should be proper public scrutiny of and parliamentary control over such developments. There are a number of possible examples. Perhaps the most significant development is the move from enabling individuals to identify themselves to enabling the state to identify individuals.

155. The draft Bill effectively establishes a national fingerprint register covering 80% of the economically active population within five years of the scheme's implementation, and 100% once the compulsory stage has been reached. It is a moot point whether Parliament would currently sanction the establishment of a comprehensive fingerprint register solely for crime fighting purposes: to date only the limited extension of finger-printing and DNA sampling for those arrested has been sanctioned. Nonetheless the Minister of State confirmed the Government's intention to use the National Identity Register as a national fingerprint register to identify individuals. He suggested that there would or might be some unspecified limits on the circumstances in which the Secretary of State would permit this to take place.

156. Irrespective of the Government's intentions, we can also expect media and public pressure to use the fingerprint register ever more extensively. The establishment of a national fingerprint register has never been a stated aim of the identity card system. Whatever the merits of such a development—and there has been no debate as to whether an identification through this means would be sufficient evidence to secure a conviction for example—we believe its use should be subject to proper Parliamentary scrutiny and decision and not developed through executive action.

157. It is also likely that that facial recognition technology will develop to the point where an individual captured on a CCTV camera could potentially be identified from the National Identity Register. Again, we doubt whether the pressure to use the system in this way could be resisted forever by future governments.

158. Whatever the merits or otherwise of such developments, their potential should be recognised. It is essential that they do not develop incrementally or by executive action but are subject to full Parliamentary scrutiny. These issues are at least as significant as the decision to make cards compulsory.

159. In a similar way, identity cards are not planned to be a single card for all public services, but it clearly is possible, and perhaps desirable, for a successful identity card scheme to develop in this direction. But this should be a decision of Parliament, not of the executive.

INFORMATION ON THE DATABASE

Addresses

160. The Home Office have said that "the information that is proposed to be held on the National Identity Register is simply that information which is required to establish a person's core identity".[126] There are important practical issues arising from the information held on the database. For example both the Local Government Association and the Police Federation wanted to have addresses on the database, even though they accepted that this would create problems. As Councillor Vernon-Jackson of the Local Government Association said "In London, 40% change address every year. In Southampton, 25% change every year. That is a huge number of cards that would have to be reissuedand that is only having to register once a year. With people who are serially moving, it will be an extremely difficult process to make sure that things are accurate."[127] The Chair of the Commission for Racial Equality similarly pointed to the "impact on different kinds of communities particularly, for example, the 100,000 or more gypsies and travellers in this country who are not settled. If there is an address on the card and they have to keep this up to date every time they move, we can begin to see chaos, catastrophe and all sorts of problems there".[128]

161. The Information Commissioner raised the question of whether an address would be printed on the face of the card. He said that it was currently acceptable for a driving licence, but he was uncomfortable with the idea of an address being on an identity card or a passport.[129]

162. Address information is most obviously necessary in areas such as housing benefit fraud and credit card fraud. One might question whether having addresses on the National Identity Register adds significantly to the security against double claiming at different addresses offered by the effective use of biometrics.

163. The functions of the Register entail establishing an individual's identity in a number of different circumstances. For some of these, such as interaction with local authorities, addresses may be necessary. There is therefore a case for including them in the National Identity Register. But to do so would have significant administrative and operational consequences, since the Register would need to be updated frequently; the extra work could lead to mistakes which would be disastrous if not properly handled. The Government should be more explicit about the case for including addresses and demonstrate that the advantages of doing so outweigh the problems that would be created. The Government should also clarify whether addresses would be only on the Register or whether they would be legible on the surface of the card itself.

164. In many parts of Europe, including Sweden and Germany, where there is a requirement to register addresses, it is a legal requirement for landlords to register their tenants. We recommend that this be adopted if the Government decides to include addresses, since it would help alleviate the problem of frequent changes of address.

The individual number

165. A related issue is whether it is necessary to have an individual number, and of what sort. Some of our witnesses took the view that this was not necessary. Professor Thomas, representing the UK Computing Research Committee, argued that US example showed the dangers of a single life-long identifying number, which, because it was inevitably widely known, gave others the ability to access an individual's personal information or impersonate him or her.[130] He and Professor Anderson, of the Foundation for Information Policy Research, both pointed to the German model, in which an individual's number changes each time the card is reissued, as one that avoided these drawbacks.[131]

166. The current practice in the United Kingdom is that when a Child Benefit claim is made, the Department for Work and Pensions generates a Child Reference Number for each child. This is in National Insurance Number format and is converted to a National Insurance Number when the individual is around the age of 15 years and 9 months and allocated automatically by the Inland Revenue just before the individual's 16th birthday. The purpose of a National Insurance Number is to act as a unique identifier to link an individual to their National Insurance contributions record, and for use as a reference number by Department for Work and Pensions for social security benefits and by the Inland Revenue for tax credits. Policy responsibility for National Insurance Numbers is held jointly by Inland Revenue and Department for Work and Pensions. The possibility thus already exists for every British citizen to have a unique number from cradle to grave, on the Swedish model.

167. The nature of the individual number and its relationship to other identifying numbers used by the state are more decisions that are crucial for the design and development of the system. The Government must be clear and open about the issues involved and enable informed parliamentary and public scrutiny of any decisions.

Biometrics

168. The Home Office described a biometric as "a digital record of a particular physical characteristic that is unique to each individual, such as fingerprints or the shape of a person's face".[132] The Government's case is that biometrics will link individuals to their card and the National Identity Register so that, as the Home Secretary told us, "the moment someone presents the same biometric but with a different identity, a different name and presentation, that would automatically show up as already existing on the database".[133] The techniques being considered for the identity cards scheme are face recognition, iris recognition and fingerprint recognition. It should be noted that the feasibility study commissioned by the UK Passport Service, DVLA and Home Office on the use of biometrics said that "Biometric methods do not offer 100% certainty of authentication of individuals" .[134] The study also suggested that about 1 in 10,000 people did not have an iris that can be used for iris recognition, while over 1 in 1000 fingers were missing or have no readable fingerprint. Even facial recognition, the nearest to universal, would have cases where automatic enrolment fails. From the experience of one of our Members at the UK Passport Service's pilot enrolment project, covered in paragraph 189, we know that the process of registering biometrics can be uncomfortable, even when successful. We are also concerned that there are medical conditions that would preclude the use of iris scanning. Furthermore, biometrics are not constant over a life-time, since, for example, fingerprints can be worn away by manual labour.

169. In addition to "failure to acquire rates", any biometric has both a false non-match, or false rejection, rate (in which legitimate card holders are not matched with their recorded biometric) and false match rates (in which someone is matched to a biometric that is not their own). These are linked, so that a reduction in the false match rate leads to a rise in the false rejection rate, and vice-versa.

170. Figures for performance by biometrics depend on whether the check is one-to-one-checking that someone is who they say they areor one-to-manychecking whether someone is already on the database. The National Physical Laboratory's feasibility study noted that in one-to-one checks good fingerprint systems were able to achieve a false match rate of 1 in 100,000, with a false non-match rate of approximately 1 in 100. Iris recognition could achieve a false match rate of better than 1 in 1,000,000 with a false non-match rate of below 1 in 100. Under ideal lighting conditions, with subjects directly facing the camera and using photographs 1 or 2 months old, facial recognition achieved a false match rate of 1 in 1,000 and a false non-match rate of 1 in 10.

171. The study concluded that for one-to-many checks to be workable with a database of more than 50 million people at least four, and preferably all, fingerprints should be collected, and for iris recognition, both irises. Face recognition was not reliable for a database of this size.

172. However, other assessment of fingerprint matches, such as one by the US General Accounting Office,[135] show far higher error levels. The critical difference between these studies appears to be the assumption about the circumstance in which the fingerprints are taken and checked against the central register. Fingerprints taken to police standards have a high reliability rate; fingerprints taken by less well trained personnel will have higher false matches and false rejections.

173. We assume that almost all one-to-many checks would be carried out by specially trained staff, either on enrolment or by law enforcement personnel in the circumstances outlined in paragraph 89. It is not clear how well the public will tolerate false rejection in other circumstances (for example when using the identity card to open a bank or investment account). The importance of reliable verification again underlines the importance of the Government making clear what level of security they expect the identity system to require in difference circumstances.

174. There are media reports of successful attempts to evade commercially available biometric software, by, for example, using a digital photograph of another person's iris with a hole cut in it for the criminal's eye. Professor Anderson, Chair of the Foundation for Information Policy Research, told us:

    "As things stand I am afraid that iris scanners, like fingerprint scanners, are liable to be defeated by sophisticated attack if they are used in an unattended operation. Attended operation is different, of course, if you train the staff properly they can feel people's fingerprints, they can look carefully at the eye and check there is no funny business."[136]

The Director of the Home Office's Identity Cards Programme told us that the UK Passport Service pilot project, mentioned in paragraph 189, is not intended to test the robustness of biometric technology or how well the system would work on a larger scale.[137]

175. The security and reliability of biometrics are at the heart of the Government's case for their proposals. We note that no comparable system of this size has been introduced anywhere in the world. The system proposed would therefore be breaking new ground. It is essential that, before the system is given final approval, there should be exhaustive testing of the reliability and security of the biometrics chosen, and that the results of those tests should be made available to expert independent scrutiny, perhaps led by the Government's Chief Scientific Adviser.

Medical information

176. The British Medical Association did not want medical information recorded on identity cards, since they "want the public to be reassured that other people who had access to their identity card were not able to access personal health information"[138] and because the information would not be updated sufficiently frequently. For the same reasons they argued in favour of keeping the National Identity Register separate from the planned national electronic health record. We agree with the BMA: it would not be either useful or appropriate to keep medical details on the Register. But it would be sensible for the identity card to be the mechanism that enables individuals to access their NHS records.

The Citizen Information Project and other Government databases

177. The identity card scheme proposes a National Identity Register (see paragraph 45) covering the population of the United Kingdom over the age of 16, resident foreigners and others. Several other Government proposals are in the pipeline for databases covering the entire population, or significant proportions of it. In this section we review these proposals and examine how far they overlap with or duplicate the proposal for a National Identity Register.

178. The General Register Office, part of the Office for National Statistics (ONS), published a consultation document Civil Registration: Delivering Vital Change in July 2003.[139] This explained the legal changes necessary to deliver proposals for modernising the civil registration service in England and Wales, which had been set out in the White Paper Civil Registration: Vital Change published on 22 January 2002.[140] Central to these proposals is the establishment of a central electronic database of key life events (such as birth, marriage and death), administered by the Registrar General.

179. On 6 January 2004 the Registrar General for England and Wales, Len Cook, published a feasibility study into developing a UK population register, with the working title of the "Citizen Information Project" (CIP). The aim of the register was to pave the way for more responsive and personalised public services. He also announced that, with ministerial agreement, a team based within the General Register Office would carry out the detailed development work over the following 18 months. Ministers would then decide whether or not to create a register for use by the public sector.

180. The Registrar General's written evidence for the Committee said that the joint ONS/Treasury feasibility study concluded that:

    "it should be possible to build a population register for use by public services across the UK. This would bring together basic contact information, about people who are usually resident in the UK, such as:

      name (with alternatives, for example, women may use their maiden name and their married name in different circumstances)

      address(es)

      date and place of birth (to distinguish between people with the same name)

      date of death

      sex

      a unique reference number."[141]

181. The Registrar General's note for the Committee also said that the CIP "would have many administrative and statistical benefits. It would become the authoritative source of name and address information for use across the public sector. This would support joined-up delivery and more efficient and effective transaction and back office services. The feasibility study shows the population register has considerable potential for improving public services and for making it simpler for people to update their name and address details held by government".[142]

182. The Registrar General told us that one benefit of the Citizen Information Project would be that: "there will be one place where you could change your address with effect across all the systems where address change could happen".[143] He added "I am completely convinced, without doing any more work, that the statistical benefits of this project will be quite immense for the UK, but they are not at all relevant in a decision as to whether a project such as this should go ahead".[144]

183. We note however that the Registrar General also told us "it will not be possible to eliminate all multiple identities either in the existing data or in future data held on the population register", but argued that "Once an individual is authenticated through the identity cards system, it will be possible to flag the record in the CIP population register that corresponds with that name and address as having been 'authenticated'".[145] We also note that the Secretary of State for Education and Skills understood the purpose of the CIP to be purely statistical.[146] We suspect that this misunderstanding is widely shared.

184. The identity card scheme and the Citizen Information Project are not the only large-scale government databases or card systems in preparation. In December 2003, the Secretary of State for Health announced the award of contracts to run the NHS Care Records Scheme, which would provide all 50 million NHS patients with an individual electronic NHS Care Record, detailing key treatments and care within either the health service or social care.[147] The Department for Education and Skills is working on introducing a Unique Learner Number, and on a common identifier with the Home Office. A database of all children is proposed under the Children Bill. The Office of the Deputy Prime Minister recently completed a National Smart Card Project which, inter alia, set out a framework for local authority smart card implementation and development. Similarly, the Scottish Executive is supporting the development by local authorities of youth and citizen's entitlement cards. These initiatives are in addition to existing databases, such as those of the Inland Revenue, UK Passport Service and DVLA. There are doubtless others of which we are not aware.

185. We doubt that the Citizen Information Project will provide "a strong and trusted legal basis for holding personal contact information" if the information on it has to be confirmed by another, separate identity card Register. There is a very large degree of overlap between the Citizen Information Project and the National Identity Register. The Registrar General mentioned the options of "comprehensive legislation to oversee information matching which in itself was conducted by individual agencies but which improves the quality of individual registers without actually going to the next step of creating a register" and of "common standards for register management in the British government": each of these would be more worthwhile than the Citizen Information Project as it is currently planned.

186. We are concerned by the proliferation of large-scale databases and card systems, since we have seen little to suggest that they are being approached in a co-ordinated way. While we have not taken detailed evidence on current proposals, other than the Citizen Information Project, we have the impression that each government department is continuing with its own project in the hope that it is not going to be significantly affected by other projects. The format of registration on different databases should be coherent and consistent.

187. We believe that the Government must tackle this proliferation of databases, examining in each case whether the number, identifier or database is needed, what its relationship is to other existing or planned databases, how data will be shared or verified and other relevant issues. For this action to be effective, it must be co-ordinated at the highest levels of the Civil Service.

188. We do not think that there should be a central database with all information available to the Government on it. But an identity card should enable access to all Government databases, so that there would be no need for more than one government-issued card.

Registration and enrolment

189. The proposed system aims to cover the adult population of the country, and the Director of the Identity Cards Programme in the Home Office told us that their estimate was that between 10 and 17 million cards would be issued a year in the initial stage, roughly equivalent to the volume of passports, drivers licences and other identity type documents currently being issued in the UK.[148] A pilot project, run by the UK Passport Service, has been set up to study the enrolment process. Originally scheduled to begin in January 2004, this ran into "a series of hardware, software and ergonomic problems"[149] which delayed its start until April. The trial is to cover some 10,000 people and is expected to run for six months; its results will be assessed by Dr Tony Mansfield of the National Physical Laboratory. Dr Mansfield was the co-author of a feasibility study on the use of biometrics commissioned by the UK Passport Service, DVLA and Home Office. Nigel Sedgwick, a Director of Cambridge Algorithmica Ltd, argued to us that it was therefore open to doubt whether Dr Mansfield could be viewed as independent of those specifying, designing and implementing the UK Passport Service's system and evaluating the trial of biometric technology.[150]

190. Witnesses made the point that enrolment is also crucial to the security of the whole system. If forged or fraudulent documents are not spotted, individuals will be able to enrol with false identities (even if biometrics mean that they will not be able to re-enrol in their real ones). Another means of creating a false identity would be if a corrupt official, for example in a British consulate overseas, were able to register applicants with no or minimal checks. There will therefore need to be a high degree of confidence on the reliability of the staff involved in enrolment, and enrolment processes will have to be designed so that through double checking, the scope for a corrupt official to register a false identity is minimised.

191. The Home Secretary conceded that it would be possible for an individual to have created a false identity by the time the scheme was implemented, and to be registered under that identity. But he argued that they would have to keep that identity for the rest of their life in the United Kingdom, since any attempt to re-register under their real identity would be pickled up through their biometrics.[151]

192. An enrolment process that involves biometrics increases the significant practical questions that any large-scale registration process will have to deal with. These include, but are not limited to, registration of the housebound and of vulnerable groups such as the mentally ill. There will also be occasions on which supporting documentation is in a foreign language or script, or even hand-written.

193. The integrity of the enrolment and registration processes are central to both the smooth running of the system and to its security. Without data of investigative or evidential quality, few of the objectives of the scheme can be achieved. Issues the Government must consider include: the number of mobile units to enrol the housebound, the elderly and those in remote locations; how sensitive the equipment is to the environment; the training of personnel; and the need to minimise opportunities for corruption and fraud. More study of these aspects is needed.

Cards

194. If the Government's scheme proceeds, decisions will have to be taken on what information is legible on the face of the card. As noted in paragraph 161, the Information Commissioner told us that there should be no more information on the card that was necessary for its intended purpose: thus, while he did not object to addresses on the current driving licences, he was not comfortable with them being legible on passports or identity cards. He also argued that information on the chip should be either encrypted or on a contact chipwhich needs to be inserted in or run through a reader rather than held near it. (The International Civil Aviation Organisation recommends unencrypted contactless chips.) The Home Office said that no decision has yet been taken on what would be printed on the face of the card nor on what will be encrypted. Decisions would be taken "in the light of work underway on the feasibility of different technologies and analysis with key user groups of the business requirements of the scheme" and in accordance with international rules on the content of chips, some of which are still under discussion.[152]

195. As set out in paragraph 46, the Government's proposals envisaged a family of different documents each of which can be used as an identity card. (We also note the suggestion that Post Office Account Cards might be added to the family of identity cards.[153]) The Government's plan appears to be cheaper than producing a stand-alone identity card, since, according to the Government's figures, a plain identity card would cost about £35, while the additional cost of passport and driving licences identity cards was only £4. However neither passports nor driving licences can, for the time being, be reduced to cards, since the first will need space to record visas and the second endorsements. In addition, there are no plans to issue passport identity cards at British posts abroad, so British citizens resident overseas will continue to use passport booklets. Using passports and driving licences as identity cards may also give rise to difficulties, such as, for example, an individual trying to access a public service while their passport identity card is waiting for a visa at a foreign consulate. The Home Office said that when the identity card is used to access public services they will ensure that procedures are in place to help those whose card has been lost or stolen, especially in cases of emergency. They argued it could also be possible for service providers to check the National Identity Register directly without the need for a card to be produced.

196. Witnesses held a range of different views on the type of card to be used. Richard Haddock, CEO of LaserCard, took the view that all the data on the National Identity Register should also be on the card, so that the Register was accessed only to issue or replace cards.[154] On the other hand, Neil Fisher, Director of Security Solutions at QinetiQ, argued that biometrics could be reduced to a two-dimensional barcode, which the individual could keep on a memory device and print off as necessary.[155] The question of the type of card used is linked to the number of biometric readers needed for the system. Mr Fisher believed that these would be "very widespread",[156] but Mr Haddock argued that technology implementation would be slower than planned and that there would be a continuing requirement for a physical card with a picture, name and fingerprints.[157]

197. The type of card to be used is a decision of the same order of importance as the architecture of the database, since it has consequences for issues such as how the card will be used and the number of readers and the infrastructure needed, both of which have significant implications for costs. Some choices, such as the nature of the chip, seem to follow a decision to use the passport as an identity card (and therefore follow ICAO) rather than any independent assessment of what would be most appropriate for an identity card. We are concerned that the Home Office appears to be taking these key decisions without any external reference, technical assessment or public debate.

198. The Government's figures on how much cards would cost compare them to 10-year passports and driving licences. The Government has not, however, confirmed explicitly how long the validity of identity cards would be. It must do so before the Bill is published.

Readers and infrastructure

199. Whatever the type of card used, it is clear that there will need to be large numbers of machines able to read the chip on the card, the biometrics on the card or an individual's biometrics. The only public estimate of the numbers needed was given by the Parliamentary Under-Secretary at the Department of Work and Pensions, who told us his Department estimated they would need some 4,500 readers; he was unable to be more precise than to say that they expected to need only "a small number of full biometric readers".[158] No other Department, including the Home Office, has given any public indication of the number of readers it expects to require, although the Health Service, for example, will need many thousands if every practice and hospital is to have at least one. We note in addition that the Police Federation expect to be able to use mobile readers.[159] There is also the question of readers for use by the private sector: if the scheme is as successful as the Government hopes, there may well be many thousands of these; the Finance and Leasing Association told us that they had distributed 28,000 ultraviolet lights for use with the existing driving licences, but expected that cost would be the deciding factor for individual businesses.[160] (The cost of readers is discussed in paragraph 208.)

200. Little attention has yet been paid to the physical infrastructure that will be needed to support the system, including issues such as how biometric readers and enrolment equipment will be connected to the central database. These issues are vital: if, for example, the infrastructure is insecure, the viability of the whole system will be compromised. Similarly, the costs are a significant factor: it is clear that they are not covered in the central budget and the presumption therefore is that each Department will have to pay. It is not clear that they are committed to doing so; as a result there is a clear danger that Departments will tend to rely simply on presentation of the card, without checking its validity or the biometrics against the central database. This would undermine the value of the whole scheme.

201. We are deeply concerned that the Government has published so little information about the number, type, distribution and cost of card readers and the infrastructure necessary to support this. This information is not only essential to proper costing of the scheme, but also to an assessment of how effective the scheme will be.

202. We are also concerned that the Home Office may be leaving it to other government departments, local government and the private sector to decide what level of investment to make in card readers and infrastructure. There is an obvious danger that each organisation will opt for a low level of security, relying on others to raise the level of security in the system as a whole. If this happens the value of the identity card system will be significantly undermined. We also expect the Home Office and other Departments to give at least broad estimates of the numbers of readers they expect to need of each type and what level of provision other organisations are expected to make.

Multiple cards

203. Press for Change recalled the long-standing common law principle that a person is free to use more than one name, provided that there is no attempt at fraud or the avoidance of an obligation. As examples of legitimate use of more than one name they gave people living trans-gendered lives, singers, writers and actors with stage or professional names and married women using their maiden names for professional purposes. They therefore argued that what they saw as legitimate flexibility should be maintained by explicitly allowing the issue of multiple identity cards to a person and the explicit provision of privacy for the link between those identities.[161] The Home Office envisaged building on procedures for issuing passports and driving licences when issuing an identity card in an alternate name or gender. They said that exact arrangements for issue of identity cards to transgendered persons have yet to be decided, and noted the need to take into account the Gender Recognition Bill.[162] We support the issue of multiple identity cards to an individual in cases where there is a legitimate need, and welcome the Home Office's expression of flexibility on this issue.

Security

204. We were told in Germany that the German identity cards were very secure. Witnesses from card manufacturers, such as Richard Haddock of LaserCard, were confident about the technological security of their products, while stressing the importance of proper security and audit procedures in the production of cards. Industry representatives also argued that the security of the system and of the database could be managed in such a way as to make the risk of fraud "minuscule".[163]

205. Other witnesses, such as the representatives of the UK Computing Research Committee and the Foundation for Information Policy Research, argued that even if the card or the database were effectively secure against attack (which they thought improbable), those seeking to create false identities would simply try to subvert the enrolment and issuing process, for example at a British Consulate abroad.[164]

206. It is obvious that no system or card will be entirely infallible. It is also clear that the enrolment process, if not put in place with security in mind, may provide the easiest way of subverting the system. But if the card and the system are set up properly, any successful attack on them would have to be determined and technically sophisticated. Professor Anderson of the Foundation for Information Policy Research argued that "the main determinant of levels of fraud is not the card technology that you use but how diligent you are at checking online whether a transaction is valid or not".[165] This suggests that systematic use of card readers by service providers would also be a deterrent to fraud.

207. We believe that an identity card system could be created to a sufficient level of security. We stress, however, that the security of the system depends as much on using the proper procedures with the appropriate level of scrutiny to verify the card in use as it does on the integrity of the card issuing process or the identity register.

Costings

208. The Home Office's public estimates for the cost of the proposals remain the £1.3 to £3.1 billion set out in the first consultation paper of July 2002.[166] The Government's intention is that the costs of running the scheme will be met from charging: these figures therefore include revenue from the cost of the cards to individuals. Officials giving evidence to us cited commercial confidentiality for saying nothing more than they were "moving away from the lower end" of this £1.8 billion range.[167] The Home Secretary told us that this figure does not include the costs of biometric readers and other equipment to be installed in other Departments, such as the Department for Work and Pensions, or the National Health Service.[168] The overall costs of these would, in any case, be difficult to establish as there is little if any certainty about the numbers and types of readers required.

209. The Cabinet Office Regulatory Impact Unit has issued guidelines on Regulatory Impact Assessments (RIAs), which state that a partial RIA is required to be issued alongside any public consultation (the final RIA is laid before Parliament alongside the resulting Bill) and that it should cover thirteen different points, including describing and quantifying the scale of the risk (or problem being addressed), identifying who is affected, including the business sectors affected, estimating the benefits and costs and identifying the key risks associated with each option and flagging up any potential unintended consequences.[169]

210. The draft Bill was not accompanied by a RIA, since the Home Office argued that it imposed no regulations on any private sector or voluntary organisations, and did not mandate which public sector organisations would be involved in the issuing of cards, and that it was not therefore appropriate to publish a RIA for the scheme. They said that since the Bill, if it became law, would lead directly to contracts for the provision of various services needed to deliver the card scheme, they needed to ensure that any published RIA would not compromise their ability to secure value for money for the scheme and added that provided that they could be assured of this their intention was to publish an RIA when the substantive Bill was introduced.[170]

211. It is intended to subsidise the cost of cards for those on low incomes, and that the subsidy will be funded by the fees paid for other cards and charged for verifying identities. It has been suggested that this is effectively taxation by stealth: the Home Office described these cross-subsidies as service charges and instanced cross-subsidisation of passports, where adult charges cover cheaper passports for children.[171]

212. The Home Office have provided us with details of the assumptions on which their costings have been based, on a confidential basis. We are not convinced that the level of confidentiality applied is justified. Cost information is an essential element in determining the value for money of any project. It is of prime importance where expenditure is funded from the public purse and of particular relevance with regard to public sector IT projects which have a history of poor performance and cost-overruns. We are also concerned that the least robust cost estimates appear to relate to the assumptions with the greatest cost-sensitivity, such as the length of enrolment time, the anticipated number of applications requiring further investigation, the cost of card production and the criteria for subsidised cards. Changes to any one of these factors could cause significant increases to the cost of the programme.

213. The failure to attach a Regulatory Impact Assessment to the draft Bill, or to provide any detailed information on estimated costs and benefits, significantly weakens the basis for pre-legislative scrutiny and the public consultation exercise. This secrecy is all the more regrettable since the case for an identity card system is founded on whether its benefits are proportionate to the problems it seeks to address: a proper cost-benefit analysis is an indispensable element of this. The excuse of commercial sensitivity should not be used to avoid publishing a full Regulatory Impact Assessment with the Bill. We recall that the 1996 report of the Committee made exactly the same point about the importance of a detailed analysis to pre-legislative scrutiny (see paragraph 10).

Procurement

214. The Government's record on large-scale IT projects is not encouraging. Some critics, such as Professor Thomas, representing the UK Computing Research Committee, believed that the difficulties inherent in a public procurement project of this scale were insuperable. He also argued that penalty clauses can only guard against financial loss, and not against the unavailability of a public service. The Home Office said however that their adherence to the Office of Government Commerce's Gateway Review system provided reassurance; Intellect believed, based on their two years of work with the Home Office on the issues, that the Home Office had a better understanding of the capabilities of the technology. Other witnesses from the IT industry were unanimous about the importance of taking decisions about the infrastructure and the basic shape of the system early on. Mr Haddock, of LaserCard, held up the process followed by the Italian government as a model:

On 26 May the Home Office announced the choice of PA Consulting as its private sector development partner, to help determine the best way of designing and implementing the scheme. The company will work on the design, feasibility testing, business case and procurement elements of the identity cards programme.[173]

215. We welcome the Home Office's efforts to overcome their record on IT procurement. We do not believe that it is impossible for them to deliver the project on time, to specification and to cost.

216. But we are concerned about the closed nature of the procurement process which allows little public or technical discussion of the design of the system or the costings involved. We do not believe that issues of commercial confidentiality justify this approach. Any potential gains from competing providers providing innovative design solutions are likely to be more than offset by the unanticipated problems that will arise from designs that have not been subject to technical and peer scrutiny.

217. Nor do we believe that the Government's OGC Gateway process has yet demonstrated the robust track record on procurement projects that would allow it to be relied upon for a project of this scale.

218. The Home Office must develop an open procurement policy, on the basis of system and card specifications that are publicly assessed and agreed. The Home Office should also seek to minimise risk, including, as appropriate, by breaking the procurement process down into manageable sections. We have already recommended that the Chief Scientific Officer be invited to oversee the development of the biometric elements of the scheme. We recommend that individuals or groups with similar expertise be invited to advise on the scrutiny of other aspects of the scheme.

Conclusions

219. Identity cards should not be ruled out on grounds of principle alone: the question is whether they are proportionate to the aims they are intended to achieve. Identity cards could make a significant impact on a range of problems, and could benefit individuals through enabling easier use of a range of public services. This justifies, in principle, the introduction of the Government's scheme. But the Government's proposals are poorly thought out in key respects: in relation to the card itself, to procurement and to the relationship of the proposals to other aspects of government, including the provision of public services. These issues must be addressed if the proposals are to be taken forward. It is important that the Government clarifies the purposes of the scheme and makes them clear through legislation.

220. We now examine the draft Bill in the light of these general conclusions.


45   Entitlement Cards and Identity Fraud, Cm 5557 Back

46   HC Deb, 11 November 2003, col 171 Back

47   Identity cards: The Next Steps, Cm 6020 Back

48   Q 609 Back

49   Note to Q 146 Back

50   Qq 179 and 836 Back

51   Home Office, Legislation on Identity Cards: A Consultation, Cm 6178, April 2004, para 2.72, p 29 Back

52   Ev 199; under section 8 of the 1996 Act employers are required to check eligibility to work; there is a range of documents which can be shown as proof. Back

53   Ev 273 Back

54   Qq 184-185 Back

55   Home Affairs Committee, Second Report of Session 2003-04, Asylum Applications, HC 218-I para 247 Back

56   Q 613 Back

57   Q 612 Back

58   Home Affairs Committee, Second Report of Session 2003-04, Asylum Applications, HC 218-I para 75 Back

59   Q 616 Back

60   Q 16 Back

61   Ev 149 Back

62   Q 618 Back

63   Ev 149 Back

64   Ev 239 Back

65   Q 274 Back

66   Ev 240 Back

67   HC Deb, 29 April 2004, col 1300W Back

68   HC Deb 25 May 2004, col 1554W Back

69   Privacy International, Mistaken Identity; Exploring the Relationship Between National Identity Cards & the Prevention of Terrorism, April 2004, www.privacyinternational.org Back

70   Q 618 Back

71   Cabinet Office, Identity Fraud :a study, July 2002, para 2.17 Back

72   Ev 150 Back

73   Ev 191 Back

74   Ev 200 Back

75   Q 549 Back

76   Q 823 Back

77   Ev 271 Back

78   Q 548 Back

79   Ev 173 Back

80   Q 539 Back

81   SP OR 13 November 2003, col 3254  Back

82   See para 184 for this and other card schemes. Back

83   Ev 258-259 Back

84   Official Record of the National Assembly for Wales, 2 December 2003, OAQ30113, p 11 Back

85   Ev 270 Back

86   HC Deb, 16 June 2004, col 965W Back

87   Ev 201 Back

88   Q 570 Back

89   Q 571 Back

90   Ev 188 Back

91   Q 315 Back

92   Q 312 Back

93   www.yougov.com Back

94   www.detica.com Back

95   www.privacyinternational.org Back

96   Q 626 Back

97   Ev 215 Back

98   Ev 222 Back

99   Q 210 Back

100   Ev 203 Back

101   Q 693 Back

102   Ev 285 Back

103   Ev 308 Back

104   Q 6 Back

105   Q 98 Back

106   Ev 307-308 Back

107   Ev 272 Back

108   Ev 240 Back

109   Q 834 Back

110   Q 161 Back

111   Q 845 Back

112   Ev 299; PFC explain "Transsexual people identify themselves as members of the sex opposite to that assigned at birth, and may undergo medical treatment known as gender reassignment. Transgender is a broader term that includes people temporarily changing their gender and appearance as well as transsexual people. These terms are not precise, so PFC uses the broad adjective trans to cover men and women in both categories". Back

113   Q 422 Back

114   Q 426 Back

115   Q 462 Back

116   Q 340 Back

117   Qq 424 and 429 Back

118   Qq 418 and 428 Back

119   Q 204 Back

120   Q 167 Back

121   Q 186 Back

122   Q 187 Back

123   Ev 314 Back

124   Q 748 Back

125   Ev 187 Back

126   Q 46 Back

127   Q 250 Back

128   Q 835 Back

129   Q 756 Back

130   Q 348 Back

131   Q 347 Back

132   Ev 199 Back

133   Q 628 Back

134   Tony Mansfield and Marek Rejman-Greene, Feasibility Study on the Use of Biometrics in an Entitlement Scheme, National Physical Laboratory, 2003, p3, www.homeoffice.gov.uk/comrace/identitycards/publications.html Back

135   US General Accounting Office, Technology Assessment: Using Biometrics for Border Security, GAO-03-174, November 2002, www.gao.gov Back

136   Q 379 Back

137   Q 671 Back

138   Q 839 Back

139   Office for National Statistics, Civil Registration: Delivering Vital Change, July 2003 Back

140   Office for National Statistics, Civil Registration: Vital Change, Cm 5355, January 2002 Back

141   Ev 245 Back

142   Ev 246 Back

143   Q 511 Back

144   Q 531 Back

145   Ev 247 Back

146   Q 607 Back

147   DoH press release 2003/0502, 8 December 2003 Back

148   Q 98 Back

149   Ev 269 Back

150   Memorandum from Nigel Sedgwick [not printed] Back

151   Q 612 Back

152   Ev 308 Back

153   Memorandum from Electronic Data Systems [not printed] Back

154   Q 430 Back

155   Q 432 Back

156   Q 499 Back

157   Q 501 Back

158   Qq 576 and 580-81 and Ev 175 Back

159   Q 245 Back

160   Qq 245 and 295 Back

161   Ev 300 Back

162   Ev 308 Back

163   Q 391 Back

164   Q 397 Back

165   Q 357 Back

166   Home Office, Entitlement Cards and Identity Fraud: A Consultation Paper, Cm 5557, July 2002, p 141 Back

167   Q 93 Back

168   Q 637 Back

169   Better Policy Making: A Guide to Regulatory Impact Assessment, www.cabinet-office.gov.uk/regulation/ria-guidance Back

170   Ev 309 Back

171   Ev 316 Back

172   Q 458 Back

173   "ID cards-Home Secretary announces private sector partner" Home Office press release 196/2004, 24 May 2004 Back


 
previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2004
Prepared 30 July 2004