05 July 2005
LSE calls for decentralised UK ID
SA Mathieson
The UK government's plans to introduce a national identity database
and card system passed their first parliamentary hurdle on 28 June,
although with the ruling Labour party's majority in the House of
Commons cut from 67 to 31.
However, the detail of the plans is set to face close examination
in the UK parliament's revising chamber, the House of Lords, following
a highly critical report published by the London School of Economics
and Political Science (LSE), one of the UK's top research universities.
It was produced with the involvement of experts from businesses,
government, academia, non-governmental organisations and professional
bodies.
The LSE report starts by stating that a national identity scheme
could create ‘significant, though limited, benefits for society’.
However, it heavily criticises the UK's current plans, which it
concludes are ‘too complex, technically unsafe, overly prescriptive
and lack a foundation of public trust and confidence’.
The UK government plans a single national identity register holding
more than 50 pieces of data per person, including iris scans, fingerprints
and facial scans, as well as other state registration numbers and
an audit trail of use. The LSE report criticises such centralisation
as ‘technologically precarious’, saying that the system
‘could itself become a target for attacks by terrorists and
others’. It adds: ‘From a security perspective, the
approach to identity verification outlined in the Identity Cards
Bill is substantially - perhaps fatally - flawed.’
Instead, it proposes a decentralised model, with individuals storing
back-ups of their card's information with trusted third parties,
such as bank branches, police stations or solicitors. This would
allow cards to be replaced without the need for a massive central
database.
The report says each card's master identification code would be
invisible, rather than on the face of the card as in the UK's planned
system. A central database would initially hold information about
an application, but this would be deleted when the card was issued.
Only the master code, along with the applicant's name, would be
retained.
Applicants would visit an automated kiosk to apply for a card; the
report says the process should be automated as far as possible,
saving money and reducing the danger of corruption. As with the
current UK passport system, applicants would need the endorsement
of trusted citizens - those in certain regulated professions such
as doctors, as used in the current passport application system -
who already owned an identity card and who had known the applicant
for a period of time, with the application process including random
checks on "referees".
Personal interviews would be the alternative, rather than used for
everyone as in the current plans.
The LSE proposals include federated identity management, allowing
spin-off identities which could be used for other purposes without
risking the primary identity. If the trusted third-party was the
owner's bank, it would be able to include its own digital certificate
on the card to authenticate the user. The current plans require
banks and other businesses to trust the government.
The LSE says it will produce costings for its model later in the
year, but claims it will be far cheaper than the government's, due
in part to much greater automation and a much-smaller and less important
centralised database.
Media coverage of the report focused on the LSE's predictions on
the UK government's current plans, which it priced at £10.6
to £19.2 billion over ten years, compared with an official
estimate of less than £6 billion.
Charles Clarke, the UK's home secretary (interior minister), dismissed
the report as “a technically incompetent piece of work”,
adding that the study “does discredit to the LSE, and is a
mistake from their point of view”. He pinned blame on Simon
Davies, director of civil liberties group Privacy International
which opposes the plan, and also a visiting fellow at the LSE.
Howard Davies, director of the LSE, responded: “This is not
a report based on one person's view. This is a six month 300 page
study guided by a steering group of 14 LSE professors and involving
extensive consultations with more than 100 industry representatives,
experts and researchers from the UK and around the world. We stand
by all our cost estimates.”
Howard Davies said that the costings were based largely on UK government
statistics and documentation, and that the LSE would be happy to
discuss its figures with the Home Office — adding that its
efforts to do this had been rebuffed. He said that Simon Davies'
role in the report had been “primarily co-ordination and support”.
Charles Clarke told the BBC that the report's authors assumed that
individuals would need to re-enrol every five years. However, the
LSE's ‘best case’ cost assumes, as does the government,
that enrolment would only need to be repeated every ten years.
The LSE figures were given measured support by UK public-sector
IT analyst Kable, which priced the scheme at £10.2 billion.
However, the company said that the price could range from £6.8
to £15.6 billion, adding that the Home Office has not released
detailed costings.
Kable's chairman William Heath said: "It's like costing Concorde
when the aircraft industry is about to change the world by launching
the jumbo jet."
Back to news index
|
|
