APPENDIX 12
Memorandum from Dr Itiel Dror, School
of Psychology, University of Southampton
THE GOVERNMENT'S USE OF ADVICE AND EVIDENCE
IN DEVELOPING POLICY: TECHNOLOGIES SUPPORTING THE GOVERNMENT'S
PROPOSALS FOR IDENTITY CARDS
EXECUTIVE SUMMARY
Technological and scientific advances rarely
initiate governmental policy; more often policies arise from political
vision, public pressure, circumstances, and so forth. Thus "evidential
based" is often obtained and used after political, personal,
and psychological commitment, and thus its impact on policy development
is limited. There are difficulties and challenges in obtaining
good, independent, and objective advice, especially within the
time scale of policy development. This is due in part to the nature
of research, finding people who are not stake holders or affected
by the policy (directly or indirectly), and dealing with long-term
predictions that relate to the policy. Technologies and policy
do not exist in a vacuum and cannot be examined in isolation;
they need to be considered within a wider context, how they can
be integrated and work with humans (including people with disability),
their social and ethical implications, and so forth. Identification
cards and use of biometric technology is a par excellence example
of such issues.
1. The Government's proposals for identity
cards raise difficult questions concerning technological feasibility,
validity and costs as well as wider social implications, all of
which shape outcomes in terms of costs-benefits. Also, this is
a very large scale long-term project in terms of initial investment,
running, constant upgrading and adjustment to new emerging technologies
and needs. Thus, just to mention three out of many salient issues:
(i) Should the identify cards be based from
the beginning on cutting-edge technologies, with multiple biometric
measures and imbedded chips, or should more conventional cards
be used in line with the experience of other countries?
(ii) Should the identify cards be initially
used only for security purposes or should they serve as a basis
for expanding uses, such as national and local licenses and record
keeping, including driving licenses, taxation and medical records—on
the basis of a national identity number system?
(iii) How much to invest to make the cards
as proof as possible against fraudulent acquisition at time of
initial distribution and then against general falsification? Being
clear from the beginning that making them more secure will require
more complex and expensive technologies and measures, and that
no mass-used identity card can be completely immune to theft and
falsification.
2. This short memo addresses the issue of
evidence base for making decisions on these and related issues,
leaving out consideration of other issues such as administrative
processes for supplying the cards, possible integration into European
Union identity cards, and a multitude of political, legal and
ethical/value issues.
3. In writing this memo comparable experiences
with large scale technology-intense projects have been taken into
account, as well as personal work of the author in major innovative
weapons systems and in forensic technologies and their uses, especially
from a cognitive sciences perspective.
4. A fundamental dilemma must be recognised.
Unavoidably, many relevant technologies have to be developed after
a decision in principle is taken on introducing identity cards
and their main initial design. Thus, the decision is in part a
"gamble" based on guesstimating, as much of the needed
evidence is not ready on the shelves but has to be developed after
the initial decision is taken. This is the case because many salient
questions can be specified only on the basis of an initial design
and after large funds for producing relevant evidence becomes
available. Furthermore, most often the time scale of research
is much slower than that of policy development.
5. However, taking decisions on the identify
card in the absence of much of the needed evidence is risky, and
puts much of the evidence production into a "tunnel vision"
with evidence casting doubt on the initial decision being unwelcome,
to put it mildly. This not only leads to the lack of objective
and reliable evidence that can really be used for policy development,
but the optimistic nature of such assessments downplays (if not
totally overlooks) vulnerabilities and potential pitfalls.
6. Therefore, maximum efforts to collect
and process objective and independent assessment on the proposed
identity cards to produce adequate evidence for making a preliminary
decision on the identity card are recommended and missing. This
evidence should mainly include the feasibility of the policy and
its effectiveness. It should also include cost estimates, evaluation
of public reaction, and wider social and ethical implications.
Field experiments such as with preliminary versions of alternative
identity card dummy examples may also be essential. Such steps
are all the more important because once an initial design is approved
changing it becomes difficult.
7. In any case, the initial design should
be "robust" in the sense of permitting adjustment to
emerging new evidence. Given the lack of evidential based initial
decision, it might be advisable to include into the initial decision
a formal proviso that after five years the initial design will
be reconsidered totally on the basis of available new experience
and evidence with the design to be revised and even discarded
as far as may be justified. This five year review and relevant
evaluation studies should be entrusted to independent bodies not
having a vested economic, political, or bureaucratic interest
in sticking with the ID scheme in general, or with the specific
initial design.
8. In other words, introduction of the identity
card, even if it were initially based on the best and objective
evidence available at that time, should be regarded as a "proof
of concept" with a steep learning curve, provisional until
re-evaluated and re-approved with less or more radical changes.
9. In tandem with the initial decision,
large scale funds should be made available for basic research
and technology development essential for more advanced phases
of identity card design and production. Doing so will assure that
salient evidence becomes available in time when the initial design
is re-evaluated so as to reduce lag of identity cards after experience
and knowledge and minimise the need for further "gambling"
and enforced retirement of identity cards made prematurely obsolete
because of too early "freezing" of designs not based
on evidence that could have been available.
10. Introduction and usage of identity cards
is a critical choice for generations to come. From the onset proper
objective and independent research should have been carried out
so as to advise and shape the development of the policy. However,
policies are rarely initiated and motivated by evidence; they
result from political vision, public pressure, circumstances,
and so forth. Hence, "evidential based" knowledge is
almost always obtained and used after political, personal, and
psychological commitment. Therefore, advice is too often obtained
only as far as it can support the existing and on-going initiative.
Thus, its input in shaping policy is relatively limited.
11. Furthermore, it is imperative to obtain
advice and "evidential base" from capable, independent,
and objective researchers who do not have personal, political
and economical interest in the policy. Whatever the quality of
the initial design, emerging technologies together with experience
with uses of the identity card, including falsifications, will
require periodic redesign with the help of new technologies. Therefore
it is recommended to accompany introduction of identity cards
with constant research and development and evaluation by independent
bodies.
12. It is cost-effective to increase the
utilisation of identity cards for additional purposes, such as
licensing and record keeping. This raises issues of privacy on
the one hand, while permitting improvement in services to citizens
on the other. To provide options for such expanded uses without
deciding on them prematurely, the basic design of the identity
card system should include salient features such as a national
identity number to be given at birth. Here, again, evidence on
relevant experiences in other countries should be collected together
with small scale field experiments to develop most user-friendly
and publicly acceptable processes. This should be done to provide
options for the future even if no present decisions on expanded
utilisation of identity card numbers are taken.
13. Budgets for introducing identity cards
should be accompanied with appropriate budgets for result evaluation
and technology improvement. Such budgets should in part be "tactical",
applied to upgrade the design as initially adopted. But in part
it should be "strategic" and basic, to reconsider the
basic initial design. The "strategic" budget and its
allocation should be supervised by an independent body.
14. It is difficult to estimate in advance
the amount of resources that should be allocated to production
of new relevant knowledge and technologies, both basic and applied.
However, as an initial conjecture based on experience with other
large-scale Research, Development and Engineering endeavours,
it is recommended that about one to three per cent of the costs
of the identity card project as a whole is a very cost-effective
long-term investment in relevant evidence production. This includes
also social technologies, such as on making the project as a whole
very user-friendly (to all members of the public, including elderly
people and those with disabilities) without sacrificing its security
goals.
15. To move ahead in the aforementioned
directions it is suggested to appoint a Chief Scientist for the
identity card authority, making sure a highly qualified and independent
person with a scientific background fulfils this position. He
should be supported by an independent science and technology advisory
group. This Chief Scientist should participate in all high level
decision making forums and be entitled to appeal against decisions
which in his view contradict available evidence to the Minister
in charge, perhaps with access to an appropriate Parliamentary
Committee.
16. To the best of the partial knowledge
of the author, only limited steps in the directions recommended
above have been taken.
February 2006
|
