مرکزی صفحہ Public Money & Management Information technology in public services: Disaster faster?

Information technology in public services: Disaster faster?

آپ کو یہ کتاب کتنی پسند ہے؟
فائل کی کوالٹی کیا ہے؟
کوالٹی کا جائزہ لینے کے لیے کتاب ڈاؤن لوڈ کریں
فائل کی کوالٹی کیا ہے؟
Public Money & Management
April, 1993
PDF, 939 KB

To post a review, please sign in or sign up
آپ کتاب کا معائنہ کر سکتے ہیں اور اپنے تجربات شیئر کرسکتے ہیں۔ دوسرے قارئین کتابوں کے بارے میں آپ کی رائے میں ہمیشہ دلچسپی رکھیں گے۔ چاہے آپ کو کتاب پسند ہے یا نہیں ، اگر آپ اپنے دیانتدار اور تفصیلی خیالات دیںگے تو لوگوں کو نئی کتابیں ملیںگی جو ان کے لئے صحیح ہیں۔


PDF, 823 KB
This article was downloaded by: [Universitaets und Landesbibliothek]
On: 14 December 2013, At: 13:16
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK

Public Money & Management
Publication details, including instructions for authors and subscription information:

Information technology in public services: Disaster

Helen Margetts & Leslie Willcocks



Professor Patrick Dunleavy in the Department of Government , London School of

Fellow in Information Management, Templeton College , University of Oxford
Published online: 07 Jan 2009.

To cite this article: Helen Margetts & Leslie Willcocks (1993) Information technology in public services: Disaster
faster?, Public Money & Management, 13:2, 49-56, DOI: 10.1080/09540969309387763
To link to this article: http://dx.doi.org/10.1080/09540969309387763

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose
of the Content. Any opinions and views expressed in this publication are the opinions and views of the
authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should
not be relied upon and should be independently verified with primary sources of information. Taylor
and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses,
damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection
with, in relation to or arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or sys; tematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://


Information Technology in Public
Services: Disaster Faster?
Helen Margetts and Leslie Willcocks

Downloaded by [Universitaets und Landesbibliothek] at 13:16 14 December 2013

Introducing Information Technology (IT) into organizations is a high-risk,
hidden-cost process. Two government departments are used in this article to
provide examples of common risk issues being played out in public services
contexts in the 1990s. It is argued that some major trends in public
administration leave the sector particularly vulnerable to risks introduced by IT.
In the analysis, design and building of information
systems, at the levels of managerial attitude and
actual practice, issues of risk in public services are
commonly treated as peripheral. Risk is most often
regarded as added to the system when it is in
operation, and as most usually relating to systems
security, the need for back-up in the event of
technical failure, issues relating to data protection
legislation and/or the prevention of fraud. However,
the processes involved in designing and introducing
large-scale databanks and computer networks into
organizations bring new hazards of which those in
positions of responsibility are often unaware. This
is especially true when these people are unfamiliar
with, suspicious, or even frightened, of the
technology itself. In addition, in the public sector
context, the prospect of information systems being
managed badly, going wrong or introducing new
risks, is not a favoured subject for academic study.
Certainly it is not a subject favoured by IT
consultants, many of whom have a strong presence
in the UK public sector.
UK Government expenditure on IT is rising
rapidly, now representing 1.7% of central
government expenditure. Tony Newton, Secretary
of State for Social Security has stated, correctly, that
in his department, 'the future is beginning to arrive
courtesy of I T (Newton, 1989). But what kind of
future? Is the traditional, cautious resistance to
innovation in the public sector being eroded by the
perceived necessity for enormous financial
investment in large-scale computer systems? Is this
an intentional desire to bring public administration
into the 1990s, or are politicians and public sector
managers unaware of the risks and hazards
involved? What are the new risks generated by IT,
and are there any which are likely to be increased
when it is being introduced into a public sector
organization? This article looks at these questions
with reference to examples of public sector
computerization projects.

The Public Sector
A range of factors commonly regarded as
distinguishing the public sector from the private
sector would seem to impinge on processes of
© Public Finance Foundation, 1993
Published by Blackwell Publishers. 108 Cowley Road, Oxford OX4 1JF and
238 Main Street, Cambridge, MA 02142, USA.

adopting IT in public sector environments. These
• Projects are large scale (usually endeavouring to
deal with a national, or a regional, population).
•Inclination to move into unproven technologies
(because they are underwritten by the State).
•Non-interactive decision-making (there is no
customer comeback).
• No 'bottom-line' decision-making: new resources
can always be raised rather than admitting
mistakes; no bankruptcy fear; no fear of
management takeover.
•Separation of policy-making and administration.
•Statutory and parliamentary regulation.
•Responsiveness to political masters and short
political time-horizons.

Helen Margetts is
currently researching the
US Budget with
Professor Patrick
Dunleavy in the
Department of
Government, London
School of Economics.
Leslie Willcocks is
Fellow in Information
Management, Templeton
College, University of
Oxford. He is also
Executive Editor of the
'Journal of Information

A quick glance at these factors suggests that there
may be additional risks diat are specific to or
exacerbated by virtue of the organization
implementing the technology being in the public
sector. Arguably also, the listed factors themselves
may have been changing over the last decade as
government has sought radical redefinition of the
parameters and processes constituting the public
sector. As one example, privatization of IT services
in some local boroughs and some regions in the
National Health Service (NHS) may have made
fear of management or external takeover more
real. Furthermore stricter financial disciplines
may introduce more 'bottom-line thinking' in public
sector IT projects. Equally, central government
initiatives in 1991 would suggest greater attention
will need to be given to customer service issues,
whether or not IT-related, in public sector
organizations. A later argument will be developed
that suggests that such 'new public management'
initiatives, particularly where (as they frequently
are) underpinned by developments in public sector
use of IT, themselves constitute considerable risk.
However, before this set of issues is addressed, and
to provide some concrete examples, risk issues are
first analysed in the context of two public sector
computerization projects, one in the Foreign and
Commonwealth Office (FCO), and the other in the
Department of Social Security (DSS).



Downloaded by [Universitaets und Landesbibliothek] at 13:16 14 December 2013

During 1989 the FCO introduced a new
computerized accounting system to make payments,
record transactions and produce their four annual
appropriation accounts. However, it did not work
properly, and final accounts, which should have
been signed in August 1989, were not produced
until the end of November 1989. These accounts
were based on ledgers which did not balance. The
Comptroller and Auditor General qualified all four
accounts; this was the most serious qualification of
an appropriation account which the present
Comptroller and Auditor General had ever made.
The accounts eventually produced in November
showed a net imbalance of £5.3M, and included
£26.4M which the FCO has posted to 'dump
accounts' in an attempt to resolve the discrepancies.
In February 1988, the FCO appointed Price
Waterhouse to study their book-keeping
requirements and to identify and evaluate a suitable
software package to replace their six-year-old
system. Price Waterhouse identified seven possible
packages and after diese had been evaluated against
various criteria, including their financial
backgrounds, it was agreed that Memory
Computers Ltd should be selected.
The FCO's aim was to run the old and new
systems in parallel from Autumn 1988. However,
because of delayed deliveries by Memory, extensive
parallel running did not start until November
1989. Although most of the necessary software was
in operation, it was not possible to reconcile the
output from the two systems. In February 1990
Memory delivered the final parts of the software,
but shortly afterwards the company went into
liquidation. In the same month the old accounting
system broke down irreparably: the hard disk
shattered. There was no back-up in case of system
failure. After questioning at the Public Accounts
Committee (PAC), it appeared that the old system
had been without back up in the eight years in
which it had been in operation.
Payments made to consultants were calculated
by the National Audit Office (NAO) at £937 000,
including £285 000 paid to Price Waterhouse to
deal with the problems caused by Memory, the
company they had recommended. The implications
from the problems caused were considerable; all
accounting information had to be reconstructed
from raw data. At one stage there was a difference
of £485M between the Parliamentary Grants
recorded as drawn and the sum shown in records
as released from the Consolidated Fund.
This example illustrates one of the most obvious
risks in computerization projects: if information
and systems are not backed up, large amounts of
resources have to be devoted to restoring them if
the hardware fails. It also illustrates one of the risks
of senior managers leaving technology entirely in
the hands of technical experts. Questioning by the
PAC in March 1991 to discuss the qualification of
the FCO accounts reveals a worrying lack of
involvement of senior civil servants in either the
choice or implementation of the new computer


system or the disaster that resulted. The Permanent
Under Secretary of State was unable to state who
had been responsible for the original decision to
employ Memory, going against the advice of the
Central Computer and Telecommunications
Agency (CCTA) (the government agency with the
remit to advise agencies on their use of IT).
Furthermore, he had never met Price Waterhouse
to discuss the problems with them, although
problems with Memory had first started to
materialize in 1989. The reliance on consultants
was such that the PAC suggested that the FCO was
almost ancillary to Price Waterhouse rather than a
prime mover in the project.

Computerization of the DSS: the
Operational Strategy
The Operational Strategy, the project to automate
fully the DSS, provides an example of
computerization on a much greater scale. The DSS
is one of the largest government departments,
consuming 30% of public spending and employing
around 10% of central government staff. The
Operational Strategy was a plan involving the
construction of large-scale computer systems and
the installation of 40 000 terminals in 1000 local
offices and unemployment offices. It is claimed to
be the largest programme of computerization ever
undertaken in Europe. Full national coverage was
expected in 1991.
Work carried out on the strategy falls into
three stages. The planning and design phase lasted
from 1982 until June 1985, designated by Dyerson
and Roper (1990), as 'die endless planning years'.
At the PAC in 1984, concern was raised at the
apparent lack of control by department managers
over the programme's progress. In 1985, the
Government announced its plans for the reform of
social security in a two-volume green paper. This
had far-reaching implications for the systems in the
strategy and new plans had to be made to account
for the subsequent restructuring of benefits. The
third stage started in 1987 when it became clear
that many of the projects had slipped behind their
original target dates. The speed of implementation
after 1987 was fast and furious, and subsequent
target dates for individual projects were largely
met. More detailed descriptions of the history of
the Operational Strategy have been given in
Willcocks and Mason (1987) and Margetts and
Willcocks (1991). In the present context what is
more interesting is: how far and which objectives
have been achieved by the beginning of 1992?

The costs of the project have risen from the proposed
£700 to £2600M (and are still rising). Between
1982, when the first broad estimates were made
and 1988, the estimated costs of the Strategy rose
from £713M to £1749M in real terms (an increase
of 145%) (NAO, 1989). The net savings fell from
£915M to £414M in real terms (a fall of 55%). The
achievement of the net 1989 value of £175M
© Public Finance Foundation, 1993

Downloaded by [Universitaets und Landesbibliothek] at 13:16 14 December 2013


depends crucially on the level of staff savings to be
achieved. A 17% shortfall in the estimate could
reduce that value to nil and put the financial
viability of the Strategy at risk (NAO, 1989). Since
then the estimates of costs have risen to £2.6 billion.
The ratio of work received to work processed for
Income Support has remained fairly constant;
'manpower requirements for assessment have not
decreased to any noticeable degree as compared
with clerical methods' (DSS, 1989).
The NAO found that 'from the outset the
Department aimed to avoid compulsory
redundancies and to absorb savings through natural
wastage' (NAO, 1989). However, by early 1988 it
became clear that, because of increases in necessary
projected savings and regional variations in wastage
rates, other options such as transferring staff
voluntarily to offices with staff shortages or
transferring work to areas with likely staff surpluses
would have to be considered. Eventually a policy
of no compulsory redundancies was proposed
provided that 'the unions co-operate fully with the
deliveryofthestrategy'(PAC, 1989). More recently,
the line between compulsory and voluntary
redundancies has become more blurred, with the
DSS considering compulsory transfers to other
DSS offices and government departments (Dyerson
and Roper, 1990). The cost of such actions is
estimated at between £12 and £35M which the DSS
made no allowance for in its estimates (PAC, 1989).

to claimants, for example VDU screens fordaimants
to work out their own entidement before official
The 'whole person concept' was always open to
a wide number of interpretations, ranging from an
objective of computer systems design (DHSS, 1982)
to that of a key plank in policy formulation. Adler
and Sainsbury (1990b) point out that 'in spite of
numerous rhetorical references to the "whole
person concept" the Operational Strategy
document contained litde evidence that much
thought had been given to what this might mean'.
Adler and Sainsbury (1990a) have produced a twovolume report funded by the DSS on die whole
person concept but this takes the form of a survey
of opinions as to what the most desirable
organization of local offices would be, and there is
litde sign diat the DSS are able or willing to invest
what would be needed to implement the
recommendations. It is clear that the main elements
of LOP, the Income Support System (ISCS) and
the Pension System (PSCS) have been developed
and implemented independendy and there are
few links between them. The claimant is treated as
a 'whole person' on arrival at a local office, where
initial information is given (sometimes from a VDU)
at a single reception point, but processing is still
carried out in separate areas depending on the
benefit being claimed. Staff within local offices still
deal widi one particular benefit and diere are no
plans to train general benefit advisers in local

Job Satisfaction

The accuracy of the new systems has been
cause for public concern. Problems with software
handling the social fund caused die Comptroller
and Auditor General to qualify the 1988-1989
account. One problem included a £200 000 shortfall
in the system's£1.6Mcalculation. InOctober 1990,
the accounts for the 1989-1990 financial year were
again questioned after the head of die National
Audit Office estimated diat one in seven income
support claims, and one in five claims for family
credit, were miscalculated (Independent, 27 October

Dyerson and Roper (1990) suggest that 'job
satisfaction in the short run has increased because
new skills are being acquired and the refurbishment
of the offices has improved the work environment',
but long-term effects look more problematic. A
1988 DSS report on job design concluded that a
wider, flatter hierarchy would result from
implementation, with most duties transferred to
the clerical officer (LOII) level. It is now widely
perceived that this and other assumptions made in
the report were mistaken. Furthermore, the tasks
undertaken by LOI Is appear to have become more
rather dian less routine: 'Rather than being liberated
from calculation in order to spend more time with
clients, LOIIs felt that they had become mere text
inputters' (Dyerson and Roper, 1990). The DSS
Pilot Evaluation Report (DSS, 1989) states that: 'as
far as job design and organization is concerned, it
is still early days', but, given diat increased job
satisfaction for Local Office staff was one of the
primary objectives of the strategy, this might have
been considered a major priority before this stage
in implementation.

The quality of service has improved in some ways.
For example claimants do not have to wait while
case-papers are found. However, most of die
smaller projects that have disappeared from the
plans were diose that were to enhance the service
© Public Finance Foundation, 1993

Although unstated, that die system should work
was presumably an implicit objective of the plans
for computerization. The speed of implementation
after 1987 undoubtedly caused numerous and
major errors in die early stages. In February 1990
the weekly average number of faults officially
reported averaged 80 per week. One office had
logged 1860 faults up to 5 March 1990 (NUCPS,
1990). However, such errors can be expected to
decrease as the system becomes more stable.
According to die DSS, the system is running at
98% availability, but diese figures do not include
down-time agreed by die department to allow
work on the system by technical divisions, or
unavailability affecting one or more staff within an
office and unless a problem affects all offices it is not
recorded (NUCPS, 1990). When the full



complement of local offices are online it seems
possible that such failures could become more

Common Risks in IT Projects

Downloaded by [Universitaets und Landesbibliothek] at 13:16 14 December 2013

Risk is involved in all IT projects. In research
across a range of government departments
Willcocks and Margetts (1991) found a range of
critical variables influencing levels of risk. As one
generalization the higher the level of
'innovativeness' the I T represents to the
organization, the higher the risks being undergone.
Other significant factors encountered include:
•Organizational readiness for change.
• IT experience and history.
• The 'strategic maturity' (ability of die organization
to operate strategically).
•Number of functions/departments being
•Size of IT project.
• Industrial relations context.
Another way of assessing risk is through utilizing
innovation theory. Thus Harrow and Willcocks
(1992) adopt from Rogers and Kim (1985) five
attributes of innovations that affect adoption rates
in public sector contexts. In terms of IT projects,
innovations with the following features are more
readily and rapidly adoptable than others:
•Greater relative advantage (over preceding
• More compatibility (with potential adopters' needs
and values).
•Less complexity.
•More 'trialability' (testing possible on a limited
•More observability (results visible to others).
All these factors can be quite easily read back into
the histories of the FCO and DSS to indicate risk
levels in the respective IT projects. Willcocks and
Mark (1989) have also shown them coming strikingly
together in the rush to implement IT designed to
provide management information crucially
supporting the NHS general managers appointed
following the Griffiths proposals. Afurther finding
by Willcocks and Margetts (1991) is that even
situations diagnosed as low risk (that is where many
of these factors tend to support IT adoption), risk
can be reintroduced byfinancialand time constraints
imposed by central government that do not reflect
the realities of project managing robust usable
systems into the organization. To build on the
argument so far, the following represent some of
the most common risks, as illustrated by the above

Risk of Costs Rising out of Control
In the case of the DSS Operational Strategy, the
original proposal, published in 1980 but on which


work had started under the previous Labour
Government in 1977, appeared to assume that
reductions in cost and increases in service and staff
satisfaction could all be achieved without
acknowledging possible tension between them. A
foreword by Patrick Jenkin, however, warns that:
we shall, as time goes on, have to make chokes, trading
off objectives like improved service to the public and
greater job satisfaction for the staff against each other
and against resource costs. (DHSS, 1980.)

It appears that such choices were indeed made.
Since 1987 one objective has been given priority;
that the Strategy should be implemented as quickly
as possible to achieve the planned staff savings.
However, given die improbability of such savings
ever being achieved, what was the point in finishing
on time? The spiralling costs of the project mean
that DSS management have failed to achieve the
objective to which they gave the highest priority.

Risk that the System will not Reflect User
In 1989 the NAO considered that 'there must be
some risk that projects shordy to be implemented
may not adequately reflect the requirements of
users'. Bearing in mind the NAO's predilection to
phrase criticisms euphemistically, this must be seen
as a reasonably serious problem. For example the
lack of a notepad facility, case-papers contain a
form on which all details of telephone calls or visits
from the claimant are kept and which is regularly
referred to; there was no equivalent method of
recording such information on the new computer
system. Although senior users were recruited to a
plethora of user committees and project steering
groups involved in planning the Strategy, users
below higher executive officer level were not
involved until implementation began in 1987. The
discrepancies between user requirements and the
system itself have narrowed since then, pardy due
to the efforts of regional implementation teams
(Dyerson and Roper, 1990). However, the
functionality of a computer system depends crucially
on the extent to which the systems design involved
understanding the work processes involved, and it
was at this stage that low-level user involvement was
at a minimum.

Risk that the Organization will not
Possess Sufficient IT Expertise
The Operational Strategy has struggled with staff
turnover rates of 45% (Computer Weekly, 22 March
1990). The NAO have identified shortage of IT
staff as the reason for slippages in several
computerization projects in the Ministry of Defence
(MoD), die NHS and the Inland Revenue. Shortage
of information technology staff is not only a problem
for governmental organizations. It has been
described as 'the most important single issue
inhibiting the UK ITindustry'. The situation in the
future will be acutely aggravated by the
demographic pattern.
© Public Finance Foundation, 1993

Downloaded by [Universitaets und Landesbibliothek] at 13:16 14 December 2013


The Trade and Industry Committee in 1989
stated that 'the best solution to the worsening IT
skill shortage is increased service training'. One
witness spoke of the 'terrible legacy of apathy'
about training in the IT user sector in the UK. In
the civil service unions' response to the Trade and
Industry Committee's first report on IT, a series of
recommendations were made for the recruitment,
training and retention of IT staff within the Civil
Service. These include a recommendation that
there should be ring-fenced trainingbudgets which
are planned in advance to be adequate and cannot
be cashed or traded in for consultants, as at present.
These proposals, along with most of the 52
recommendations made by the Trade and Industry
Committee in the First Report on Information
Technology were largely ignored by the
In fact, the DSS dealt with the problems of skills
shortages almost exclusively via the use of
consultants. At the outset, the DSS planned on the
basis that their own staff would undertake most of
the computer development work and that outside
consultants would supplement this as required.
However, consultants were involved right from the
start; these were principally Arthur Anderson, CAP
and Computer Sciences Corporation (CSC). In
April 1984 the major tender for the development
of the system software went to a consortium made
up of ICL, Logica and three universities: Imperial
College, Lancaster and Surrey. InNovember 1984
the Government rejected a plan proposed by the
PAC to overcome data-processing skill shortages
through special treatment for computer staff
(Computing, 29 November 1984). In 1986-1987,
the DSS employed around 150 consultants at a cost
After 1987 when Eric Caines took over as
Director of Operational Strategy, the number of
consultants used on the project increased
significantly. Half the internal programmers
working on the project were moved to other areas
because they were seen as a strike risk. This eroded
the skills base that had started to build up at the
development centre in Lytham. CSC were dismissed
and Arthur Anderson became the principal
consultants used. Price Waterhouse were employed
to recalculate the projected savings of the project.
In 1989 the entire Livingstone Computer Centre
was contracted out to Electronic Data Systems (a
company owned by Ross Perot), a non-unionized
firm, for a preliminary period of five years, after
three firms had been invited to tender. A
memorandum from John Moore, then Secretary
of State, said the decision was part ofhis programme
for 'making my department's computer installations
less vulnerable to disruption from industrial action'
(Guardian, 20 November 1989). The price of the
EDS bid was £15M, less than diose of the other
firms but half a million pounds more than an inhouse bid. There are similar plans for Norcross,
the area computer centre in North Fylde, although
in this case the tender was awarded without
competition. The number of consultants employed

© Public Finance Foundation, 1993

in 1987-1988 was 235, at a cost of £22M, nearly five
times the cost of equivalent in-house staff (NAO,
1989). This is in line with a wider trend in
government computing; in 1987-1988,
approximately one-third of the Government's IT
staff costs was spent on external support (Trade &
Industry Committee, 1989).

Risk of Consultant Bankruptcy
The use of consultants in the DSS undoubtedly
contributed to the spiralling costs of the project.
However, there is another risk involved in the use
or wrote the system going into liquidation. This
was illustrated by the FCO and also recently in the
NHS(PAC, 1991). The only safeguard against this
is to investigate the financial soundness of the
organization before placing contracts: this is very
difficult because of the unpredictability of the
computer consultancy market. In fact, this was the
function that Price Waterhouse performed,
unsuccessfully, for the FCO; even such an
experienced and well-established organization was
unable to safeguard against this risk.

The Separation of Policy-Making and IT
To some extent diere is an unavoidable split between
policy and administrative decisions. It is neither
politically acceptable for technology to dictate policy,
nor financially feasible for policy to be made
independent of the administrative apparatus
available. The Trade and Industry Committee
concluded in 1989 that 'the most effective users of
Information Technology are those where theboard
has a strategy for the company and an IT strategy
which flows from that'. They also recorded a
comment from DTI officials:
the commitment of senior management and by that one
does not mean senior data-processing managers but the
top managers of an organization — is crucial to the
effective implementation of Information Technology.

(Trade and Industry Committee, 1989.)
Obviously these preconditions for success will always
be difficult to achieve in a government organization
where policy is decided by elected representatives
and administration carried out by appointed
officials. Yet during the Operational Strategy the
policy and management domains seemed in some
ways to be unnecessarily divided. The social security
reforms of 1985-1988 had far-reaching effects on
the project. Litde development work had taken
place on the main projects at die time the reforms
were introduced, but die changes meant diat project
plans had to be altered within a relatively brief
period of two years up to April 1988. According to
many sources those involved in planning the
Operational Strategy were not consulted until the
green paper was published. Eric Caines indicated
at the PAC in 1989 that the urgency of modifying
supplementary benefit software to income support
software in time for the introduction of the new


Downloaded by [Universitaets und Landesbibliothek] at 13:16 14 December 2013


benefit was the cause of many of the problems
experienced. Many of the smaller projects, largely
those designed to improve quality of service, were
dropped from the plans. It appears that a rulesbased benefit such as income support was easier to
computerize than supplementary benefit, which is
discretionary. However, the lack of consultation
gives cause for concern.
While those in charge of administration appear
unconcerned with policy issues, it also seems that
politicians are reluctant to equip themselves with
an understanding of IT. As the Trade and Industry
Committee pointed out in 1989, most government
departments now have an IT strategy committee.
Ministers do not attend these meetings. The Trade
and Industry Committee (1989) suggested that,
although these committees could be classified as
Civil Service rather than Ministerial: 'the absence
of Ministers from their meetings shows either that
ministers are not as closely involved in running
their departments as might be expected or that
they do not attach high priority to the use of
Information Technology'.
The separation of policy-making and the use
of IT means that public sector organizations may
be particularly vulnerable to risk of the
requirements of information systems changing
abruptly when policy changes are made. In The
Netherlands this is dealt with by compelling
legislators to produce reports on the implications
for technology whenever legislation is changed;
there is no equivalent system in the UK.
Furthermore, flexibility in policy-making may be
lost when those implementing technology are
unaware of the kind of policy changes that might
be planned or desired in the future. For example,
the possibility of merging the income tax and
social security systems after computerization was
included in the plans for the Strategy. It has been
mentioned as a possibility since then by, for
example, Nigel Lawson in his budget speech in
March 1985. It appears that policy-makers think
that because both systems are computerized the
problems of merging them cannot be
insurmountable. However, John Kenworthy,
Head of the Information Technology Services
Agency (ITSA), has been quoted as saying it
would in fact be impossible; there are technical,
physical and conceptual barriers (Civil Service
College Seminar, June 1991).

Risk of Crash or Failure
Crash or failure can lead to loss of information but
also loss of resources in recovering the damage.
This is demonstrated by the FCO and potentially
illustrated by the DSS. There is little evidence to
suggest that public sector organizations like the
DSS are investing in the disaster-recovery systems
which banks and other large companies in the
private sector know from bitter experience are
necessary if large-scale computer networks are to
be implemented and relied upon with confidence.
In a recent CCTA survey reported to the PAC on


the contingency plans of government departments
for standby hardware and software systems in case
of disaster, the DSS were one of the 10% of
government departments that failed to reply (PAC,
1989). It appears that their contingency plan is
based on the ability to switch work between area
computer centres, a facility originally introduced
to reduce the risk of stoppage due to industrial
disputes. It is unlikely that this contingency has
ever been tested and given that the DSS, like most
implementors of any technology from computer
systems to the M25, have underestimated
transaction loads perhaps they need a disaster in
order to learn how to have one.

Isolation of IT
The Operational Strategy was the first attempt to
introduce computers into the heart of die DSS, by
incorporating their use in all aspects of the
department's work. All the problems encountered
seem to point to one major feature of the DSS's use
of IT. While succeeding in bringing the computers
themselves into the heart of the organization, all
expertise in IT has been kept firmly on the
periphery. While much of the design and
specification of the system has been carried out by
consultants, those with experience of actually using
the DSS administrative systems were largely
excluded. The politicians and policy-makers have
made litde effort to understand the implications of
the technology. Senior managers appear to regard
it as a cost-cutting tool, which can be administered
as far as possible in isolation from the rest of the
organization. The resistance of DSS management
to the training of in-house staff means that
integration of IT into management, improvement
of IT project management skills and the overcoming
of shortages in IT specialist staff may never happen.
The isolation of IT could be increased by the
introduction of the agency structure to the UK civil
service. Further marginalization of computer
expertise in the DSS has seemed likely after April
1990 when all the DSS's information technology
staff were moved to ITSA, deepening the division
between IT and clerical staff. The annual budget
of ITSA is around f 400M. April 1991 also saw the
birth of the Benefits Agency, consisting of all staff
who worked on NI contributions. The management
of ITSAare proud that in the future their worth can
be measured by the competitiveness of the computer
consultancy market; if the Contributions Agency or
the Benefits Agency are dissatisfied with the service
they get then they can go elsewhere. However, it is
far easier for users of computer services to define
what they want when they possess an understanding
of what is possible. If all technological expertize has
been moved to the periphery of the organization,
it seems unlikely that the DSS will know whether
they are dissatisfied or not.

The Future of IT in Public Administration
This final section investigates whether larger trends
in public administration are laying the future public
© Public Finance Foundation, 1993

Downloaded by [Universitaets und Landesbibliothek] at 13:16 14 December 2013


sector especially open to the risks introduced by IT.
Hood (1990) has identified five 'megatrends' in
public administration today. These are:

policies in the future; for example it would now be
prohibitively expensive to implement a system
based on means tested benefits.

•Attempts to check the growth of government.
•The internationalization of public administration.
•Automation in public administration.
•The privatization of public sector provision.
•The rise of the 'new public management'.

Malversation: Problems which Ik in abuse or potential
for abuse of public office rather than in excess costs or
incompetence, paid for in terms of loss of trust and
effective political rights for citizens

As Hood has pointed out, these trends are neither
jointly exhaustive nor mutually exclusive. All might
be facilitated by, or exacerbated by, the widespread
introduction of IT. For example, the Operational
Strategy appears to demonstrate all of the trends.
The objective given the highest priority, to cut
costs, was to be achieved by the reduction in staff
facilitated by the new system. It makes possible
John Kenworthy's (ITSA) aim that UK social security
systems should link up with odiers in Europe.
Germany and Austria already share data. It has
automated the local offices. It was achieved by the
employment oflarge numbers of private consultants
and contract staff in preference to the training of
in-house staff. And the management techniques
used satisfy the criteria of what Hood terms 'the
new public management', a term that he (and
others) have used to indicate a shift away from
issues of policy to issues of management involving
a strong emphasis on cost-cutting and efficiency
criteria. The term also involves a disaggregated
approach to the management of the public sector,
involving the break up of traditional public
bureaucratic structures and attempts to increase
'contestability' and competition in public services
Hood has highlighted three problem areas
likely to arise from these trends. All three can be
illustrated by examples from computerization
projects in the public sector.
Waste: Avoidable incompetence and inefficiency

The new public management places great emphasis
on reduction of waste. The Operational Strategy
appears to provide further evidence for Hood's
thesis that 'it is capable of creating waste too. Its
tendency to denigrate most kinds of technical
expertize other than economics or accounting can
leave it vulnerable to implementation fiascoes'. Its
focus on the short term and the measurable can
leave it vulnerable to the pattern of achieving
transient success followed by collapse.
The Operational Strategy also illustrates the
different types of waste to which the public sector is
vulnerable. IT can be used as an 'instrument of
policy', to facilitate policy-making hitherto
impossible and to make the implementation of
existing policies easier.
However, this means taking a more strategic
approach to the purchase and use of technology,
rather dian viewing it narrowly as a cost-cutting
tool. The failure of policy-makers and politicians to
familiarize- themselves with such approaches has
actually narrowed the possibilities for social security
© Public Finance Foundation, 1993

This is illustrated by the extensive private sector
participation in public management, which has
helped to give it the characteristics of a 'policy
boom' (Dunleavy, 1989), and the increased use of
consultancy, franchising and contracting making
the area of overlap between the public and private
sectors far larger than before. As demonstrated
above, the continued use of private sector
consultants has meant many government
departments becoming involved in expensive
long-term relationships with specific firms that
are virtually impossible to break.
The potential for invasion of privacy by the
existence of the DSS central index also opens up
possibilities for 'malversation'. It has been used as
a strong argument against the privatization of
ITSA: as John Kenworthy pointed out at a Civil
Service College Seminar in 1991 - 'You are all on
my computer now!'. Certainly it raises questions of
democracy, especially with the departmental
exchangeofdata via the Government Data Network.
The possibility of Europe-wide social security
systems indicates the need for a European directive
on data protection at the very least.
Catastrophe: Socially created disasters, the capacity of
government or other human action to create physical
damage on a large scale as a result of system failure

In a discussion of the risk involved in large-scale
computerization, Angel and Smithson (1989) show
how the failure of prediction (more difficult with
information technology than other large-scale
projects) grows when 'what should be a wideranging investigation of the implications of the
new system turns out to be a much narrower cost
analysis', entailing a n u m b e r of critical
assumptions that cannot be justified without a far
deeper examination. Ironically, by being
presented with IT as a mechanism primarily for
reducing costs rather than a potential disaster
area, public sector managers are being encouraged
to undertake enormous financial risks.
Furthermore, Hood suggests that the resilience
of the public sector may be endangered by a break
up of the public sector into a fragmented structure
ofseparate managerial units. Recent reorganization
of the DSS is a good example of how IT can form
partofarethinkingofoverallorganizational design.
In 1989 the 'Moodie report' (DSS, 1988b) proposed
the relocation of administrative functions in
Glasgow, Belfast and Wigan for 21 offices that were
deemed tobe delivering an unsatisfactory service.
These local offices will remain as reception points
with terminals for viewing only while the assessing
of claims will be carried out in the social security
centres. Further reorganization will divide the 500


Downloaded by [Universitaets und Landesbibliothek] at 13:16 14 December 2013


local offices into 200 clusters consisting of two or
three branch offices which will be service delivery
points linked to a main administrative office. The
Moodie Report has initiated a move towards
customer serving local offices supported across the
wires by locationally separate back offices.
This illustrates how fragmentation is facilitated
by, and would perhaps be impossible without, the
introduction of large-scale computer networks. In
addition it shows how the break up can lead to
contradictions between organizational and
technological forms. While the public sector is
being broken up into non-integrated units, the
managers of these units are being provided with
centralized, standardized systems over which they
have no control. This becomes a particularly
dangerous situation when units who are supposed
to share information via such systems are actually
set up in competition with each other, as is
happening with hospitals who are opting out of the
NHS under the current UK health service reforms.

By making economy the primary objective of
computerization, to some extent ignoring the
human factors and losing the wider benefits that IT
can bring, the DSS experienced problems that
were costly and, ironically, took them further from
their aim. Computing is not in itself a powerful and
influential force within organizations. Rather, it is
the actions (or lack of action) of senior management
that have a vital effect on the style of computing that
emerges. The development and implementation
of the Operational Strategy appears to illustrate
many of the trends observable in the management
of public bureaucracy in the 1990s. At the same
time, it highlights the dangers. IT as an instrument
of policy is not an uncontentious or unproblematic
issue. It cannot be used as a straightforward costcutting tool in isolation from the rest of the
administration. If it is viewed as such by 'new'
public managers, then the public sector will become
increasingly vulnerable to risk.

Adler, M. and Sainsbury, R. (1990a), Putting the whole
person concept into practice - Final report: Parts 1
and 2. (Report to the DSS).
Adler, M. and Sainsbury, R. (1990b), The social shaping
of information technology: computerization and the
administration of social security. Paper at the Social
Security Workshop at the London School of Economics.
Angel, I. and Smithson, S. (1989), Managing Information


Technology: A Crisis of Confidence. LSE Department of
Information Technology, Working Paper No. 20.
London School of Economics, London.
DHSS (1980), A Strategy for Social Security Operations.
Working paper. HMSO, London.
DHSS (1982), Social Security Operational Strategy: A Framework for the Future. HMSO, London.
DSS (1989a), Operational Strategy Pilot Evaluation Report;
Management Summary and Main Report. HMSO, London.
DSS (1988b), The business of service. The report of the
regional organisation ocrutiny. Internal document,
Dunleavy, P. (1989), The architecture of the British
central state, parts I and II. Public Administration
Dyerson, R. and Roper, M. (1990), Computerization at the
DSS 1977-89: The Operational Strategy. LBS Technology Project Papers Number 4. London Business
School, London.
Harrow, J. and Willcocks, L. (1992), Management, innovation and organizational learning. In Willcocks, L.
and Harrow, J. (Eds.), Rediscovering Public Services
Management. McGraw Hill, London.
Hood, C. (1990), Beyond the public bureaucracy state.
Extended text of Inaugural lecture, London School of
Economics, London.
Margetts, H. and Willcocks, L. (1991), Information
technology as policy instrument? Trends and prospects in the UK social security system. Paper given to
the EGPA Conference: Information in the Public
Sector. The Hague, the Netherlands.
NAO (1989), Department of Social Security, Operational
Strategy. Report by the Comptroller and Auditor
General. HMSO, London.
Newton, T. (1989), Keynote address at Investing in
Quality: Managing the Technological Change of the
90s. RIPA/Hoskyns Joint Symposium, London.
NUCPS (1990), DSS Local Office Computerization and
Administration. Briefing Document prepared by
NUCPS, London.
PAC (1989), Foreign and Commonwealth Office: Qualification of Accounts. House of Commons paper 275i,
Session 1990/91.
Rogers, E. and Kim, J. (1985), Diffusion of innovations
in public organizations. In Merritt, R. and Merritt, A
(Eds.), Innovation in the Public Sector. Sage, Beverly
Trade and Industry Committee (1989), First Report on
Information Technology; and Minutes of Evidence, Session
1988/9. HMSO, London.
Willcocks, L. and Margetts, H. (1991), Informatization
in UK public services: from implementation, through
strategy to management. Paper at the EGFPA Conference: Informatization in the Public Sector. The Hague,
the Netherlands.
Willcocks, L. and Mark, A. (1989), IT systems implementation: research findings from the public sector. Journal of Information Technology, 4, 2.
Willcocks, L. and Mason, D. (1987), The DHSS Operational Strategy, 1975-1986. Business Case File in Information Technology. Van Nostrand Reinhold, London.

© Public Finance Foundation, 1993