TENTH REPORT
The Committee of Public Accounts has agreed to
the following Report:
PAROLE
INTRODUCTION AND SUMMARY OF CONCLUSIONS AND RECOMMENDATIONS
1. Parole is the early release of prisoners into
the community under the supervision of the Probation Service.
It was first introduced in England and Wales by the Criminal Justice
Act 1967 and was part of a package of measures designed to "keep
out of prison those who need not be there". The present system
of parole is based largely on the provisions of the Criminal Justice
Act 1991, and is available only to those prisoners serving determinate
sentences of four years or more. Prisoners become eligible for
parole when they have completed half their sentence, known as
their parole eligibility date.[1]
2. The Parole Board is responsible for examining
applications and deciding whether or not to grant parole. It is
headed by a Chairman and has two full-time members and 103 part-time
members. Members are recruited from around the country although
the Board sits in London. The Board's part-time members are drawn
from a variety of walks of life, including barristers, solicitors,
teachers and people with experience of business and industry.
Parole applications are normally considered by a three member
panel.
3. In reaching its decisions, the Parole Board is
directed by the Home Secretary to consider primarily the risk
to the public of the prisoner committing a further offence while
on parole, and to balance this risk against the benefits to the
public and the prisoner of early release under supervision, which
might help rehabilitation and reduce the risk of re-offending.[2]
4. Unwarranted delay in the parole process places
an additional strain on available prison accommodation and is
also expensive - each week's delay in releasing a prisoner on
parole costs in the region of £450. Ensuring that applications
are considered in an efficient and effective manner requires the
close co-operation of a number of organisations, particularly
between the Prison Service and the Parole Board. On the basis
of a report by the Comptroller and Auditor General, the Committee
took evidence from the Home Office as the sponsoring Department,
the Prison Service, and the Parole Board. We considered Parole
Board decision-making, whether the processing of parole applications
is timely, and the arrangements made for preparing prisoners for
release.
5. Three main points emerge from our examination:
- Over the last three years, more effective joint
management of the parole process by the Prison Service and Parole
Board has reduced delays in processing parole applications significantly.
Between 1996-97 and the first six months of 2000-2001, for example,
the Prison Service increased the proportion of prisoners' parole
dossiers submitted to the Parole Board on time from 40 per cent
to 83 per cent; whilst the Parole Board increased the number of
decisions made by the due date from 43 per cent to 92 per cent.
We are pleased to see this significant improvement in performance.
- A lack of information means that the Prison Service
and Parole Board are not able to demonstrate fully that parole
applications are dealt with in a consistent and equitable manner,
for example according to the type of offender, educational attainment
and region of origin. Figures from the Parole Board suggest that
parole rates vary considerably between ethnic groups. Whilst recent
research conducted for the Home Office concluded that most of
the variation could be explained by the type of offences committed
and other relevant factors, the Parole Board and Prison Service
should monitor the relative success rates of different prisoner
groups on a regular basis to ensure parole applications are being
handled consistently and fairly.
- Parole plays an important part in helping prisoners
resettle into the community. Participation in offending behaviour,
drug and education programmes whilst in prison can be beneficial
for the prisoner in obtaining parole and in reducing re-offending.
Currently there are places on offending behaviour programmes for
only about two thirds of suitable prisoners.
6. Our more specific conclusions and recommendations
are set out below.
Parole Board Decision-making
(i) Around 60 per cent of applications for
parole are rejected. Whilst independent research commissioned
by the Home Office into the Parole Board's decision-making suggests
the degree of risk posed by some prisoners may be overestimated,
the Parole Board should have full regard to the need to protect
the public when considering prisoners for parole. The Home Office
should, however, review periodically whether the right balance
continues to be struck between granting parole and keeping prisoners
in prison. (paragraph 13)
(ii) Neither the Prison Service nor the
Parole Board could provide a breakdown of the success rate of
parole applicants according to category of offence, educational
attainment or their regional background. To better target advice
and support on parole matters, the Prison Service, in conjunction
with the Parole Board, should examine the success rates of different
prisoner groups in gaining parole. (paragraph 14)
(iii) Whilst research commissioned by the
Home Office suggested that much of the variation in success rates
between different ethnic groups could be explained, the Home Office
and Parole Board should review the success rates of different
ethnic minority groups within offender categories to demonstrate
that there is no bias when considering applications from prisoners
with similar criminal records. (paragraph 15)
Timely processing of parole applications
(iv) In 1998-99, the Parole Board's target
for parole decisions to be notified two weeks before the prisoner's
parole eligibility date was met in only 58 per cent of cases,
leading to extra costs for the Prison Service, estimated at some
£2.5 million in 1998-99. Additionally, keeping parolees in
custody longer than is legally necessary can be less than fair
to the individuals concerned and their families. (paragraph 25)
(v) By the first six months of 2000-01 92%
of prisoners were notified of the parole decision by the target
date, and we are pleased to recognise the efforts made by the
Prison Service and the Parole Board to reduce delays. To maintain
the momentum of recent improvements, the Prison Service should
monitor the release dates of parolees against their parole eligibility
dates, and publish the results in its Annual Report together with
an estimate of the cost of releasing prisoners late. For lifers,
who have no entitlement to parole but who may be released on licence
if assessed as safe, there is scope for the Prison Service to
further improve performance by increasing the proportion released
within three months of the expiry of their tariff. (paragraph
26)
(vi) Currently some 20 or more documents
are often required for the prisoner's dossier sent to the Parole
Board to inform their review. Whilst improvements have been made
in the process for compiling dossiers, the Prison Service, in
conjunction with the Parole Board, should consider whether all
the reports and documents are necessary, such as separate reports
on the prisoner by both the Probation Service and the Prison Officer.
(paragraph 27)
(vii) Transfer of a prisoner to another
prison while under consideration for parole can seriously delay
the process. The Prison Service should consider whether more prescriptive
guidance is needed for prisons, based on good practice, in handling
the transfer of prisoners during the parole process. (paragraph
28)
(viii) The establishment of a specialist
deportation unit within the Prison Service has increased the proportion
of foreign nationals notified of their parole application results
by the due date, from 27 per cent in 1998-99 to 59 per cent in
the first nine months of 2000-01. However, it is still well below
the figure of 91 per cent for prisoners in general. The Prison
Service should give due priority to the timely processing of applications
for parole from foreign nationals subject to deportation. (paragraph
29)
(ix) Prison governors should notify the
Immigration Service promptly when they receive into custody a
prisoner recommended for deportation by the courts, to provide
the Immigration Service with the maximum time to make preparations
for deportation. The Prison Service should also monitor the effectiveness
of communications between its deportation unit and the Immigration
Service in achieving timely deportation of such prisoners. (paragraph
30)
Preparing prisoners for release
(x) The Prison Service's programmes aimed
at tackling a variety of prisoners' educational and other needs
are evolving with emerging findings suggesting a positive impact
in reducing reconviction rates. The Prison Service should monitor
the number of prisoners who cannot be accommodated on relevant
programmes, and review its planned programme provision in the
light of this information. (paragraph 36)
(xi) Prisoners' sentence plans, which are
intended to provide a programme of action to help prepare prisoners
for their release, do not fully reflect the criteria used to judge
prisoners' suitability for parole. The Prison Service should monitor
the quality of sentence plans through reviews by the Prison Service's
Standards Audit Unit. Prisoners who are eligible to apply for
parole should, periodically, be given a written assessment of
their performance against the targets in their sentence plans
so that they can make informed decisions about their parole application.
(paragraph 37)
(xii) The Prison Service's plans to address
a lack of awareness of parole criteria amongst prisoners by means
of a booklet on parole are helpful. The Prison Service should
consider whether other initiatives are also needed, particularly
as the levels of literacy amongst many prisoners are low. (paragraph
38)
PAROLE BOARD DECISION-MAKING
7. In 1999-2000, 41 per cent of applicants for parole
were successful, broadly the same level as in previous years.
In measuring the quality of their decision-making the Parole Board
and Prison Service consider the number of parolees recalled to
prison because of breaches in their licence conditions or re-offending.
In 1999-2000, 10 per cent of parolees had to return to prison:
6.2 per cent because they had breached one or more of their licence
conditions or behaved in a way which suggested that there was
a risk of them re-offending; and 3.8 per cent because they had
committed a further offence.[3]
8. The Home Office told us that research carried
out on its behalf by Professor Hood of the University of Oxford
Centre for Criminological Research and Dr Stephen Shute of Birmingham
University, published in May 2000, had compared the Parole Board's
decisions against an actuarial calculation of prisoners' re-offending.
This work had suggested that the Parole Board had tended to err
on the side of caution in granting parole.[4]
9. On whether there was a risk of bias in the parole
process, particularly between different ethnic groups, the Home
Office told us that its analysis of parole decisions in 1999-2000
showed a wide range in the success rates of different ethnic groups
(Figure 1). Professor Hood and Dr Shute found a similar variation
amongst their sample of 400 cases. Their analysis indicated that
Asian prisoners were more likely to be in the lowest risk band
because they had, on average, less serious criminal histories;
were more likely to have no adjudications against them; were more
likely to have finished any courses started; and, in consequence
were more likely to be recommended for parole by their probation
officers. The researchers concluded that the different success
rates of the various ethnic groups reflected the extent to which
the prisoners in those groups met the criteria for parole used
by the Parole Board.[5]

10. We also sought information on the success rates
amongst different social groups and the type of offence. However,
the Prison Service could provide no reliable breakdown. The Director
General of the Prison Service told us that in his experience white-collar
criminals were probably more likely to gain parole, probably because
of the nature of their offence, the greater likelihood of them
having a home to go to, and possibly a job as well. Educational
background might also be a factor, particularly in enabling prisoners
to understand the parole process.[6]
11. The Home Office believed that there were a number
of safeguards against institutionalised bias in approving parole
applications. It was unlikely that a particular parole panel of
three members would meet together more than once. Each panel considered
24 cases but the cases were not allocated by prison or prison
area. Whilst it would not be practicable to collect data on individual
Board members' decisions, it would become apparent if a panel
member consistently went against the views of colleagues.[7]
12. The Parole Board looks at whether prisoners have
shown, by attitude and conduct, that they are willing to address
offending behaviour. We asked about the prospects of parole for
those prisoners who refuse to admit guilt. The Parole Board said
that it had to assume that the prisoner had been properly convicted,
but acknowledged that a prisoner who refused to admit their guilt
might be less likely to get parole. For some prison programmes,
particularly sex offender programmes, admission of guilt is an
important pre-requisite for getting a place on a course, and where
offending behaviour had not been addressed the Board would almost
certainly conclude that the prisoner was still dangerous.[8]
Conclusions
13. Around 60 per cent of applications for parole
are rejected. Whilst independent research commissioned by the
Home Office into the Parole Board's decision-making suggests the
degree of risk posed by some prisoners may be overestimated, the
Parole Board should have full regard to the need to protect the
public when considering prisoners for parole. The Home Office
should, however, review periodically whether the right balance
continues to be struck between granting parole and keeping prisoners
in prison.
14. Neither the Prison Service nor the Parole Board
could provide a breakdown of the success rate of parole applicants
according to category of offence, educational attainment or their
regional background. To better target advice and support on parole
matters, the Prison Service, in conjunction with the Parole Board,
should examine the success rates of different prisoner groups
in gaining parole.
15. Whilst research commissioned by the Home Office
suggested that much of the variation in success rates between
different ethnic groups could be explained, the Home Office and
Parole Board should review the success rates of different ethnic
minority groups within offender categories to demonstrate that
there is no bias when considering applications from prisoners
with similar criminal records.
TIMELY PROCESSING OF PAROLE APPLICATIONS
16. The parole process starts with the prison compiling
a dossier of reports and documents on the prisoner, which is used
by the Parole Board in reaching its decision on the application
for parole. The parole timetable requires the dossier to be with
the Parole Board 10 weeks before the prisoner is eligible for
release on parole, and for the Parole Board to notify its decision
at least two weeks before that date. These timescales should ensure
that successful applicants are released on time.[9]
17. In 1998-99, prisons submitted just 49 per
cent of dossiers to the Parole Board on time. As the Parole Board
cannot always make up delays arising from the late submission
of dossiers, it notified only 58 per cent of its decisions
by the due date in 1998-99. The National Audit Office reported
that late notification had occurred in 853 (39 per cent) out of
the 2,214 cases where parole was granted. The late release of
prisoners resulting from these delays was estimated by the National
Audit Office to have cost the Prison Service some £2 million.[10]
18. Since 1998-99, the Prison Service and Parole
Board have reported a significant improvement in performance.
Between April 2000 and September 2000, the latest figures available
at the time of our inquiry, the percentage of dossiers arriving
at the Parole Board on time had risen to 83 per cent and the percentage
of decisions made by the Parole Board by the due date was 92 per
cent (Figure 2).[11]
Figure 2: Performance against key targets in the
parole timetable 1996-2001
|
1996-97
%
|
1997-98
%
|
1998-99
%
|
1999-2000
%
|
2000-20013
%
|
Percentage of parole dossiers arriving at the Parole Board on time
| 401 | 2
| 49 | 72
| 83 |
Percentage of Parole Board decisions made by due date
| 43 | 46
| 58 | 84
| 92 |
Notes:
1No reliable data available but estimated by Prison Service at around 40 per cent
2No reliable data available
3 April to September only
|
Source: Prison Service
19. The Prison Service acknowledged that parole had
not been well managed in the mid-1990s. This had partly reflected
management's focus at that time on maintaining security following
some high profile escapes, possibly at the expense of other prison
activities. The Prison Service had put in place a number of initiatives
to improve performance. Area Managers now received monthly data
on the performance of their prisons against the parole timetable.
A central help desk for staff involved in parole had been established.
Annual conferences were held to discuss parole issues and a quarterly
parole newsletter issued. The Prison Service's Standards Audit
Unit audited the parole process. Of the seven prisons audited
last year all had passed the standards set for parole, including
that for providing complete dossiers to the Parole Board on time.[12]
20. The prisoner's parole dossier comprises at least
14 and frequently 20 or more reports and papers. Difficulties
in obtaining these documents can cause serious delay in processing
the parole application. On whether all the documents were necessary,
the Director General of the Prison Service told us that, in his
view, the report from the Prison Officer was generally not of
a good quality, duplicated some information already reported elsewhere
and was probably not very useful to the Parole Board. The Prison
Service was working with the Probation Service to produce an offender
risk assessment system and expected this system to provide a continuous
log of the prisoner's progress through their sentence. In time
this would make it possible to combine at least two of the current
reports required for the dossier.[13]
21. In response to the National Audit Office survey,
90 per cent of prison parole clerks said that they had problems
obtaining police reports, 84 per cent in obtaining the field
probation officer's report, 82 per cent in obtaining relevant
papers from the sentencing court and 51 per cent in obtaining
a list of previous convictions. Part of the difficulty had been
poor communication between the Prison Service and other parts
of the criminal justice system. The Home Office said that it had
taken steps to bring together the different partners at national
and local level to identify and remove blockages in the passage
of information; and to raise awareness that achieving the different
objectives of the criminal justice system was a shared responsibility.
For example, the Prison Service had now established links with
the Police National Computer to enable lists of previous convictions
to be downloaded to prisons immediately. However, the Prison Service
acknowledged that its own Inmate Information System, which supports
the parole process was under pressure. It had recently agreed
an information technology partnership with EDS, known as Quantum,
which should enable it to link it up electronically with other
criminal justice agencies, including the Probation Service.[14]
22. Delays in processing a prisoner's application
for parole can be due to their transfer to another prison during
the parole process. The receiving prison may have to wait for
reports on the prisoner to arrive from the prisoner's previous
establishment, and the interview between the prisoner and a Parole
Board member may be delayed. In the eight months ended June 1999,
prisoner transfers accounted for nearly half of cancelled interviews.
Different prisons adopted different policies towards the transfer
of prisoners during the parole process, and the Prison Service
does not monitor the number of transfers, or whether they are
justified. However, some transfers were necessary to relieve overcrowding
or to maintain security. A new Parole Standard, issued by the
Prison Service, required prisons to put in place their own systems
for managing transfers. Performance against the Standard would
be audited.[15]
23. The National Audit Office reviewed the arrangements
for early release of prisoners serving life sentences. Lifers
have no entitlement to parole, but may be released on licence
which remains in force for life. The overall aim of the lifer
review system is to ensure that as many life-sentenced prisoners
as possible are released on or shortly after the expiry of their
tariff, the minimum period they must serve to satisfy the requirements
of retribution and deterrence. Only a minority of prisoners who
are assessed as safe are released on the expiry of their tariff
or within a few months of it (Figure 3). The Home Office told
us that as the Home Secretary's changes to the timetable for lifer
reviews take effect, it should be possible for all lifers judged
safe to be released within three months of their tariff expiry.[16]
Figure 3: Release of life sentence prisoners in
relation to tariff
Release date in relation to tariff
| Releases in 1996-97
| Releases in 1997-98
| Releases in 1998-99
| Releases in 1999-2000
|
On tariff expiry | 2 (2%) |
1 (1%) | 0 (0%) | 5 (4%)
|
Within 3 months | 10 (11%) |
3 (3%) | 15 (15%) | 10 (8%)
|
Within 6 months | 9 (9%) |
5 (5%) | 12 (11%) | 14 (10%)
|
Between 6 and 12 months | 16 (17%)
| 23 (21%) | 15 (15%) | 27 (20%)
|
Over 12 months1 | 58 (61%)
| 77 (70%) | 61 (59%) | 78 (58%)
|
Total released in year | 95 (100%)
| 109 (100%) | 103 (100%) |
134 (100%) |
(i) Note1 These figures include prisoners who have made insufficient progress to be assessed as suitable for either a move to open conditions or release on licence, therefore taking them beyond their tariff.
|
Source: Lifer Review Unit
|
24. The National Audit Office also found that in
1998-99 only 27 per cent of applications for parole from
foreign nationals subject to deportation were notified of the
result of their application on time, and that delays had cost
in the region of half a million pounds. Around 8 per cent of parolees
spent an additional 100 days in custody. The Prison Service had
set up a specialist deportation unit to tackle the problem of
delays, and the proportion of decisions notified to the prisoner
on time had since increased to 59 per cent. However,
in some cases the Prison Service was still not informing the Immigration
Service of a potential release until the start of the parole process,
which was often too late to make arrangements for deportation
by the release date.[17]
Conclusions
25. In 1998-99, the Parole Board's target for parole
decisions to be notified two weeks before the prisoner's parole
eligibility date was met in only 58 per cent of cases, leading
to extra costs for the Prison Service, estimated at some £2.5
million in 1998-99. Additionally, keeping parolees in custody
longer than is legally necessary can be less than fair to the
individuals concerned and their families.
26. By the first six months of 2000-01 92% of prisoners
were notified of the parole decision by the target date, and we
are pleased to recognise the efforts made by the Prison Service
and the Parole Board to reduce delays. To maintain the momentum
of recent improvements, the Prison Service should monitor the
release dates of parolees against their parole eligibility dates,
and publish the results in its Annual Report together with an
estimate of the cost of releasing prisoners late. For lifers,
who have no entitlement to parole but who may be released on licence
if assessed as safe, there is scope for the Prison Service to
further improve performance by increasing the proportion released
within three months of the expiry of their tariff.
27. Currently some 20 or more documents are often
required for the prisoner's dossier sent to the Parole Board to
inform their review. Whilst improvements have been made in the
process for compiling dossiers, the Prison Service, in conjunction
with the Parole Board, should consider whether all the reports
and documents are necessary, such as separate reports on the prisoner
by both the Probation Service and the Prison Officer.
28. Transfer of a prisoner to another prison while
under consideration for parole can seriously delay the process.
The Prison Service should consider whether more prescriptive guidance
is needed for prisons, based on good practice, in handling the
transfer of prisoners during the parole process.
29. The establishment of a specialist deportation
unit within the Prison Service has increased the proportion of
foreign nationals notified of their parole application results
by the due date, from 27 per cent in 1998-99 to 59 per cent in
the first nine months of 2000-01. However, it is still well below
the figure of 91 per cent for prisoners in general. The Prison
Service should give due priority to the timely processing of applications
for parole from foreign nationals subject to deportation.
30. Prison Governors should notify the Immigration
Service promptly when they receive into custody a prisoner recommended
for deportation by the courts, to provide the Immigration Service
with the maximum time to make preparations for deportation. The
Prison Service should also monitor the effectiveness of communications
between its deportation unit and the Immigration Service in achieving
timely deportation of such prisoners.
PREPARING PRISONERS FOR RELEASE
31. Prisoners' failure to address their offending
behaviour is often a reason for a parole application failing.
The Prison Service runs a variety of programmes designed to tackle
educational needs, drug misuse and offending behaviour. Successful
completion of an appropriate course can enhance a prisoner's chance
of gaining parole and assist their resettlement into the community.
For example, prisoners who complete an appropriate offending behaviour
programme are more than twice as likely to get parole than those
who do not. However, the programmes required to meet an individual
prisoner's needs are not always available. We were told that in
2001-02 there will be 6,000 places available on offending behaviour
programmes whereas the Prison Service estimates that demand for
these programmes is probably closer to 9,000.[18]
32. On educational needs, currently 65 per cent of
prisoners have literacy levels below that of a 14 year old, thereby
seriously impairing their chances of employment. In 1999-2000
42,000 prisoners achieved a basic qualification in literacy and
numeracy.[19]
33. The Prison Service told us that these programmes
were still quite new, would take time to set up and needed to
be properly delivered with the help of specialist support. Emerging
findings on the impact of programmes was encouraging, with reductions
in reconviction rates of between 10 and 14 per cent for prisoners
classed as medium risk.[20]
34. The Prison Service acknowledged that parole was
not yet fully integrated into the sentence planning process. An
assessment of prisoner needs is normally made when they are first
inducted into prison life. An important part of this process is
the preparation of a sentence plan, which should set out the prisoner's
targets for addressing his needs. The National Audit Office found
little evidence of the actions needed to meet parole requirements
being set out clearly in the plans. The Prison Service expects
that a planned prison-probation offender assessment system should
provide a better basis for planning a prisoner's time in prison.[21]
35. Research conducted by the Home Office suggests
that prisoners' awareness of the parole system is poor, including
awareness of factors taken into account by the Parole Board in
assessing a prisoner's suitability for parole. As a result, prisoners
may develop an optimistic view of their chances of parole, and
react badly if applications are rejected. The Prison Service has
been working with the Prison Reform Trust to produce a booklet
on parole for prisoners, due to be published in early 2001. However,
the Prison Service acknowledged that this project in itself would
not wholly overcome prisoners' lack of understanding of the parole
process.[22]
Conclusions
36. The Prison Service's programmes aimed at tackling
a variety of prisoners' educational and other needs are evolving
with emerging findings suggesting a positive impact in reducing
reconviction rates. The Prison Service should monitor the number
of prisoners who cannot be accommodated on relevant programmes,
and review its planned programme provision in the light of this
information.
37. Prisoners' sentence plans, which are intended
to provide a programme of action to help prepare prisoners for
their release, do not fully reflect the criteria used to judge
prisoners' suitability for parole. The Prison Service should monitor
the quality of sentence plans through reviews by the Prison Service's
Standards Audit Unit. Prisoners who are eligible to apply for
parole should, periodically, be given a written assessment of
their performance against the targets in their sentence plans
so that they can make informed decisions about their parole application.
38. The Prison Service's plans to address a lack
of awareness of parole criteria amongst prisoners by means of
a booklet on parole are helpful. The Prison Service should consider
whether other initiatives are also needed, particularly as the
levels of literacy amongst many prisoners are low.
1 C&AG's Report, HC 456, 1999-2000, para 1.1 Back
2 ibid,
paras 1.2, 1.22 Back
3 Evidence,
Qs 64, 71 Back
4 Evidence,
Q2 Back
5 C&AG's
Report, paras 1.21-1.22 and Evidence, Appendix 1, pp 14-17 Back
6 Evidence,
Qs 99-103 and Evidence, Appendix 1, pp 14-17 Back
7 Evidence,
Qs 80-82 and Evidence, Appendix 1, pp 14-17 Back
8 Evidence,
Qs 11, 44-50 Back
9 C&AG's
Report, paras 1.15, 2.5 Back
10 ibid,
paras 2.6, 2.8, 2.10 Back
11 Evidence,
Appendix 1, pp 14-17 Back
12 Evidence,
Qs 3, 16, 87-88 Back
13 Evidence,
Qs 13, 24 Back
14 Evidence,
Qs 1, 39-43, 59-63, 91 Back
15 Evidence,
Qs 105-106; C&AG's Report, paras 2.21-2.26 and Figures 11
and 12 Back
16 Evidence,
Q9; C&AG's Report, paras 2.13 and 24; Evidence, Appendix
1, pp 14-17 Back
17 Evidence,
Q5; C&AG's Report, para 2.55 and Figure 15; Evidence, Appendix
1, pp 14-17 Back
18 Evidence,
Q11 Back
19 Evidence,
Qs 103-104, 107; C&AG's Report, paras 3.4-3.5 Back
20 Evidence,
Qs 11, 37 and Appendix 1, pp 14-17 Back
21 Evidence,
Qs 37, 62; C&AG's Report, para 3.10 Back
22 Evidence,
Qs 92-96 Back
|