Expenditure%20trends%20and%20spending%20priorities%20for%20the%20Department%20of%20Correctional%20Services

1
Expenditure trends and spending priorities for
the Department of Correctional Services Report
to the Portfolio Committee on Correctional
Services 20 April 2011 ISS, CSPRI, NICRO
2
Purpose of the report

• Response to request by Chair of Portfolio
  Committee last year
• Civil society initiative to provide more detailed
  input
• Process funded by the OSF and involved ISS, NICRO
  and CSPRI and utilised skills of economist with
  experience in working with treasury

3
What the report does

• Provides an analysis of the DCS budget
• Identifies the drivers of cost in the DCS
• Attempts to cost rehabilitation programmes were
  they to be rolled out nationally and be based on
  international best-practice
• Demonstrates that within the current budget and
  facilities provision of rehabilitation programmes
  is not possible
• Calls for an overhaul of policy regarding
  incarceration

4
Key issues

• Report highlights two issues
• Policy shifts are required in order to improve
  the return on investment to society that
  results from incarceration of those who are
  convicted of crime
• The policy shifts that are required involve all
  departments in the criminal justice system and
  require dramatic re-assessment of the focus of
  the DCS
• Require a radical re-thinking of the role of
  incarceration in society

5
Keeping it in perspective

• In 2011/12 SA will spend R16,8 billion keeping
  160 000 offenders in prison
• In the same year we will spend R13,8 billion
  educating all learners in SA from Grade R to
  matric
• Incarceration is expensive and getting the
  policy right to ensure minimal but effective use
  of incarceration as a sentencing option and
  ensuring that DCS delivers the best service at
  the lowest cost is vital

6
Presentation outline

• Basic facts regarding spending levels and
  patterns in DCS
• What drives costs in the DCS
• Impact of the White Paper on the work of the DCS
  and financial implications of full implementation
• Policy choices and implications
• Conclusions and recommendation

7
Basic facts about spending on corrections

• 2010/11 DCS budget was R15,2bn 37 higher than
  2007/8
• DCS budget will rise by a further 23 between
  current year and 2013/2014
• This is a 69 growth since 2007/8
• But this growth must be contextualised

8
Basic facts about spending on corrections
9
Contextualising the growth in the DCS budget

• If the DCS budget had grown at the same rate at
  the police budget it would be some R1,1bn higher
  than it is now, or R700 million larger if one
  removed the cost of the 4 new PPPs
• DCS budget as a share of national budget is
  projected to fall from 2,1 in 2007/8 to 1,8 of
  national budget in 2013/14.
• Not a bad thing, but just worth noting

10
Interpreting the budget

• Functional breakdown of the budget by programmes
• Administration core administrative and financial
  support
• Security pays for the guarding of prisoners
• Corrections pays for the staff needed to develop
  sentence plans
• Care provides for the physical and psychological
  care of offenders, including purchasing food for
  them
• Development provides for training, recreation
  and other services for prisoners
• Social Reintegration manages the release of
  prisoners and the monitoring of parolees and
• Facilities, which provides for the physical
  infrastructure needed by DCS

11
Programme allocations

• 34 of the budget each year is spent on Security
• 26 is spent on Administration
• 12 is spent on Facilities
• 11 is spent on Care
• 9 is spent on Corrections, and
• 3 each on Development and Social Reintegration

12
(No Transcript)
13
Interpreting the budget

• Another, perhaps more useful way to breakdown the
  budget is by economic classification
• about 66 personnel costs
• 28 goods and services (incl leases, property
  payments, medicines, food and outsourced
  services)
• 7 on physical infrastructure (but this varies -
  between 5 and 11 between 2007/8 and 2013/14)

14
(No Transcript)
15
Misleading analyes

• Discussions about costs sometimes frames on basis
  of daily cost incurred by accommodating a
  prisoner
• 2010/11 average cost was R259
• 2007/8 average cost R198 (26 lower)
• This may mistakenly be used to make assumptions
  about cost effectiveness
• BUT, the DCS budget varies depending on the
  infrastructure spending

16
Misleading analyses

• Such an analysis is mistaken
• Impact of infrastructure spend can distort the
  figures
• Consider the effect that the R455 million that
  would have been allocated in 2013/14 for the PPP
  facilities would have had on the calculation of
  average daily cost of incarceration per prisoner

17
What drives costs in DCS?

• Three factors driving costs in DCS
• Prisoner numbers
• Range and quality of services DCS provides to
  prisoners (including accommodation, nutrition,
  medical treatment, safetyand security and so
  on)Changes in characteristics of prisoners may
  change the costs e.g. ATDs require more
  administrative handling, more escorts too and
  from courts
• Cost of services provided (fundamentally this
  relates to personnel costs)

18
What drives prisoner numbers?

• Four variables
• Crime rate
• Quality of policing and prosecution services
• Laws and court practices regarding granting of
  bail
• Sentencing regime
• While parliament and other role players can have
  an influence over these, the influence of DCS
  over these variables is limited
• So, while the number of prisoners is a key
  variable in determining whether its budget is
  adequate or not, this something that DCS has much
  control over

19
Prisoner numbers

• Rapid growth in prisoner numbers between 1995 and
  2003
• Since 2003 prisoner numbers have remained stable
• Since 2006 the daily average prisoner population
  (including ATDs) has varied between 159 000 and
  165 000

20
Cost drivers prisoner numbers
21
Why are prisoner numbers stable?

• Perhaps surprising if one considers the high rate
  of crime, and violent crime in particular
• Also massive growth in police numbers - by 66
  000 since 2003
• So, either fewer criminals making their way
  through the justice system OR they are receiving
  shorter sentences
• Latter, not true (see the following graph)

22
Black line shows the average sentence length
which has risen from 4.4 to 7.3 years since 1995
23

• This is a consequence of minimum sentencing
  legislation
• So, DCS fortunate that the rest of the criminal
  justice sector is not functioning optimally
• 15 year review by the Presidency found that the
  number of convictions in SA courts fell from 96
  000 in 2006 to 75 000 on 2008 (see the following
  graph).

24
(No Transcript)
25
Same research found that the conviction rate for
serious crimes committed between 2001 and 2006
varied between 20 of all murders reported to
police and 2 of car thefts
26
SAPS ARREST TRENDS 2003 - 2010

• Total arrests increased by 24.
• Arrests for priority crimes increased by 9

27
(No Transcript)
28
Implication for DCS

• The number of newly admitted prisoners has fallen
  by over 44 since 2004, so
• Number of prisoners has stabilised
• Turnover of prisoners has reduced
• Prisoners tend to stay in prison for much longer
• Were the CJS functioning optimally, DCS would be
  in serious crisis

29
What drives costs in DCS?

• Personnel costs are another driver of costs in
  DCS
• Personnel costs determined by
• Number of staff
• Remuneration of staff

30
Personnel costs

• DCS, like other government departments has little
  control over agreements reached in Public
  Services Bargaining Chamber
• Rapid increases in average remuneration of civil
  servants in the past few years
• 7-day week should have resulted in savings but
  there is no certainty that it is effective,
  practical or results in actual savings

31
How do we know the DCS budget is under pressure?

• Vacant posts frozen after 2007 wage agreement
  resulting in 2 decline in staff numbers (from 41
  000 in 2007/8 to 40 260 in September 2010)
• Reduction in spending on a range of non-essential
  issues (e.g. fleet services, use of outside IT
  consultants, advertising etc)
• This is not unreasonable DCS should have done
  this anyway but WHY is the budget under pressure?
• Not an increase in prisoner numbers
• May be that there are insufficient funds to foot
  the increased salary bill, but the key question
  is whether DCS has improved or increased the
  range of services provided to prisoners?

32
Has DCS increased the quality or quantity of
services to prisoners?

• DCS committed to delivering services in terms of
  the White Paper
• Has been an increase in sentence plans from 9836
  in 2007/8 to 95 000 projected for 2013/14 (but 40
  000 completed in 2010)
• No substantial increase in literacy training
  projected (annual 4 500)
• No significant increase in skills development
  programmes (6 800 a year)

33
Has DCS increased the quality or quantity of
services to prisoners?

• Also, no significant changes in staff complement
  so broad commitments not been reflected in
  staffing allocations (see the next graph)

34
(No Transcript)
35
The same point can be made by looking at
projected growth between different programmes.
Those that grow the slowest are Development,
Corrections and Social Reintegration
36
Services?

• 2011 DCS budget shows that Security will grow
  faster than Development, Corrections, Social
  Reintegration and Care
• Also, consider that only prisoners with sentences
  of over 24 months qualify for sentence plans
  which represents a reduction of services since
  2008 (amendment of the Correctional Services
  Act).
• Impact of this decision ? 70 of those released
  from prison annually have not had sentence plans
  or access to services (programmes or counselling)

37
(No Transcript)
38

• Consequence of this is that DCS acts as a
  revolving door for prisoners coming in, spending
  short periods of time in prison and then leaving
  - at a high administrative cost
• So, we are not actually seeing an increase or
  improvement in services thus the biggest driver
  of cost, since prisoner numbers have remained the
  stable is the cost of staff remuneration
• This has driven DCS to the point where it cannot
  realistically plan and deliver high-quality
  rehabilitation interventions
• Increases in prisoner numbers if the CJS becomes
  more effective will deepen the problem

39
Implementing the White Paper

• So, lets try to establish what implementing the
  White Paper will cost
• Based on international literature, effective
  rehabilitation programmes have several key
  features in common

40
Features of effective programmes

• Community-based rather than prison-based
• Intensive (at least 6 months long)
• Focus on high risk individuals
• Risk levels determined by sophisticated
  instruments
• Cognitive behavioural techniques used
• Matching of therapist and substance of programme
  to specific offenders
• Offenders would be offers vocational training and
  job-enhancing opportunities
• Would include intensive, mandatory after after
  care component
• Include stringent ME

41
Factors influencing cost

• The number of prisoners who qualify to receive
  such interventions
• The number of months over which the typical
  intervention will take place
• The ratio of recipients of these interventions to
  the number of professionals providing the
  services (which, for ease of expression, we will
  call class size)
• The proportion of the day that each class will
  take up for the professional concerned, with the
  proviso that anything more than 40 a day will
  make it impossible for an individual professional
  to be involved in more than one class at a
  time and
• The qualifications (and, therefore, cost) of the
  typical professional involved in the delivery of
  these services

42
Calculation

• Possible to calculate
• How many classes annually
• Number of professionals involved
• Cost of professionals
• Number of prisoners receiving interventions60
  000 (based on interventions only to those
  sentenced to more than 24 months)
• Intervention classes will last 6 months
• Class size average at 10
• Professionals can only give one class a day

43
Estimated cost of rehabilitation
44
Other costs (not calculated)

• Increased administrative capacity required (add
  25)
• Development of risk assessment and predictive
  tools
• Infrastructure costs
• Additional resources
• These could double the costs presented in the
  previous table

45
Implications of the costing

• If DCS were to deliver services to 66 000
  prisoners, the least possible cost would be R1bn,
  i.e. a R1bn increase to the existing budget
  which doesnt sound unreasonable
• But, are such programmes effective and practical
  AND is this the right policy approach?
• Current challenges relate to housing all
  prisoners humanely and ensuring safety and
  security in prisons

46
Additional questions

• Will or can these really reduce the amount of
  crime committed by prisoners after their release?
  
• How certain can we be of that this will occur and
  that the reason for the decline will be the
  programmes the DCS offers rather than simply the
  amount of time prisoners spend in prison and the
  fact theyve aged in the process?
• Is it even possible to rehabilitate prisoners
  living in inhumane conditions?
• Is it possible to find sufficient dedicated and
  qualified professionals to provide the
  interventions that DCS and the White Paper
  envisage?
• These are not easy questions to answer, so it is
  not clear that DCS ought really to prioritise a
  large-scale rehabilitative intervention.

47
Recommendations

• Policy revision, to include
• Is there value in passing custodial sentences of
  24 months or less, given that this category of
  prisoner spends a relatively short time in
  prison, receives no services during that time,
  but probably costs the department a great deal in
  terms of administration and makes up 70 of all
  releases annually.
• Whether minimum sentencing legislation is
  appropriate and can achieve the objective of
  deterring crime.
• Whether the DCS should be offering rehabilitation
  services, or whether it should prioritise raising
  the level of accommodation, security, nutrition
  and care to levels consistent with humane
  detention.
• Whether to focus whatever rehabilitative
  resources are available on younger offenders,
  particularly those who are first-time offenders
  and those with substance abuse problems.
• Whether open prisons could serve the purpose of
  reducing overcrowding and easing re-entry into
  society at a lower cost than rehabilitation
  programmes can.
• Establishing a system of post-release support for
  prisoners re-entering society and a
  community-based programme for those serving
  non-custodial sentences.

48
? 27 12 346 9500
? www.issafrica.org
? cgould_at_issafrica.org
4848