Summary: The Government claims that Public Service Agreements
have increased accountability and transparency to an 'unprecedented
level'. However, some outcomes have been reported in misleading
terms, reducing accountability; and some useful targets
have been dropped, diminishing transparency. In one case,
the Home Office concealed a fall of 58,000 offences brought
to justice by describing it as 'slightly ahead of trajectory'.
The declared aim of Public Service Agreement (PSA) targets
is to increase public accountability. Targets, according
to the Treasury, 'have become increasingly outcome-focused',
and are now supported by 'rigorous performance information'.(1)
Both accountability and transparency have 'increased to
an unprecedented level'.(2)
It would be more true to say the complete opposite, that
accountability and transparency have decreased to an unprecedented
level. First, some results have not been presented with
total accuracy. And second, when the number of targets was
reduced in 2002 and 2004, some useful benchmarks were abandoned
altogether and others made less demanding.
The intention of the Home Office, according to its 2004
annual report, is to narrow the 'justice gap' by increasing
the number of offenders brought to justice. The aim was
expressed more exactly in the 2002 PSA target: 'Improve
the delivery of justice by increasing the number of crimes
for which an offender is brought to justice to 1.2 million
by 2005-06; with an improvement in all criminal justice
system (CJS) areas, a greater increase in the worst performing
areas and a reduction in the proportion of ineffective trials'.
The target was amended in July 2004 and the deadline extended:
'Improve the delivery of justice by increasing the number
of crimes for which an offender is brought to justice to
1.25 million by 2007-08.'
Initial reporting of progress towards the 2002 target began
with the normal honesty expected of a public service. The
Home Office Targets Delivery Report (February 2003) declared
the baseline to be 1.104m in 1999-2000 and admitted that,
at the end of March 2002, the figure was 1.025m, some 79,000
below the starting point.
However, Sir John Gieve, Permanent Secretary at the Home
Office, in the Departmental Report 2003 said: 'Our latest
figures show that 1.046m offences were brought to justice
in the year to September 2002, slightly ahead of the trajectory
to the target of 1.200m by 2005-06'.(3) It turns out that
what he meant by 'slightly ahead of trajectory' was 58,000
below the starting point of 1.104m.(4)
The Autumn Performance Report of December 2003 continued
to be upbeat by summarising progress as follows: 'Number
of offences brought to justice is increasing'.(5) But again,
the reported achievement of 1.074m, in the year ending June
2003, continued to be 30,000 below the baseline figure of
1.104m in 1999-2000, a fall of 2.7%.
The Home Office Departmental Report 2004 (6) was notable
for the discrepancy between the summary (p. 12) and the
body of the report (p. 51). The summary reported that 1.096m
offenders had been brought to justice in the year up to
November 2003, and the findings were shown in a diagram
that gave an impression of steady progress. (p. 12) No exact
figure was given for the 'baseline', but it was shown in
the diagram to be midway between the 1.000m and 1.050m marks,
and is presumably the March 2002 figure (1.025m). But it
is not the baseline, as the February 2003 report had made
clear. The achieved figure was still below the baseline
(1.104m) and far below the target of 1.200m. In the body
of the report (p. 51) the text only refers to the increase
of 3% over the 12 months to December 2003 (not November
as in the summary), but there is a diagram on the same page
showing that the increase was below both the starting point
and the target. This trail of incompatible statements and
diagrams makes it very difficult for an interested citizen
to work out what exactly is going on.
here to view the misleading Home Office chart and how
it should have looked
The PSA target in 2002 was: 'Reduce crime and the fear
of crime; improve performance overall, including by reducing
the gap between the highest crime Crime and Disorder Reduction
Partnership areas and the best comparable areas; and reduce:
vehicle crime by 30% from 1998-99 to 2004; domestic burglary
by 25% from 1998-99 to 2005; robbery in the ten Street Crime
Initiative areas by 14% from 1999-2000 to 2005; and maintain
It was weakened in July 2004 to: 'Reduce crime by 15%,
and further in high crime areas by 2007-08.' But, the most
serious concern is not the watering down of the target but
the accuracy of progress reports.
When the results give a good impression of the Home Office,
they have usually been presented in a straightforward way.
For example, the favoured measure of total crime is the
British Crime Survey (BCS). The annual report for 2004 declares
the baseline for overall crime to be the 2002 BCS, 12,563,000
crimes. It goes on to report that interviews for the year
to December 2003 found 12,079,000 crimes, a fall of about
The technical notes on the PSA targets published in March
2003 (updated in July 2003) say that vehicle crime and domestic
burglary are measured by the BCS, whereas robbery is measured
by police records. These were the sources used by the Autumn
Performance Report 2003 (published December 2003) and the
Home Office Departmental Report 2004 (April 2004).
The baseline for vehicle crime is the 2000 BCS (crime in
1999): 2,941,927 crimes. The target is a reduction of 30%
to 2,059,349 crimes by 2004-05. BCS interviews in the year
to June 2003 found 2,319,000 vehicle crimes (down 21%).(8)
By December 2003 there were 2,263,000, down 23%.(9)
The baseline for domestic burglary is also the 2000 BCS:
1,261,364 crimes. The target is a 25% reduction to 946,023
by 2005-06. BCS interviews to December 2003 found 949,000
domestic burglaries, down 25%.
So far so good, but when the robbery figures are presented,
the attitude changes. The baseline for robbery is police-recorded
offences in 1999-2000: 68,782 crimes in the ten street crime
initiative areas. The target is a reduction in those areas
of 14% to 59,153 crimes. The Autumn Performance Report for
2003 honestly reports that, in 2002-03 the police recorded
83,661 robberies in the ten areas, an increase of 22%.(10)
After that date, presentation of the findings seems to have
fallen into different hands.
The latest figures are not given as a separate total and
have to be calculated from Crime in England and Wales 2003-04.
There were 76,777 robberies in the ten areas (a fall of
7% compared with the previous year) but 11% above the baseline
and 30% above the target. An honest observer would have
reported that robberies were 11% up, in the manner of the
Autumn Performance Report, 2003. Instead, the Home Office
Departmental Report 2004 does not give the total number
of robberies, merely saying there was a 17% reduction from
2001-02 to 2002-03 and that further 'substantial reductions'
had been made in 2003-04.(11)
PSA target 4, aimed to reduce the economic cost of crime,
an important consideration for householders facing higher
insurance bills and bearing the cost of installing security
devices in cars and homes. A reduction of 20% by 2000-01
was reported, followed by an increase in 2001-02, which
still left a 5% reduction, compared with the 1999-2000 baseline.
The closing date for the target was 2004, but no further
outcomes have been reported after the annual report in 2003
and the target has now been dropped altogether.
The aim of reducing the reconviction rate has also been
dropped. Moreover, the results achieved under the 200o and
2002 targets have never been fully reported. The aim was
to achieve a 5% reduction in the actual reconviction rate
compared with the predicted rate by 2005-06. According to
the Autumn Performance Report in December 2003 the baseline
was reconviction rates for the last quarter of 1999-2000
- information which was not available at the time.
However the report claims that reconviction rates based
on an earlier baseline were lower. The one-year rate for
juveniles who ended their sentences in 2001 was said to
have been 22% (in Online Report 18/03, published in February
2003). There has been no further report. The result for
adults in 1998/99 (based on 1997 baseline data) found that
3.2% fewer offenders were reconvicted. The Departmental
Report 2004 declared the baseline for young offenders and
adults to be the first quarter of 2000. However, in April
2004, the results were still not out and were not expected
until September 2004.(12)
In July 2004 the target was dropped and it seems unlikely
that the results of the 2002 target will ever be reported.
The routine reconviction data, published by the Home Office
in Prison Statistics, show a high rate of reoffending. Fifty-nine
per cent of all prisoners discharged in 1999 were reconvicted
for a standard list offence within two years of discharge.
For young males (under 21 at the time of sentence) the proportion
was 74%.(13) The trend over the last few years does not
tell us very much. In 1987, 57% of all prisoners were reconvicted
within two years. The proportion fell to 51% in 1992, only
to increase again to the current 59%.
The overall figure for those commencing community sentences
in 1999 was 56%. For males aged 10-17, 76% were reconvicted
within two years. For some groups, the rate was extremely
high. Ninety-five per cent of all offenders aged under 21
with eleven or more previous convictions who commenced a
community sentence in 1999 were reconvicted within two years.
For all aged under 21, it was 69%.(14)
Privately, Home Office officials acknowledge that attempts
to reduce reconvictions have failed. High hopes were placed
on offending behaviour programmes, but they are now considered
unsuccessful and nothing has been found to replace them.
Rather than admit failure, which might have stimulated public
debate and led to a re-doubling of efforts and a reappraisal
of current methods, the Home Office has tried to cover up
its bad performance.
The lesson is that political parties cannot be trusted
to control public access to the information citizens need
to evaluate government performance. Politicians view information
as ammunition that could be used against them by opponents.
The remedy is a truly independent information service, accountable
to Parliament, not to the government of the day.
David G. Green
1. Public Service Agreements 2005 - 2008, Spending Review
2004, July 2004, p. 1.
2. Public Service Agreements 2005 - 2008, Spending Review
2004, July 2004, p. 2.
3. Home Office Departmental Report 2003, Cm 5908.
4. Home Office Targets Delivery Report, February 2003, Cm
5. Home Office Autumn Performance Report 2003, p. ii.
6. Departmental Report 2004, p. 12.
7. Home Office Departmental Report 2004, Cm 6208, April
2004, p. 11.
8. Home Office Autumn Performance Report 2003, Cm 6057.
9. Home Office Departmental Report 2004, p. 11.
10. Home Office Autumn Performance Report 2003
11. Departmental Report 2004, pp. 11 and 39.
12. Departmental Report 2004, p. 13.
13. Prison Statistics 2002.
14. Probation Statistics 2002, Table 9.10.
The new Director of Public Prosecutions, Ken Macdonald,
has attacked the Government for swallowing American Government
rhetoric about crime and punishment and following its policies.
If only his accusation were true.
American experience has shown that increasing the prison
population sharply reduces crime. Twenty years ago Americans
were more likely to be victims of burglary. Today it is
the other way round. The US burglary rate was more than
double that in England and Wales in 1981, but by 1999 the
burglary rate here was nearly two-thirds more than America's.
Why? Because the risk of being caught, convicted, and sentenced
to custody increased in the United States but fell in England
and Wales. Take burglary: in 1981 for every 1,000 alleged
American burglars 5.5 were in jail. By 1996 the proportion
had increased to 8.8. In England and Wales the trend was
in the opposite direction: 6.9 out of every 1,000 alleged
burglars were in jail in 1981 falling to only 4.5 in 1999,
having been as low as 2.3 in 1994.
But Mr Macdonald is right to point out that the Government
is saying one thing and doing another. In its white paper
of July 2003, Justice for All, the Government boasted that
it had already increased prison capacity by 18%, but then
went on to say that: 'Our aim is not to increase the prison
Yet, the prison population continues to rise. It is now
74,000, up from 61,000 in 1997 when Mr Blair came in. The
increase partly reflects the ambiguity of the Government's
policies, but the biggest influence has been the judges.
The Government has been trying to reduce the prison population
by letting prisoners out early under curfew, but the judges
keep on sending them down. A higher proportion of convicted
criminals are being sentenced to custody and sentences are
In 1996 10% of sentences passed by magistrates were for
immediate custody. In 2001 it was 15%. The Crown Court figure
was also up from 60% in 1996 to 63% in 2001. Average sentence
length imposed by Crown Courts for indictable offences increased
from 23.6 months in 1996 to 26.0 in 2001, substantially
up on ten years earlier in 1991, when it was 20.5 months.
The real story is not that Mr Blunkett is in thrall to
American tough-guy rhetoric but that the Government is trying
to abandon one of the few policies that is working. Mr Blunkett's
policy may be only a pale imitation of American practice,
but increasing the prison population has cut crime. Yet,
the government has embarked on a policy of replacing prison
with 'intensive supervision' in the community.
What is the evidence that increasing the use of prison
reduces crime? By 1993 the rising crime rate had become
a major public concern and towards the end of that year
the earlier anti-prison policy was reversed by Home Secretary,
Michael Howard. Between 1993 and 2003 the average number
of people in prison rose from 45,600 to 74,000, an increase
of over 62%. Even if no deterrent effect is assumed, the
incapacitation effect of imprisoning another 28,000 criminals
has been substantial. How can we work out the incapacitation
effect? The Home Office report, Making Punishments Work,
estimated that the average offender carried out 140 offences
per year. The variation was large, and offenders who admitted
a drug problem, were committing an average of 257 offences
While in jail offenders cannot break into your house, whereas
when on a community sentence they still have the free time
to steal. We can make a rough calculation of the incapacitation
effect of jailing 28,000 offenders. If each prisoner carried
out the average number of offences identified by the Home
Office, then 3.9 million offences against the public would
be prevented by 12 months in jail.
No wonder Mr Blunkett has reluctantly continued Michael
Howard's policies but, if the Government wants to keep crime
falling, it should build more prisons.
How many more prison places do we need? The Home Office
has estimated that about 100,000 persistent offenders carry
out about half of all crime. The Social Exclusion Unit thought
that about 20,000 of them were in jail at any one time.
A good starting point would be to begin building the accommodation
for the other 80,000 as soon as possible.
But isn't prison a notoriously unsuccessful way of reforming
offenders? The Home Office estimates that nearly 60% of
prisoners are reconvicted within two years of release and
for young offenders it's closer to 75%. But what is the
alternative? Official figures gives the impression that
reconviction rates for prison and community sentences are
roughly the same, suggesting that they are equally ineffective
methods of rehabilitating offenders. But there is a difference.
The official figures do not capture the incapacitation effect.
The two years during which reconvictions are measured start
at different times. For criminals sentenced to jail the
clock starts ticking on release, missing out the incapacitation
effect of prison, but for community sentences it starts
at the beginning of the sentence. This practice has been
criticised by an official review but misleading information
continues to be published.
We should not give up hope of trying to change offenders
for the better, but a wise policy should aim at both public
protection and rehabilitation. Prison protects but does
not change offenders all that much. Community sentences
don't change offenders either, but offer minimal protection.
The search continues to be on for better ways of rehabilitating
offenders but, while we are waiting, the public is entitled
to protection. In that sense, prison works.