Annex A
Note by the Home Office
INTERNATIONAL COMPARISONS ON THE USE OF CUSTODY
INTRODUCTION
1. The Committee requested a note on international
comparisons on the use of custody and the extent to which this
reflected the level of criminality. The Committee was also interested
in any information on comparative reconviction rates with other
countries for custodial and community disposals.
SUMMARY
2. The relationship between the prison population
and crime rates is complex. Statistics show that the prison numbers
per 100,000 population in England and Wales are high in comparison
with most other Western European countries, but very significantly
less than in the USA and Russia. The increase in the prison population
in England and Wales between 1992 and 1997 is also high compared
with most Western European countries, although considerably lower
than in the Netherlands and Portugal. The proportion of prisoners
serving sentences of less than one year is low, those between
one and five years high and about average for five years and over
compared with most other Western European countries. No useful
conclusions can be drawn about the relationship in other countries
between changes in recorded crime and the prison population. No
reliable information is available to enable international comparisons
to be made between reconviction rates.
PRISON POPULATION
3. The prison population mainly reflects:
— the extent to which crimes are cleared
up
— the extent to which accused are remanded
in custody
— the length of pre-trial detention
— the extent to which courts impose
custodial sentences
— the average time served
— the extent to which custodial sentences
are suspended.
Unfortunately good comparative information is
missing on many of these factors.
4. The latest information (from 1 September
1997) shows that after Portugal, England and Wales, at 120 prisoners
per 100,000 population, had one of the highest per capita prison
populations in Western Europe (Chart 1). However, this is below
the Czech Republic (209), the USA (645) and Russia (687) (not
shown in Chart 1) and Poland, New Zealand and Hungary. In most
countries the prison population has risen substantially in recent
years. The increase in England and Wales between 1992 and 1997
is also high compared with most Western European countries, although
considerably lower than in the Netherlands and Portugal (See Chart
2).
Unconvicted prisoners
A 1996 Council of Europe Survey indicated that
while only 15 per cent of the prison population in England and
Wales were unconvicted, in France the proportion was 36 per cent,
Germany 29 per cent and the Netherlands 34 per cent. Thus excluding
unconvicted prisoners from Chart 1 would increase the gap between
England and Wales and other Western European Countries.
Juveniles in prison
In England and Wales custodial sentences are
available at an earlier age than many other countries reflecting
different administrative methods for handling young offenders,
in particular the supervision of such offenders in some countries
by Social Welfare departments. The 1996 Council of Europe Survey
showed that 18 per cent of the prison population in England and
Wales is under 21 years of age, compared with France 10 per cent,
the Netherlands 8 per cent, Portugal 6 per cent, Sweden 4 per
cent and Switzerland 4 per cent.
THE USE
OF CUSTODY
AND SENTENCE
LENGTHS
5. Comparisons between countries in their
use of custody are limited by the absence of good data. This partly
reflects different justice systems and legal definitions, in different
countries. However, it also reflects different statistical methods
and the lack of national sentencing data in many federally organised
countries. Comparisons on the use of custody need to take account
of the availability of alternative methods of handling offenders.
For example, in many Continental countries the prosecutor has
the power to award a sanction if the defendant admits their guilt.
The Council of Europe has carried out two surveys covering 19901
and 1990/96 (unpublished) which collected information on the use
of custody relative to other disposals from both the prosecutor
and the court.
6. Results from the first survey (1990)
for theft offences indicate that England and Wales made comparatively
little use of custody for theft offences, but when custody was
used the proportion of sentences given of one year or more was
comparatively high.
Table 1
Percentage of offenders sentenced to custody
and length of sentence for theft offences
| Use of custody (%)
| Proportion of sentences
1 year and over (%)
|
| Norway | 41 | 8
|
| Netherlands | 35 | 5
|
| Scotland | 31 | 10
|
| France | 27 | 12
|
| Switzerland | 27 | 9
|
| Northern Ireland | 16 |
5 |
| England and Wales | 13 |
22 |
| Sweden | 11 | 9
|
1 European Sourcebook of Crime and Criminal Justice Statistics,
Draft Model (Council of Europe 1995).
7. Although more recent information from the Council
of Europe on the use of custody is not yet published, it is known
that there have been sharp rises in the use of custody in many
other countries (eg France, Netherlands) as well as England and
Wales since 1990.
8. Information collected by the Council of Europe on
the prison population (rather than prison admissions) in 1996
confirms that courts in England and Wales have a lower proportion
of short prison sentences compared with France, the Netherlands
and the Scandinavian countries. There is also a high use of middle
range sentences (one year and under five years) while for longer
sentences (five years and over) England and Wales is more in line
with other European countries (See Charts 3,4,5). In most countries
time spent in pre-trial detention is deducted in full from the
length of a prison sentence.
9. Information collected by Farrington and Langan2 confirms
a much higher use of custody and sentence length for most offences
in the United States than in England and Wales.
IMPRISONMENT AND
THE EFFECT
ON CRIME
RATES
10. Any effect of imprisonment on crime is seen to operate
via:—
(a) general deterrence (dissuading potential offenders
from committing crime);
(b) specific deterrence (dissuading those in prison from
committing further crimes); and
(c) incapacitative effects (with those in prison being
unavailable to offend in the community).
There is little firm information as regards (a) and (b), although
some estimates have been made of the size of the incapacitation
effect in this country and the USA.
11. Tarling3 looked at the incapacitation effect of custody
in England and Wales between 1975 and 1986. He concluded that
imprisonment reduced recorded crime by between 6 per cent and
9 per cent from its potential value; figures in line with those
in US research. The small effect on recorded crime is because
only a small proportion of offenders are given custodial sentences
and many receive short sentences. Based on this analysis Tarling
also estimated that a 25 per cent increase in the prison population
in this country would reduce recorded crime by 1 per cent. Estimates
from the US suggest a rather stronger incapacitative effect (due
to the larger US prison population). Tarling's figures depend
slightly on the types of crime affected by changes in sentencing
and whether the increase is achieved through longer sentences,
or more short ones.
12. Due to the lack of internationally comparable data
on the use of custody, comparisons between countries are usually
made between changes in the prison population and recorded crime
rates. Simple comparisons between the prison population and the
level of recorded crime are possible but may be misleading. These4
tend to show England and Wales similar to France, Germany and
the Netherlands but significantly below the rate of Eastern Europe,
Portugal and Spain. However, differences in the crimes included
in the total amount of recorded crime make such comparisons difficult.
Some suggestive results are available from the last International
Crime Victimisation Survey which measured levels of crime experienced
in 1995 by householders in England and Wales and ten other industrialised
countries. Levels of victimisation in this country were generally
high compared to the other countries. While this may explain some
proportion of our relatively high prison population, sentencing
decisions also play a part.
13. An alternative approach is to compare changes in
recorded crime with those in the prison population. Two different
comparisons have been made:
(a) Main OECD countries
2Crime and Punishment in America and England, D.P. Farrington
and P. A. Langan (unpublished).3Analysing offending—Data,
Models and Interpretation, Roger Tarling (1993).4Criminal statistics,
England and Wales, 1996, p23.
Table 2
PERCENTAGE CHANGE IN RECORDED CRIME AND PRISON POPULATION,
FOR THE MAIN OECD COUNTRIES 1987-96
| Change in prison population | Number of OECD countries
| Average increase in prison population (%)
| Average increase in recorded crime (%)
|
| Decrease | 4 | -11.8
| 17.5 |
| 0-20% increase | 5 | 12.0
| 12.2 |
| 20% and over | 9 | 58.0
| 13.6 |
| Total | 18 | 29.7
| 14.1 |
Table 2 looks at changes in recorded crime and the prison population
in 18 OECD countries for the period 1987-96. Crime increased in
all countries, with the biggest increase in the four countries
where the prison population fell. While this is consistent with
an "imprisonment effect", the average increase in crime
in other countries bore no relationship to the size of the increase
in the prison population. A second comparison for 26 countries
over the much shorter period 1992-96 showed no clear differences
in the changes in crime rates for different changes in the prison
population.
Table 3
PERCENTAGE CHANGE IN RECORDED CRIME AND PRISON POPULATION,
FOR THE MAIN OECD COUNTRIES 1992-96
| Change in prison population | Number of OECD countries
| Average increase in prison population (%)
| Average increase in recorded crime (%)
|
| Decrease | 7 | -8.7
| -0.4 |
| 0-10% increase | 10 | 6.1
| 0.3 |
| 10% and over | 9 | 32.1
| -1.5 |
| Total | 26 | 11.1
| -0.5 |
(b) US States
A similar comparison has been carried out for US States.
Figures for individual US States are available for a rather longer
time period and will not reflect cultural disparities.
Table 4
PERCENTAGE CHANGE IN RECORDED CRIME AND PRISON POPULATION
FOR US STATES 1982-96
| Change in prison population | Number of States
| Average increase in prison population (%)
| Average change in recorded crime (%)
|
| Over 220% increase | 10 |
275 | -19 |
| 167-214% increase | 10 |
191 | 5 |
| 146-165% increase | 10 |
155 | -9 |
| 92-142% increase | 10 | 122
| 1 |
| 53-92% increase | 10 | 53
| 5 |
| US total | | 177
| -6 |
Note: Based upon 50 States. In comparing changes in prison
populations and the UCR crime index, each State is counted of
equal weight. Prison numbers are those sentenced to one year or
more in federal and state prisons. Jail inmates and incarcerated
juveniles are omitted. The crime figures do not include drug offences,
whereas many of the "new" US prisoners are imprisoned
because of these—albeit often involved in "conventional"
crime.
The prison population increased in all US States over this
period and the nine States with the highest increase experienced
the sharpest fall in crime. In other groups, however, there was
no entirely clear pattern whereby more use of imprisonment was
accompanied by correspondingly more favourable crime rates.
In view of the variation between comparisons it cannot be
concluded that there is any clear relationship between recorded
crime and changes in the prison population.
RECONVICTION RATES
14. Information from the earlier Council of Europe study
covered reconvictions and suggested that the new survey results
are unlikely to yield any useful comparisons.
June 1998
| 
