Regional variation in sentencing: The incarceration of aggravated drink drivers in the New Zealand District Courts
| DOI | 10.1177/0004865813483295 |
| Published date | 01 December 2013 |
| Date | 01 December 2013 |
| Author | Wayne Goodall,Russil Durrant |
| Subject Matter | Articles |
Article
Australian & New Zealand
Journal of Criminology
46(3) 422–447
Regional variation in
! The Author(s) 2013
Reprints and permissions:
sentencing: The incarceration
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0004865813483295
of aggravated drink drivers
anj.sagepub.com
in the New Zealand District
Courts
Wayne Goodall
Institute of Criminology, Victoria University of Wellington, New Zealand
Russil Durrant
Institute of Criminology, Victoria University of Wellington, New Zealand
Abstract
Consistency of sentencing outcomes is a key aspect of the achievement of justice. There is
uncertainty and disagreement about whether consistency of sentencing outcomes is achieved
between locations in New Zealand. This study uses administrative data to examine the degree to
which similar offenders convicted of aggravated drink driving receive similar sentences for similar
offending in the different circuits of the New Zealand District Courts. Logistic regression is used
to control for offence seriousness, criminal history and offender demographics in order to isolate
as far as possible the influence of location on sentencing decisions. Results show that while
offence seriousness and criminal history combined have the greatest influence on sentencing,
location is also influential, with substantial differences between circuits in the use of incarceration.
Keywords
aggravated drink driving, disparity, focal concerns, sentencing, sentencing variation
Introduction
The idea that offenders who commit similar offences should be subject to similar sen-
tences is an important principle in modern Western criminal justice systems. Although
the notion of judicial discretion remains important, and there exists a ‘perennial and
endemic tension between the principle of consistency and individualised justice’
(Freiberg and Krasnostein, 2011: 73), there is widespread agreement that a fair and
just criminal justice system should limit unwarranted disparities in sentencing
(Johnson, 2011). In New Zealand section 8(e) of the Sentencing Act 2002 provides
Corresponding author:
Wayne Goodall, PhD student, Institute of Criminology, Victoria University of Wellington, PO Box 600, Wellington,
New Zealand.
Email: 4ward@clear.net.nz
Goodall and Durrant
423
that judges ‘must take into account the general desirability of consistency with appro-
priate sentencing levels and other means of dealing with offenders in respect of similar
offenders committing similar offences in similar circumstances’. In short there is an
expectation that variability in the sentencing of similar offences (and offenders) be
kept to a minimum.
There is a substantial body of empirical research devoted to the investigation of senten-
cing variability, although the vast majority of this work has been carried out in the United
States (see Johnson (2011) and Ulmer (2012) for reviews). Perhaps unsurprisingly the most
consistent finding from this work is that legal sentencing factors (particularly the serious-
ness of the offending and the offender’s prior criminal history) are the principal predictors
of sentencing outcomes (e.g. Crow and Gertz, 2008; Dixon, 1995; Ostrom et al., 2004;
Spohn, 2000). Many studies have focused on the effect of so-called ‘extra-legal’ factors
such as race, ethnicity, age and social background on sentencing outcomes (e.g. Blackwell
et al., 2008; Jeffries and Bond, 2009; Kautt, 2009). The findings from this research are
mixed, and the importance of these factors appears to vary by the type offender, the
offence committed, and the context of offending. A smaller body of work has explored
regional disparities in sentencing. Most of these studies tend to find significant differences
between geographical locations (e.g. Johnson, 2006; Johnson et al., 2008).
Although research on sentencing variability has generated a number of significant
findings, recent reviews highlight the need for further research (Johnson, 2011; Ulmer,
2012). Importantly, as Ulmer (2005: 1501) notes, ‘the most glaring gap in the literature is
that almost all of the research on sentencing disparity is limited to the contemporary
North American – particularly U.S. – context’. The lack of international research is of
particular note given significant cross-national differences in sentencing policies and
practices. Notable exceptions include research in the Netherlands (Johnson et al.,
2010) and Australia (Jeffries and Bond, 2009; Snowball and Weatherburn, 2007).
However, with the exception of several relatively small-scale studies (Jeffries et al.,
2003; Law Commission, 2006; Spier, 1989) there has been no systematic research on
sentencing variability in New Zealand. There is disagreement about whether there is
evidence of real variation or just a lack of appreciation of the detailed reasons for
sentence. The primary purpose of the present study is to report on the extent of variation
in the use of incarceration between District Court circuits in New Zealand for the offence
of aggravated drink driving. The focus on a single offence enabled the development of a
detailed model taking into account a wider range of factors in greater depth than is
possible with more general models. The model also explored the extent to which both
legal and extra-legal factors played a role in sentencing outcomes.
The modelling is the first stage in a wider programme of research covering separate
analyses for three large volume offence types: aggravated drink driving, burglary, and male
assaults female, and the decisions on whether to incarcerate and the length of incarceration.
Explaining sentencing variability
The complexity, uncertainty and time pressure said to characterise sentencing decisions
open up the opportunity for variability (Albonetti, 1991; Ostrom et al., 2004;
Steffensmeier et al., 1998). Historically, judges were entrusted with wide sentencing dis-
cretion to address both the complexity and uncertainty arising from the wide variation in
424
Australian & New Zealand Journal of Criminology 46(3)
offending and offenders and to arrive at sentences according to the merits of each
offender. The continuous flow of offenders to be sentenced and the limited time available
to weigh each case against the law and resolve the uncertainties increases the challenge.
In short, the task demands of sentencing tend to militate against the optimal and time-
considered balancing of factors that might result in optimal and completely consistent
outcomes.
Simon’s (1976) rejection of the classical notion of unbounded rationality and recog-
nition of significant limits on the ability of humans to reach optimal decisions when
faced with complex decisions made under time constraints can be applied to sentencing
(Albonetti, 1991). Limited computational capability, incomplete information and time
constraints preventing the gathering of more information present challenges for decision
makers. Faced with an imperfect assessment of alternatives and incomplete information
decision makers accept outcomes that are ‘good enough’ or ‘satisficing’ (an outcome
hoped to be close to the best possible answer) but which may not be optimal. Heuristics
are strategies that assist decision makers in the face of uncertainty and under time
pressure (Gigerenzer, 2006; Todd and Gigerenzer, 2007). Heuristics simplify problems,
reduce the time required, and do not require an optimal decision, just one that it is good
enough (Gigerenzer, 2006).
The regularly repeated axiom that sentencing is an art rather than a science suggests
that judges and commentators recognise sentencing as a search for ‘good enough’ or
‘satisficing’ decisions. The discretion entrusted to judges, the complexity, uncertainty
and time pressure affecting sentencing, and the relative isolation of judges in courts
and circuits combine to suggest that sentencing is likely to be subject to the use of
heuristics (Gigerenzer, 2006; von Helversen and Rieskamp, 2009).
Sentencing researchers have drawn on the theory of bounded rationality and heuris-
tics to develop theories explaining sentencing disparity. Albonetti (1991) developed a
causal attribution theory that addressed both racial and gender disparity. According to
Albonetti (1991), when judges are uncertain about whether imprisonment is warranted
they focus on the likelihood of re-offending. Albonetti (1991) theorises that judges resort
to causal attributions based on stereotypes to resolve the uncertainty with the result that
African Americans are more likely to be imprisoned because they are perceived to be
more dangerous and more likely to re-offend, whereas females are treated leniently
because they are perceived to be less likely to re-offend.
Steffensmeier et al. (1998) developed a more nuanced theory suggesting that judges
address three ‘focal concerns’ when determining sentence: blameworthiness, protection
of
the
community,
and
the
practical
implications
of
sentencing
decisions.
Blameworthiness takes account of the offender’s culpability and the harm done by the
offending. It is associated with retributive sentencing or just deserts, and focuses on the
severity of the offending. The judgement about severity may be modified by offender
factors. Prior offending may increase blameworthiness, whereas personal circumstances
(physical and mental health, an abusive upbringing) may decrease it. The second ‘focal
concern’, community protection, draws on similar factors but is conceptually distinct
because it focuses on the need to protect the community by either incapacitating...
To continue reading
Request your trial