Sentencing in the lower courts: a statistical analysis

Criminology Research Council grant ; (20/75)

This study was based on information derived from the court records of 14,311 offenders convicted of driving with above the prescribed concentration of alcohol (PCA) in New South Wales in 1976. The general aim of the research was to develop a set of statistical models which could be used as a basis for explaining the sentencing process.

It was found that there were variations in penalties from court to court which were not explicable in terms of variations in offender or offence characteristics. The evidence therefore points to the existence of disparities between magistrates. Although it is possible to classify magistrates according to their general degree of toughness or leniency, a more efficient classification involves, in addition, the notions of individualisation of penalty through the use of restricted licences and the notion of a tariff approach which attempts to scale the punishment to the perceived seriousness of the crime. In more concrete terms, the aspects of the penalty which most clearly illustrated these disparities were the level at which they set fines, the use of restricted licences, the use of bonds and the relative balance of fine and disqualification period.

Secondly, it was found that extra-legal factors, particularly age and occupational status, exerted an influence on the penalty over and above the contribution of factors such as previous convictions and blood alcohol count. Age effects were not a simple reflection of driving experience, and the very young (18 and 19 year old offenders) received particularly severe penalties. The youngest offenders were at greater risk of going to prison or receiving a bond, and also received longer disqualification periods and heavier fines. Offenders not in the workforce, particularly the unemployed and those classified as pensioners, were penalised more severely than those in the workforce, while the occupational status of the employed was inversely related to penalty severity. The small number of professional status offenders received particularly light penalties.

Previous DUI convictions emerged as the single most important determinant of penalty severity (followed closely by the magistrate) but the effect of previous convictions varied to some extent with the age and occupational status of the offender. Thus, for example, variations in penalties between status groups was less marked for second offenders than for first offenders. Surprisingly, BAC exerted a relatively minor influence on the penalty, despite the fact that it is widely regarded as an objective indicator of offence seriousness.

Legal representation emerged as a minor but significant factor in the determination of most aspects of the penalty. Representation was a significant factor both in the statistical sense and in the sense that it made a real difference to the penalty received. For example, unrepresented offenders received disqualification periods up to 50 per cent longer than those imposed on represented offenders. However, there was an important interaction between representation and magistrate, with tough magistrates apparently being more influenced by a solicitor than lenient magistrates.

The representation/magistrate interaction illustrates perhaps the most theoretically interesting finding of the analysis: that magistrate sentencing style has an influence not only in its own right but also an influence on how other factors, particularly previous DUI convictions, are perceived and weighted in the determination of a penalty. Thus, for example, there was a bigger difference between lenient and tough magistrates for first offenders than for second offenders. As a second example, tough magistrates were particularly harsh on 18 year old offenders, while among lenient magistrates there was little difference in their treatment of offenders in the 18-22 year old age range.

The full results of this research are to be published in the form of a book.