Summary
Aim
To examine the effect of the Bail Act 2013 (NSW) and subsequent 'show-cause' amendments on trends in the number and proportion of defendants being refused bail.
Method
Descriptive analysis of the number of defendants, proportion of all defendants and proportion of ‘bail eligible’ defendants refused bail each month in all NSW courts between February 2011 and May 2016. Kendall’s tau is used to test for significance in trends in the pre- and post-intervention periods (i.e. before and after the Bail Act reforms).
Results
The number of defendants refused bail showed a significant increasing trend of 2.95 defendants per month (p < .01) for the pre-intervention period of February 2011 to May 2014, and a mean number of defendants of 1,042.57. The mean number of defendants rose to 1,264.19 defendants for the post-intervention period of January 2015 to May 2016. No significant trend was found for the post-intervention period (p =.06). The rise in mean postintervention was higher than what would have been expected due to the increasing trend pre-intervention.
The proportion of all defendants being refused bail showed no significant trend either pre- or post-intervention (p =.06 pre-intervention and p =.23 post-intervention). There was a slight difference in mean proportion between the two periods (.098 pre-intervention vs .108 post-intervention). However, it seems likely that this slight difference can be attributed to the very slight (although statistically insignificant) trend pre-intervention. Taking both periods together, there appears to be a very slight but significant increasing trend (p < .01) across the whole period, with a mean rise of .0002 per month.
The proportion of 'bail eligible' defendants also showed a significant increasing trend pre-intervention by .001 per month (p<.01). However, the post-intervention proportion showed no significant trend (p = .84). The mean proportion per month increased from .276 to .326 between the pre- and post-intervention periods. Again, the increase in mean proportion between the two periods is higher than what would have been expected given the pre-intervention trend.
Conclusion
The Bail Act 2013 (NSW) and subsequent amendments appeared to have an effect on the number and proportion of bail eligible defendants refused bail. However, they appear to have had little to no effect on the proportion of all defendants refused bail. This suggests that defendants who had previously been released on bail are now having bail dispensed with or bail refused.