Summary
Aim
To estimate the impact of a 10 per cent increase in arrests on (a) the number of male and female offenders sentenced to full-time prison terms (FTP) and (b) the number of offenders given community-based supervised orders (SO).
Method
Three pairs of time series (arrests and SO, male arrests and male FTP, female arrests and female FTP) were constructed from counts of male and female arrests, male and female FTP and SO over the 153 months from January 1998 to September 2010. Autoregressive distributed lag (ARDL) models were used to examine the impact of changes in the number of arrests on the outcome (SO, male FTP and female FTP).
Results
A 10 per cent increase in police arrests results in a 2.3 per cent increase in the number of SOs one month later. If the 10 per cent change persists, the increase in SO is estimated to be 4.1 per cent in the long term. A 10 per cent increase in the number of male arrests produces an immediate (same month) 3.3 per cent increase in the number of male FTP. If the 10 per cent change persists, the increase in male FTP over the long term is estimated to be 4.0 per cent. A 10 per cent increase in female arrests produces a 4.6 per cent increase in female FTP one month later. If the increase in female arrests persists, the increase in female FTP over the long term is estimated to reduce to 3.7 per cent. The short-run costs of a 10 per cent increase in police arrests are $2.6 million (supervised orders), $18.7 million (male FTP) and $2.2 million (female FTP), spread over 11 months for supervised orders and male full-time imprisonment and over 8 months for female full-time imprisonment.
Conclusion
Changes in arrest rates have significant and rapid impacts on demand for correctional services. Central agencies need to pay close attention to the downstream impact of policies that are likely to result in an increased arrest rate. Correctional agencies need to closely monitor arrest rate trends.