Final report and evaluation 16 April 2018

Project approach

Global intertemporal data envelopment analysis was conducted for all South Australian councils over the period 2012 to 2016 inclusive. Analysis of scale for the same period was also undertaken. Finally, censored regression was undertaken to identify the determinants of efficiency.

Project achievements

The analysis of efficiency levels in South Australian local government revealed that:

• The relative technical efficiency of the typical council in South Australia has reduced somewhat during the period of the study and the spread of efficiency results has been increasing over time. Rural councils typically have relatively higher technical efficiency (with respect to the output proxies that we employ) when compared to urban councils and the spread of efficiency scores is narrower for rural councils than urban councils.

The analysis of scale in South Australian local government revealed that:

• The majority of councils in South Australia might be expected to suffer efficiency losses from any increases to their size, with relatively few councils expected to enjoy efficiency gains from increasing their size.

Based upon the findings of our analysis of the determinants of efficiency, we observed that:

• Efficiency might be expected to decrease as population increases up to 75,183 persons, after which efficiency might increase again (to partially mitigate earlier losses).
• Increases in population density are associated with an increase to efficiency, though the magnitude of this influence is relatively small.
• Demographics of local government populations are generally associated with the level of efficiency, consonant with the corpus of scholarly literature.
• There is no evidence of an association between the relative level of financial assistance grants and efficiency.
• The participation of councils in shared services was statistically associated with reduced efficiency.

Project outputs

The outputs that were produced by the project:

• Data envelopment analysis efficiency scores and scale estimates for each and every council for the period 2012-2016. Econometric estimates of the mean response associated with various determinants of efficiency.
• Councils are instructed in the report to contact the author directly for their specific scores, should they wish to receive them, to ensure that the information remains privileged (LGA SA advised that scores for all councils should not be reported – therefore we only report aggregate trends and have initiated the aforementioned procedure should councils seek more specific information).

Project output distribution

The report is published on this page (above).

A copy of the report is also available on the UTS website.  We also have undertaken to send the author of the report to Adelaide to present the findings (at our cost) at a time that suits LGA SA.

Performance measures

As undertaken, when we present the report we will provide auditors with the opportunity to complete a short survey.