Even compared with the scant literature on amalgamation of councils, very little research has been directed at the water and sewerage functions of local councils, with the notable exceptions of the Allan Report (2006), Byrnes (2008), Byrnes, Dollery, Crase and Villano (2008a; 2008b) and Dollery (2008). Serendipitously, almost all of the very limited work that has been done in this area has focused mainly on an empirical evaluation of the efficiency of water and wastewater provision in the non-metropolitan regions of NSW and Victoria as a case study of a ‘natural experiment’. As we noted earlier, while Victoria has ‘regionalised’ water and wastewater services through forced amalgamation, NSW has yet to embark on this course of action, which has afforded researchers an excellent opportunity to compare the relative efficiency of these two approaches.

With respect to wastewater, in addition to the seminal earlier work in the Allan Report (2006), Byrnes, Dollery, Crase and Villano (2008a) considered the efficiency characteristics of local wastewater services in non-metropolitan NSW and Victoria by examining 14 Victorian and 42 NSW water utilities over the period July 2000 to June 2004. Two major conclusions can be drawn from this analysis. Firstly, scale economies do not exist in wastewater services and other factors predominant in determining relative efficiency, especially the proportion of residential users compared with industrial users and the composition of governance boards. The main policy implication flowing from this analysis was that ‘bigger is not better’ and thus ‘regionalization’ could not hope to gain any economic benefits from scale economies in wastewater. Secondly, the Allan Report (2006) recommended cooperation between councils on the acquisition and sharing of scarce skilled employees, while Byrnes, Dollery, Crase and Villano (2008b) showed that the composition of water-utility governing boards should be changed to include more people with specialized expertise. In policy terms, it follows that wastewater-utility reform in NSW should concentrate on enhancing council cooperation in skills sharing and the composition of governing boards rather than on structural change.

With respect to water services, very similar conclusions emerge. In an interstate efficiency study, Byrnes, Dollery, Crase and Villano (2008b) examined 52 water utilities from regional NSW and Victoria in order to measure the relative productive efficiency of these water utilities over the four year period 2000 to 2004. The policy implications of this work are as follows: there are no significant scale economies in water services and other factors determine relative efficiency, such as groundwater availability, conservation measures, the proportion of industrial users, and especially the composition of governance boards. It follows that in local water services ‘bigger is not better’ and thus ‘regionalization’ could not hope to reap any economic benefits from scale economies. In addition, as in the case of wastewater services, the Allan Report (2006) recommended that adjacent councils share skilled employees and Byrnes, Dollery, Crase and Villano (2008b) argued that the composition of water-utility governing boards should reflect more people with specialised expertise. In common with wastewater, it would thus seem that water-utility reform in NSW should concentrate on enhancing council cooperation in skills sharing and modifying the composition of governing boards.

In a pre-emptive defence of the Regional Aggregations proposed under preferred option 1, the authors of the Final Report made special reference to a lengthy submission to the Inquiry by Dollery (2008), in the form of an independent report produced for the United Services Union, which considered the problem of the empirical magnitude of scale economies in water and wastewater services. Dollery (2008) drew heavily on the Allan Report (2006), Byrnes (2008), and Byrnes, Dollery, Crase and Villano (2008a; 2008b) in order to argue that not only did no scale economies exist in water and wastewater services, but that existing work, which compared the Victoria and NSW industries, showed that diseconomies of scale were apparent in Victoria in the aftermath of that state’s radical restructuring program.

In an effort to rebut this line of criticism, the authors of the Final Report (2008: 35) observed that ‘the United Services Union’s [Dollery 2008] submission suggests that diseconomies of scale may be generated by “regionalisation”’ on grounds that ‘research into the relative efficiency of NSW and Victorian water utilities referred to in Chapter 3 [of the Final Report] indicates that while the Victorian water utilities tend to be more technically efficient, the efficiency gap is reduced by the diseconomies of scale affecting the Victorian water utilities’. The Inquiry thus accepted that ‘there are additional costs to being “too big”’ in water and wastewater services. In other words, the Final Report fully endorsed the arguments on scale diseconomies advanced in Dollery (2008), which present serious problems for any proposals recommending structural change aimed at increasing the size of water utilities, such as the Regional Aggregation option supported in the Final Report.

The thrust of this empirical argument is parried in the Final Report (2008: 35) using the following line of argument: ‘The 13 Victorian non-metropolitan water utilities serve an average 46 000 connected properties (range 14 000 to 127 000 properties)’ whereas, by contrast, ‘the size of groups proposed in Option 1 would vary between approximately 8 300 – 60 000 connected properties’. In essence, the authors of the Final Report contend that scale diseconomies will not afflict the proposed Regional Aggregation option groupings of water and wastewater services because they are, on average, smaller than their Victorian counterparts, when size is measured in terms of connected properties, rather than some other factor, such as gross spatial area serviced.

This argument is problematical. Quite apart from the obvious point that the size of the proposed groupings for non-metropolitan NSW overlap considerably with the amalgamated Victorian utilities, with some recommended NSW groupings exceeding the size of their Victorian equivalents in terms of connected properties, it also ignores important qualifications provided by Byrnes (2008), Byrnes, Dollery, Crase and Villano (2008a; 2008b) and Dollery (2008) as well as well-known arguments in the literature on the efficiency measurement of local service provision (see, for instance, Worthington and Dollery 2000a).

In the first place, for both sewerage and wastewater services, Byrnes (2008), Byrnes, Dollery, Crase and Villano (2008a; 2008b) and Dollery (2008) were at pains to stress that factors other than the scale of operations and the composition of governance boards were at play in determining efficiency. For example, in the case of wastewater, the treatment of sewage depends (in part) on where effluent is discharged, a factor often beyond the control of local councils. Similarly, Byrnes, Dollery, Crase and Villano (2008a) argued that the proportion of residential users compared with industrial users, which councils typically cannot influence substantially, had a significant impact. In an analogous vein, for water services, Byrnes, Dollery, Crase and Villano (2008b) found that external factors, including access to groundwater, drought-induced conservation measures, and the proportion of industrial users, affected efficiency.

These observations on the comparative efficiency of water and wastewater utilities echo a much broader theme in the international empirical literature on performance monitoring (Worthington and Dollery 2008) and efficiency measurement in local government service provision (Worthington and Dollery 2000a; 2000b), which has an Australian local government strand (Worthington and Dollery 2001; 2002). It is well-known that numerous ‘non-discretionary’ factors, which cannot be controlled by local authorities, can have a decisive impact on the economic efficiency of service provision. Given the vast differences in climatic, environmental, topographical and other natural factors, as well as water endowments, between the different local government areas in non-metropolitan NSW, it is hardly surprising that differences in council performance in service delivery have emerged. However, because a substantial, and largely unknowable, proportion of differences in performance are directly attributable to non-discretionary external factors, public policy must take this into account. Structural change enacted on the assumption that councils have complete discretion over operational efficiency is thus unwise.

A further problem with the Regional Aggregation groupings proposed in the Final Report revolves around the case for a uniform minimum size for these groupings of 10 000 connected properties. In their analysis in Chapter 3 of the Final Report, its authors argued that existing small water utilities did not perform well compared to their larger counterparts on the implementation of the Best Practice Management of Water Supply and Sewerage Guidelines, as measured in terms various specific compliance indicators. While this is certainly true at first sight, a well-known and significant problem with reporting frameworks, such as the NSW Best Practice Management of Water Supply and Sewerage Guidelines, is that they rely on partial performance indicators, expressed in absolute terms. Comparing the ‘performance’ of different utilities is thus of severely limited value, since a given utility may be the benchmark for one indicator, in the medium range for a second indicator, and ‘bottom of the class’ for a third (Worthington and Dollery 2008). Ranking thus becomes meaningless.

A final noteworthy feature of the 32 groupings proposed under the Regional Aggregation option is that not all groupings meet the six criteria for structural aggregation set out in the Final Report. In this regard, the authors of the Final Report (2008: 35–6) observe that ‘a majority of the 32 proposed groups satisfy the aggregation criteria; however, not all do’. By way of illustration, the Final Report provided four examples of cases which did not fully met the criteria:

These considerations suggest that the 32 Regional Aggregate groupings should be regarded as indicative rather than mandatory. Put differently, the problems we have identified indicate that sewerage and water-utility restructuring should depend on the particular circumstances of specific councils rather than a blanket prescription for all councils.