Print PDFPrevious PageTable Of ContentsNext Page

The Size of Local Government in a Three-Tier Federal Government System with Special Reference to New South Wales, Australia

Peter Abelson1
Applied Economics,
The Governance of Local Communities
, 259-276 Nova Science Publishers, New York, 2016


This chapter discusses the size of local government in a 3-tier federal government system. The discussion is in the context of the New South Wales (NSW) Government’s policy to increase the size, and reduce the numbers of, local governments; but the principles are general. The chapter discusses the core principles relating to the size of local government. It then identifies and discusses four major criteria for evaluating the size of local government: strategic capacity to work with higher levels of government; financial capacity and efficiency; the delivery of local services; and effective local democracy and social capital. The chapter concludes that in a 3-tier federal system like Australia’s, where the role of local government is relatively minor, there is a strong case for small local governments.  

The optimal size of local government is a universal and ongoing issue. This chapter reviews the issues of council size in a federal government system with three levels of government. The discussion is in the context of the NSW Government’s actions to increase the size, and reduce the numbers of, local councils in metropolitan and rural areas. However, most of the discussion and conclusions have general application at least for 3-tier federal government systems.

In Australia, government comprises a central government, a second level of government made up of six states and two territories, and some 560 local governments (councils) with an average population of just over 41,000. All local governments are creations of the state governments. The national Constitution does not recognise local government.

In the last 20 years, the three eastern Australian states of Victoria, Queensland and NSW have significantly reduced the number of local councils. In the mid-1990s, Victoria slashed the number of local councils from 210 to 79. In 2008, Queensland reduced the number from 156 to 73, but with four de-amalgamations in 2014, the number is now 77. In May 2016, the NSW Government amalgamated 43 councils, which reduced the number from 152 to 129 and, at the time of writing, plans to amalgamate another 19 councils subject to the outcome of legal proceedings

Australian local governments are large by most international standards. As reported by Callanan, Murphy, and Quinlivan (2014), among “Western” countries, only the UK, Ireland, Denmark and New Zealand have larger local authorities (as measured by population). Relevantly, we may also observe that each of these four countries is essentially a two-tier system of government. On the other hand, the average population per local authority is 37,500 in the Netherlands, 32,500 in Sweden, 9,000 in Canada, 6,500 in Germany, 4,500 in the United States and only 1,500 in France.

In this chapter, Sections 1 and 2 discuss the core principles relating to the size of local government and four criteria for evaluating the size of local government. Sections 3 to 6 discuss these criteria in more detail: the relationship between state and local government and the related concept of strategic capacity; financial capacity and efficiency; the delivery of local services; and social capital, the role of democracy and local communities.

It is generally agreed in the economics literature that the prime function of local government is the provision of local public goods (Abelson, 2012). Many public goods; such as local roads, waste disposal services, recreational facilities, and local libraries, provide exclusively, or at least mainly, local benefits and can be provided in varying levels to different communities. Efficient provision of these goods requires that they satisfy local household preferences.

It is also widely accepted in principle, if not always in practice, that the allocation of public goods to different levels of government should be determined by the political ‘principle of subsidiarity’. The subsidiarity principle states that public decisions should be made as closely as possible to the citizens affected by the decisions by the smallest, and least centralized, competent authority.2

Local government may also have some responsibility for local social services. While, central government has the prime responsibility for distributional transfers because it is better able to reduce inequalities over the whole community, many local communities wish to support the less well-off households within their community and some redistribution has positive local externalities. In addition, local conditions may vary and local communities are often better at perceiving local needs and more efficient at supplying them. Thus, local governments often have a significant role in supporting less well-off households subject to the constraint that redistribution does not significantly distort business or household location.

Related to these points, different communities often have different preferences and different needs. Where this applies, the economic Theorem of Decentralisation (Oates, 2006) shows that service differentials are better achieved small local governments that can provide differentiated services than by larger government units that tend to supply uniform services across the whole area. If preferences for local public goods vary, and there are no economies of scale or externalities, decentralized provision increases welfare by equilibrating output to demand (Musso, 1998). Uniform services provide too few services to those who want more or higher quality services and too much taxation on those who want fewer services.

Figure 1 illustrates the benefits of decentralisation. This shows a local public good produced at constant marginal cost. The demand curves DA and DB show aggregate demand for the local public good in two communities, A and B. The efficient quantities would be QA and QB respectively. If a single government provided a uniform output (QU) to the two communities, area A would receive too much of the good and area B too little. The welfare gains from decentralised provision would equal areas ABC and CEF respectively.

Applied Economics

Figure 1 Deadweight loss due to uniform provision of a public good

Other arguments for local provision of local public goods include: local government is usually better informed about local preferences; in small communities, costs are more transparent and easier to control; and competition between jurisdictions increases the efficiency of service delivery. On the other hand, if there are economies of scale in service provision or spillovers (externalities) between local governments, larger local governments may provide local public goods more efficiently than do smaller local governments. Thus, analysts often frame the assignment of functions to levels of government as balancing demand heterogeneity against economies of scale. However, as we will see, larger bureaucracies often fail to achieve economies of scale.

A particularly important externality is exclusionary zoning (sometimes described colloquially as nimbyism). By restricting development, exclusionary zoning may increase housing costs across the whole of an urban area. More generally, there is the issue of whether small local governments have the capacity and scale to work constructively with the second (or first) level of government. These have been core issues of contention in NSW. This is essentially a problem of harmonizing state or city governance and local governance, not about externalities between local governments.

For completeness, we should note the classic Tiebout (1956) theorem that a competitive system of essentially small local governments would provide an efficient outcome where households would choose the community that provides their desired package of services and taxes and move if another community offers a preferred package. The Tiebout model, sometimes described as ‘voting by the feet’, implies that local public goods can be provided as efficiently as private goods in perfectly competitive markets.

The Tiebout model is attractive, but based on some critical assumptions. These include: households are mobile and can relocate without cost; no significant economies of scale; no neighbourhood spillovers; the costs of local public services are shared equally across all households; the local community can enact exclusionary zoning laws to prevent low-income households entering the community, which would lower the average property price and drive up the tax rate. Some of these conditions may occur in the United States where there are a large number of local jurisdictions and where households are relatively mobile. The Tiebout model has less application in Australia where household movement is infrequent and expensive and local authorities supply less important services.

In summary, the subsidiarity principle and the economic theorem of decentralisation point to small local governments. Possible economies of scale and capacity to work with higher levels of government point to larger government units.

Form should follow function. If local government provides major health or education services, the optimal size of local government may be larger than if it is providing, say, a local library or childcare service. On the other hand, the more levels of government there are, the smaller will be the functions of the lowest level of government and the smaller its appropriate size. Thus, on the one hand, the optimal size of local councils depends on their role. On the other hand, the roles of local government will be a function of their size. So, we have a “chicken and egg” argument. Which comes first: the role or the size of local government?

Ideally, we would optimise jointly the scale of local public services and community size. For example, consider a local hospital. For any given size of community, hospital size (measured say by number of beds) is efficient when the marginal cost of an extra hospital bed per member of the community equals the marginal benefit. A larger community requires a larger hospital. However, as hospital size increases, at some point the marginal cost of increased size exceeds the marginal benefit. Schedule QQ in Figure 2 shows notional optimal hospital size as a function of various community sizes.

On the other hand, for any given hospital size, there is an optimal community size. For a hospital with a fixed size, per capita cost falls with size of community. On the other hand, service quality falls as crowding increases. The size of the community is optimal when the fall in the marginal quality of service with an extra member equals the marginal fall in per capita cost. In Figure 2, the NN curve shows the locus of optimal community sizes for various hospital sizes. At point E, both hospital and community size are optimal.

This model of community formation implies that optimal community size and service quantity may vary for each public good: for health services, police forces, fire services and so on. Assuming no joint costs or scope economies, there would be a large number of local authorities of different sizes. Of course, this outcome has significant practical limitations and we need to consider how other factors that might also influence optimal size of local government.

Applied Economics

Figure 2  Optimal community size and service level

In practice, in Australia the state governments create the roles for local councils and then create the councils. Doubtless, this occurs in other countries also. Thus, while arguably the role and size of local governments should be optimised jointly, in this chapter we work from the premise that the roles are set first.

In Australia, while the roles of local councils vary from state to state, local councils generally provide and maintain inter alia:

● local infrastructure (roads, footpaths and drainage),

● urban parks and bushland,

● planning and development approval services,

● building services, including inspections and enforcement

● waste collection and management

● administration of public spaces, parking facilities and street parking

● public health services, such as water and food inspection, public toilets and animal control,

● cultural facilities, such as libraries, art galleries and museums,

● active recreational facilities, such as sports fields, swimming pools, camping grounds

● social/welfare services for youth, families and seniors

● water and sewerage services in some areas in some states

Accordingly, the issue is what size of government is best suited to providing these essentially local services? Our purpose here, it should be noted, is to discuss how to answer this question, not to provide numerical answers. Accordingly, we ask what key criteria should determine the size of local government. And, what kind of answers would applications of these criteria suggest?

The discussion above, and indeed the debate in NSW, suggests four main, to some extent inter-related, criteria.

1. A strategic capacity to work with higher levels of government

2. Economic efficiency, sometimes described as financial capacity

3. Effective provision of preferred local services

4. Effective local democracy and social capital

The first criterion has had a high profile in the debate in NSW. Justifying the recommendations of the Independent Local Government Review Panel (ILGRP, 2013a; 2013b) that local councils should be amalgamated into units of around 200,000 people, the Chair, Graham Sansom (2015, p.65), stated that “The ILGRP’s concerns were with the effectiveness of local government as an arm of metropolitan governance.” The rationale is that local councils should have the capacity to assist with metropolitan planning and the provision of adequate transport and housing infrastructure.

The second, more common, economic efficiency criterion has also been high profile in NSW. For example, ILGRP (2013a, p.6) stated that “it is also clear that the financial base of the sector is in urgent need of repair … put simply there are too many councils chasing too few resources.” ILGRP (2013b, p.72) stated, “NSW simply cannot sustain 152 councils.” This somewhat unspecific concept of financial sustainability morphed into the more conventional one that larger local government units can achieve savings through economies of scale. Formally this reflects technical economic efficiency (the provision of a certain quantity of goods and services at least cost).

The third criterion, the supply of services that the local community seeks, is of course related to the second one. The more technically efficient is any agency, the more services it can deliver for a given budget. Nevertheless, the provision of local services also embodies allocative efficiencythat is, providing the services that the local community wants. An ability to provide these services implies both knowing local preferences and having an incentive to meet these preferences.

The fourth criterion, effective local democracy and social capital, has political and economic implications. Most people want to have some control over, or at least inputs into, decisions about their local services and local environment. The larger the governing unit, the weaker is this control or input likely to be. In addition, indeed, the weaker is the sense of community and identity. Economic impacts may also occur as a result of a decline in social capital and volunteering in a larger governing unit.

In NSW there has also been much rhetoric about the need for “scale” in local government and indeed several agencies, including the ILGRP, the NSW Independent Pricing and Regulation Tribunal (IPART) and the NSW government, adopted scale (defined as a minimum size of population) as an independent and over-riding criterion. The ILGRP (2013a; 2013b) considered that at least 200,000 people are necessary for strategic and financial capacity. Likewise, IPART (2015) found that the majority of councils in NSW were “not fit for the future” because they lacked the deemed requisite scale of some 200,000 people.

However, it is hard to justify how scale (minimum population size) has any intrinsic importance and neither ILGRP nor IPART provided evidence for this criterion. If scale matters, it must be as a means to some important objective. If a local government satisfies the four criteria above, scale should have no separate standing as a criterion. Scale may be an output of analysis, but not a predetermined criterion.

In the following sections, we discuss the application of these four criteria to local government size. As we will see, there are often no clear metrics. Moreover, to determine preferred local government size, we will need to have some way to weight these criteria. The weighting problem is addressed in Section 6 below.

Clearly, there needs to be consistency and complementarity between the various levels of government, and especially between state and local government. Below, this chapter discusses first some major governance issues for large urban areas (urban planning, housing, transportation, and costs of administration) and then discusses some micro criteria for strategic capacity advanced by ILGRP and IPART.

A major motivator of the state government’s desire to halve the number of councils in Sydney is the desire to improve metropolitan governance, address the problems of high house prices, and transport congestion. House prices in Sydney are 40% higher than the average in other Australian capital cities (Reserve Bank of Australia, 2015). Local councils in inner and middle areas of the city tend to oppose higher residential and commercial densities and some strongly oppose new transport infrastructure, especially motorways. The state government views this opposition not only as creating unaffordable housing and traffic congestion, but also as holding back economic growth, jobs and real incomes. In addition, some government agencies find dealing with large numbers of councils is time consuming, irritating and resource intensive.

There are two large issues here. Is this state government view of the problems and their causes correct and, if so, what is the appropriate response? These are complex issues, for which a brief three summary responses will have to suffice here.

First, the issues of high house prices and traffic congestion are endemic and are not readily resolved. As discussed by Gao and Melser (2016), high house prices are an equilibrium welfare outcome reflecting household trade-offs of incomes, house prices, travel costs and amenity. In particular, the high house prices in Sydney reflect the high amenity premium and to a less extent the income premium of residence in Sydney. Increasing the supply of new housing will have only a marginal impact on the house price premium. The evidence (Abelson et al., 2005) suggests that the price elasticity of housing with respect to total stock is -around -3.5. Since increasing the new housing by 50% in any one year increases housing stock by about 0.66%, this would reduce real house prices by just 2%.

Secondly, the state government has responsibility for economic development, housing and transport and the power to facilitate these developments within topographical and amenity constraints. The NSW Department of Planning and Environment (2014) has developed plans for Sydney based on demographic projections and an understandable desire to assist less well-resourced areas. However, the state government does not have an explicit and credible cost-benefit methodology for allocating housing and jobs to sub-regions, for example which would be based on market forces representing household and business preferences subject to transport and environmental constraints. Nor, until recently, has it had much appetite for major transport projects.

Moving forwards, consolidation of local government areas may reduce administration costs for both state government and formal local opposition to major developments. However, the state government has the power to determine and achieve sub-regional housing and transport targets and policies without forced amalgamations. It can determine local planning (zoning) regulations. It also has the power to determine property take for, and land uses around, major transport projects. In addition, councils can cooperate in sub-regional planning via sub-regional organisations.

Turning now to more micro criteria for strategic capacity, the ILGRP (2013a, p.8) defined strategic capacity by the following 9 sub-criteria, all endorsed by the state government, and explitly endorsed by IPART (2014, p.4).

1. More robust revenue base and increased discretionary spending

2. Scope to undertake new functions and major projects

3. Ability to employ a wider range of staff

4. Knowledge, creativity and innovation

5. Advanced skills in strategic planning and policy development

6. Effective regional collaboration

7. Credibility for more effective advocacy

8. Capable partner for State and Federal agencies

9. Resources to cope with complex and unexpected changes

These nine criteria include three which focus on councils’ capacity to work with a higher level of government (criteria 6, 7 and 8), two which involve financial capacity (criteria 1 and 9), and five which are aimed primarily at improving services to the local community (criteria 2, 3, 4 and 5). Working with the state government has been discussed above and financial capacity is discussed in the next section.

A fundamental difficulty with all these criteria is that they lack definition and metrics. Accordingly, they cannot be readily, or objectively, validated.

Three further points may be made. First, arguably, competition and motivation spur creativity and both increase with a larger number of small councils. Second, small local governments can cooperate with other councils through shared services. There are many examples of this in Sydney. Third, councils can buy in any specialist skills that they may need. Access to skilled workforce is entirely a function of financial capacity, not of size.

Given the above observations, it may be inferred that the arguments for strategic capacity are more about facilitating state government objectives, especially for metropolitan governance, than about increasing the internal capacity of local councils.

A major claim of proponents of larger councils is that they produce significant cost savings. In this section, we consider the arguments, and evidence, for or against this claim.

First a few introductory observations, Australian local governments are local public service monopolies supplying a large variety of unpriced and specialized services. There are no market metrics of prices, profits, market shares, or unit costs. Typically, we observe expenditures, which are a product of the quantity of various services supplied and their average unit costs. Thus, ideally we need to sort out unit costs from the quantities supplied. Moreover, the use of widely different depreciation rates across councils complicates expenditure and cost comparisons. These points should caution, but not disqualify, any conclusions drawn from empirical analysis.

General issues affecting size and efficiency

The key argument for the claim that larger councils produce cost savings is that they achieve economies of scale. This is typically assumed to occur with administration overheads and backroom operations, such as rate collections and IT services. This may be based on observed duplications of functions in local councils or by analogy with private commercial operations.

These possible technical disadvantages of small scale can be responded to, at least to some extent, by shared service arrangements (see Spicer and Found, 2016, Commentary for Canada) and competitive out-sourcing of service provision. In addition, larger geographical areas create travel costs for consumers.

More importantly, there may be significant costs (diseconomies) associated with large bureaucracies with unclear metrics. Large monopolies are behaviourally less constrained. Costs are often less clear and internal contracts are often less precisely defined. These are key management challenges for staff and councillors. Behavioural economics indicates that the key to productivity is Intrinsic Motivation. As Frey (2013, p.172) writes, “many people intrinsically interested in the work they do have been demotivated by bureaucratic interferences.”

There is substantial literature on the costs of bureaucracy (Niskanen, 1994; Northcote Parkinson, 1958). As Parkinson observed, efficiency often falls in large bureaucratic bodies. One reason for cost inflation is that managers in bureaucracies have a greater incentive to increase the number of employees reporting to them (as salaries are often linked to the number of people that they manage) than they have to reduce costs.

However, the key message that comes out of this brief discussion is that we must look at the evidence on costs and council size rather than rely on generic, high level, assertions about possible economies or diseconomies of scale.

Evidence for economies of scale

There is, actually, remarkably little evidence of economies of scale in local government jurisdictions. Indeed, nearly all the evidence is in the reverse direction. So, what have Australian advocates of amalgamation pointed to?

As noted by Bell, Dollery, and Drew (2016) the ILGRP (2013a; 2013b) did not cite any empirical evidence of cost savings from amalgamation.

IPART (2014) attempted to rectify this omission by citing a significant inverse correlation between population size and expenditure per capita. According to IPART (ibid. p.39), the data showed that “around 30% of the variation in [operating expenditures] per head amongst the councils of Greater Sydney is inversely associated with their population and that [operating expenditures] per head [are] lower the larger the population of the LGA.”

However, this finding failed fatally to account for the substantial inverse correlation (of –0.49) between local council population and income per capita. Abelson and Joyeux (2015) showed that expenditure per capita is largely explained by average taxable income per capita and a dummy variable for major business centres. After differences in income, levels and business activity are allowed for, Abelson and Joyeux found no evidence for economies of scale in larger local councils in Sydney.

The NSW TCorp (2013) report on the financial outlooks of councils foreshadowed the finding that financial weakness is a function of income rather than size. Drawing on household income from the 2011 Census data, Abelson and Joyeux (2015) found that the average taxable income of the 30 council areas in metropolitan Sydney deemed by TCorp to have a moderate or strong financial outlook was $61,237. On the other hand, the average taxable income of the seven council areas in metropolitan Sydney deemed to have a financially weak outlook was $42,366. No area with an average taxable income of over $48,000 had a projected weak financial outlook.

In its assessment of council mergers proposed by the NSW Government, IPART (2015) reported a financial analysis by Ernst and Young (EY). EY assumed that mergers would result in 15% savings in total operating costs of the amalgamated councils (excluding the costs of the largest council) less some initial costs of the amalgamation. EY based this projection on the financial targets of competitive corporates engaged in designed and selected takeover targets! As Bain and Company has shown, “The open secret about M&A is that most deals fail to [achieve] the synergies companies expect…Most merging companies…don’t have a clear understanding of the level of synergies they can expect through increased scale” (Miles, Borchet, & Egan Ramanathan, 2014, paras. 1-2). Further to the point, in IPART (2015, Annex, p.30), Ernst and Young acknowledged, “the available empirical evidence on the extent to which local council amalgamations will yield net savings in costs is mixed and tends to vary with activity. Econometric analysis does not provide strong support.”

Finally, in putting out its council merger proposals, the NSW Government commissioned the professional services firm KPMG to provide “independent” advice on the financial impacts of the proposals. KPMG (2016) reported, inter alia, that they assumed 3% saving on 80% of expenditure on materials and contracts, 7.4% overall staffing efficiencies in metropolitan areas but none in rural areas, and various savings in information, communications and technology investments. These savings were partly offset by some assumed merger transition costs. Typically, the KPMG projections produced net savings of about 2.5% of expenditures over some 30 years.

Apart from minimalist nature of the outcome, there are several problems with these estimates. First, as revealed in KPMG (2015), these assumptions were based largely on forecasts of savings in reports on earlier amalgamation proposals in Queensland and Auckland.3 Second, KPMG assumed that if the savings were technically achievable by a merged entity, they would occur. This assumes away bureaucratic risk and the evidence on costs cited below. Third, the analysis fails to account for some key costs such as a major new head office or the integration of different workforce systems, for example, where one council provides most services in house, and another one outsources significant services. Nor does it account for the travel time costs that the merger would impose on residents.

There is a well-known story in economics:

A physicist, a chemist, and an economist are stranded on an island with nothing to eat. A can of soup washed ashore. The physicist says “Let’s smash the can open with a rock.” The chemist says, “Let’s build a fire and heat the can first.” The economist says, “Let’s assume that we have a can opener!”

The phrase “the can opener” is used to describe modelling which employs particularly unrealistic assumptions to reach conclusions.

Evidence against economies of scale

In contrast to the lack of evidence on scale, economies there is extensive peer reviewed published work on the costs of amalgamations. Dollery, Grant, and Korrt (2012) provide detailed and extensive evidence in Australia and internationally that forced amalgamations have not produced any cost savings. They concluded (p.134): “Advocates of local government amalgamation have based their arguments on the premise that substantial efficiency gains and cost savings would flow…It seems clear that the empirical literature does not support this position”.

In another major review, Dollery, Grant, and Korrt (2013) cited 15 international studies from the United States, Canada, and Europe all of which throw doubt on claimed economies of consolidated local councils. They also examine eight Australian inquiries into the financial sustainability of local councils over the past decade and found (p.215) that “with one exception, these inquiries are sceptical of the ability of forced amalgamation to improve local authority financial viability.”

In further studies, in a cross-section study of councils in Sydney, Drew and Dollery (2014) showed that there is no relationship between scale (population) and financial capacity in these councils. And Bell, Dollery and Drew (2016) showed that, in 2014, NSW councils amalgamated between 2000 and 2004 were no more financially sustainable than councils that had not been amalgamated.

Similar conclusions have been reached in other countries. Faulk and Grassmueck (2012) found that council mergers had not achieved expected savings in the United States. Reese (2004) and Found (2012) report that council expenditure has risen after council mergers in Canada. Callanan et al. (2014) found limited, if any, economies of scale in larger local governments in Ireland.

Conclusions on financial capacity and population size

Like the Victorian and Queensland Governments before it, the NSW Government has reiterated repeatedly the mantra that larger councils will produce significant saving and therefore better services. However, the corporate business world is not an appropriate model. Public bureaucracies are monopoly service providers; they are not competitive businesses. The larger a bureaucratic monopoly the greater is the potential for waste. Most critically, many empirical studies in Australia and other countries have shown that amalgamated councils do not produce savings.

As we have seen, the provision of local services to meet local preferences is at the heart of the role of local councils. We should note here that, in discussing the criterion of effective delivery of local services, the burden of proof is rather different from the burden for the first two criteria discussed above. It is a natural presumption that larger local governments are likely to meet the criteria of strategic and financial capacity better than smaller ones. Thus, the onus is on advocates of small governing units to show why this may not be the case.

On the other hand, the decentralisation theorem suggests that communities with different demographics and preferences demand differentiated services, and hence should be serviced by separate local agencies. The burden of proof now lies with advocates of larger government units to show how, for an equivalent financial capacity, larger governing units can meet consumer preferences as well as, or better than, smaller units. In practice, the ILGRP, IPART, and the NSW Government assumed simply that services are a function of assumed financial savings and totally ignored the issue of service heterogeneity and quality.

To start the discussion, we observe that increasingly many local government services are personal services (social support, cultural and recreational), provided to individuals or small groups of individuals. Over the last 30 or so years, there have been major increases in elderly single people living alone, divorce rates, domestic violence, women at work, and mental health problems in all parts of society. These are social challenges, not technological. Prima facie, one would expect that personal services would be best provided in relatively small communities and that smaller governing units (and communities) would be able to respond to individual problems better than large ones.

More generally, the delivery of desired of local services depends on four factors:

(i) the capacity to understand local needs

(ii) the motivation to meet local needs and

(iii) the ability to deliver services and

(iv) financial capacity (previously discussed and not discussed below)

Regarding (i), understanding local needs, personal interactions, communications and consultations between local residents and businesses and decision makers (elected representatives and senior council officers) are highly helpful, even critical, to understanding preferences. Regarding (ii) motivation, as Frey (2013) and other researchers have shown, productivity and meeting service demands is critically a function of the intrinsic motivation of the workforce (i.e. workers identifying with the objectives of the organization). This again is easier to achieve in small organisations than in large ones. The capacity to deliver services (iii) is a function of clearly defined tasks and delegations for council staff. It is also a function of volunteer services. In many areas, volunteers play a key role in supplementing service delivery. In larger councils, there is a real risk that potential volunteers will feel less connected to the values of their ‘local’ community and therefore less motivated to contribute.

These topics warrant more empirical investigation. However, there is good reason to expect that all these three key service factors will be stronger in small local government units.

On the other hand, where resident preferences vary, it is hard to produce differentiated services across a large area. Differentiated services, say more or less library or cultural facilities, imply different levels of resource allocation and therefore (in the absence of user payments) they involve cross subsidies. Inevitably, decision makers will tend towards uniform provision of services over cross-subsidies and the related potential charges of favouritism or even corruption associated with differentiated services.

As it happens, the writer has conducted a survey of service standards (as measured by communications between elected councillors and local residents) in local government units of varying size. Abelson (1981) showed that households in small local government areas in Sydney receive much higher levels of service from their elected councillors than do households in larger areas. Measured by hours of work per representative per citizen, households in local jurisdictions with 40 000 persons received twice the services that households received in jurisdictions with 120 000 persons. Also, households in the smaller areas made 66 per cent more telephone calls, sent 25 per cent more letters to their representatives and received twice the number of visits from their representatives than did households in areas with 120 000 persons. These differences were not explained by economies of scale in the larger local government areas. While there have been extraordinary changes in communication technologies, it is reasonable to assume that per capita communications between residents and elected representatives fall significantly as the size of the governing jurisdiction rises.

As we have observed above, metrics of service delivery, and especially of preferred services, are difficult to obtain and inter-council comparisons hard to draw. This is doubtless why the ILGRP, IPART, and the NSW Government have largely ignored the effective delivery of local public services. However, councils' customer satisfaction surveys and related data can provide essential evidence on these issues. Much more could be, and should be, done to gauge resident satisfaction with the provision of local public services and their causal factors.

Effective local democracy means having some control over your local services and local environment. The Organisation for Economic Co-operation and Development (2009, p.103) defines social capital as ““networks together with shared norms, values and understandings that facilitate co-operation within or among groups… these networks and understandings engender trust and so enable people to work”. This chapter contends that local democracy and social capital (a sense of community) are strongly linked.

In his classic, best-selling, book, Bowling Alone, Robert Putnam (2000) showed that while Americans have grown more affluent, their sense of community has withered. Many cities and traditional suburbs have become anonymous places where people sleep, work, and do little else. Putnam showed that as people spend more time working in their office, commuting to work, and watching TV alone, they spend less time joining community groups and voluntary organisations and socialising with neighbours.

In 2010, Andrew Leigh published Disconnected, a book exploring the decline in community in Australia. Leigh found that, between the 1980s and the 2000s, Australians had fewer friends they could drop in on uninvited and fewer friends they could call on for a favour. In his recent book, The Art of Belonging, a leading Australian social analyst, Hugh Mackay (2014, p.1) wrote, “We rely on communities to support and sustain us and, if those communities are to survive and prosper we must engage with them and nurture them”. Indeed.

A major finding of the 2012 Nobel Prize winner in Economics, Elinor Ostrom (1990) is that small communities generate much greater trust and social capital. It is not credible that communities of 300,000 people, or even 150,000, can generate as much inter-personal trust and mutual social capital as communities of say 30,000.

Most people want some empowerment over their own lives, their local services, and their local environment. A larger council does not reduce the amount of government that people experience. Instead, by disempowering citizens, it increases the sense of being governed.

Elected councillors provide an essential link between the community and the council officers and administration. Under the NSW Local Government Act 1993 (Section 232, p. 109), a key role of a councillor is to “facilitate communication between the community and the council.” In small local government areas, citizens are represented by local councillors, all or most of whom have extensive local knowledge, emotional connection, and availability. Substantially reducing representation per head of population significantly weakens accountability and hence service levels. This was a major concern expressed in resident submissions to the NSW Government’s merger proposals.

We have already noted the critical role of volunteers in providing community services, both through councils and outside it. Volunteering is fundamentally a function of social capital and thus is likely to fall as the size of local government increases.

In our view, criterion 4, the democratic will of the local population is the most critical criterion. If a well-informed community as a whole prefers a smaller local governance model, this would generally indicate that they thought they were getting better value for money than they would expect in a larger local government area. As far as the larger unit would provide savings and/or improved services, the local community is still saying that it would willingly bear this cost for the sake of effective local democracy and social capital.

Of course, some may argue that, for the greater good of the urban or metropolitan area, the capacity to work with a higher level of government (criterion 1) should override local preferences. Given the arguments presented above that the state government has sufficient power to deal with broader geographical issues, at least in Sydney, in our view criterion 1 should rarely override local democracy.

Ideally, the role and size of local government should be determined jointly. This rarely happens. Rather, roles are allocated first and size is determined afterwards. Critically for our discussion, in a three-tier level of government, the roles typically allocated to local government are relatively minor at least in financial terms, though not necessarily for quality of life. This is the case in most of Australia.

This chapter has discussed four main criteria for assessing size of local government: strategic capacity to work with higher levels of government, economic efficiency (financial capacity), and effective provision of preferred local services’, as well as effective local democracy and social capital.

In our view, the subsidiarity principle of democratic government  public decisions should be made as closely as possible to the citizens affected by the decisions  should be the default position. In the famous words of Baron de Montesquieu (1748), “In a small republic, the public good is more strongly felt, better known and closer to the citizen”.

The chapter argues that small local government areas generally empower households, maximize social capital, and are the best vehicles for provision of local services and protection of the local environment. The strong evidence is that larger councils do not produce financial savings.

However, the second level of government (the NSW Government in the case of this chapter) often has a different perspective. It places a high priority on the role of local government as an effective partner in government. Some readers could interpret this view more as a desire to control local government than to work with it.

While councils need “strategic capacity” in one or other way, a council needs also to be able to provide desired local public services to its local community. Thus, we conclude that it is inappropriate to elevate weakly defined strategic capacity over other equally, or more important, objectives and capacities.

However, these views should be tested rather than taken for granted. Voluntary mergers may work where adjacent areas have similar socio demographics, resident demands and priorities, and easy access without topographical barriers. Councils' customer satisfaction surveys and related data provide essential evidence on these issues.

The chapter concludes that the single most important criterion for enlarging local government area should be informed local community preferences as tested by polls or large-scale surveys.

Abelson, P. (1981). Some benefits of small local government areas. Publius, The Journal of Federalism, 11(1) 129-40.

Abelson, P. (2012). Public economics: Principles and practice (3rd Ed.). New York, NY: McGraw-Hill.

Abelson, P. & Joyeux, R. (2015). Smoke and mirrors — Fallacies in the New South Wales government’s views on local government financial capacity. Public Money and Management, 35(4), July.

Abelson, P. Joyeux, R., Milunovich, G and D. Chung (2005). Explaining house prices in Australia: 1970 to 2003. Economic Record, 81, S1-S8.

Bell, B., Dollery, B., & Drew, J. (2016). Learning from experience in NSW? Economic Papers, 35(2), 99-111.

Callanan, M., Murphy, R., & Quinlivan, A. (2014). The risks of intuition: Size, costs and economies of scale in local government. The Economic and Social Review, 45(3), 371-403.

Dollery, B., Grant, B., & Korrt, M. (2012). Councils in cooperation, shared services and Australian local government. Sydney: Federation Press.

Dollery, B., Grant, B., & Kortt, M. (2013). An evaluation of amalgamation and financial viability in Australian local government. Public Finance and Management, 13, 215-238.

Drew, J. & Dollery, B. (2014). The impact of metropolitan amalgamations in Sydney on municipal financial sustainability. Public Money and Management, 34(4), 281-288.

Faulk, D. & Grassmueck, G. (2012). City-county consolidation and local government expenditures. State and Local Government Review, 44(3), 196-205.

Found, A., 2012, Economies of Scale in Fire and Police Services in Ontario, OMFG papers in Municipal Finance and Governance 12. Toronto: University of Toronto, Munk School of Global Affairs, Institute on Municipal Finance and Governance.

Frey, B. (2013). How should people be rewarded for their work? In A. Oliver (Ed.), Behavioural Public Policy (pp. 165-183). Cambridge: Cambridge University Press.

Gao, G. & Melser, D. (2016). Revealed preference measures of quality of life in Australia’s urban and regional areas. Economic Record, 92(Special Issues), 15-29.

Independent Local Government Review Panel (ILGRP). (2013a). Future direction for local government, Twenty essential steps. Retrieved on September 21, 2016 from:

Independent Local Government Review Panel (ILGRP). (2013b). Revitalising local government, Final report. Retrieved on September 21, 2016 from:

Independent Pricing and Regulatory Tribunal (IPART). (2014). Review of criteria for fit for the future. Retrieved on September 21, 2016 from:

Independent Pricing and Regulatory Tribunal (IPART). (2015). Methodology for assessment of council for fit for the future proposals. Retrieved on September 21, 2016 from:

KPMG. (2015). Options analysis. Local government reform. Model design. Draft Methodology Paper.

KPMG. (2016). Outline of financial modelling assumptions for local government merger proposals. Technical Paper prepared for NSW Department of Premier and Cabinet.

Leigh, A. (2010). Disconnected. Sydney: University of New South Wales Press.

Mackay, H. (2014). The Art of Belonging. Sydney: MacMillan.

Miles, L., Borchet, A., & Egan Ramanathan, A. (2014, August 13). Why some merging companies become synergy overacheivers. Bain Brief. Retrieved on September 21, 2016 from:

Montesquieu, C. (1784). The Spirit of the Laws. See 1989, p.83. A. Cohler, B.C. Miller & H.S. Stone (Eds.). (A. Cohler, B.C. Miller & H.S. Stone, Translators). New York. Cambridge University Press.

Musso, J. A., (1998). Fiscal federalism as a framework for government reform. In F. Thomson & M. T. Green (Eds.), Handbook of Public Finance (pp. 347-395). New York: Marcel Dekkar.

Niskanen, W. A. (1994). Bureaucracy and public economics. Edward Elgar.

Northcote Parkinson, C. (1958). Parkinson’s law: Or the pursuit of progress. London: John Murray.

NSW Department of Planning and Environment (2014). A Plan for Growing Sydney.

NSW Local Government Act 1993,

NSW TCorp. (2013, April). Financial sustainability of the New South Wales local government sector.

Oates, W. (2006). On the theory and practice of fiscal federalism. Institute for Federalism and Intergovernmental Relations, Working Paper No. 2006-05. Retrieved from:

Organisation for Economic Co-operation and Development (OECD). (2009). OECD Insights: Human Capital. What is social capital?

Ostrom, E. (1990). Governing the commons: The evolution of institutions for collective action. Cambridge: Cambridge University Press.

Putnam, R. (2000). Bowling alone: The collapse and revival of American community. New York: Simon & Schuster.

Reese, L. (2004). Same governance, different day: Does metropolitan reorganisation make a difference? Review of Policy Research, 212(4), 595-601.

Reserve Bank of Australia (2015). Submission to the Inquiry into Home Ownership. House of Representatives Standing Committee on Economics.

Sanson, G. (2015). Debate: The case for local government amalgamations in Sydney: fact and fiction. Public Money and Management, 35 (1), 65-66.

Spicer, Z. and A. Found, 2016, Thinking Regionally: How to Improve Service Delivery in Canada’s Cities, Commentary No. 258, C.D. Howe Institute, Toronto.

Tiebout, C. (1956). A pure theory of local expenditure. Journal of Political Economy. 64, 416-424.

Previous PageTop Of PageNext Page