The first principle that deserves mention is that of subsidiarity, which has been interpreted in a federal system of government as implying that provision of goods and services should be administered at the lowest level feasible within the national interest (see Brown 2002b; Brown, Podger, Head this volume; Access Economics 2004; Twomey and Withers 2007). The rationale appears to be that this permits such provision to most closely match the preferences of the people.

The second principle to emerge is that of correspondence, which argues that where consumption or use of a particular good or service is limited to the boundaries of a particular jurisdiction, then its provision should be allocated to a sub-national government whose boundaries are defined by the spatial benefit (or market area) boundaries associated with this good or service (Oates 1972, 2005; Warren 2005; Williams 2005b). The resulting allocation generates economic efficiency since it allows for a matching of local demand and supply, with voters able to move between jurisdictions in search for an optimal mix of provision and associated taxes and charges given their individual needs. An obvious difficulty confronted when putting flesh around this principle is that, carried to the extreme, each good or service provided by governments could conceivably have a different set of spatial benefit (or market area) boundaries leading to a need for a multitude of overlapping levels of government (Access Economics 2004; Productivity Commission 2005, 2006). Clearly, common sense is needed when interpreting this principle if only three (or at most four) levels of government are being considered.

The third principle involves giving due recognition to economies of scale in the provision of goods and services, with a case generated for movement of provision to a higher level where it costs less if produced or provided by single jurisdiction rather than separate smaller ones (Access Economics 2004; Williams 2005b). It has been argued more recently that possibilities for separation of production from provision should be exploited where feasible, since this may permit provision to be retained with smaller units, while at the same time allowing them to take advantage of mutually agreed co-operative production arrangements at a scale sufficiently large to generate maximum cost savings (Dollery 2005 b,c).

The fourth principle recognises the constraints imposed by existing jurisdictional boundaries and argues for the need for a mechanism to resolve inter-jurisdiction spill-overs or spill-ins of benefits (and/or costs) of a particular good or service. In the absence of such mechanisms, economic inefficiency in the form of under- (or over-) provision of such goods or services would result. The case is thereby created for responsibility for these goods or services to be either transferred to a higher level of government, or for it to remain at the lower level but for intervention by a higher level through system of tied grants aimed at providing ‘compensation’ to those lower level jurisdictions disadvantaged by the nature of the observed spill-overs (or spill-ins). The tied grant solution is preferred where location-specific or individual-specific cost differences occur between jurisdictions and/or where economies of scale are exhausted at comparatively low population levels and/or geographic spreads (Oates 1972, 2005; Productivity Commission 2005, 2006).

The fifth principle recognises that inter-jurisdictional differences in the nature, cost and/or level of provision of particular goods or services could generate negative impacts on the mobility of factors of production, resulting in economically inefficient locational choices. The literature suggests that where significant transaction costs are imposed on labour or firms from movement across jurisdictions, then a case can be made for minimum standard setting at higher levels and the introduction of a system of compensating grants for lower level jurisdictions for which this imposes cost increases (National Commission of Audit 1996; Productivity Commission 2006).

The final principle suggests that accountability is strengthened if responsibility for a particular function is tier-specific (Brogden 2006; National Commission of Audit 1996; OECD 2006; Selway 2001). For many important functions this is not a realistic option, with most analysts acknowledging that assignment of responsibilities will most often resemble a marbled not a tiered cake (see Wiltshire this volume). What is necessary, then, is to ensure that the nature of this marbling is not randomly generated but rather emerges from a process of careful deliberation (and is the subject of periodic renegotiation as economies of scale and other characteristics of production or provision change over time). If expenditure responsibility in a broad area (e.g. education or health) is necessarily shared, then respective roles in segments of this area need to be agreed and mutually understood, with cooperative arrangements put in place to ensure appropriate ongoing coordination between these segments to ensure they mesh well together at key transition points (Warren 2006; Wilkins 2004).

The principle of fiscal equivalence implies that each level of government should finance its assigned functions with funds it raises itself. However carried to extreme this would lead to significant inefficiencies in tax collection from many revenue sources and distortions to locational choices of individuals and firms. The latter concern has also led to arguments suggesting that taxes on highly mobile tax bases should be allocated to higher levels of government, as should taxes on tax bases that are uneven across jurisdictions (Access Economics 2004; Dahlby 2001; Productivity Commission 2006).

The resulting lack of alignment of expenditure responsibilities and tax assignments leads to a situation of vertical fiscal imbalance – where revenue raising capacity of at least one level of government exceeds its expenditure needs whilst the reverse is the case for the other levels. In general there is support for tax sharing rather than reassignment of tax bases and rate schedules in this situation, implying the need for a system of intergovernmental transfers or grants. Such a system needs to be managed carefully – in particular, a high level of autonomy over expenditure priorities and service management is required by the recipients in order to ensure that the efficiency benefits of competitive federalism are not constrained (Allen Consulting Group 2006; Grewel and Sheehan 2003; Hancock and Smith 2001; Saunders 2002; Warren 2006).

The current level of vertical fiscal imbalance in Australia has been a cause for concern, with the Commonwealth Government’s revenue collections exceeding its expenditure needs by up to 40% (House of Representatives Standing Committee 2003). While transfer by the Commonwealth of GST revenues to the States has made significant inroads into addressing the latter’s own-source revenue shortfalls, a similar arrangement remains elusive for local government (Access Economics 2004; Productivity Commission 2006).

Australia’s current system of intergovernmental grants includes horizontal fiscal equalisation (HFE) payments administered by the Commonwealth and various State Grants Commissions (Commonwealth Grants Commission 2006a, b; Local Government Grants Commission South Australia 2005, New South Wales Local Government Grants Commission 2005, Northern Territory Grants Commission 2005, Queensland Local Government Grants Commission 2005, Tasmania State Grants Commission 2005, Victoria Grants Commission 2005, Western Australian Government Grants Commission 2005). These payments are aimed at ensuring that jurisdictions at each sub-national level of government receive funding from the Commonwealth, such that if each made the same effort to raise revenue from its own sources and operated at the same level of efficiency, then each would have the capacity to provide services of the same standard to their constituents (Morris 2002; Williams 2005a).

These HFE payments currently involve correction for disadvantage on both revenue and expenditure sides, but some concern has arisen over the efficiency effects of the latter (Hancock and Smith 2001; McLean 2004; Petchey and Levtchenkova 2004; Usher 1995). Nevertheless, various attempts at estimating the size of such efficiency effects have concluded that they are small, and possibly worth incurring for the gains they generate in terms of equity (Dixon et al 2002, 2005; Harding et al 2002; Williams 2005b).

Clearly there is a need to evaluate the implications of various change options for the magnitude and nature of both vertical and horizontal equalisation payments – with a recognition that the order of magnitude of such payments may need to involve both a one-off compensation for past inadequate service provision in some newly emerging jurisdictions, and a steady state set of payment schedules for such jurisdictions in the future.