Working paper

Subsidiarity: Implications for New Zealand (WP 02/03)

Formats and related files

Authors: Kevin Guerin

Abstract#

Subsidiarity requires taking decisions at the level of government best placed to do so, but does not say what that level is. Rather, it gives a broad framework within which to have the debate. Implementing subsidiarity means (1) allocating roles appropriately between levels of government, (2) co-ordinating implementation of decisions, and (3) managing accountability and participation. Subsidiarity does not, however, tell us how to achieve these goals. It is therefore more about how a decision is made than about what the specific decision is. Europe, the United States and Australia have adopted varying solutions to these issues. New Zealand’s ability to influence the trans-Tasman outcome is likely to be limited. The main implications for New Zealand are in designing trans-Tasman institutions and allocating responsibilities between central and local government.

Acknowledgements#

This paper constitutes part of the future research agenda identified in Guerin (2001). Detailed and frank feedback from Joanna Gordon, Murray Petrie and Mark Sowden from the Treasury was much appreciated, along with comments on this and/or earlier work from Gareth Chaplin, Megan Claridge, Mary-Ellen Fogarty, and Kirsty Hutchison from the Treasury, and Peter Mumford from the Ministry of Economic Development.

Disclaimer#

The views expressed in this Working Paper are those of the author(s) and do not necessarily reflect the views of the New Zealand Treasury. The paper is presented not as policy, but with a view to inform and stimulate wider debate.

1  Introduction#

This paper shows how the concept of subsidiarity can be applied in the New Zealand debate on the form of government, domestically and in the trans-Tasman context.

The Concise Oxford Dictionary defines subsidiarity as “the principle that a central authority should have a subsidiary function, performing only those tasks which cannot be performed at a more local level.”

This definition can be recast as being the process of determining “at what level of governance decision-making should be carried out” (Claridge and Box, 2001, p4) which can be local, regional national or trans-national. This approach allows for explicit upward or downward allocation of functions.

In summary, subsidiarity can be seen as being about making sure that decisions are taken at the most appropriate level; for example by those most directly affected, by those best informed and by those best placed to deal with any consequences. This leaves unresolved, of course, the highly subjective issue of what is “appropriate”.

The paper begins in Part 2 by attempting to identify standard criteria that can be used in the application of subsidiarity.

It then moves on to review in Parts 3 and 4 how the concept has been reflected in overseas experience, focusing on the EU where the term has achieved greatest prominence, the US where constitutional constraints on government have been explored to the greatest degree and Australia, where the principles behind the concept of subsidiarity are of direct relevance to the increasing network of trans-Tasman governmental relationships.

Part 5 then looks at implications for governmental structures within New Zealand, with concluding remarks in Part 6.

2  General principles for allocation and co-ordination of functions#

This section addresses various principles underlying any discussion about how governmental functions should be allocated initially and co-ordinated thereafter.

These include balancing between national consistency and preserving local diversity, co-ordinating policy processes and ensuring accountability and participation, and the use of constitutional frameworks to achieve these principles.

It concludes by discussing decision criteria for making decisions on allocation of roles in a manner consistent with the principle of subsidiarity.

2.1  Consistency vs diversity – allocation of powers#

The issue of how best to allocate powers between national and trans-national bodies, and between levels of government within a nation (taking into account the concept of subsidiarity) must be addressed from both private (individual and firm) and public policy perspectives.

Ideally this would result in simultaneously facilitating diversity in preferences, conditions and values, maintaining the ability to live, work or run a business anywhere within a locality, region, nation or trans-national grouping, having consistent regimes that facilitate the activities of individuals and firms anywhere within a harmonised set of jurisdictions, and reflecting the principles of subsidiarity through allocating decision-making powers to the most appropriate level of governance.

This is theoretically feasible, in that the decisions which most affect the freedom to live, work or do business in different places could be made at the higher levels of governance, with the level of decision-making declining as the impact of the decisions on individuals’ or firms’ freedom declined.

It is also consistent with theories of fiscal federalism which assert that the jurisdiction of decision-making should correspond to the jurisdiction of effects (Claridge and Box 2000, p69).

However, it requires careful balancing of conflicting objectives, especially around the importance of minimising transaction costs. There may be situations where the advantages of maintaining distinctive policies outweighs the results of a straight benefit/cost analysis.

In this context, it is interesting to note that even Switzerland (not a member of the EU, or most other trans-national bodies) still finds it difficult to both maintain the ability of sub-units (cantons) to experiment with differing policies and open up domestic markets to achieve international harmonisation.

Competition can have advantages – the Tiebout effect suggests individuals migrate to match preferences for the combination of taxes and public goods they prefer, forcing governments to optimise the quantity and “price” of public goods, and protecting individuals from coercive government (van den Hauwe, 1999, p609).

At the same time, however, consistent rules across jurisdictions facilitate exchange, and support the role of central governments in dealing with intra- and inter-national externality effects, including through making and enforcing international treaty commitments. It is crucial to determine where uniformity matters and where it does not. There may be situations where inter-jurisdictional competition offers net national benefit as opposed to encouraging rent-seeking or cost-shifting behaviour.

How the rules are set for allocating different policy areas to each category is a much more difficult and complex issue where we can draw on principles such as those identified below by Kerr, Claridge and Milicich (1998). The public good nature of maintaining armed forces, for example, along with the potential for free riding, argues strongly for centralised provision.

Constitutional design can be seen as a matter of determining which voting rule or choice mechanism would be specified for each state activity, with the best decision rule for each activity being the one that minimised “interdependence costs” - “perceived for each individual as being the sum of the external costs levied on that individual if not part of the decision set and the decision making costs for that individual if part of the decision set” (van den Hauwe, 1999, p612-3).

This approach of linking involvement in decisions to the effect of the decisions is closely related to the concept of subsidiarity in the design of state institutions.

As well as defining what each level of government can do, a constitution may (and perhaps should) also define what they cannot do, for example, a constitution may ban any level of government from infringing on certain rights of citizens, prevent lower levels of government undertaking certain activities reserved to higher levels (or vice versa), or prohibit one lower level government from taking actions that impinge on activities in a locality for which another lower level government is responsible. Subsidiarity is one reason for such bans.

Federal constitutions often define the number of lower tier governments, their respective responsibilities and local representation in national government. The degree of decentralisation will be on a spectrum between the two extremes of fully centralised or fully decentralised responsibilities (neither of which is likely to be practical). Federal constitutional structures are therefore one means of implementing or achieving subsidiarity.[1]

2.2  Inter- and intra-governmental co-ordination#

A crucial part of resolving the above issues is to determine how much co-ordination should occur between levels of government, and what form it should take; i.e. hen and how often, who should take part, and whether it would be for information only or as part of the decision-making process. Co-ordination can take the form of consultation, delegation, or joint planning and/or implementation.

In a steadily more globalised policy environment, national governments need to consider whether existing internal arrangements for intra-governmental co-ordination are adequate to optimise the outcomes from their international relationships, and how to achieve transparency, public access and accountability in any structures established to manage international relationships (Bermann, OECD, 1994).

Effective co-ordination long-term requires institutionalisation of contacts, reciprocal notifications of regulatory actions and transparency, participation and accountability regimes. These are particularly important for cross-border regulation, since as regulations become more closely harmonised, there is increasing pressure to integrate planning and decision-making across jurisdictions and to do so at earlier stages of the policy process. (OECD, 1994, p60):

Subsidiarity provides one approach for starting a discussion on where any particular policy responsibility should sit. It would normally support placing the responsibility at as low a level of government as practicable given other factors (see the discussion below on decision criteria).

Notes

  • [1]The New Zealand situation with a unitary parliament and no written constitution is less formally defined, but the implications for flexibility and certainty are debateable. The unwritten constitution exerts considerable influence and can be very resistant to change. It can also shift rapidly (as with the arrangements for transitional governments after the 1984 election), and does not always result in clearly separate central and regional/local roles.

2.3  Accountability and participation#

Another factor to consider is the relative effectiveness of governance and accountability arrangements at each level of government. Optimally any such problems would be dealt with in the institutional design. If this is not achieved, then any decisions on the allocation of responsibilities will inevitably reflect those actual or perceived problems. Failing to address governance and accountability issues can therefore block institutional change, regardless of its desirability.

In the trans-Tasman situation, one example would be a joint regulatory agency which in theory would need to (not an exhaustive list) consult in both countries, be answerable to Ministers in both countries, and possibly the states and territories, report to both national Parliaments, and be supervised by both sets of ombudsmen and covered by both official information regimes.

Such a set of arrangements would be difficult to design, and require significant effort by the agency, Ministers and Parliaments to make them work effectively in practice.

At the local government level, delegation or joint provision of powers creates the same problems regarding accountability to the public and opportunities for public input. The recent proposal for a trust to manage water supply in Wellington raised concerns about precisely these issues.

2.4  Subsidiarity - decision criteria#

To allocate responsibilities between levels of government in the optimum manner; i.e. to make subsidiarity effective in practice; requires clear criteria to guide decisions.

Such decision criteria could include (Kerr, Claridge and Milicich, 1998, p16) where the relevant expertise and/or knowledge exist (which may differ), where effective accountability arrangements exist (for example, close to those affected by the policy, but avoiding small group domination), and how best to make trade offs.

Table 1 – Some Criteria for Considering Subsidiarity
Goal for Decisions Assumptions Local control Regional control Central control
Balanced Allocation of powers is optimal if decisions are made by those who feel the effects and bear the costs. no externalities exist; externalities can be internalised within region; significant externalities exist;
Informed Whether the necessary information is objective or subjective (or specific/idiosyncratic and non-specific/general) affects who is best-placed and most capable to make decisions.

info is held locally or is unimportant;

preferences varying and important;

regional scale data is important;

info is held centrally and is important or complex;

preferences homogenous or unimportant;

Cost-effective Economies of scale in decision-making may exist, particularly if preferences are homogenous. low decision-making costs; medium decision-making costs; high decision-making costs;

Table 1 illustrates how factors such as whether or not externalities exist, the costs of gathering the information needed to support a decision and the degree to which preferences vary by jurisdiction affect the preferred level for making decisions that are simultaneously balanced, informed and cost-effective.

In applying such criteria it may also be useful to focus on why local governance is valued, for example (Bermann, 1994, p339-344) facilitating effective accountability of decision-makers to those affected, allowing self-determination and political liberty, achieving flexibility to take account of local circumstances and preserving identities and diversity.

The key balance in trade-offs is therefore between meeting the specific needs of one community; and delivering consistency between jurisdictions. Local identity and national consistency both have pluses and minuses. In particular, minimising transactions costs is important (through measures such as removing customs barriers, and having consistent provisions in areas such as occupational licensing and health and safety rules) while reflecting local conditions and culture increases acceptance of the regulatory regime. It is also important to ensure that inter- and intra-national commitments are met.

Some high-level examples of possible outcomes of the decision criteria described in Table 1 are listed in Table 2.

Table 2 – Central vs Regional/Local Provision
Function Balance Information Costs External Commitments Provision
Defence Costs and benefits are highly dispersed. Significant externalities. National government holds crucial information which is highly subjective. Significant economies of scale. Essential. Central.
Environment Costs or benefits can be dispersed. Externalities can be significant. Held at all levels. High. Can be essential in some cases. Mixed central, regional and local.
Road names Highly localised. Local. Low. Low. Local.

The same criteria are likely to apply in general across national borders as within them. Claridge and Box (2000. p73) suggest that trans-national decision-making may be optimal where externalities cross national borders, international public goods exist, objective scientific information is important and hard/expensive to access, international preferences are homogenous and decisions could be more cost effectively made by or with others.

These criteria are consistent with the general approach taken by the EU and the various definitions of subsidiarity cited in this paper. Not all of them need apply in any given case (and often won’t), but the more that do apply, the stronger the argument will be for regulation through a trans-national mechanism.

An example of where responsibility for regulating different aspects of a single area of policy can usefully be assigned to multiple levels of government, which each have appropriate knowledge and powers, is maritime safety where international organisations set minimum standards and co-ordinate policy setting processes, national governments can set more stringent standards for their own waters, provide rescue services and inspect visiting vessels and local governments can control inshore activities such as jet-skis and ferries.

3  Overseas experience#

While subsidiarity has been given most attention in Europe, the issues involved have also been addressed, not always explicitly, elsewhere. This section compares US, European and Australian approaches.

3.1  US experience#

The concept of democratic federalism involves assigning each responsibility according to how the assignment best ensures personal, political and economic liberties (Inman and Rubinfeld, 1998, p10). Other priorities are of course possible.

In the case of the United States of America, the powers of the Federal government are technically strictly defined by the Constitution. In practice the definition can be more free ranging due to varying Court interpretations of these powers and as a consequence of the States seeking powers and/or resources from Congress.

Extensive debates occur about imposing responsibilities on the States without any supporting resources, although the cost of unfunded mandates must be reported (not the case in New Zealand), and about limited or no State discretion in meeting Federal targets. Nonetheless, actual challenges to Federal competence in recent times are rare.

The general approach of the US Supreme Court is that the legislative process itself is the protection for State autonomy The Court will not review a Congressional decision to pre-empt State regulation, or subsidise the States adopting a standard regulatory model (Bermann, 1994, p423-427). The subsidiarity principle is not as central to debates about the location of governance in the US as in the EU. Effectively the division of powers and the representation of the States in Congress have come to fulfil the role for which the principle of subsidiarity has developed in Europe.

3.2  EU experience#

The US constitution was intended as such from the beginning. The EU (European Union) constitutional framework, if it can be described as such, evolved from an international agreement. This may explain why it does not deal comprehensively with issues such as enumeration of EU powers, EU pre-emption of national law and implied powers. This has left significant scope for judicial interpretation by the European Court of Justice (Bermann, 1994, p348-362).

EU decision-making is based on determining at what level to regulate, based on whether there is direct assignment to the EU or exclusive powers at EU level in the relevant area or the subsidiarity test, and how to regulate at the EU level, based on proportionality (with implementation and enforcement usually done at national level (Stauffer, 1999)).

The origin of subsidiarity as discussed in the EU context is drawn from 20th century Catholic philosophy, in particular from the following text (Pope Pius XI, 1931): “The supreme authority of the State ought, therefore, to let subordinate groups handle matters and concerns of lesser importance, which would otherwise dissipate its efforts greatly … in observance of the principle of "subsidiary function."

It has been suggested that subsidiarity in the EU was initially advanced by Christian Democrat politicians as a justification for enlarging the Commission’s competencies, i.e. increasing the efficiency of government now demanded achieving certain goals at the EU level (Kersbergen and Verbeek, 1994, p216). However the argument runs both ways, and now subsidiarity appears to be one of the main arguments used by national and sub-national governments to limit the Community’s role.

Subsidiarity in Europe has been seen in the above terms as a concept of the relationship between the state and society, or (particularly in Germany) as a lasting constitutional principle of distribution of powers or an ongoing counterbalance to “creeping federalism” (Kersbergen and Verbeek, 1994, p226).

The subsidiarity principle is now embodied in the Treaty with two facets, the EU must act where the objectives can be better attained at Community level and the EU must not act outside its exclusive competence where objectives can be sufficiently achieved by Member States acting individually (European Communities, 2000).

The subsidiarity principle is closely linked to the proportionality principle (a common basis for judicial review in European law) which generally requires that the measure must bear a reasonable relationship to the objective, the costs must not manifestly outweigh the benefits, and the measure must be the least burdensome option.

In practice, EU constitutional arrangements can be described as a form of decentralised federalism, with continual pressures towards greater centralisation of responsibilities in many areas, for example to achieve advantages of scale or avoid “destructive competition” between regulators, the “race to the bottom” theory (van den Bergh, 1996, p364).

Greater integration in some areas, such as open borders, increases those pressures. One consequence has been weakening of the majority requirements for joint decisions in a number of policy areas.

This may be a major reason for the increasing contrary pressure to increase the role of regions as opposed to national and community governments and put substance behind the principle of subsidiarity, so that as well as determining whether it can exercise powers, the EU must evaluate whether it should do so. The latter is a matter for judgement which will often be highly subjective.

The EU has released a White Paper on European Governance for consultation until March 2002. The specified work areas are broadening and enriching debate on European matters, the process of producing and implementing rules, decentralisation, promoting coherence and co-operation, contributing to world governance and integrating policies across the continent.

Issues identified for public comment include, whether the allocation of competences between levels of government is “per se” or as a means for achieving objectives, how dynamic factors (for example, crises, technology etc) cause shifts in policy responsibilities and whether principles such as the protocol to the Treaty of Amsterdam are adequate; and how to organise interaction between the various levels of government.

These questions could presumably equally be asked with respect to the trans-Tasman relationship and relations between New Zealand central, regional and local government. It is not clear, however, that they will be given the lack of formal institutions for intergovernmental relations across Australasia or within New Zealand.

3.3  Australian experience#

Within Australia, subsidiarity arises primarily as a Commonwealth vs States and Territories issue. The degree of Commonwealth authority varies widely, often due to explicit or implicit constitutional constraints, with implementation details frequently a matter of state discretion.

Some bodies (such as the Australian Competition and Consumer Commission) draw on both federal and state laws, while the national electricity grid is owned by a company set up by the States (Wilkins, 2000).

Fiscal pressures create the strongest drive towards centralism due to the dependence of the States on the Commonwealth’s wider funding powers. This is a classic issue of efficiency in revenue collection and debt raising encouraging centralisation while efficiency in expenditure encourages decentralisation. Control of the purse can negate the intent of strict constitutional allocation of roles.

Within states, the form of the debate is less clear and the division of powers appears to be substantially at the discretion of the state government – a situation more parallel to that of local government in New Zealand.

4  Trans-Tasman arrangements#

The implications of subsidiarity for assignment of roles and design of arrangements for accountability and participation need to be considered when attempting to harmonise regulatory regimes across borders (Guerin, 2001), given the interactions required between jurisdictions. For New Zealand, the key cross-border arena is the trans-Tasman one.

Managing different depths of harmonisation along the length of the regulatory chain, from policy through to enforcement, is complicated enough in itself, but becomes even more so when different levels of jurisdictions can be involved at different stages. This is already an issue with trans-Tasman harmonisation, for example food standards regulation where state and territory government are directly represented.

4.1  The purpose of trans-Tasman arrangements#

It is crucial before getting into the details of designing cross-border allocations of functions within Australasia to determine the objective(s). These could include (this is a non-exhaustive list) achieving economies of scale and scope, avoiding destructive regulatory competition, closer economic or political integration or creating a world recognised centre of regulatory expertise.

In addition, we need to determine what we are trying to optimise – the outcome for New Zealand, for Australia, for Australasia, for APEC or for the world?

The goal(s) of co-regulation in each policy area should be determined before a new structure is established and the more that the answers for Australia (or for each state of Australia) and New Zealand differ, the more significant will be the implications for institutional design. When the objective is a solution that is optimal at a national or local/regional level, subsidiarity would suggest that a trans-national solution is not ideal. Failure to address these issues risks compromising any new structure which could therefore be ineffective or worse than the status quo.

The relevant outcomes and goals will depend on the particulars of each proposal, the interdependencies between them and the longer-term strategic direction of the relationship (for each country).

This makes it difficult to predict broad outcomes. Much more work is required at both the detailed practical level and the higher political level before anything approaching firm forecasts for the extent of future Australasian regulatory arrangements can be developed.

4.2  Potential for gains#

New Zealand’s ability to achieve gains from harmonisation with Australia will depend critically on the nature of the problem being addressed, the specific features of the sector concerned and existing constitutional and institutional arrangements in Australia.

It will be crucial to determine at an early stage how much initial and ongoing leverage New Zealand will have and what alternatives exist. Where Australia’s approach is already constrained by constitutional or other inter-governmental arrangements, harmonisation may either be impossible (for example because of differences between Australian states and territories) or constrained to an existing Australian pattern.

It will become increasingly important whether the Australian Commonwealth or States hold the authority to regulate in the relevant sector. The states may frequently be in a position to set the boundaries for the scope and depth of harmonisation. This may more frequently lead to situations where the depth of harmonisation within a particular sector varies significantly (for example trans-national standards, national testing and local enforcement).

It will therefore become increasingly important for New Zealand Ministers, officials and other stakeholders to be aware of the particular Australian constitutional and institutional arrangements in specific sectors prior to engaging seriously in any harmonisation exercise.

In Australia, subsidiarity is usually managed through a combination of strict constitutional frameworks and understandings of varying explicitness established though joint Ministerial and Officials groupings. These bodies, in which New Zealand has full membership and voting rights in relation to any decision involving the Trans-Tasman Mutual Recognition Agreement operate at several levels, Head of Government meetings, Council of Australian Governments (COAG), Premiers’ Conferences, Loan Council, Treaties Council, Ministerial Councils; and other Ministerial fora with four or less jurisdictions involved (Australian Government, 1999).

These complex decision-making arrangements, while lengthy and highly political, nay well be necessary to achieve a consensus that will allow an effective policy to be implemented. They can also, however, serve to make it difficult to alter any joint Australian position once reached.

The lack of any true supranational institutions within the trans-Tasman relationship also leaves scope for difficulties to develop, as any agreements must take the form of either rigid treaty provisions where the choice in any dispute may be to accede to the other party’s wishes or abrogate the entire arrangement, or legislation which is ultimately under the control of only one of the parties.

As far as application of subsidiarity is concerned, Australia’s approach is strongly shaped by constitutional limitations. New Zealand’s ability to shape trans-Tasman institutions from a different perspective is likely to be limited. If it is restricted simply to retaining a degree of national discretion, rather than having true influence on development and implementation of a joint approach, then the trade-offs between joint and independent action may be quite stark.

5  New Zealand – central/regional/local government#

This section addresses the roles of the levels of Government in New Zealand and mechanisms for co-ordination between and within levels.

5.1  Central control vs local discretion#

As far as relationships between central and local/regional governments are concerned, the fundamental constraints include the scope within which local and regional governments can operate and the degree to which central government can intervene to modify outcomes.

Some key issues are whether local democracy is the key principle or can local government only be “trusted” within certain limits, does central government remain accountable to voters/taxpayers for the appropriate exercise of powers which it has delegated to local/regional governments, and what happens when the exercise of local powers has national implications; for example, on transport links?

It can be argued that “policy-making powers should be assigned to local government, except when this would likely be ineffective, inefficient, inimical to others, or demonstrably unnecessary” (Pelkmans and Sun, 1994).

The answer in New Zealand appears to be that central government retains ultimate control and accountability over all levels of government. In the area of funding, where central government taxes income and expenditure, while local government taxes land (the archetypal immobile asset), even that separation and the very ability of local government to tax at all, is dictated by parliament and can only be changed by parliament.

The allocation of rules and accountabilities between levels of government in New Zealand is determined not by a written constitutional framework, but by the processes of central government policy formulation within a Westminster-style unitary parliament.

This was demonstrated early on in the development of New Zealand democracy when Parliament chose in 1876 to abolish provincial governments (which had their own parliamentary structures), and in the 20th Century to abolish the second chamber of the national parliament (1 January 1951), abolish special purpose local authorities and compulsorily merge territorial local authorities and create regional councils[2].

Local government in New Zealand tends therefore to operate (to varying degrees) as an agent of national government, administering or interpreting rules set centrally.[3] The range of discretion allowed to local government in making and implementing policy varies from nil to almost complete. Local government activities can usefully be divided into three categories of (1) prohibited (policy set and implemented nationally), (2) mandatory (policy set nationally and implemented locally) and (3) discretionary (completely local) activities.

Land transport policy is a mixture with funding arrangements operating at both national and regional/local level and sometimes overlapping. The Resource Management Act covers the full spectrum as it prescribes procedures and criteria, leaves the actual decisions at the regional/local level, but provides for central government guidance or call-in; for example, through national policy statements.

Such approaches are consistent with the discussion above on the design of federal institutions. Fundamentally, however, there does not appear to be any likelihood that the allocation of powers between central and local government in New Zealand will be other than at the discretion of central government for the foreseeable future. To do otherwise would require a comprehensive constitutional review, the discussion of which is outside the scope of this paper.

It is also not clear whether there is a way within existing constitutional constraints to simultaneously achieve, at national, regional and local levels, clear separation of roles, direct linkages between powers and funding, true accountability for each role, and effective co-ordination of delivery of governmental functions.

Within Maori society there are also multiple potential levels of governance – whanau, hapu, iwi and pan-Maori organisations. The particular issues raised as a result are not discussed in detail in this paper which is confined to state governance issues.

5.2  Inter- and intra-governmental co-ordination#

In New Zealand, transfer of functions between, or joint provision by, local governments is already provided for to some degree, and likely to be more so once the current reform process is complete so that the accountability transfers along with the functions.

Between central and local government, co-ordination is much less frequent and less structured, which is likely to make firm planning and commitments more difficult. Within the New Zealand constitutional structure, any true negotiation between central and local government is likely to be limited, given the underlying reality of where the power lies.

The existence of a power of compulsion is relevant here – in the EU context it has been suggested that too much dispersion of power within a state can lead to the EU choosing to retain powers itself because national governments cannot guarantee effective action (Bermann, 1994, p343). In this case, regional empowerment within a nation can lead to national disempowerment, and presumably eventually regional disempowerment as well.

Within central government, co-ordination is more formally required and can in effect be imposed at the stage of a Cabinet decision if it has been lacking to that stage. However, when international interactions are involved, this can be more difficult as effective commitments can be entered into well before the time comes to seek formal decisions at a national level. Traditionally in most countries, co-ordination is handled through a single department responsible for foreign relations, but this is coming under more strain as the range and depth of international co-ordination increases.

The solutions available are to either create much more effective inter-departmental co-ordination mechanisms. or invest significant greater resources into departments to facilitate direct multi-departmental participation in international fora.

It can be argued that the former would have been done before if feasible at low cost, but conversely existing arrangements are likely to contain the customary degree of bureaucratic inertia. The latter is clearly undesirable, from the point of view of costly duplication alone, even without considering the likelihood of mixed messages undermining New Zealand’s negotiating position.

Beyond departmental co-ordination, there is a greater trend for direct Ministerial participation in trans-Tasman decision-making bodies, and proposals for joint parliamentary oversight of trans-Tasman regulatory bodies. This increases the need to ensure New Zealand is not committed to a position that would not have been reached if consultation between departments or Ministers had occurred. These types of regulatory co-operation are likely to require closer than ever co-ordination within New Zealand to achieve the best national outcome.

Such joint regimes are also likely to increase the pressure to bring even closer together the underlying principles of regulation between countries. The trans-Tasman regimes tend to be driven by the COAG principles which are very similar to the New Zealand Regulatory Impact Statement regime (Australian Government 1997). The New Zealand establishment of a central compliance cost unit in the Ministry of Economic Development also follows to some degree the Australian model. Further narrowing of the gaps seems likely as regulatory regimes become more harmonised.

Notes

  • [2]Except for some aspects of land use rules under the Resource Management Act, the powers of regional councils are complementary to city and district councils, rather than controlling; i.e. it is not a clear central/regional/local government hierarchy.
  • [3]Decentralisation involves central government making decision locally through branch offices. Devolution involves transferring decision rights to regional/local government, but the degree of discretion and the scope for local differences to be reflected in decisions can vary widely.

6  Conclusion#

Subsidiarity as a concept provides a broad framework within which to debate the allocation of governmental responsibilities; i.e. the best compromise between availability of information, location of costs and benefits of the decision, accountability of decision-makers, and cost-efficiency and effectiveness.

That allocation of responsibilities must be supported by arrangements to minimise duplication of functions, ensure that all necessary functions are performed somewhere[4], co-ordinate input by different arms of government to decisions with a wide impact, and ensure adequate transparency, accountability and participation.

Subsidiarity does not tell us how to make the necessary compromise or what the supporting arrangements will look like, but does suggest that following these processes is essential to achieving an optimal outcome.

Europe, the United States and Australia have adopted varying solutions to these issues. Both the American and Australian approaches are primarily constitutional, relying on texts and structures to protect individual and community interests, but the Australian version gives less primacy to the centre. The EU approach emphasises the principles of subsidiarity and proportionality but gives little detailed guidance on their application. This seems closer to English and New Zealand constitutional experience

Opportunities to apply the concept of subsidiarity in the New Zealand context are rare due to the limited interactions between governments, the still small number of trans-Tasman governmental structures and New Zealand’s limited ability to influence the trans-Tasman outcome.

When those opportunities do arise, however, considering the issues raised by the concept may be of considerable value in focusing efforts on the key aspects of allocating functions and ensuring their optimal performance.

Notes

  • [4]Although this restricts the ability of lower levels of government to determine whether performance of the function is in fact necessary in the first place – thereby putting the honesty of the original allocation of responsibility in question.

References#

Australian Government (1999) “Commonwealth-State Ministerial Councils - A Compendium.”

Australian Government (1997) “Council of Australian Governments - Principles and Guidelines for National Standard Setting and Regulatory Action by Ministerial Councils and Standard-Setting Bodies.”

Balaster, Peter (2000) “Liberalising Interstate Commerce – The Swiss Experience.” Paper presented to OECD International Forum on Regulatory Reform, Puebla, Mexico, 25 September.

Bermann, George A. (1994) “Taking Subsidiarity Seriously: Federalism in the European Community and the United States.” Columbia Law Review March (2):331-456

Bermann, George A. (1994) “Managing regulatory rapprochement: institutional and procedural approaches.” in Regulatory Co-Operation for an Interdependent World (OECD): 73-91

Claridge, Megan and Box, Sarah (2000) “Economic Integration, Sovereignty and Identity: New Zealand in the Global Economy.” New Zealand Treasury Working Paper 00/22. http://www.treasury.govt.nz/publications/research-policy/wp/2000/00-22>

European Communities (2000) “The ABC of Community Law.”

European Union (2001) “White Paper on European Governance.”

Guerin, Kevin (2001) “Regulatory Harmonisation – Issues for New Zealand.” New Zealand Treasury Working Paper 01/01, January. http://www.treasury.govt.nz/publications/research-policy/wp/2001/01-01>

Inman, Robert P. and Daniel L. Rubinfeld (1998) “Subsidiarity and the European Union”, NBER Working Paper 6556, May.

Kerr, Suzie, Claridge, Megan and Milicich, Dominic “Devolution and the New Zealand Resource Management Act.” New Zealand Treasury Working Paper 98/7. http://www.treasury.govt.nz/publications/research-policy/wp/1998/98-07>

Kersbergen and Verbeek (1994) “The Politics of Subsidiarity in the European Union.” Journal of Common Market Studies 32(2) June.

OECD (1994) Regulatory Co-operation for an Interdependent World, OECD Public Management Studies

Pelkmans and Sun (1994), “Towards a European Community regulatory strategy: lessons from “learning-by-doing.” Regulatory Co-Operation for an Interdependent World , OECD: 179-199,

Pope Pius XI (1931) Papal Encyclical Quadregismo Anno (1931), http://www.vatican.va/holy_father/pius_xi/encyclicals/documents/hf_p-xi_enc_19310515_quadragesimo-anno_en.html>

Stauffer, Thomas (1999) “Subsidiarity as Legitimacy.” Intergovernmental Fiscal Relations and Local Financial Management, CEU/OS/WB Summer Course of the Central European University

van den Bergh, Roger (1996) “Economic Criteria for Applying the Subsidiarity Principle in the European Community: The Case of Competition Policy.” International Review of Law and Economics 16: 363-383

van den Hauwe, Ludwig (1999) “Public Choice, Constitutional Political Economy and law and Economics.” Encyclopedia of Law and Economics. http://encyclo.findlaw.com/>

Wilkins, Roger (2000) “Promoting Regulatory Quality Between Levels of Government: Australia.” Paper presented to OECD International Forum on Regulatory Reform, Puebla, Mexico, 25 September.