Formats and related files
This document was commissioned by the New Zealand Treasury. However, the views, opinions, findings and conclusions or recommendations expressed in it are strictly those of the author(s), and do not necessarily represent and should not be reported as those of the New Zealand Treasury. The New Zealand Treasury takes no responsibility for any errors or omissions in, or for the correctness of, the information contained in this paper.
A key choice in regulation is whether to centralise or delegate decisions
New Zealand’s Resource Management Act 1991 (RMA) integrates environmental management issues and local planning. The regime has moved away from a prescriptive approach to one based on balancing multiple outcomes, and delegates that balancing process to local government. Central government sets the legislative framework and can provide guidance, such as policies, standards and information. However, there has been only one national policy statement (NPS), legislatively prescribed on coastal policy. Until 2004 there were no national environmental standards (NESs).
New Zealand has delegated many environmental decisions to councils but given them limited help.
This leaves open why guidance has not been provided, and whether this can change without fundamental alterations to the regime. These questions are particularly topical given that a recent review of the RMA has proposed increased guidance and an effort to improve local government capability, but does not consider fundamental alterations necessary. Implementation issues aside, the Act’s design, by making national policy statements (NPSs) primarily another policy input rather than a binding rule, inherently limits their impact and therefore the willingness of central government to commit resources to them. Therefore public expectations of large volumes of guidance may have been unrealistic.
The best way to move forward is to accept the design constraints embedded in the RMA and work within them. The current proposed legislative changes will help.
Given central government cannot currently require specific outcomes other than standards on quite narrow issues, its options are firstly to accept the status quo, and secondly to change to a new regime that allows for central direction. This second option is risky, expensive and contrary to commitments to local decisions. There is a third option to provide guidance with acknowledged limits, backed up with capacity building. This is effectively the approach of the Government’s current proposals. The third option is the most attractive, and key to its success is a realistic appreciation by public and private stakeholders of both the extent and limits of central government’s scope to act.
New Zealand has a delegated environmental regulation regime.
This paper explores the perceived and actual role of central government in environmental regulation in New Zealand. It focuses on the provision of guidance to local decision-makers, and why this guidance has not materialised to the extent expected. Guidance here is taken to mean policy direction and standard setting from a national perspective, particularly through national policy statements (NPSs) and national environmental standards (NESs) but not limited to those tools.
New Zealand, through the Resource Management Act 1991 (RMA) and the Local Government Act 2002 (LGA), has a legislative framework that provides for integrated regulation (covering local planning and environmental effects) at regional and local levels. That planning was intended to occur within a context of guidance from central government that has largely not materialised.
Central government has not fully used its available tools for providing guidance – why?
This paper does not directly propose a different approach, for example complete local delegation of decisions, or the more centralised planning controls and funding arrangements seen in England. Rather the core question addressed is how existing guidance tools have been used and whether this is likely to or should change. This may lead to wider conclusions about the viability of the existing structure, but that question and the development of alternatives are a different debate.
Previous papers have explored general issues of delegation and devolution in the New Zealand context (Guerin 2002, Claridge and Kerr 1998, Kerr, Claridge and Milicich 1998). Those discussions are relevant to this more focused paper but are not reviewed in depth here.
The paper first outlines the current structures for environmental regulation at national and local government levels in New Zealand and the limited linkages between them. It then explores the reasons for making decisions at each level, the causes of tension between them, and the inherent constraints that the design approach of the RMA places on how central government can influence outcomes. This then allows some tentative conclusions about guidance options going forward.
Current central and local government stru ctures for environmental regulation#
“Environment” is a term that can be given very wide or narrow meaning depending on circumstances and intent. A narrow meaning can relate solely to ecological functions while wider meanings can include social or economic aspects.
The RMA is comprehensive and principle based.
The drafting of the RMA reflects this ambiguity. Part II specifies matters to be provided for, have regard to, or be taken into account. These matters cover everything from natural character to public access, Maori culture, historic heritage, efficient use and development, amenity values, trout and salmon habitat and the Treaty of Waitangi. Balancing between these matters is at the core of the application of the RMA.
In defining and achieving environmental outcomes under the RMA, both national and local government go through similar stages. Central government has more up-front discretion in defining outcomes through legislation. The stages are to firstly define the outcomes that are to be sought. The second stage is to determine the tools that will be used for the purpose and how they will be combined. The third stage specifies who will make the initial decisions on their application to specific circumstances. These are initial decisions since all such decisions are subject to court review. Decisions can be by elected persons (Ministers or councillors) or independents (hearing commissioners or boards of inquiry).
Objectives specified in Part II of the RMA, eg:
Sustainable Development Programme of Action:
Growth and Innovation Framework:
Legislation or regulations
NPS or NES
Water conservation orders
|Local||Growing Healthy Communities||
Plans (COP, LTCCP, RPS, RP)
Table 1 gives examples of outcomes that have been specified in New Zealand and the methods that legislation, mainly the RMA and LGA, has provided for achieving them.
The central government role#
Central government has to define what outcomes are significant and how they are to be achieved. This includes the division of responsibility between levels of government, and the extent to which the exercise of those responsibilities is empowered or constrained. Methods for achieving national outcomes can be direct, through prescriptive regulation or funding of projects, or indirect, through a local government process that can be guided by principles and supported by guidance or resourcing.
Central government can state its policy approach, and set standards on technical matters.
Outcomes are set at the national level primarily through legislation, primarily the RMA. These outcomes must then be interpreted by local government and the courts. There is some scope for elaboration of the legislative text. This can occur for the RMA through either NPSs setting out policy principles, or NESs, which are effectively regulations providing more prescriptive controls on narrower issues. Regional policy statements and plans must give effect to provisions in an NPS that affect the statement or plan, but the NPS does not override or replace Part II of the RMA. Part II controls what such statements or plans must contain and sets the primary criteria for decisions on individual applications for resource consents.
After more than a decade of experience with the Act, the system for setting and achieving environmental outcomes in New Zealand should presumably be settling down. This would also suggest a sufficient understanding of its strengths and weaknesses to have identified where guidance is required, and for central government to have provided it.
In practice central government has made little use of its tools.
However, the extent to which this settling down has occurred is variable. Considerable experience has been built up in operating the RMA, but that experience has also exposed gaps in available instruments. There is a debate underway on the extent to which those gaps represent fundamental flaws in a devolved system or are simply errors in implementation that can be fixed (eg, through additional national support and guidance). Experience has also highlighted issues such as a focus on prescriptive regulatory approaches and of course the influence of local politics.
That experience has also highlighted where issues of capability arise within local and central government in the Act’s application. This includes a simple lack of the scientific understanding and data needed to deliver on some outcomes, whether due to funding issues or simply the constraints of a small country.
Further, external pressures are forcing all levels of government to confront issues where national coordination or resourcing is necessary or where local and national interests can conflict. A prime example is freshwater. Shortages require allocation between uses, such as irrigation and electricity generation, which have differing local and national impacts. Meanwhile the increased run-off from more intensive agriculture is a factor in falling water quality affecting national health and environmental outcomes. At what point does the national interest justify overriding a local interest in weaker or stronger standards, or prioritising one outcome above others?
These questions are inherently difficult to answer, and highlight a reason why the level of guidance has not met public expectations. There has been only one NPS (on coastal policy) and there were no NESs until 2004.
The New Zealand Coastal Policy Statement (NZCPS) is mandatory, was gazetted in May 1994 and is currently under review. The context for the NZCPS is that the Minister of Conservation approves it, as well as regional coastal plans, restricted coastal activities (the Minister is the consent authority and has an appointee on the hearing committee), vesting of reclaimed land and tendering of space in the coastal marine area. This context gives the NZCPS a quite distinct status.
Anecdotal reasons for the lack of guidance may not tell the full story.
Anecdotal reasons for the lack of other guidance often cited include spending priorities within central government, the fact that the “call-in” mechanism changes the decision-maker but not the decision framework, and the cumbersome nature of the NPS mechanism. It is also difficult to define such guidance in the absence of experience with application of the Act (some plans are only just coming into force) and a reluctance to constrain local communities. For all these reasons, and probably others, the extent of central government involvement in environmental policy is a very live issue.
The current proposed changes to the RMA (see Table 2) will increase the Government’s options but do not themselves help with decisions on when and how those options should be taken up. Later sections of this paper explore whether there are deeper reasons for the lack of such action to date and if so what it means for the future.
The board of inquiry now required for each proposed policy statement will be optional. Instead, the Minister could engage in consultation.
It will be possible to specify that certain provisions can be included in council planning documents without the need for normal local planning processes.
NESs in some cases will be able to apply absolutely (without scope for council variation), or to require councils to show that it is necessary if they want to set standards that are stricter.
Call-in powers for the Minister for the Environment will be extended to applications for private plan changes, designations and heritage orders (ie, not just resource consents). The board of inquiry hearing the consent application will be required to have certain skills and will be selected from a standing body of commissioners. Appeals will be allowed only on points of law.
A ”menu” of alternatives to call-in will include:
- For example, the RMA was not really designed to handle allocation of natural resources held in common, but rather focused on management of activities on privately owned land, and did not compel councils to cooperate on issues with cross-boundary impacts.
- Call-in allows the Minister to establish a board of inquiry to consider a resource consent application.
The local government role#
It is important to review briefly the framework in which local government carries out its share of the tasks under the RMA, and the structures for managing central and local government interactions.
Local councils are bound by multiple Acts and community preferences.
Whatever happens at the central level, once that is in place, councils must work through their own processes at the local level as defined by the RMA and LGA. This includes determining what they are aiming to achieve, both in terms of implementing the national requirements and under their own responsibilities on behalf of their communities.
The RMA provides for regional and district plans and regional policy statements, which relate to achieving the objectives of the RMA itself. This addresses most environmental outcome-setting issues for local government. It still leaves local government the more operational task (not addressed in this paper) of processing specific consent applications.
Councils are established, however, under the LGA and draw a number of powers and responsibilities from that Act, as well as specific provisions of many other Acts. This Act requires councils to undertake Community Outcomes Processes to identify social, economic, environmental and cultural outcomes for community wellbeing. These outcomes should then flow into LTCCPs by July 2006 covering council actions, links with others and the monitoring and measurement of progress. Managing the interfaces between these two regimes complicates the task for councils, which also of course must perform many non-environmental functions under other specific legislation.
In performing these tasks, and their other functions, councils interface with central government to identify where national approaches are binding or discretionary, obtain relevant guidance and draw on resources and information as available.
Although central government defines the structures of local government and prescribes many of its functions, there are no formal links between national and local government policy processes. This can make it difficult for local government to know what its obligations are and where information can be found.
Links between central and local government are loose.
However, a number of mechanisms have developed to manage the inevitable linkages. These include at the political level a series of forums between Ministers and senior local politicians. There are also direct service provision arrangements between national and local government eg, on food safety, and increasing involvement of local government representatives in policy development processes, although this is case specific and not mandated.
Another approach has been to establish networks of local stakeholders (including councils) and central government agencies to address regional issues. Attempts in this direction have been made, particularly in the social policy area, but success is variable. Key constraints include limited central government resources to engage with multiple processes while delivering against other outcomes, and differing timetables at local and national levels. There can also be conflicts between accountability and flexibility in national agencies.
The alternative of designing one set of rules is generally faster and has a lower cost. However, that design process does not have to be directive. It can involve consultation, facilitate information exchange and allow for joint development by central and local government.
Reducing differences can involve standard rules, or a standard approach to making rules.
One option that may help is to create a standard set of principles to apply to the development of central government policy that affects local government. This option has been raised by local government. The purpose would be to increase the extent to which policy takes local concerns and implementation issues into account. This issue may be further developed through the forums noted above.
- Such standard principles should be able to be linked directly to the existing regulatory impact statement (RIS) regime, which already outlines a model policy development process (Guerin 2003). The RIS requirement is also linked to the Cabinet Office manual, which sets out consultation requirements and is binding on central government agencies but has no legislative status. The value of such a set of principles depends on its own content and on the effectiveness of enforcement mechanisms within central government itself.
Problems and responses#
The preceding discussion set out briefly the framework within which central and local governments interact in general. Within the specific circumstances of New Zealand’s environmental regulatory regime there is, in theory at least, considerable scope for decisions to be made at either level. The extent to which central government becomes directly involved depends on the specific issues and circumstances in question, political factors, and the inherent design principles of the RMA.
Who should make decisions?#
Few general principles exist for central vs local government actions.
There is little if any explicit guidance for decisions within central government on when it is appropriate to act directly, versus leaving an issue to local action, and if the latter whether to then constrain, guide or provide support for that local action. Influences can include the legislative powers relevant to the issues (eg, whether central government can require specific actions, set decision-making principles, or simply offer guidance). Other influences include the nature of the information needed to make decisions, the degree to which impacts are location specific, and the capability of the levels of government.
An important factor is balancing knowledge versus independence. Local decision-makers are likely to have a better understanding of local circumstances and community views. However, they are more subject to pressure from affected parties and likely to place greater weight on local rather than national interests. National or independent decision-makers trade off less awareness of local conditions for greater independence, but the process for their appointment or the referral of the decision to them can itself colour perceptions of the outcome. Those perceptions can be as significant as the decision itself in affecting community responses to specific decisions and the overall regime. Uncertainty around the right approach may lead to a desire for diversity in approaches between councils as an experimental means of determining the best regulatory option.
General rules about who is best placed to exercise powers, and how best to match powers and accountabilities, are difficult to apply in specific cases.
Governance structures and accountability mechanisms are relevant to the allocation of decision-making powers and vice versa. For example, local government responds to a narrower community, which may lead to different priorities. On the resourcing front the provision of funding or other support from central government inevitably creates pressure on local government to comply with national wishes. Failure to provide such central support can lead to concerns about “unfunded mandates” imposed on local communities.
The degree of national interest in an issue and its relevance to competition between local governments can affect roles (Gerber and Taske 2000). Delegation of powers can also be used as an indirect means of limiting restrictive regulation or subsidies by placing the responsibility for such measures on a level of government that is constrained in its resources and facing competition from comparable jurisdictions. For example, local government subsidies to State firms in China are constrained by limited local resources and pressure to use those resources to attract foreign investment (Qian and Weingast 1997). A converse view is that “Interest groups pragmatically desire the regulatory level whose outcome they like best and the relation of group strengths on the … levels then determines the preferred levels of regulation” (Noam 1982, p279). The impact can differ widely, with one company preferring local regulation because of contacts or a resourcing advantage over a local council, and another preferring national intervention to override locally centred objections.
The above standard arguments affect more the overall design of a legislative regime such as the RMA, than specific decisions within it. More relevant to the RMA, and the specific issue of when central government should provide guidance, is the problem of differences between national and local interests or priorities, or regional variations in values.
In practice the debate focuses around differences between national and local interests.
Common reasons for such differences include a particular concentration of one cultural or sectoral perspective in an area, or that the impacts of a choice differ by area. For example, there may be a national interest in preserving bush environments or reducing pollution, but a local interest in jobs. Conversely, the national interest may call for transmission or road infrastructure that impinges on locally important scenic or historic features, or creating jobs nationally through hydro electricity generation may prevent job creation locally through irrigation.
In all of these situations a mechanism is needed to resolve such differences. In the case of the RMA it is generally lacking as the legislation does not explicitly address the national interest (although it can be read into Part II, particularly section 5) and little national guidance has been provided.
- An example is the management of natural disasters, which can often be regarded as an insurance issue in terms of a trade-off between preventive measures and building up reserves against an event. The magnitude and geographical spread of impacts can determine responses and there are clear legislative powers for both central and local government.
- The application of theoretical constructs such as principal-agent theory to such situations is not straightforward as local authorities typically answer to multiple principals eg, central government, regional councils (for district or city councils) and local voters; while central government is itself an agent of voters nationwide. “A simple dyadic model does not capture the dynamic interactions that occur between principals and agents” including “the interactions of various types of principals and agents” which may have similar characteristics (Waterman 1998, p36). “Agents can side with principals that most closely represent their policy perspectives or they can play one principal off against another” (Waterman 1998, p18). All these possibilities exist in the application of the RMA in New Zealand.
Working within the RMA framework – what is possible?#
The primary tools available to central government for managing environmental policy decisions in New Zealand are in the RMA. These range from the general to the specific and can relate to specific decisions or general parameters.
Central government guidance is limited in part because the RMA purposely limits central control.
Those tools and their scope for changing outcomes are constrained by the fundamental design features of the RMA. The Act has a clear (though overlapping) set of principles and desired outcomes. It requires an integrated approach to planning and consent approval, avoiding ad-hoc or partial decisions and covering local planning issues as well as the environmental effects of an activity. Finally it delegates most decisions to local government and requires independence of decision-making where there is potential for a conflict of interest.
As a consequence of these features, the New Zealand system does not generally provide any scope at local or national level for overriding standard processes to either permit or prohibit a specific activity. Central government can “call-in” a decision but this only moves the decision to an independent body, rather than changing the criteria for the decision.
All decisions are subject to the criteria in the RMA and reviewable by the courts against those criteria. For example, independent commissioners recently decided against a street race in Auckland that was supported by the council. It could be argued that New Zealand has chosen to trade off the inability to make such specific interventions for the longer-term benefits of an integrated consensus on planning outcomes. Conversely the argument could be made that the outcome correctly reflected local preferences.
This of course assumes that such a consensus can be reached and that it has such benefits. The particular example cited may also simply reflect local variations rather than a fundamental problem – such a race does not need a resource consent in Wellington because the plan there already provides for such an event. In the Auckland case the commissioners had to apply the general provisions of the Act to a specific application within a plan that did not contain such a provision. This may have been deliberate or simply an oversight.
Central government’s options are limited to prescribing general outcomes or issuing decision-making guidance. .
A fundamental consequence of the lack of an intervention power in specific cases, however, is that central government can only intervene by either prescribing outcomes or refining the criteria to guide decision-making.
The ability to prescribe outcomes is also inherently limited because of the costs of constraining activities through ex-ante evaluations, the information requirements of doing so and the practical limits of such techniques given variations in real world circumstances. A further problem is that prescribing outcomes in a lasting manner requires a degree of national or social consensus on what is to be achieved and how to do so. The lack of such a consensus is inherent in Part II of the RMA, which provides for the balancing of so many different outcomes. This is both easier and harder at the local level, in both cases because of being closer to the action.
The role of guidance#
This leaves central government relying in most situations on issuing guidance. However, its ability to provide guidance in turn is limited (although to a lesser extent) by the same problems as the prescriptive solution. The development of guidance must therefore reflect the reasons why the RMA is structured as it is.
Guidance cannot ensure particular decisions are made. It can only be successfully developed where the issue in question is susceptible to clarification, and there is sufficient consensus on the nature of the problem and possible solutions. Even then there is a balance of risks around changing the status quo where the new approach must be applied by so many different parties at separate levels of government and then interpreted by the courts, with a substantial time lag before the results are apparent. A prudent central government will not rush to such solutions without a convincing case for net gains.
The above discussion does not completely exclude the potential for clarification of the RMA’s application by central government, but may help to explain why such guidance has been, and is likely to remain, relatively limited (beyond the basic issue of funding).
Guidance has its role but its limitations need to be recognised and expectations managed.
Ultimately, the issue may come down to how you make outcome-based (as opposed to prescriptive) regulation work, solving the problem that “regulations are born principled but die detailed” because of a lack of trust in how the regulation will be interpreted. There is a role for guidance but it can only achieve so much. Beyond that point, the answer might be not to “focus on trying to improve the clarity or precision of the rule or principle itself” but “building the capability of the community that has to interpret and apply it” ie, improve councils’ capability and access to information (Skidmore, Chapman and Miller 2003, p81-82). In reality a combined approach of better guidance and more capacity building may be the best available within the fundamental RMA framework. The current proposals for “improving the RMA” are consistent with such an approach.
- This refers to the current framework and that which would apply if the current amendment proposals go through. Those amendments aim to reduce some of the problems that have limited central guidance (eg, cumbersome procedures or the inability to require joint council processes) but do not change the fundamental constraints on central government’s role.
- This applies whether the intervention is through statute or regulation (including an NES), or by using NPSs to amend regional policies or plans.
The RMA requires the balancing of a number of factors and a range of impacts.
The current legislative framework for defining and achieving environmental outcomes in New Zealand is an integrated one that forces the balancing of multiple outcomes. That requirement means that decisions must be based on greater information and incorporate more explicit trade-offs than might otherwise occur.
The recent changes to the RMA recognise some of these issues. Changes have included emphasis on increasing the capability of local government, providing more options for central government involvement, and committing to giving more guidance on policies. However, the changes do not directly address questions such as how conflicts between national and local interests should be resolved, or whether changes are required in when and how central government becomes involved. This may reflect a deliberate decision not to question the underlying manner in which the design of the RMA constrains what central government can do.
Delegated principle-based decisions limit central control.
The RMA is designed on the presumption that decisions should balance multiple outcomes, be as integrated as possible, be made at the local level wherever possible and not be ad-hoc. It inherently and deliberately limits the nature of national guidance or control. This approach may not be practical for resource issues with non-local impacts, for example aspects of climate change policy or oceans policy. The limit on guidance reduces the incentive for its preparation. However, the existence of that limit and the rationale for it are not widely understood. This can result in frustration locally when guidance does not materialise and centrally when it cannot achieve what is desired.
Moving forward, the current reforms should reduce some of the problems that have limited central guidance, for example cumbersome procedures or the inability to require joint council processes. The reforms will not change the fundamental constraints on central government’s role. They will give central government some additional options, and provide for increased capacity building in central and local government as well as wider education on the RMA, but not alter the limits on what guidance can achieve.
This leaves three paths open. One is to maintain the status quo legislatively and on central government involvement. This risks undermining the credibility of the current regime through increased frustration at the lack of guidance and foregoing any opportunities that may well exist for better guidance within the existing limits.
A second path is to replace the Act with a regime that provides for more direct intervention. This would be contrary to Government commitments to local decision-making and it would be a radical change of approach. It could have been put in place in 1991 if desired. Doing so now would require another decade of expensive adjustment and change without any guarantees of improved outcomes overall. Replacing the Act should be deferred until clearly necessary as it is an option that can be taken up in future.
The third path is to move ahead, identify what guidance can usefully be provided and actually do so. Such an approach requires undertaking the basic debates necessary to define strictly what the guidance is seeking to achieve, and then rigorously testing whether those outcomes can and should be addressed at a national level. This may limit guidance primarily to broad policy statements of the national interest and technical standards (backed up by information provision), since central government would probably be reluctant to express firm preferences on specific outcomes that could be set aside by councils, but that is presumably better than no guidance at all.
Once its limits are recognised, central guidance in conjunction with capacity-building remains a key component in the overall concept of the RMA.
This third path recognises that the RMA restricts the nature of central government guidance for valid reasons, such as the difficulty of prescribing in advance how multiple outcomes should be balanced in particular cases. If central government wishes to prescribe specific outcomes in specific circumstances, the current RMA is probably not the Act it wants.
However, the RMA represents a major investment by the Government, business and the community. The conflicts that are now highlighted after more than a decade of its application are fundamental issues for society rather than by-products of the Act itself. Getting the right solutions to make the RMA work effectively will be far more important than doing so quickly. Making another effort to make this system work effectively, as signalled by the recent review, appears preferable to the cost of what may be unnecessary, and would certainly be major, changes to a system now so embedded in economic and social frameworks.
Claridge, Megan and Suzi Kerr (1998) "A case study in devolution: The problem of preserving kiwi habitat in the far north." Wellington, New Zealand Treasury, Working Paper No 98/7a. http://www.treasury.govt.nz/workingpapers/1998/twp98-7-a.pdf>
Gerber, Brian and Paul Teske (2000) "Regulatory policymaking in the American states: A review of theories and evidence." Political Research Quarterly 53(4): 840-886.
Guerin, Kevin (2002) "Subsidiarity: Implications for New Zealand." Wellington, New Zealand Treasury, Working Paper No 02/03. >
Guerin, Kevin (2003) "Encouraging quality regulation: Theories and tools." Wellington, New Zealand Treasury, Working Paper No 03/24. http://www.treasury.govt.nz/workingpapers/2003/03-24.asp>
Kerr, Suzi, Megan Claridge and Dominic Milicich (1998) "Devolution and the New Zealand Resource Management Act" Wellington, New Zealand Treasury, Working Paper No 98/7. http://www.treasury.govt.nz/workingpapers/1998/twp98-7.pdf>
Noam, Eli (1982) "The choice of governmental level in regulation." Kyklos 35: 278-291.
Qian, Yingyi and Brian Weingast (1997) "Federalism as a commitment to preserving market incentives." Journal of Economic Perspectives 11(4): 83-92.
Skidmore, Paul , Jake Chapman and Paul Miller (2003) "The long game: How regulators and companies can both win." London, Demos.
Waterman, Richard Rouse, Amelia Wright, Robert (1998) "The venues of influence: A new theory of political control of the bureaucracy." Journal of Public Administration Research and Theory 8(1): 13-38.