CONSTITUTIONAL TRADITIONS AS BOUNDARIES IN STANDARDISING ADMINISTRATIVE RULEMAKING THROUGH TRADE AGREEMENTS

Abstract Pioneered by the US, recent mega-regional trade agreements such as the CPTPP have incorporated ‘regulatory coherence’ provisions—mirroring the US Administrative Procedural Act's core designs—to balance between domestic regulatory autonomy and international cooperation. Building upon existing literature that traces the trajectories of the diffusion of regulatory coherence across jurisdictions, this article analyses how Australia's constitutional tradition could effectively condition the development of regulatory coherence in a Westminster-based model of governance. It is argued that the global entrenchment of regulatory coherence is contingent upon the inherent boundary defined by the political dynamics and constitutional structures within a jurisdiction.


I. INTRODUCTION
The implications of the rules of the World Trade Organization (WTO) for national sovereignty and policy space have been a recurring debate among policymakers, practitioners and academics. According to conventional wisdom, international trade law goes beyond 'trade' as such and hinges on a matrix of 'non-trade' concerns, including public health, environmental protection, intellectual property, human rights, cultural traditions, animal welfare, and national security. However penetrating these impacts are, they seem more or less limited to specific issues, rather than having a cross-cutting effect on a nation's legal system as a whole. While the WTO has had an incremental effect on Members' domestic laws in various discrete areas, the political and economic momentum in the context of the emerging 'mega-regional' negotiations has propelled more innovative (and arguably, more intrusive) institutional designs. These new-generation agreements seem to go beyond the traditional concerns of free trade agreements by introducing what is known as 'regulatory coherence' to impose a common set of procedural safeguards that could arguably overhaul a nation's entire regulatory system.
One example is the Trans-Pacific Partnership (TPP) Agreement-now rebranded as the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP). Under US leadership, the TPP negotiators agreed on a stand-alone chapter concerning regulatory coherence. 1 In many ways, the chapter mirrors administrative rulemaking in the US by including, amongst other things, notice and comment, public consultation, inter-agency coordination, regulatory impact assessment (RIA), cost-benefit analysis, and judicial as well as administrative review. 2 For instance, '[r]ecognising that differences in the Parties' institutional, social, cultural, legal and developmental circumstances may result in specific regulatory approaches', Article 25.5.2 of CPTPP requires the Parties to follow the steps when it comes to regulatory impact assessments: (a) assess the need for a regulatory proposal, including a description of the nature and significance of the problem; (b) examine feasible alternatives, including, to the extent feasible and consistent with laws and regulations, their costs and benefits, such as risks involved as well as distributive impacts, recognising that some costs and benefits are difficult to quantify and monetise; (c) explain the grounds for concluding that the selected alternative achieves the policy objectives in an efficient manner, including, if appropriate, reference to the costs and benefits and the potential for managing risks; and 1 Trans-Pacific Partnership Agreement (signed 4 February 2016). On 11 November 2017, the 11 remaining parties replaced the TPP with the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (signed 8 March 2018, effective on 30 December 2018). Chapter 25 on regulatory coherence remains intact and thus will not affect our analysis. For the purpose of this article, we use the term TPP and CPTPP interchangeably. The TPP was negotiated under the Obama administration and never ratified by Congress, as former President Trump withdrew the US from it. For international agreements' impacts on the US domestic legal order, including ratification, self-executing and non-executing treaties, see S Mulligan, 'International Law and Agreements: Their Effect upon US Law' Congressional Research Services (2018). Australia ratified CPTPP on 31 October 2018. As a general rule, Australia's treaty-making process involves six steps: (i) the lead minister to seek a mandate to negotiate the treaty; (ii) negotiations and finalisation of text; (iii) all treaty actions must be approved by the Federal Executive Council (ExCo) and the Prime Minister must also be informed of the matter; (iv) once ExCo approves, the treaty may be signed by Australia's representative; (v) following the signature, treaties are tabled in both Houses of Parliament for consideration by the Joint Standing Committee on Treaties (JSCOT); and (vi) finally, once all domestic procedures have been completed, agreements can be made for entry into force. See generally Australia Government, Department of Foreign Affairs and Trade, 'Australia's Treaty-Making Process' <https://www.dfat.gov.au/international-relations/treaties/ treaty-making-process>. For a recent account of regional integration from an Asian perspective see P Hsieh, New Asian Regionalism in International Economic Law (Cambridge University Press 2021). 2 See art 25.5 of CPTPP.
regulations of general application will be consistent through common procedural norms agreed upon at a mega-regional level. In this light, the CPTPP is not a traditional free trade agreement (FTA) but a new species that constitutes a meta-norm marker reconfiguring the infrastructure of administrative rulemaking systems-at least for the Asia-Pacific region and potentially beyond. Despite the withdrawal of the US from the TPP after President Trump took office in early 2017, Chapter 25 remained unchanged when the rest of the 11 Pacific Rim countries decided to proceed without the US. Beyond the TPP/ CPTPP, trade pacts spearheaded by the European Union (EU) have, in some ways, accepted the same approach. The EU's negotiating proposal for the Transatlantic Trade and Investment Partnership (TTIP) and the recently finalised text of the EU-Japan Economic Partnership Agreement (EU-Japan EPA), for instance, both have included provisions on good regulatory practices. 7 The overarching nature of regulatory coherence will arguably have an even deeper impact in a domestic constitutional sense. The intellectual argument underpinning regulatory coherence goes beyond 'rationality' from a purely economic perspective and hinges upon accountability, legitimacy, and democracy. This cuts two ways. First, authoritarian regimes that are wary of its democratisation ramifications have been reluctant to embrace regulatory coherence in the context of international trade negotiation. The Chinese strategic approach to regulatory coherence in its FTAs is a salient example. 8 The second, and the more interesting point is its implications for administrative law systems even in democratic countries. For democracies in the developing world, the lack of capacity remains an institutional hurdle to the full implementation of elements of regulatory coherence (eg, public consultation, RIA), which require considerable technical, financial, and administrative resources. For developed countries, this can be equally challenging when treaty obligations on regulatory coherence may not fit squarely into a country's existing administrative law and, at times, constitutional traditions.
This article explores this latter aspect, with a specific focus on the Australian context. For decades, the US has been promoting regulatory coherence (or, good regulatory practices) through informal arrangements, soft law, and trans-governmental networks, including the Organization for Economic Cooperation and Development (OECD) and the Asia-Pacific Economic 7 For a survey of regulatory coherence related clauses in the preferential trade agreements

892
International and Comparative Law Quarterly Cooperation (APEC). 9 For instance, it is reported that the US was one of the two members that already embraced impact assessments after the OECD issued the Recommendation on Improving the Quality of Government Regulationfollowed by the 1997 OECD Policy Recommendations on Regulatory Reform. 10 Years on, the notion of regulatory coherence is nothing new for many Western countries. Be that as it may, there were concerns when the EU included a legally binding chapter on regulatory coherence in its recent mega-regional trade agreements. 11 Given the diverging economic, legal, and political systems between the EU and the US, it is no surprise that importing regulatory coherence-a notion with its root in the US Administrative Procedure Act (APA)-is regarded as somewhat problematic on the other side of the Atlantic. Commenting on the chapter on regulatory coherence under TTIP, Anne Meuwese, for instance, suggested that if good regulatory principles and practices were included in the final text of the EU and US trade deal, 'this would be the most formal endorsement of these OECD principles by the EU and the clearest step toward codification of better regulation principles and tools'. 12 However, these recommendations indicate that 'net maximization of benefits' is a principle of good regulation-which in many cases is 'at odds with the precautionary principle' and 'other objectives for regulation prioritized at a constitutional level in the EU'. 13 Yet this does not necessarily mean that transplanting this same idea even to jurisdictions with a similar legal culture is trouble free. Take Australia, the focus of this article. Although both the US and Australia are former British territories with established common law traditions, Australia follows the UK's Westminster model that manages administrative rulemaking in a rather different fashion to the US. Australia's divergent path complicates both the way and the extent to which it implements the core elements of regulatory coherence-which are the inherent limits for global normative development of good regulatory practices. These elements might not be observed in 9 See eg T Bollyky, 'Regulatory Coherence in the TPP Talks' in C Lim et al (eds), The Trans-Pacific Partnership: A Quest for a Twenty-First Century Trade Agreement (Cambridge University Press 2012) 171, 176-8. Admittedly, there is no direct evidence showing that it is the US that took the lead in including the regulatory coherence concept in the OECD. Some may consider the OECD as an instrument in the diffusion of policy-and there is no exception for tools like RIA. Others may relate the diffusion of good regulatory practice to the fact that the explosive growth of regulatory agencies came hand in hand with privatisation in the 1980s and onwards. While these observations are all well made, the evidence indicates that many of the elements of regulatory coherence have already been adopted in the US-and such experience has been transferred to other nations such as Canada, long before the OECD issued its 1995 recommendation. It suffices to say at least that the US was a key player in shaping good regulatory practice and sharing its experience with others-including developed countries in terms of these practices. See  Australia at the same level as was contemplated by US trade negotiators when they attempted to embed key features of its APA in the FTAs. To explore and unpack the hidden limits to regulatory coherence as an emerging global norm, this article contrasts the US and Australian administrative rulemaking processes from a historical perspective. It follows Andrew Edgar by distinguishing the two systems as reflecting respectively 'deliberative democracy' and 'parliamentary democracy' in relation to the rulemaking process and argues that different traditions constitute an inherent limitation to full implementation of regulatory coherence in the Australian context. The remainder of this article proceeds as follows.
Section II traces the development of the APA, a compromise between competing interests and ideologies at that time. Deliberative democracy as embedded in the APA represented a response not only to concerns about arbitrary administrative powers but also the increasing demands of accountability. Australia did not take such a path. Although the Australian constitution features certain American flavours such as federalism, it follows the Westminster tradition and subjects administrative rulemaking almost exclusively to Parliamentary control. As Section III reveals, there was a sea change from the 1980s onwards, when Australia began to reorient itself towards the US-style rulemaking process at both the Commonwealth and the state levels. Many key elements of regulatory coherence seen today, such as public consultation and sunset review, have been incorporated into the Australian administrative rulemaking process. These changes predated the recent initiatives seen in many mega-regional trade agreements. In addition to the pressure from trading partners under the banner of neoliberalism, these changes also responded to the insufficiency of Parliamentary oversight as a controlling power given the explosive growth in scope and quantity of delegated rulemaking. Notwithstanding these shifting sands, Australian administrative rulemaking has remained aligned with the Westminster tradition rather than adopting a US-style approach.
There is a growing body of literature on the emergence of regulatory coherence in the context of mega-regionalism. However, its normative compatibility with different domestic constitutional frameworks is yet to be understood. Such a framework determines the basic contours of the separation of powers and thus the way each jurisdiction holds administrative organs accountable. As promising as regulatory coherence is, this notion has its root in the US and may not fit squarely into other jurisdictions. This holds true not only for authoritarian countries like China, which has been examined elsewhere, but also for other liberal democracies, including Australia. Tracing the historical trajectory of its constitutional paths permits reflection upon the US's penetrating power along with globalisation and international trade, and of course, their inherent boundaries.
Three caveats are in order. First, the global normative diffusion of regulatory coherence through institutions such as the OECD, APEC, and TPP/CPTPP has 894 International and Comparative Law Quarterly been documented elsewhere and will not be repeated. 14 Rather, this article focuses on how internal dynamics impact upon the importation of regulatory coherence by examining the role of the Westminster tradition in shaping the Australian administrative rulemaking processes. While the notion of regulatory coherence has been diffused through non-legally binding commitments, this does not mean that soft law approaches do not have an effect on a nation's regulatory environment. 15 A broader point-and the main message of this article-that it is the divergent social, economic, and political conditions of each country that makes it appropriate to take a 'softer' approach to bake these tools into trade agreements. How far, as a matter of practice, these tools can be implemented in each country hinges on local conditions-as in the case of Australia.  17 In Australia before federation, the Privy Council on several occasions held that the colonial legislatures could delegate their legislative powers and that the doctrine of delegates non potent delegare did not apply. 18 Similar positions were taken by the High Court after federation. Both the US and Australia subject delegated legislation to judicial review as a controlling mechanism. What distinguishes these two common law jurisdictions in this regard lies in the presence of additional safeguards to ensure deliberative legitimacy in the administrative rulemaking process, which turns on the different constitutional paths taken: deliberative democracy and the Westminster tradition.

A. The US Deliberative Democracy Model
Historically, the administrative rulemaking process in the US developed in a piecemeal fashion. There were some early attempts to deter the growth of administrative rulemaking through the non-delegation doctrine in the 1930s, and Congress managed to maintain certain procedures for particular agencies and acts. 19 It was not until the passage of the APA in 1946 that the US had a more unified body of administrative law. 20 At one level, the advent of the APA represented a critical bipartisan compromise to rebalance the powers between legislature and executive. 21 For a long period before the New Deal, administrative rulemaking in the US featured what Richard Stewart described as a 'transmission belt model', seeing agencies as merely implementing policies authorised by Congress. 22 Although agencies engaged in certain quasilegislative activity, it was considered an exercise of discretion necessary to carry out Congress's statutory mandate. Such discretion was not a delegation,  International and Comparative Law Quarterly but a necessary action taken only to 'bring […] to fruition Congress's political choice' without offering 'some competing conception of what is best'. 23 Hence, there should be no legitimacy deficit concerns provided such discretion did not involve 'significant exercises of policy discretion', 24 and Congressional oversight accompanied with judicial scrutiny would safeguard individuals against the abuse of power and reduce arbitrariness by the executive to a minimum. Although this transmission belt account seemed to accord with the nondelegation requirement and echoed the contractarian theory by Hobbes and Locke-'consent is the only legitimate basis for the exercise of the coercive power of government' 25 -it failed to capture the dynamic American politics in the 1930s. At the time, in order to stimulate the economy in response to the Great Depression, many legislatures passed little, if any, guidance to control the exercise of administrative discretion and therefore no longer offered an effective safeguard to individual liberty. 26 The New Deal supporters who believed science and economics would save the nation from persistent economic downturn therefore turned to the so-called 'expertise model' to address legitimacy concerns. 27 In their view, these agencies are essentially 'politically dis-interested entities comprised of professionals' driven by their 'professional knowledge and training'. 28 Thus, provided the exercise of regulatory authority via the rulemaking process did not hinge upon 'subjective value judgments that belong in the political realm', 29 their exercise of discretion should not be controversial. The fact that expertise centred on competence, rather than procedure, attracted harsh criticism. 30 It was these concerns that in part gave birth to the APA, a compromise between the traditional and expertise models. . 27 ibid 472-3 (observing that the expertise model also 'achieved distance from politics by creating a realm of administrative decision-making in which insulation from politics was both explicable (ie, political judgment simply was not implicated) and justifiable (ie, political judgment would only serve to disrupt technocratic judgment)'. 28 Seidenfeld, 'The Role of Politics' (n 24) 1404. 29 ibid. 30 Schultz Bressman, 'Beyond Accountability' (n 17) 472 (who noted that these agencies 'too often decided matters without a hearing, on the basis of matters outside their purview, with preformed biased, and without regard to the combination of functions that might impugn their impartiality'). When the court was asked to address these procedural issues, however, they often refrained from intervening in a way that restricted an agency's discretion, tending rather to grant the agencies freedom in choosing their procedures. See  The APA broadly described the term 'rule' as taking the form of a quasilegislative act of general application 32 and building on the legislative model, set out basic procedural principles for a rulemaking process that required: General notice of proposed rulemaking shall be published in the Federal Register … shall include (1) a statement of the time, place, and nature of public rulemaking proceedings; (2) reference to the legal authority under which the rule is proposed; and (3) either the terms or substance of the proposed rule or a description of the subjects and issues involved. 33 To make sense of these transparency mandates, the APA went on to require: After notice required by this section, the agency shall give interested persons an opportunity to participate in the rule making through submission of written data, views, or arguments with or without opportunity for oral presentation. After consideration of the relevant matter presented, the agency shall incorporate in the rules adopted a concise general statement of their basis and purpose. 34 Notwithstanding this development, the US rulemaking process continued to evolve with the changing political ideology of the post-APA era. By the 1970s, the expertise account had lost traction, as commentators began to reflect upon not only the ability of regulatory agencies to serve the public interest but also, more fundamentally, 'the very existence of an ascertainable "national welfare" as a meaningful guide to administrative decision'. 35 The entire focus has since moved beyond preventing the arbitrariness of administrative powers to focus on addressing political accountability. 36 The main concern was around the 'affirmative side' of government-that '''has to do with the representation of individuals and interests" and the development of governmental policies on their behalf'. 37 In line with the emerging 'majoritarian paradigm' of that time, 38 the prevailing account thus shifted to the 'interest group representation model'-through which the decisions of 'agencies' would garner legitimacy 'based on the same principle as legislation'. 39 On this approach, the legitimacy of administrative agencies is 32 Administrative Procedure Act section 551(4) (US). 33 ibid section 553(b). Broadly, there are two types of rulemaking processes under the APA: formal and informal. In cases where 'rules are required by statute to be made on the record after opportunity for an agency hearing', administrative agencies must follow the formal rulemaking process, which involves a courtroom-type hearing. In most other cases, agencies promulgate rules through the informal process governed by section 553 of the APA. Unless otherwise specified, the term 'rulemaking' throughout refers to the informal one. On formal rulemaking process, see ibid, section 556-557. 34

898
International and Comparative Law Quarterly effectively grounded on a more inclusive and deliberative decision-making process. Yet such an account was hardly immune from criticism. There was scepticism about the self-interest of administrative agencies, 40 which could from time to time become 'entangled with private interests' 41 resulting in problems of regulatory capture. According to Schultz Bressman, rejecting the technocracy and representation models required a more active role for the President-described as the 'presidential control model'. 42 During the Nixon, Ford, and Carter administrations, there were various initiatives working towards this goal by reshaping administrative rulemaking. 43 An important development occurred in the 1980s when the Reagan administration issued Executive Orders 12,291 and 12,498. To implement its deregulation agenda, part of President Reagan's election campaign of small government, these Executive Orders imposed vigorous oversight of regulatory activity. 44 The former required agencies to take into account relevant cost and benefits, and more importantly, that major proposed rules must be accompanied with a regulatory impact analysis for the review of the Office of Management and Budget (OMB). 45 The latter went one step further 40  by creating a 'regulatory planning process' which provided an early vantage point for the OMB to shape and coordinate the regulatory agencies' agenda for the next year. 46 These two instruments were subsequently replaced by Executive Order 12,866 issued by President Clinton. Executive Order 12,866 featured the basic structure to supervise administrative rulemaking, while addressing in particular the interagency coordination and compatibility via the Office of Information and Regulatory Affairs (OIRA) within the OMB. 47 Moreover, Executive Order 12,866 imposed on agencies the duties to conduct periodic review of existing significant regulations. 48 Together, these efforts served to ensure the legality and rationality of any proposed regulation, which was then subject to judicial review based upon the 'arbitrary and capricious' standard as specified under the APA. 49 This contrasts with the pre-APA era, when an 'agency needed no evidence, no record, and no statement of reasons to support a rule; rules were rarely challenged; and challenges were rarely successful' as the result of the deferential standard of review. 50 Since the Supreme Court's ruling in Motor Vehicle Manufacturers Association of the United States, Inc. v State Farm Mutual Automobile Insurance Company in 1983, 51 courts have been more willing to take a rigorous approach by applying 'hard look review' to ensure that regulatory agencies have taken into account the relevant issues and supported their rules with evidence and reasons. In this light, courts have continued to play a pivotal role in deliberative-type rulemaking process.
As things stand, contemporary US administrative rulemaking has combined different models and ideologies over recent decades. The driving forces behind this deliberative-type rulemaking process were not only the concerns about arbitrary exercise of administrative powers, but also the source of accountability to legitimise agencies' decision-making processes in the constitutional order. 46

900
International and Comparative Law Quarterly

B. Australia's Westminster Tradition
The Australian rulemaking process reflects a different constitutional pathway. The Australian Constitution is an hybrid document built upon both the US and the UK models. 52 As Professor Harold H Bruff put it, the Australian Constitution shares many similarities with its US cousin in the aspects of its written form, federalism, and judicial review, among others. 53 This is so because in the drafting process, the Australian framers-notably, Andrew Inglis Clark, Henry Parkes, Samuel Walker Griffith, and William McMillan -followed the US Constitution closely through their personal experience and readings. 54 Despite the US Constitution's profound influence on the Australian government structure, the Australian Constitution was, according to Sir Owen Dixon, a former Justice of the High Court, essentially 'a redraft of the American Constitution of 1787 with modifications found suitable for the more characteristic British institutions and for Australian conditions'. 55 A key feature embodied in the US Constitution that Australia intentionally did not follow was the separation of legislative and executive powers of the government. 56 In contrast to the US President who occupies a separate branch of government, the Governor-General in Australia by convention acts upon the advice of the Prime Minister. Ministers are responsible to the Parliament. 57 This arrangement reflects a different path from that enshrined in Article 1 Section 6 of the US Constitution which provides that 'no Person holding any Office under the United States, shall be a Member of either House during his Continuance in Office'. 58 This divergence was in part, according to Bruff, the product of a historical accident. The Westminster system of responsible government was not yet established in the UK when the US Constitution was in the making. Given that the UK monarchy had from time to time attempted to project greater influence over the Parliament by offering lucrative positions to its members, 59 the founders of the US Constitution decided to avoid such a situation by keeping the executive and legislative branch separate. The rigid division of powers adopted in Article 1 Section 6 of the US Constitution was a direct response to protect legislative independence from such 'corruption'. 60 In contrast, by the time Australia shaped its Constitution, the UK had evolved to place the Parliament, rather than the Crown, at the centre of political control. 61 Taking into account the developments in the US and the UK, the drafters of the Australian Constitution carefully reflected upon and selected different models that best fitted its local context.
The fact that Australia consciously opted for the Westminster model was in part due to two major factors. First, the US Constitution grew out of the War of Independence and thus insisted on the complete separation of powers. Australia designed its Constitution with no desire to break the umbilical cord with the British. 62 The Australian Constitution is an enactment of the British parliament and a product of a long-term, peaceful transition that is, in Cane's term, 'still not complete'. 63 The drafting of the Australian Constitution was not pursuing independence from the UK, but rather, creating the necessary mechanics for 'fortifying defence capacity and smoothing trade relations' of the colonies. 64 Second, in light of their experience with unelected, autocratic Governors-General, the Australian Constitution-makers were opposed to conferring too much power on one person and rejected a presidential executive in the US. 65 Consequently, the Australian government structure followed the Westminster system which kept the Executive in check through what is known as 'responsible government'. 66 This British inheritance was reaffirmed by the High Court in the 1930s, which in turn, had an effect on Australia's approach to administrative rulemaking. In the early years after federation, debates on administrative rulemaking and the non-delegation doctrine resurfaced. This enabled the High Court to revisit the principle of separation of powers embodied in the Constitution and its relation to delegated legislation. In Victorian Stevedoring & General Contracting Co Pty Ltd & Meakes v. Dignan, a landmark case involving a challenge to broad delegated power in the Transport Worker Act 1928 (Cth), the High Court 60 H Bruff, Untrodden Ground: How Presidents Interpret the Constitution 15 (2015) ('When the Convention turned to the relationship of Congress to the executive, they exuded fear of what they called "corruption," a term with a special historic meaning. English kings had developed a technique that jeopardized legislative independence. They "corrupted" Parliament by granting lucrative offices to its members, in a successful effort to sway their loyalties and maximize power.') [hereinafter Bruff, Untrodden Ground]. 61

902
International and Comparative Law Quarterly interpreted this doctrine against the historical context of the drafting of the Australian Constitution. 67 Commenting on major constitutional grounds invoked therein-namely Sections 1, 68 61, 69 and 71 70 -Justice Owen Dixon spoke of the US influence on the Australian Constitution and held that 'these provisions both in substance and in arrangement, closely follow the American model upon which they were framed'. 71 While Justice Dixon took the view that these provisions 'logically or theoretically make the Parliament the exclusive repository of the legislative power of the Commonwealth', he nevertheless drew the attention to the Westminster tradition by concluding that Parliament's power to authorise subordinate legislation depends more on the history and usage of British legislation and theories of English law than juristic analysis. 72 Justice Evatt echoed this view by holding that the Commonwealth form of government is underpinned by the British system, where an Executive is responsible to Parliament unlike under the United States Constitution. 73 In his view, the principle of separation of powers should be read in the context of the close relationship between the legislative and executive branches of the Commonwealth. 74 The High Court in Dignan explicitly reaffirmed the Westminster tradition as the government structure chosen by the founders, despite certain resemblances to the US model. 75 Under this Westminster tradition, administrative rulemaking in Australia is generally subject to parliamentary control through notification in the government gazette, scrutiny by a parliamentary committee, and a potential disallowance process, among other methods. While these mechanisms ensured some degree of transparency, there was no general statutory requirement of public consultation as set out under the APA except in certain specific statutes. 76 Beginning in the 1980s, there were various reforms at the federal, state, and territory levels seemingly moving towards the US-style rulemaking process. Despite these developments, the Westminster tradition can be seen as a key structural factor when it comes to the implementation of deliberative democracy embedded in the US rulemaking model. While Australia had intentionally departed from the US model and retained control of the executive branch by the legislature under a Westminster framework, recent decades have witnessed a sea change that seems to move towards the US model of rulemaking under the 'administrative state' rationale. Such a trend towards convergence comes from two sets of driving forces-external harmonisation (or more bluntly, Americanisation), pressure and internal governance demands. Externally, the US has been exporting its APA-style administrative rulemaking process under the name of 'good regulatory practices' or 'regulatory coherence' at the regional and multilateral levels, through institutions like the OECD, APEC, WTO and TPP. These efforts of global normative diffusion have resulted in certain developments in Australia towards the US model. Internally, in light of the practical limits of Parliamentary control over rulemaking processes, there has been an increasing demand for administrative rationalisation and accountability in Australia. Due to the growing trend towards deregulation in various international forums, Australia has gradually shifted to the US-style of institutional design of the rulemaking process.

A. Australia's (Re)Modelling on the US-Style Rulemaking Process
While there was some sectoral legislation in the 1970s that introduced elements of deliberative democracy such as public consultation, it was not until the 1980s that Australian jurisdictions began to see a shift in the landscape of administrative rulemaking processes. 77 Among others, Victoria's Subordinate Legislation (Review and Revocation) Act 1984, as amended and inserted as part of the Subordinate Legislation Act 1962, was the first to introduce public consultation, regulatory impact analysis, and the option of a sunset clause in delegated legislation. 78 This Act resembled US-style rulemaking in several aspects: the Attorney-General must consult the Legal and Constitutional Committee when formulating guidelines for the preparation and content of regulations ('statutory rules') to ensure 'consultation, coordination and uniformity' in the process. 79 Such guidelines shall be published in the government gazette, and available to all Ministers and stakeholders 'involved in the preparation of statutory rules'. A regulatory impact statement was generally required unless the Premier provided otherwise. 80 Such a statement was required to include the reasons for the proposed rules, the objectives, a summary of the results of the regulatory impacts, and was required to be open for public comments and submissions for no less than 21 days following its publication. 81 Moreover, these mechanisms were accompanied by a ten-year sunset provision to ensure a fresh review of the existing rules regularly. 82 The 1984 Act was replaced by the Subordinate Legislation Act 1994, though the core elements outlined above remained unchanged. 83 Just a few years after Victoria's Subordinate Legislation (Review and Revocation) Act 1984 (Vic), New South Wales adopted a similar scheme by virtue of the Subordinate Legislation Act 1989 (NSW), 84 and included whether the proposed rule 'may have an adverse impact on the business community' and whether 'the objective of the regulation could have been achieved by alternative and more effective means' as part of the parliamentary scrutiny committee's review process in the Legislation Review Act 1987 (NSW) (formerly, the Regulation Review Act 1987 (NSW)). 85 Shortly thereafter, Tasmania and Queensland followed suit through their Subordinate Legislation Act 1992 (Tas) 86 and the Statutory Instruments and Legislative Standards Amendment Act 1994 (Qld), respectively. 87 In the Australian Capital Territory, the Legislation Act 2001 imposed a regulatory impact statement without providing for consultation or a sunset provision. 88 There were similar reform initiatives at the federal level, too. The Administrative Review Council had, since 1992, proposed a new federal regime akin to those in Victoria and New South Wales. 89 These initiatives failed to yield fruitful results until the Legislative Instruments Act 2003 (Cth) was passed. This Act introduced public 79 ibid, section 11. 80 ibid, sections 12(1) and (3). 81 ibid, section 12(1)(d). 82 ibid, section 3A. 83 See eg Subordinate Legislation Act 1994 (Vic), section 5 (Automatic revocation of statutory rule), section 5A (Extension regulations), section 6 (Consultation), section 7 (Regulatory impact statement), section 11 (Comments and submissions). 84 See eg Subordinate Legislation Act 1989 (NSW), section 5 (Regulatory impact statements), section 6 (Regulatory impact statements not necessary in certain cases), Pt 3 (Staged repeal of statutory rules). 85 Legislation Review Act 1987 (NSW). 86 See eg Subordinate Legislation Act 1992 (Tas), section 5 (Regulatory impact statements), section 6 (Regulatory impact statements not necessary in certain cases). 87 See eg The Statutory Instruments and Legislative Standards Amendment Act 1994 (Qld), Pt 5. 88 Interestingly, the ACT seemed to downgrade the legal rigour of regulatory impact analysis for not imposing a public consultation requirement in this context. Legislation Act 2001 (ACT) Ch 5. See also Allars (n 77) 181 (observing that this is a 'curious provision given that Part 5.2 imposes no consultation requirement'). 89 Administrative Review Council, Report to the Attorney-General: Rulemaking by Commonwealth Agencies (Report No 35) (1992). consultation and sunset schemes and borrowed from the US by establishing the Federal Register of Legislative Instruments. 90 However, the Legislative Instruments Act 2003 omitted certain crucial aspects that had been recommended by the Administrative Review Council: it did not impose regulatory impact assessment, nor did it require agencies to notify the public of a proposed rule. 91 Later, along with the promulgation of the Acts and Instruments (Framework Reform) Act 2015, the Legislative Instruments Act 2003 (Cth) was renamed the Legislative Act 2003, though the central arrangements of the rulemaking process generally remained unchanged. 92 In addition, there were a few other detailed non-statutory guidelines that served supplementary functions. Victoria, for instance, set up the Victorian Competition and Efficiency Commission in 2004, now known as the Office of the Commissioner for Better Regulation (OCBR), a gatekeeper to monitor agencies' compliance with the Victorian Guide to Regulation and ensure that 'impact assessment presents a credible, transparent and evidence-based analysis' suitable for public consultation and decisionmaking'. 93 Similar initiatives can be found elsewhere at the state and federal levels, including the Guide to Better Regulation of New South Wales, 94 the Queensland Government Guide to Better Regulation, 95 the Tasmania Legislation Impact Assessment Guidelines, 96 and the federal Australian Government Guide to Regulation and the Best Practice Regulation: A Guide for Ministerial Councils and National Standard Setting Bodies. 97 Even Western Australia-a state that left key elements like regulatory impact analysis, public consultation, and sunset schemes out of its Interpretation Act 1984 (WA)-followed suit by adopting the 'Regulatory Impact Assessment Guidelines'. 98 On the one hand, the fact that Australia has reoriented itself in some respects towards the US-style rulemaking process was driven by an increasing demand for a rationalised and accountable framework in response to the proliferation of delegated rulemaking activity after the 1980s and, on the other, it reflected neoliberal political and economic ideologies.
The explosive growth of delegated legislation over the past few decades underscores the weakness of the fragile reliance on the parliamentary control model. Contrasting the landscape of administrative rulemaking at the federal level before and after the 1980s, Professor Douglas Whalan, the former legal adviser to the Senate Standing Committee, pointed out that the Regulation and Ordinances Committee examined a total number of only 192 instruments in 1949 (142 of them were statutory rules while most of the rest were ordinances or regulations under ordinances). 99 By the late 1980s, however, the number of instruments reviewed by the Committee had risen to 1,352. Even more problematic was that most of the instruments reviewed came in a variety of forms other than traditional forms such as statutory rules and were not previously indexed. 100 While this explosive growth reflected the dynamic evolution of society, the proliferation in terms of volume and variety complicated the legislature's burden in holding executive bodies to account. 101 Apart from the sheer number of administrative rules, the nature of parliamentary control further fuelled the demand for extra safeguards in this context: a robust parliamentary review would require substantial resources and time, which was the very reason for such delegation in the first place. 102 Of course, the rise of delegated legislation also raised concerns about the quality of the drafting. 103 Moreover, the traditional Westminster framework was attacked for its transparency deficits: these rules were made secretly, for they typically remained out of public view until they became operative. 104 A more cynical view against the traditional model was that the growth of administrative rules was 'part of a deliberative plan to avoid the unwelcome attention of the parliament'. 105 In a way, these concerns had been factored in when Victoria lawmakers drafted the 1984 Act. In its report, the Legal and Constitutional Committee began by recalling the caution of Lord Gordon Hewart, the former Chief Justice of England in his seminal book 'The New Despotism', in relation to the tendency of using delegated legislation and referring to concerns raised by the British Committee on Ministers' Powers: Their true bearing is rather that there are dangers in the practice; that it is liable to abuse, and that safeguards are required. The problem which the critics raise is essentially one of devising the best safeguard. 106 The Committee then canvassed similar concerns in the Australian context and remarked on the drawbacks of the current regime: It has been suggested that a system of regulations assessment is necessary because currently there is little or no public involvement in the regulation making process … Some people contended that subordinate legislation is, however, made by bureaucrats who are not answerable to the public; it is made without publicity, without consultation, and without opportunity for interested parties to air their views. 107 While the Committee noted that consultation processes were already conducted by certain departments at the time, it pointed out that the lack of an overarching framework applicable to all government agencies in a systematic manner was far from satisfactory. The Committee linked the Victorian government's growing interest in regulatory reforms to the relevant developments in North America, explained at length the key features of the US model, and underscored the APA as well as the regulatory reforms during the Carter and Reagan administrations, including Executive Orders 12,044 and 12,291. 108 It bears noting that when the Victorian government considered reorienting its framework towards deliberative democracy, it emphasised, in particular, the experience of Canada, another Westminster-style government that had been undergoing regulatory reforms since the 1970s. 109 Quoting a report by the Victorian Chamber of Manufactures, the Committee cautioned that 'it is important to be aware of the different operating environments' in the US and Canada. 110 The Committee highlighted that 'the regulatory system and associated problems in the United States are quite different in nature and magnitude from those existing in Australia: the complex regulatory system existing in the United States is lacking in Australia.' 111 The Committee further contrasted the different regulatory models in North America, pointing out that the reforms in the US were largely driven by the ideology of 'deregulation' while the main objective pursued by Canada was more about 'improved cost-effectiveness rather than deregulation'. 112 Considering the 'many similarities between the federal systems in Canada and Australia', the Committee was convinced that 'the Canadian experience in regulatory reform may have greater relevance to the Australian situation'. 113 Thus, while Australia gradually moved towards the deliberative rulemaking model, it nevertheless took into account different regulatory environments and traditions in the reform processes. It should come as no surprise that Australia would address attributes of deliberative democracy-and similarly elements of regulatory coherence, as analysed below-in a different manner to the APA.

B. Testing the Inherent Boundary
It is clear from the above discussion that the Australian approach to administrative rulemaking has witnessed a sea change over the past three decades by adding a flavour of deliberative democracy inspired by the US model. Yet a closer examination reveals its Westminster heritage makes Australia's system different from the APA-type rulemaking setting. Several aspects of this are worth emphasising.
A helpful starting point is parliamentary oversight. For instance, at the federal level, delegated legislation must be registered in the Federal Register of Legislation, tabled before each House of Parliament, and is subject to disallowance by the legislature. 114 Specifically, the Senate Committee on Regulations and Ordinances scrutinises these legislative instruments to ensure they (i) are made per the statutes; (ii) do not trespass unduly on personal rights and liberties; (iii) do not make rights and liberties of citizens unduly dependent on administrative decisions that are not subject to independent merits review; and (iv) do not contain matters more appropriate for an Act of Parliament. 115 In contrast to the APA, the current scheme under the Legislation Act 2003 (Cth) seems to be retrospective rather than prospective. 116 Despite several amendments, the basic structure of parliamentary control over delegated legislation remained unchanged. 117 Secondly, while Australia has seemed to borrow some APA characteristics, the nuanced way it incorporated these elements has shaped the rule-making process quite differently from the US model. Not every Australian jurisdiction includes key features of deliberative democracy and regulatory coherence in relevant statutes. Public consultation, for instance, is not referred to in the legislation of the Australian Capital Territory, the Northern Territory, South Australia, and Western Australia. A regulatory impact statement is not a legislative requirement in the Northern Territory, South Australia, Western Australia, and the Commonwealth. Sunset clauses are not required in the Australian Capital Territory, Northern Territory, and Western Australia. As mentioned above, however, some jurisdictions recommend these approaches in government handbooks or guidelines. 118 Further, even though these elements are part of the statutes or handbook-type documents, they are generally unenforceable. In the Commonwealth, for instance, while the Legislative Act 2003 states that 'rule-makers should consult before making legislative instruments', it makes clear that 'the fact that consultation does not occur does not affect the validity or enforceability of a legislative instrument'. 119 According to Gabrielle Appleby, public consultation requirements in the legislation are 'narrow and weak, amounting to little more than a recommendation to consult'. 120 In a similar vein, the Australian Government Guide to Regulation picks up what is omitted in the Legislation Act 2003 by requiring all Cabinet submissions to include a RIA; non-compliance will only be published by the Office of Best Practice Regulation (OBPR) and does not affect the validity of the rules.
Indeed, there are some instances in Australia where deliberative democracy elements have been subject to judicial review. Yet these cases are more an exception rather than a rule and typically involve legislation that makes these elements mandatory. The most salient example in this respect is the New South Wales Environmental Planning and Assessment Act 1979 (NSW), Section 57 of which requires that the relevant agency 'must consult the community in accordance with the community consultation requirements for the proposed instrument' while making a local environment plan. 121 The courts have applied this mandate rigorously in cases like South East Forest Rescue Inc. v. Bega Valley Shire Council. There, the Court found that the agency failed to consider all submissions made by 'members of the public objecting to the proposed development' as required by the Act. 122 The same Act has another key feature: it enables courts to review the rules by removing legal hurdles to standing. 123 In general, only those 'regulated persons'-whose 'rights and interests' are adversely affected by an administrative decision-can bring an action for administrative review. 124 This then allows members of the public objecting to the development to seek review under 'procedural fairness' per section 5(1)(a) of the Administrative Decisions (Judicial Review) Act 1977 (Cth). This ensures the members of the public access to the Court in contesting the proposed development. This should be distinguished, according to Kioa v West, from those policy or political decisions that affect a person 'as a member of the public or a class of the public'. 125 By adding an open standing clause together with a three-month time frame for challenging the rule, this Act relaxes the traditional restrictions on procedural fairness and thus facilitates deliberative democracy. There is other similar case law; most, however, relates to sectoral legislation that makes public consultation compulsory. These examples can only be viewed as an exception to the Westminster model. 126 All in all, although Australian administrative rulemaking has come a long way since sectoral legislation began in the 1970s followed by various reforms at state and Commonwealth levels, and notwithstanding criticisms against weak parliamentary oversight, 127 the current system is distinct from the US rulemaking process. This underscores the possible limitations on the continued projection of a US-style approach to further promote regulatory coherence as a global norm in the age of mega-regionalism. US-style reforms around transparency, cost-benefit analysis, and participation may be in some deep sense incompatible with Australian constitutional institutions, or at least very difficult to graft organically in a context rooted in parliamentary supremacy rather than deliberative rulemaking. The fact that Chapter 25 of the TPP/CPTPP is unenforceable through dispute resolution may be an acceptable middle ground, at least for those without a sufficiently accommodating environment. 128 Similarly, the Peru-Australia Free Trade Agreement (PAFTA), signed between the two CPTPP members, embraces a nearly identical 'soft' design-a full-fledged regulatory coherence chapter (Chapter 24) but at the same time a carve-out from the dispute settlement mechanism. 129 Australia's FTAs with most Southeast Asian countries, on the other hand, seem to reflect a less ambitious status quo in terms of regulatory coherence. Recent trade pacts such as the Singapore-Australia Free Trade Agreement (SAFTA), the Thailand-Australia Free Trade Agreement (TAFTA), the Malaysia-Australia FTA (MAFTA), the Korea-Australia FTA (KAFTA), the Japan-Australia Economic Partnership Agreement (JAEPA), the China-Australia Free Trade Agreement (ChAFTA), and Australia-Hong Kong Free Trade Agreement (A-HKFTA) contain no regulatory coherence chapters whatsoever. 130 Australia's most recent accession to the Regional Comprehensive Economic Partnership (RCEP), not surprisingly, shows little clarity of path. Such ambivalent stances and mixed approaches reflect the tensions or frictions and test the boundary between the penetration of global regulatory coherence and domestic constitutional tradition in the Australian context. 131 Australian institutional design and political dynamics continue to influence, if not define, how and to what extent the elements of US-style regulatory coherence may be embraced in a democracy with different constitutional traditions.

IV. CONCLUDING REMARKS
Recent decades have witnessed the rise of regulatory coherence as a new global norm alongside the emergence of the new plurilateralism. Pioneered by the US in multiple international arenas such as the OECD and APEC, recent megaregional trade agreements have incorporated core elements of regulatory coherence as a means of balancing regulatory autonomy and international cooperation. In some countries, the embrace and development of regulatory coherence has also been driven by domestic demand for reform to rationalise the regulatory environment and rulemaking process. However, the global entrenchment of regulatory coherence is contingent upon the political dynamics and constitutional structures in a jurisdiction.
This article examines the case of Australia and evaluates the normative ramifications for other democratic countries by contextualising the embrace (and, of course, rejection) and development of some key elements of regulatory coherence in the Australian setting. The analytical focus of this article is whether, when, and how the notion of regulatory coherence interacts with the Australian constitutional tradition-and particularly parliamentary 129 Art 24.9 of Peru-Australia Free Trade Agreement (PAFTA). 130 It should be noted that some of these FTAs follow the more conventional path of incorporating a 'transparency' chapter in the agreement. See eg China-Australia Free Trade Agreement (ChAFTA) Ch 13; Korea-Australia FTA (KAFTA) Ch 19; Malaysia-Australia FTA (MAFTA) Ch 17; and Australia-Hong Kong Free Trade Agreement (A-HKFTA) Ch 16. 131 To the best of the authors' knowledge, at the time of writing, no changes had been made at Commonwealth and state levels in response to the fact that regulatory coherence was included as part of some of Australia's trade agreements.

912
International and Comparative Law Quarterly supremacy. Given that the Australian constitutional structure keeps the executive in check through the notion of responsible government, in which Parliament is the exclusive repository of the legislative power of the Commonwealth, administrative rulemaking in Australia has generally been subject to Parliament's control. While there have been a few waves of reform in recent decades towards the US APA model premised upon deliberative democracy, the Australian constitutional structures and political dynamics have persistently served as inherent limits to the further embrace of some elements of regulatory coherence and have resulted in uneven normative development. Beyond Australia, this article sheds light on the discourse of the future development of global regulatory coherence by pointing out how constitutional tradition as an overarching path-dependent anchor frames and conditions the way in which, and the extent to which, elements of regulatory coherence may become entrenched in a democracy.