Economic Evidence in Regulatory Disputes: Revisiting the Court-Regulatory Agency Relationship in the US and the UK

This article examines the issue of the appropriate scope of review of economic evidence enshrined in the discretionary assessments of utility regulators in the US and the UK. It advances a balance of institutional competencies approach to the question of the degree of deference owed to the regulatory agency’s economic assessments. In doing so, it revisits the doctrinal positions advanced in the US and the UK for the substantive review of administrative discretion, so as to become attuned to the challenges posed by economic evidence. Drawing on insights from political science and economics, the suggested approach illuminates the institutional disadvantages of the courts that may warrant a high degree of deference. At the same time, however, it remains sensitive to the polycentric elements of regulatory disputes as well as to a number of institutional realities that may attenuate the weight of such comparative institutional disadvantages.

better able to implement those goals than another institution, comparative institutional analysis stresses the primary role of institutional choice: 'the decision of who decides '. 18 Komesar approaches these choices in 'participatory terms' (the direct or indirect participation in the decision-making of the affected stakeholders). Which institution brings out the diversity of affected interests on a given issue, while avoiding situations of both 'minoritarian bias' (the over-representation of minority interests seeking rents) and 'majoritarian bias' (the over-representation of majority interests, such as consumers)? 19 Significant scholarly attention has focused on the relative institutional advantages and disadvantages of courts in assessing different types of evidence (i.e. scientific, economic, forensic) and in various areas of law (i.e. securities regulation, competition law). 20 This article seeks to contribute to this body of literature by contextualising some of the institutional competencies claims in the, so far, unexplored area of economic evidence in regulatory disputes. That said, it does not attempt to provide a 'golden rule' concerning the appropriate degree of deference to the regulators' economic assessments. The ambition is much narrower and critical: to stress that any determination of the appropriate scope of review of economic evidence should avoid single-factor considerations (e.g. that judges does not possess expertise in economics) in favor of a richer set of institutional 18 Ibid 3. 19 Ibid 27. 20  The article is structured as follows. Sections 2 and 3 present the context in which the analysis of the suggested comparative institutional competencies approach will be conducted. Section 2 discusses the relevance of economic evidence in regulatory disputes and explores its implications for the review of regulatory agencies' decisions. Section 3 provides an overview of the different avenues for challenging regulatory decisions in the US and the UK. Section 4 examines the doctrinal positions advanced in the US and the UK for the substantive review of regulatory agency discretion and section 5 advances the balance of institutional competencies approach to deference. Finally, in Section 6, some concluding observations are offered on the implications of the substantive analysis for the current debate on regulatory appeals in the UK.

Economic Evidence in Regulatory Disputes: Scope and Implications
It would be impossible to investigate the relationship between the court and the regulatory agency against the backdrop of economic evidence, without first venturing into the legal relevance and implications of economic evidence in regulatory disputes.
Doing so, will help to better appreciate the significance of comparative institutional competencies considerations in determining the degree of deference to the regulators' economic assessments. dimension of the law invites the assistance of an economic expert. 26 Scot Brewer summarises this phenomenon in the term 'epistemic deference '. 27 Second, economic evidence frequently informs policy decisions which involve technical aspects (e.g. the design of the appropriate pricing methodology) and value judgements. For example, the implementation of the unbundling provisions of deregulatory acts in the telecommunications sector -those provisions which require the incumbent to share the components of its network (e.g. switches, trunks, loops) with a rival party -involves a huge trade-off between the objectives of static 28 and dynamic 29 efficiency. 30 Mandating access to all components of the network can potentially diminish the incentives of the new entrants to build their own network infrastructure, and those of the incumbent to keep up and improve its property. 31 Denying access, on the other hand, may increase the incumbents' market power, thus impeding competition from new entrants. In such cases, the courts are confronted with a policy choice and judges are required to assess whether the economic methodologies employed by the regulatory agency are fit for purpose and whether the inferences drawn from economic analysis are sufficiently robust to support the regulator's policy decision. 26 27 Ibid. 28 Static efficiency occurs when marginal production costs are minimised (productive efficiency) or when the goods and services are allocated to the uses in which they have the highest value (allocative efficiency). 29 Dynamic efficiency relates to demand in investment and innovation. 30 The mandatory unbundling provisions are enshrined in Section 251(d) of the US Telecommunications Act of 1996. In the EU see Regulation (EC) 2887/2000 (n 22). 31 For a discussion see AT&T Corp. v Utilities Board, 525 US 366 (1999).

B. The Case of Polycentric Regulatory Disputes
The challenges posed by economic evidence merit special attention in the case of regulatory disputes that present 'polycentric' elements. 32 In the Fullerian sense, a polycentric task is one that encompasses a large and complicated web of interdependent relationships. 33 In such cases, a decision taken may affect many actors, thus leading to a fluid state of affairs if all affected interests are taken into account. However, polycentricity may as well refer to situations where a decision affecting one actor could have a different set of repercussions for other parties leading to a redefinition of the parties affected.
A perfect example of a polycentric task is how to set the appropriate regulatory access price; that is the price under which competing firms will have access to essential inputs (e.g. electricity transmission network) provided by the incumbent. In access pricing the regulator has to strike a balance between the divergent interests of the new entrants, the incumbent and ultimately the consumers. The best outcome of this balancing exercise demands comprehension of a highly complex range of cause-effect relationships. First, there is a distributional concern. Pricing at marginal cost maximises efficiency, but does not compensate the incumbent for his fixed costs related to investments in the network.
Second and relatedly, there is a dynamic efficiency concern. In other words, the regulator has to ensure that the method employed will also contribute to the long-term interest of the consumer, the encouragement of technological innovation as well as the growth and 32 L Fuller 'The Forms and Limits of Adjudication' (1978) 92 Harvard L Rev 353. 33 Ibid 394. viability of the sector at issue. Little, however, guidance is provided in the agencies' governing statutes in such cases, hence allowing the regulators a large measure of policy discretion to balance the competing interests at stake. 34 Polycentric disputes, Fuller argued, are inherently unsuited for adjudication and thus best resolved through managerial discretion or through negotiation and contract 35 or left to the forces of the market. 36 Nonetheless, Fuller accepted that polycentric elements are present in all problems resolved by adjudication. What becomes important is knowing when 'the polycentric elements have become so significant and predominant that the proper limits of adjudication have been reached'. 37 It follows that the polycentric dimension of regulatory disputes does not necessarily imply the removal of the court's supervision altogether (non-justiciability), but may warrant a special weight to be accorded to the economic assessments of the regulatory agency. This, however, as we shall see, presupposes a case-by-case assessment of the factors that may attenuate the polycentric dimension of the dispute in question, thus making it amenable to the adjudicative process. 38 If the regulatory and the adjudicative institutional processes are imperfect alternatives, as this article contends, the challenges posed by economic evidence in regulatory disputes should be addressed by a comparative institutional analysis of these imperfect alternative choices to select the least imperfect one. However, this is not as simple as it may sound. 34  The examination of the different avenues for challenging regulatory decisions in the US and the UK, undertaken in the following section, will reveal that a plethora of imperfect decision-makers are involved in the regulatory and adjudicative process.

Decisions in the US and the UK
The US and the UK radically differ with respect to the procedures available for the review of economic evidence enshrined in regulatory decisions. First, in the case of the US, federal regulatory agencies' decisions are reviewed by means of judicial review only.
In contrast, in the UK, the routes of both judicial review and merits review are offered to challenge sector-specific regulators' decisions. Second, as will be shown, in the US, review of economic evidence is 'internal', meaning that 'it takes place within the institution in which the original decision-maker was located at the time the decision was made'. 39 In contrast, in the UK it is most of the times 'external' to the regulatory agency, as it takes place in a different institution, the Competition and Markets Authority (CMA). 40 Finally, while the US has resisted the idea of establishing specialised courts for the resolution of regulatory disputes, in the UK a specialist tribunal has been established,

A. The Processes for Regulatory Appeals in the US
The US has adopted an administrative litigation model for the review of economic evidence, which reflects its long tradition of administrative adjudication in the regulation of natural monopolies. To explain this further, when a decision involving economic evidence made by the FERC or the FCC is challenged or disputed, a common pattern is for the decision to be first reviewed by an adjudicative institution embedded in the regulatory agency, that of the Administrative Law Judge (ALJ). Crucially, the ALJ is functionally separated from the investigatory and prosecution arm of the agency. The administrative litigation model is closely modelled to court adjudication, 'as it involves the participation of the affected interests in the decision-making process through proofs and reasoned arguments'. 42 The administrative hearing is trial-like, with testimony and evidence presented by trial staff, such as attorneys and trial staff, who assist the ALJ and participate in the hearings as impartial representatives of the public interest. They may also serve as expert witnesses. 43 The opportunity for oral testimony by witnesses and the opportunity of cross-examination enable the agency to test its factual record.
The process is governed by the provisions of the APA; a default procedural framework governing federal regulatory policymaking. The ALJ's initial or recommended decision is final subject to a de novo review or appeal by the agency itself either by the commissioners or by members of the regulatory agency's governing body. Agency heads 42 Fuller (n 32) 365. 43 See 18 CFR 385.102(b). thus perform a 'second pair of eyes' review of the decisions and are free to substitute their judgement for that of the ALJ on questions of law, fact and discretion. 44 If the regulatory agency affirms the original decision or varies it in a way that does not satisfy the affected party, that party may seek external review of the decision in the federal courts on the various grounds of review listed in the APA. Federal courts are limited to the record produced by the agency itself and refrain from a de novo examination of the facts. It is not for the courts to substitute their views for that of the agency not to provide the reasonable basis for the agency's action. 45 Although in principle, all of the US Circuit Courts can hear petitions for review of federal regulatory agencies, the Court of Appeals for the District of Columbia (DC Circuit) merits special attention. Technically speaking, the DC Circuit is a regional court staffed by judges from all over the USA. Its location, however, right in the heart of the federal's government capital, and its exclusive jurisdiction over a variety of legal challenges to administrative law action has rendered it a 'semi-specialised' court. 46 Although the DC Circuit is the court of final resort for most of cases, there is always the possibility of further appeal to the Supreme Court. The Supreme Court enjoys certiorari jurisdiction in reviewing lower court's judgement de novo.

B. The Processes for Regulatory Appeals in the UK
In line with the principles of subsidiarity and national procedural autonomy, 47 the UK enjoys ample discretion in determining the means available for challenging regulatory decisions subject, of course, to the twin principles of equivalence and effectiveness. 48 While in the original legislation establishing the utility regulators, regulatory decisions could only be challenged by way of judicial review before the generalist High Court and on limited grounds (i.e. illegality, irrationality or procedural impropriety), today the routes of both appeal and judicial review are available. In fact, the introduction of statutory rights of appeal in the late 1990s have gradually led to the marginalisation of judicial review as the primary means to challenge regulatory decisions. 49 Furthermore, the establishment of the specialist CAT has largely replaced the High Court as the primary venue for hearing such challenges. Contrary to the ordinary courts, the CAT's bench combines legal and non-legal expertise in areas such as economics, business and accountancy. 50 Crucially, the CAT enjoys both statutory review (similar to common law judicial review) and statutory appeal jurisdiction (where it engages with the factual merits of the case). It can be argued that the Parliament dealt with the limitations of judicial review by providing broader appeal rights to the CAT, which would not be subject to the restrictions of judicial review. 47 The principle of subsidiarity is laid down in art 5, 3 TEU. The doctrine of national procedural autonomy has been developed by the CJEU in its case law starting with Case C-33/76, The processes, however, for the review of regulatory decisions are largely inconsistent. Financially significant regulatory decisions for the investors and ultimately the consumers (e.g. price control decisions, licence modifications, references concerning non-licensable activities in the gas and electricity and energy code modifications appeals) can be appealed on the merits to the CMA. 51 The CMA enjoys broad access to epistemic competence as well as access to a pool of investigators, including Utilities Panel members. 52 Furthermore, its in-depth inquisitorial approach allows for a better appreciation of the underlying issues in price control and licence modifications cases than the CAT's adversarial approach. In the energy appeals process, as a result of the introduction of the EU Third Energy Package reforms, 53  Court of Appeal on behalf of a party or anyone else with sufficient interest. 56 However, in the case where an appeal raises a price control matter this is hived off by the CAT and referred to the CMA for determination 'on the merits'. In such cases, participation in the CMA proceedings is limited to the parties to the case, the appellant, the regulator and any interveners who are admitted by the CAT.  very promulgation of the 'hard look review' in the early 1970s illustrates that federal courts had assumed a corrective role for the alleged failures of the regulatory process to regulate in the public interest. 60 In contrast, the low irrationality review in the realm of utilities regulation emerged inter alia as a reaction to the apparent incapacity of the courts to deal with the complexity of the regulators' economic assessments following the privatisation of utilities.

Substantive Review of Regulatory Discretion in the US and the UK
This section will take a closer look to the evolution of these divergent doctrinal positions and their application in regulatory disputes.

A. The Scope and Intensity of Substantive Review of Regulatory Discretion in the US
As a general rule, FERC's and the FCC's discretionary assessments are subject, at the minimum, to the 'arbitrary, capricious [or] abuse of discretion' test of the APA. 61 Nonetheless, the availability and scope of review is largely determined by whether the regulatory agency's action involves rulemaking or adjudication, according to the distinction firmly embedded into the APA.
Rulemaking resembles legislative action, whereas adjudication is similar to court proceedings. Two kinds of rulemaking are envisioned by the APA: formal and informal rulemaking. Formal rulemaking refers to 'rules required by the statute to be made on the 60  record after an opportunity for an agency hearing'. 62 Informal rulemaking, otherwise known as 'notice-and-comment rulemaking', is triggered by a notice to the public of the 'subjects and issues involved' 63 followed by an opportunity for outsiders to provide comments or objections to the proposed rule. Generally, informal rulemaking generates rules made pursuant to congressionally delegated authority. The legislative rules, if valid, operate like a statute.
Adjudication is defined broadly in the APA as a 'process for the formulation of an order'. 64 An order binds only the party to the proceedings, but some orders, such as those issued at the conclusion of the FERC's adjudication, operate as precedent. 65 Both formal rulemaking and formal adjudication require prescribed hearing procedures with the aim of creating a factual empirical basis upon which the agency will form its decision. Formal adjudication, however, affords a more limited class of persons -often the parties -to present evidence and to have the opportunity for cross-examination. 66 Although the APA envisions two kinds of routes that the agency may follow to effectuate its statutory mandate, in reality the latter is free to select from a wider set of policymaking tools. 67 The FERC, for example, is authorised to promulgate legislative rules, conduct administrative adjudication, and/or initiate judicial enforcement. It can be argued that as a general pattern the FCC relies heavily on rulemaking, 68  FERC relies on both adjudication and legislative rules. 69 Crucially, ratemaking for electric and gas utilities is classified as rulemaking, but is conducted very much like formal adjudication before an ALJ. 70 Rules made on the 'record' are subject to the 'substantial evidence standard'. 71 The same standard applies to an agency's factual findings in the course of adjudication, which will be assessed to see whether it is based on the 'whole record'. 72 It is important to stress that in such cases federal courts rarely undertake a de novo examination of the economic facts, as these have been largely established in the course of the complex fact-finding process undertaken by the ALJ. Indeed, the record of the FCC and the FERC cases illustrates that courts increasingly review agencies' action under the arbitrary and capricious test of the APA deferring to agency findings of fact, provided they are supported by substantial evidence on the record. 73 However, nor the text nor the legislative history of the APA clarify the intensity of the arbitrary and capricious test. Over time, it has given rise to both light-touch review and intrusive review of administrative action; the latter commonly referred to as 'hard look' review. Developed by the DC Circuit in the late 1960s and early 1970s, in response to agencies' turn to informal rulemaking as a means of policy making and to the beliefs in 'agency capture', 'hard look' review demanded an intrusive review of the agency's 69 Rossi (n 65). 70  action. In the legal academy, US administrative law scholars supported this position by advocating a more 'stringent judicial review' 74 of the regulatory agency's actions, which would be less deferential to political outcomes, 75 as well as a more intrusive judicial review of agency rulemaking so as to combat agency capture. 76 The doctrine also required that agencies satisfy the court that they had considered all of the dimensions of the matter before it and that it had engaged in reasoned decision making when claiming discretion. 77 Additionally, they should justify departures from past policies and allow effective participation in regulatory decision-making by a broader range of stakeholders.
The Supreme Court endorsed the doctrine in the State Farm case, 78 stating that a reviewing court ought to test the quality of the agency's reasoning process as to whether 'the decision was based on a consideration of the relevant factors' whether 'there has been a clear error of judgment'; and whether: the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs contrary to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise. 79 Hence, the Court embraced not only the procedural and but also the substantive dimensions of the 'hard look' review.
In the realm of utilities regulation, the 'hard look' review has been applied in judicial review of the FERC's and FCC's exercise of policymaking discretion in light of a technical, or economic record, such as the one that which exists in rate-setting cases. For example, in reviewing the FCC's and FERC's exercise of discretion in rate-setting cases the federal courts employ a rationality type of review to examine whether the agency took into account 'the relevant data and articulate[d] a satisfactory explanation for its action including a rational connection between the facts found and the choices made'. 80 Rate setting is an inherently complex process that involves economic, technical and policy assessments. 81 Although judges routinely hold that the implementation of the legislative requirement of 'just, reasonable and non-discriminatory rates' involves 'policy judgements that lie at the core of the regulatory mission', 82 the application of the 'hard look' review has enabled federal courts to reverse, vacate or remand a number of the FERC orders on the grounds that the latter failed to engage in 'reasoned decision making', or failed to explain their underlying policy, 83

B. The Scope and Intensity of Substantive Review of Regulatory Discretion in the UK
The early case law, in the wake of utilities privatisation, suggests that the low intensity rationality review applied to the regulators' discretionary assessments was informed by constitutional considerations as well as by the apparent incapacity of the courts to deal with the complexity of economic appraisals. Substantive decisions were challenged under the standard of Wednesbury unreasonableness. Under this test of review courts could set aside the conclusion reached by a public authority entrusted by Parliament with discretion on a matter if this decision was wholly absurd, that is a '(…) decision which is so outrageous in its defiance of logic or of accepted moral standards that no sensible person who had his mind to the question to be decided could have arrived at it'. 87 It was thought that limiting substantive review in this way would prevent courts from entering into a constitutionally forbidden territory. The deferential posture was further justified by the regulator's special knowledge and expertise in regulatory matters. 88  arguments, weigh the compelling considerations and arrive at a judgment…If (as I have stated) the court should be very slow to impugn decisions of fact made by an expert body and experienced decisionmaker, it must surely be even slower to impugn his educated prophesies and predictions for the future. 90 The judicial reluctance extended to the regulator's findings of fact, which formed the basis of his discretionary judgment. In a number of cases, English courts have held that the existence or non-existence of fact should be left to the judgement or discretion of the decision-maker. 91 The input into the courts' reasoning of the abovementioned considerations led to a long-lived harmonious relationship between the judges and the regulatory agencies: the judges would focus on the process of regulatory decision-making, whereas the agencies would focus on the substance thereof. The absence of a proportionality test for substantive review that would require the courts 'to assess the balance the decision-maker has struck' 92 paying attention to the 'relative weight of interests and considerations' 93 subjected the regulatory agencies' decisions only to the 'monolithic' Wednesbury test of extreme unreasonableness -a test that is very difficult for the claimant to satisfy. 94 The economic evidence enshrined into regulatory decisions was largely immune from judicial 90  scrutiny giving rise to a controversy over limited regulatory accountability. The judicially created boundaries of 'no-go' areas of review allowed regulators to operate largely unchecked; rarely did they state reasons or provided clear criteria for their decisions. 95 The force, however, of low intensity rationality review seems to have dwindled in recent years, especially in light of constitutional and jurisprudential developments that have allowed a more demanding judicial scrutiny than traditional Wednesbury review.
Such developments are better captured in the notion of a 'multi-streamed jurisdiction' 96 , a term employed to denote that judicial review now encompasses not only common law principles, but also relevant applications of EU law and of the European Convention of Human Rights (ECHR).
Hence, claimants wishing to challenge the regulators' economic assessments may demand a more stringent review of agency's discretion based inter alia on the 'European' The foregoing analysis revealed that although the doctrinal positions presupposed the institutional advantages and disadvantages of courts when determining the degree of deference to regulatory decisions, they have ever since evolved to include varying intensities of review. There is, however, still room for these to become aligned with the institutional and procedural framework governing the regulatory appeals processes. For instance, how would the doctrine of deference to agency's expertise, as applied by the UK generalist High Court, be accommodated in a specialist tribunal, such as the CAT? In such a cases, blanket deference on the basis of the regulators' expertise is unconvincing, as it is rooted in the anachronistic assumption that the regulator is the only one who possesses the special knowledge. Furthermore, one may question whether the generalist judge is institutionally well suited for strict policy review of such disputes, given his informational limitations and political insulation? Will a 'hard look' style of review lead to a better administrative policy, or will it discourage the agency from acquiring policyrelevant expertise? The next section will discuss how a balance of comparative institutional competencies approach to deference may overcome these issues and ensure that substantive review of regulatory discretion becomes attuned to the nature of economic evidence and the polycentric dimension of regulatory disputes.

COMPETENCIES APPROACH
At their most basic, comparative institutional competence considerations are built on the assumption that one branch of government is better at performing a specified function than the other. As used here, a balance of comparative institutional competence approach to deference envisions that when faced with the agencies' discretionary economic assessments, courts should weight their relative institutional advantages and disadvantages in resolving such issues against that of the regulators, before deciding on the appropriate amount of deference to accord to the agency's assessments.

A. The Comparative Institutional Disadvantages of Courts
As compared to the regulatory agencies, generalist courts suffer from a number of institutional disadvantages in assessing the different manifestations of economic evidence in regulatory disputes. These include: (i) lack of economic expertise; (ii) limited access to information; (iii) lack of requisite institutional legitimacy, especially in addressing polycentric disputes and questions connoting broad policy considerations, and finally (iv) susceptibility to 'minoritarian' bias.

(i) Lack of economic expertise
One of the most obvious institutional disadvantages of ordinary courts relates to their lack of expertise in regulatory matters. Judges sitting on such courts do not appear to have the special skills and expert knowledge necessary to assess economic evidence; not can they easily acquire such skills. For instance, they do not possess cross-disciplinary expertise in economics and/or business and accountancy, as members of the CAT do, and they deal with a much wider spectrum of disputes, other than regulatory disputes. The The court's lack of economic expertise provides a basis for judicial deference to the agency's economic assessments grounded on the latter's 'epistemic authority'. 112 In such a case, the judge acknowledges its 'epistemic deficit' 113 in evaluating the economic evidence enshrined in regulatory decisions and accords greater weight to the judgement of the agency, which is closer to the facts on the ground. Aileen Kavanagh refers to this kind of deference as 'substantial' 114 deference, in the course of which the judge recognises his 'institutional shortcomings' 115 in respect of an issue. It follows that failure on the part of the court to afford substantial deference to the regulator will lead to the court giving a wrong pronouncement on the issue at stake. Substantial deference, in turn differs from the 'minimal' deference that is owed to the agency because of constitutional  115 Ibid 139,192. 116 Ibid.

(ii) Limited access to information
Another comparative institutional disadvantage of the courts is their limited access to information that is relevant to the dispute in question. Judges depend for their information on the evidence that is put before them in a case. In other words, they are situated decision-makers, called to decide upon regulatory disputes on an ad hoc basis. This is especially the case in the context of adversarial proceedings, whereby the adjudicator is informed solely by the litigating parties and thus he is significantly insulated from the whole spectrum of relevant information. The regulatory agency, on the other hand, has a continuous involvement in the administration of the sector it supervises and enjoys broad access to information regarding the problems and characteristics thereof. This in turn, may warrant a high degree of deference to the regulators' assessments because of his superior institutional capacity in gathering and assessing relevant information.
There are several ways by which the regulatory agency can acquire such information.
First, during the consultation period that precedes the promulgation of a new rule. All UK utility regulators enjoy statutory information gathering powers, 117 which they may use if they believe it is appropriate to do so. Both the consultation and the pre-consultation processes allow the gathering of relevant market data, the exchanges of view with stakeholders and affected interests, and therefore enable epistemic communities representing different values to be represented in the regulatory process. This in turn allows a broader representation of the diverse set of consumer preferences, which strengthens the regulatory agency's capacity to assess the effects of its interventions. In 117

(iii) Lack of requisite institutional legitimacy
As compared to regulatory agencies, both generalist courts and specialist tribunals lack the requisite institutional legitimacy to address cases involving broad policy considerations. As already discussed, in many contexts, regulatory agencies are required to trade off efficiency with equity considerations. This is because utilities regulation has diverse legitimate aims that are not solely guided by economic efficiency. For example, conditions for access to bottleneck services for new entrants are mainly driven by efficiency considerations related to the creation of a competitive marketplace. The provision of universal service, however, seeks to create equality of outcome by supporting the service to a legislatively defined category of users. This may involve the provision of a set of services to end users at prices that depart from those resulting from normal conditions. In such cases, a trade-off may arise in meeting the objective of promoting competition, while at the same time ensuring universal access to a set of services.
The regulatory agency may be said to enjoy superior institutional legitimacy in resolving such trade-offs deriving in turn from its political accountability. This is because it is more informed on current political preferences than the judiciary, which makes it 'more likely to accurately reflect current enactable preferences than judicial estimates about what political forces would enact'. 121 The judiciary, on the other hand, is a 'counter-majoritarian' 122 institution composed of unelected judges who should not use their power of review to act contrary to the 'majority will' as expressed by representative institutions. Hence, an increased judicial interference with policy choices jeopardises the legitimacy of the institution. Especially in the UK context, institutional legitimacy considerations are strongly pronounced, in light of the principle of parliamentary sovereignty. Deference, thus, becomes relevant because there are constitutional reasons for granting authority over the judiciary to the regulatory agency.
The abovementioned considerations should not, however, imply an abdication of the judicial role and function altogether. The judge will still have to assess the economic methodologies and considerations employed to support the agency's policy decision as well as the factors relevant to the regulators' determinations. The more expert the court or the tribunal in economics the greater the scrutiny that will be applied. The court, however, cannon pronounce on the right policy to be pursued because the agency has the legitimate authority to decide on this.

(iv) Courts' Susceptibility to 'Minoritarian Bias'
Finally, contrary to what intuition would suggest, courts, and not solely regulatory agencies, can be susceptible to 'rent seeking' influences. While the initial wave of public 'new school of public choice theory' increasingly brings the courts into the framework of public choice analysis. 124 For example, Frank Cross argues that '[c]ourts are simply another venue in which influence may be brought to bear upon government policy'. 125 He further discusses the variety of forms that the lobbying of courts can take, including 'text cases and forum shopping, amicus curiae participation, provision of expert witnesses, selective settlement and other techniques'. 126 Therefore, on this basis the Olsonian tenets of interest group theory can also be ratified in the context of litigation, as narrow groups with greater economic stakes in regulatory issues will be more willing to devote time and resources to hiring the best lawyers and economic experts for their case, so as to be effectively represented in expensive litigation over regulatory disputes. 127 Ultimately, the judge might end up being captured by special interests.
Most worryingly, courts' susceptibility to 'minoritarian bias' might exacerbate the instances of tactical litigation. In other words, narrow groups may be encouraged to bring unmeritorious appeals with a view to hamper or obstruct the regulators' functions. In the UK, OFCOM has advanced concerns about tactical litigation, in the context of spectrum policy, where incumbent mobile operators allegedly attempted to hold up by the threat of litigation the auction for the 800MHz/2.6GHz spectrum. 128 Strategic litigation may also be encouraged in the aftermath of the Infinis 129 judgment in the UK, which, as already discussed, introduced a novel action for damages for unlawful state actions in violation of the ECHR rights. Companies may be from now on willing to use human rights claims to seek more favorable forms of redress in judicial review proceedings.

B. Some Implications of the Suggested Approach in Polycentric Disputes
The case of polycentric disputes epitomises the balance of institutional competencies approach in determining the appropriate scope of review of economic evidence. As you may recall, such disputes do not solely involve the assessment of complex economic appraisals, but further the difficulty of distinguishing who will be affected in the complex web of interdependent relationships.
The regulator through negotiation and consultation can acquire an informed view of the heterogeneous interests. For instance, in the course of designing the Revenue = Incentives + Innovation + Outputs (RRIO) price control mechanism OFGEM sought to enhance the engagement of heterogeneous interests, such as consumer representatives, network users and companies. The authority also set up a Consumer Challenge Group for the latest RRIO-ED1 to act as a 'critical friend' to OFGEM to help ensure that the price control settlement is in the best interests of existing and future consumers. 130 Moreover, the regulatory process may better ensure the adequate representation of those groups that have been traditionally deemed vulnerable to 'majoritarian bias', such as the disabled, the elderly or the chronically sick. In fact, the UK utilities regulators have a specific objective to promote the interests of such groups. 131 The adjudicative process, however, cannot take into account, to the same extent, the complex repercussions that may result from the establishment of a price control mechanism nor secure the meaningful participation of all affected interests in court. Hence, it is the polycentric dimension of such disputes rather than the difficulty of courts in general to deal with price setting, which may warrant a greater flexibility to the regulatory agency to assess the weight of the relevant issues at stake.
The CAT seems to acknowledge the polycentric element of certain regulatory disputes it is called upon to adjudicate. Prominent amongst these are the so-called 'termination rates disputes' (TRD) 132  consumers who may be adversely affected from the higher prices passes on to BT's customers; hence consumer representative bodies should be advised.

C. Factors Attenuating the Comparative Institutional Disadvantages of Courts
While institutional competence considerations are an influential factor in determining the appropriate scope of review of economic evidence, they are not decisive. The presence of a number of institutional realities may significantly attenuate the comparative institutional disadvantages of courts and hence the degree of deference accorded to the regulatory agency's economic assessments. This section will discuss a non-exhaustive list of such realities. Broadly, those relate to: i) the court's specialisation in regulatory matters; and ii) to the court's rules of procedure and its access to epistemic competence.

(i) Court's specialisation in regulatory matters
In the presence of a specialised tribunal, like the UK CAT, the deference accorded to the regulatory agency's economic determinations may be lower in comparison to where the assessment is made by a generalist court without any direct access to epistemic competence. The CAT itself has acknowledged that its familiarity with the statutory regime as well as 'the relevant expertise in its disposal may render the tribunal a more demanding and/or less deferential tribunal than might otherwise be the case where a court (…) is in contrast to the more normal situation where a non-specialised court is called upon to review the decision of a specialised decisionmaker. For that reason we are unpersuaded that there is necessarily a direct "readover" to section 120 from cases such as Cellcom, Interbrew, T-Mobile, and the Rail Regulator that have been cited to us. 142 The court's specialisation in economic evidence, through its repeated exposure to regulatory disputes may also enhance its relative epistemic competence vis-à-vis the regulatory agencies . For example, you may recall that the US DC Circuit's members achieve a certain degree of specialisation though their repeated exposure to regulatory cases. This kind of specialisation, which is owed to rules of venue and geographical patterns in litigation, is better captured in the notion of 'subject matter' specialisation.
The mechanisms of opinion specialisation and case assignment may also ameliorate the institutional disadvantages of courts with respect to their lack of economic expertise. For instance, in his important study on the 'Myth of the Generalist Judge', Edward Cheng sheds empirical light on the de facto specialisation of the judge through the process of opinion assignment in the federal courts of appeals. 143 Finally, judges' prior educational background or training in economics or, possibly, their prior experience in a regulatory agency of some kind may strengthen their capacity to deal with economic evidence. 142 Ibid. 143 EK Cheng, 'The Myth of the Generalist Judge' (2008) 61 Stan L Rev 519.

(ii) The court's rules of procedure and its access to epistemic competence
The court's rules of procedure may enhance the information record of the courts concerning the dispute in question and allow for the better representation of affected interests. This in turn strengthens the institutional capacity of the court to deal with polycentric disputes. For instance, the CAT's procedural rules favour an extensive -often forensic in nature -examination in respect of each aspect of the regulators' findings of fact and expert analysis. 144 Prominent amongst these is the obligation on the part of the tribunal to ensure case management through active intervention. 145 Unlike the narrow scope of the duty of case management on the High Court, 146 there is wide scope for case management under the CAT's Rules of Procedure. 147 Those rules, which are partly based upon those of the EU General Court 148 as well as on the Civil Procedure Rules, 149 crucially reserve to the CAT a central role in controlling the process, which is not inquisitorial but adversarial. In the course of the case management conference, which consists of a pre-hearing review serving a variety of functions (e.g. evidence gathering and fact-finding), the tribunal may order the disclosure of documents or require the parties to particularise their pleadings. 150 There also exists an enforcement mechanism in the form of penalties for non-compliance with the court orders. 151 Crucially, the Tribunal may consider new evidence, which was not submitted to the regulator before it made the decision which is being appealed, 152 or evidence that informed the regulator's decision during the administrative phase but that was made available to the parties either in the regulator's final decision or during the appeal process. 153 The CAT is also entitled to require the provision of evidence, which was not before the regulator, if it considers that the party who wishes to adduce it has shown good reason to justify it and that it is in the interests of justice to admit it. 154 For instance, the tribunal may consider an econometric study that had not yet been conducted when the regulator was forming his decision. Vodafone v OFCOM 155 is a case in point. On the basis of the witness statement submitted on behalf of the intervener H3G the CAT decided that OFCOM's decision to mandate direct routing of calls to ported numbers was not robust and should be overturned.
The CAT considers its ability to consider new evidence as conferring 'a broad discretion to the tribunal regarding the admission or exclusion of evidence which discretion is coloured by the nature of the appeal that is being heard'. 156 The principle by the Court of Appeal in BT v OFCOM. 157 It was made clear that no party should enjoy an unfettered right to adduce fresh evidence on appeal and that parties should present their cases during the administrative phase as fully as the circumstances permit. In sharp contrast, there is limited scope for evidence in judicial review proceedings, which was not before the original decision-maker. 158 A High Court judge will typically refuse to allow a decision-maker to adduce evidence to justify its original decision or to allow a party to challenge a decision on the basis of material that was not available to the decision-maker when the original decision was made. In such cases, the High Court will normally remit the decision for reconsideration by the original decisionmaker on the basis of the new evidence. The rationale for that restrictive approach is that the reviewing court should normally be directing its attention to the material available to the body whose decision is being reviewed rather than deciding the merits of the case de novo. 159 Finally, although the CAT is first and foremost an appellate tribunal and not the oral cross-examination of witnesses, especially when the primary facts are in dispute. 161 In fact, the tribunal has gone as far as to argue that a merits appeal 'provides…a right to call and cross examine witness[es]'. 162 Although the CAT disfavours 'prolonged crossexamination sessions that last for days and days and days', 163 in a number of regulatory disputes it has ordered the cross-examination of witnesses. In the PPC 164 case, for instance, which involved a challenge against OFCOM's determination to resolve disputes between BT and other communications providers regarding BT charges for Partial Private Circuits (PPCs), four witnesses were cross-examined on the economic assessments underlying BT's cost-orientation obligation. To take another example, in the 08-numbers case 165 involving two appellants (BT plc and Everything Everywhere Limited), OFCOM as a defendant and five interveners, 49 witness statements and expert reports were adduced in evidence, 13 witnesses of fact provided evidence and seven economic experts were cross-examined on the 24 reports analysing the impact of BT's proposed price charges. OFCOM has, indeed, praised cross-examination as a powerful mechanism in 'exposing witnesses whose evidence is not empirically supported, contradictory or misleading'. 166 In contrast, the High Court rarely hears witnesses or expert witnesses and it has never sat with an assessor in a judicial review case. In sum, the balance of comparative institutional competencies approach invites a caseby-case assessment of the differences in competence characterising the decision-making processes involved in the dispute in question. In doing so, this approach rejects those normative prescriptions on deference that are informed by single-factor explanations (i.e. the judge does not possess expertise in economics) in favor of a richer set of factors that may determine the degree of deference owed to the regulators' economic assessments. It therefore, furnishes a more pragmatic response to the challenges posed by economic evidence in regulatory disputes that is attentive to institutional realities and determinants of judicial behavior.

Regulation?
This article highlighted that the substantive review of economic evidence enshrined in the regulatory agency's discretion should become sensitive to institutional competencies considerations, if the challenges posed by economic evidence are to be effectively met.
The pluralistic nature of the regulatory and adjudicative process -where various, imperfect alternative decision-makers interact -coupled with the polycentric element of regulatory disputes, bring to the forefront factors such as fact-finding competence and evidence gathering tools that should also inform the scope of review. Through the lenses of comparative institutional analysis, the regulatory and adjudicative processes emerge as interdependent complements, as they may both be both equally defective in some aspects, in ensuring the soundness of regulatory decisions Hence, deference emerges as an institutional choice of the least imperfect alternative; an essentially competenceassessment exercise. Drawing on insights from political science and economics, the suggested approach illuminates the institutional disadvantages of the courts that may warrant a high degree of deference. At the same time, however, it remains sensitive to a number of institutional realties and procedural developments that may attenuate the weight of such comparative institutional disadvantages.
Interestingly enough, there seem to be a few instances where US and UK courts have assessed their relative institutional competence vis-à-vis the regulators. In the US, courts have started taking into account their institutional limitations, 169 although, as Cass Sunstein and Adrian Vermeule argue, this recognition remains 'episodic and occasional'. 170 In the UK, in the adjudication of regulatory disputes that involve polycentric elements (e.g. the imposition of regulatory remedies) the CAT nearly always affords a considerable degree of deference to the regulators, despite its epistemic competence and the wide possibility for third party interventions that allow the representation of a greater number of interests. For instance, in the Albion 171 case, when the tribunal was asked to specify a minimum retail margin, it refrained from doing so by engaging in an implicit comparative institutional analysis of the relative merits of the adjudicative and regulatory process, when presented with polycentric disputes: How can the Tribunal [or a competition authority] determine this margin without examining costs and demands, indeed without acting as a price-setting regulator, the determinations of which often last for several years (and are themselves subject to appeals)? […] (H)ow [should] the Tribunal, or the Authority, … respond when cists or demands change over time, as inevitably they will. The efficient margin fixed today may, through economic and business changes, become the inefficient margin of tomorrow. We do not say that these questions are unanswerable, but we have said enough to show why courts normally avoid direct price administration, relying on more appropriate methods. 172 Finally, the conclusions reached in this article may prove to be especially relevant to those countries designing or reforming the appeal mechanisms in the utilities sector, such as the UK. In June 2013, the Department for Business, Innovation and Skills (BIS) launched an important consultation on 'Streamlining Regulatory and Competition Appeals' 173 whose outcome is still pending. The proposals range widely and if implemented could hugely affect the appeal mechanisms for utility regulators in the UK.