In The Supreme Court of Bermuda

CIVIL JURISDICTION

2006: No. 216

IN THE MATTER OF SECTION 60 OF THE TELECOMMUNICATIONS ACT

1986

AND IN THE MATTER OF THE DECISION OF THE MINISTER OF

TELECOMMUNICATIONS AND E-COMMERCE DATED 20th JUNE, 2006

BETWEEN:

THE BERMUDA TELEPHONE COMPANY LIMITED

-and-

THE MINISTER OF TELECOMMUNICATIONS AND E-COMMERCE

Appellant Respondent Dates of Hearing: 14 – 21 August 2008 Date of Judgment: 6 October 2008 Jeffrey Elkinson and Ben Adamson of Conyers Dill & Pearman for the Appellant; Huw Shepheard of the Attorney General’s Chambers for the Respondent.

JUDGMENT

INTRODUCTION

1. This is an appeal under section 60 of the Telecommunications Act 1986 (‘the Act’) against a decision of the Minister of Telecommunications & E-commerce (‘the Minister’) reducing the Local Access Charge (‘the LAC’) from 5¢ per minute to 3¢ per minute. The LAC is what long distance telephone carriers1 have to pay to access the local telephone system, which in Bermuda is owned and operated by the appellant (‘BTC’). The change 1 The long distance carriers are frequently referred to as ‘Class A’ carriers, that being the designation of the licence they hold under the Public Telecommunications Service (Licence) Regulations 1998.

may not look great, but in revenue terms for BTC each cent is roughly equivalent to $1.2M p.a.

2. The appeal is somewhat stale. It was commenced by Notice of Motion on 19th July 2006, and was essentially ready for hearing by the end of that year, but it was not then brought on. It was eventually listed for hearing in October 2007, but was then adjourned due to the serious illness of counsel. When, in February of this year, the parties eventually agreed a hearing date it was not until August. I consider these delays unacceptable, not least because the ultimate decision will affect third parties, namely the long distance carriers. When the matter finally came before me I held a directions hearing, at which I ordered that the long distance carriers be given notice of the proceedings. As a result one of them, Cable & Wireless filed a short affidavit pointing out the adverse effects that a decision in favour of the appellants would have on them given the passage of time.

3. Part of the delay appears to be due to the fact that the parties completely misunderstood the nature of the appeal process, and filed voluminous affidavit evidence, including lengthy opinions from experts in the field. Both parties seem equally to blame for this, as they entered into it with enthusiasm, and even agreed that there should be cross-examination, without any reference to the Court (although I should stress that the attorneys who appeared before me on the appeal were not those responsible for its early conduct). Upon the matter coming before me I held a series of directions hearings in order to limit the scope of the evidence, and directed the preparation of a list of issues.

4. The procedure on appeals of this nature is governed by RSC Ord. 55. Such appeals will normally be conducted on the record, but additional evidence may be received if the court allows it2. Although they are said to be by way of re-hearing3, the extent of that will be curtailed where an appeal is only permissible on a point of law.4 Appeals under section 2 RSC Ord. 55, r. 7(2). 3 RSC Ord. 55, r. 3(1). 4 See the Supreme Court Practice, 1999 ed., at note 55/1/2 – “In many instances appeals may be limited by the statutory provisions creating the right of appeal to points of law, and thus the provisions of this Order 2

605 are limited to points of law or ‘mixed fact and law’ (whatever that latter expression may mean). I have dealt with all of this further in paragraphs 13 – 15 below.

BACKGROUND

5. The Act establishes the Telecommunications Commission (‘the Commission’) “Establishment of Telecommunications Commission 7 (1) There shall be established a body to be called the Telecommunications Commission which shall advise the Minister in the discharge of his functions under this Act and discharge such other functions as may be imposed or conferred upon them by or under this Act or any statutory provision. (2) The Commission shall consist of not less than five persons and not more than nine persons who shall be appointed as Commissioners by the Minister by notice in the Gazette to hold office during good behaviour for such terms not exceeding three years at a time as may be specified in their respective letters of appointment.”

6. Among the functions of the Commission are those conferred by section 24 of the Act, under which it is given the authority to review charges for existing telecommunication services6 either on its own motion or on receipt of ‘a complaint’: – “24. (5) On receipt of a complaint regarding a Carrier’s rates and charges, on the direction of the Minister, or of its own motion, the Commission may review the Carrier’s rates and charges after making such enquiry into the matter as the Commission may think fit and having regard to subsection (2), where applicable, and may give a direction changing the rates and charges for the service in whole or in part. concerning a rehearing will have no application (Green v The Minister of Housing and Local Government

[1967] 2 QB 606).”

5 “Appeals to the Supreme Court 60 (1) Any person aggrieved by — (i) a decision of the Minister under section 11, 13 or 25; or (ii) a directive of the Minister under section 15 or 22, may appeal on that account to the Supreme Court on a point of law or mixed fact and law.” 6 Section 2 of the Act defines telecommunication service as follows: "telecommunication service" means a service consisting in the conveyance of anything by means of telecommunication whether or not the circuits are provided by the service provider or others;” And ‘telecommunication’ as: "telecommunication" means any transmission, emission or reception of signs, signals, writing, images, sounds or intelligence of any nature by wire, radio, optical, or other electromagnetic system and cognate expressions shall be construed accordingly;” 3

(6) Before the Commission gives a direction under subsection (5) the Commission shall notify the Carrier of its intention and shall allow the Carrier fourteen days or such longer period as the Commission may specify to submit reasons in writing why the direction should not be given.”

7. An appeal lies from a direction of the Commission to the Minister: “Appeal to Minister against a direction of the Commission 25 (1) Any person aggrieved by a direction of the Commission under section 14, 22 or 24 may, within twenty-one days of being notified thereof, or such longer period as the Minister may allow, by notice in writing appeal to the Minister. (2) For the purposes of an appeal under subsection (1) the Commission shall, at the time of giving, or not later than three working days after the date on which it gives, a direction, give reasons, in writing, for the direction. (3) On an appeal under subsection (1) the Minister may, if he thinks fit, vary any direction of the Commission and such variation shall take effect on such date as the Minister may determine. (4) The Minister shall give his decision as soon as possible after the hearing of the appeal but in no case later than sixty days after receipt of the notice of appeal. (5) When a person gives notice of appeal to the Minister, the Minister may suspend the implementation of the direction of the Commission pending the outcome of the appeal.”

8. The background to this appeal is that, for various reasons, the Commission has been reducing the LAC since 19977, when it stood at 24¢ per minute of use. The latest round began in 2004, when the Commission embarked on a review of the LAC under section 24(5) of the Act, and eventually issued a directive of 5 November 2004, reducing the LAC from 5¢ to 4¢ per minute with effect from 1 November of that year. There was a subsequent clarification issued on 22 December 2004, which is not material to this appeal, except that as a result of it the time for appealing to the Minister was extended until 31st January 2005. There was then a successful appeal to the Minister, as a result of which the Minister, in a written decision of 1st April 2005, ordered the Commission to “conduct a more transparent and comprehensive review of BTC’s cost to provide LAC.” In that decision he also announced the appointment of an independent expert to collect and process the necessary data. 7 See the section headed “Procedural History” on p. 4 of the Reasons of 25th January 2006. 4

9. Following that the Commission did indeed reconsider the matter with the help of a consultant, Dr. Gabel. He had been involved in the original reductions to the LAC in 1997, but not in the interim. He conducted an investigation, and produced a report. This was done in two stages: a draft report was produced and given to BTC on 9th September 2005. They were given the opportunity to make comments on that, which they did on 20th and 26th September, and Dr. Gabel then issued his final report on 5th October 2005. There was then further consultation, and eventually, on 10th February 2006, the Commission issued a directive, this time reducing the rate “from 5¢ per minute to 3¢ per minute effective January 01st, 2006.” BTC again appealed, by letter of 21st April 2006, attaching submissions and three affidavits. It was BTC’s case that the rate should remain at 5¢ per minute. However, in a written ruling of 20th June 2006, the Minister rejected the appeal as being ‘without merit’ and sustained the directive fixing the LAC at 3¢ per minute. In doing so he addressed each head of the appeal in order and gave brief reasons for dismissing each of them. It is against that decision that this appeal is now brought.

THE ISSUES

10. The Notice of Motion by which this appeal was brought is dated 19th July 2006. It sought a reversal of the Minister’s decision, and declarations that the Minister had no jurisdiction to hear the appeal, with a long list of alternative remedies, the thrust essentially being that the 5¢ a minute rate be restored from 1st January 2006 and/or that the matter be referred back to the Minister or the Commission for “full and proper determination” and that in the interim the LAC be restored to 5¢ a minute.

11. Many of the original grounds of appeal were not pursued, and in any event they have been overtaken by the list of issues, which I directed.8 I directed such a list because the grounds were convoluted and repetitive; some were unparticularised9; and the expert evidence filed by the appellants ranged over a very wide area indeed. It was, therefore, necessary to identify exactly what points were really in contention. 8 See cl. 5 of my Order for Directions dated 30th May 2008 9 E.g. Ground 2(2)(c) “. . . the Directive was . . . irrational in that . . . it was apparently illogical or arbitrary, and supported by inadequate evidence and reasoning;” 5

12. The list of issues is structured under four heads, being (i) the nature of the review by the Supreme court; (ii) due process/fair hearing issues; (iii) the approach to non-traffic sensitive costs; and (iv) what remedy is available. The due process/fair hearing issues turn largely on the rejection of BTC’s data for the allocation of costs between voice and data transmissions, and the use of a model based on figures for that derived from Colorado. The issues on cost turn largely on the Commission’s assumption that non- traffic sensitive costs were covered by the monthly charge to subscribers, and its failure to mandate a proportionate reduction in the long-distance carrier’s rates. I now turn to each in detail. I have set out verbatim the appellant’s statement of each issue in bold type, and then my consideration of it.

NATURE OF REVIEW BY THE SUPREME COURT

1. In reviewing the decision of the Minister under section 60, what is the applicable standard of review?

13. I noted above that the parties filed extensive evidence and that that was inappropriate. It may be that they were misled by the use of the word “rehearing” in Ord.

55. That expression does not mean that the court will treat the matter as coming before it for the first time, and hear it anew. Its meaning was explained by May LJ in E. I. Dupont de Nemours & Co. v S. T. Dupont [2003] EWCA Civ 1368: “90. Rehearings on appeal under RSC Ords 55 and 59 were well understood not to extend to rehearings in the fullest sense of the word. The court did not hear the case again from the start. It reviewed the decision under appeal giving it the respect appropriate to the nature of the court or tribunal, the subject matter and, importantly, the nature of those parts of the decision making process which were challenged.”

14. In the case of appeals under the Act, the extent of appellate review is further restricted by section 60(1) to points of law or mixed fact and law. The latter expression, though well understood in the context of judicial review, is not a common expression to find in statutory provisions in this jurisdiction. It seems, however, that it is more 6

frequently used in Canadian statutes, and the Canadian Supreme Court has explained it as follows: “It is important to distinguish questions of mixed fact and law from factual findings (whether direct findings or inferences). Questions of mixed fact and law involve applying a legal standard to a set of facts. On the other hand, factual findings or inferences require making a conclusion of fact based on a set of facts. Both mixed fact and law and fact findings often involve drawing inferences; the difference lies in whether the inference drawn is legal.” Housen v Nikolaisen [2002] 2 SCR 235 at para. 24.

15. I do not think it particularly necessary to belabour this. I accept the appellant’s submission10 that the proper approach on an appeal such as this is: “. . . not the same as a judicial review but engages the merits of the decision . . . [it] is an appellate style review, but [one] where the Court is prevented from questioning pure factual findings. It can review the commission’s conclusions (namely that the rate was just and reasonable). In that review, the Court should give due deference to the original decision maker and only intervene if the decision is plainly wrong.”

DUE PROCESS/FAIR HEARING ISSUES

2. Was the Minister correct in concluding that the section 24 inquiry had been conducted in a fair manner? (i) Did the Commission act fairly in relation to BTC’s costs data:

1. Did the Commission reject BTC’s SME data, due to concerns about its being out of date and/or undocumented, without giving BTC an opportunity to provide updated and/or documented SME data?

2. Did the Commission pronounce a policy change, in terms of approach to SME opinion data, after rather than before the enquiry?

16. By way of background, an important issue was what it actually costs BTC to provide access to its system to the long distance carriers. There are two broad elements to this: (i) costs which are sensitive to the amount of traffic on the system; and (ii) costs which are 10 See the Skeleton Argument of Appellant of 13th August 2008, paragraphs 19 and 20. 7

not. But it is not enough to quantify these costs, because BTC’s lines are used for other purposes than voice communication, and in particular they are used for data transmission. In its submission to the Commission it had been BTC’s case that it had no means of dividing up its costs between voice and data, other than having its own in-house people (who are referred to throughout as Subject Matter Experts, or ‘SMEs’ for short) express an opinion on it. They only allocated 15% of switching investment to data, and BTC then applied that figure across the board. The Commission’s consultant, in his report of 5th October 2005, rejected this on three grounds, being that the SME assessments were (i) old, (ii) premised on a constant allocation to each portion of the system, and (iii) undocumented. The consultant called for a higher standard, citing the attitude of the US Federal Communications Commission (‘the FCC’).

17. It is BTC’s case that undocumented SME assessments had always been accepted in the past, and the Commission did not announce any change to this policy until the decision itself, when, in the undated Executive Summary11 the Commission said – “3. The Commission declares its intention, with this decision, to reduce its reliance on the opinion of subject matter experts. • Where parties have failed to provide sufficient and/or verifiable data for the Commission to properly conduct its deliberations, the commission finds that it is obligated to fill in evidentiary gaps in the record with its own data, inputs, and/or assumptions, using to the greatest extent possible, neutral and public sources.”

18. BTC argues that announcing this policy change after, rather than before, the inquiry is unfair. However, it is plain that this issue arose early in the process. The Commission’s expert, Dr. Gabel, came on the scene in April 2005. BTC complains that their request for information about the scope of his review went unanswered, but they can have had little illusion about what was at stake and the importance of accurately stating what it cost them to provide the relevant services. It seems that Dr. Gabel met with BTC on 27th April, although BTC would only proceed on a “without prejudice basis”. Dr. Gabel, in his evidence, said that from the outset of his involvement he explained his doubts concerning 11 It is not clear when that document was issued. In the core bundle it comes immediately after the direction of 10th February, but a reminder letter of 2nd February states that it was sent out on 25th January 2006. 8

the SME opinions, and he said he did so at the first meeting he had with BTC in April 2005, when he communicated to BTC’s costs analyst, Kevin Hadall, that the SME data was unacceptable12. I accept his evidence on that.

19. During the period 27th April to 9th September BTC gave Dr. Gabel “information that was attainable through our existing systems and processes13.” This included their cost model, which supported the existing 5¢ LAC. On 9th September they received a first draft of Dr. Gabel’s report, and this was discussed at a conference call on 16th September 2005, and as a result of that “BTC committed to produce further information to aid Dr. Gabel in finalizing his report” although they were only given a week to do so. There were then exchanges between Dr. Gabel and BTC’s expert, Dr. Parsons, and the deadline for providing information was extended, so that BTC filed further comments and information on 26th and 27th September 2005. Dr. Gabel’s report was then finalized on 5th October 2005, and then, by letter of 14th October, the Commission gave BTC an opportunity to make submissions on it by 21st October, which they did, while reserving their rights to object to the overall process and taking pains to remind the Commission that it must ‘apply its own mind to [Dr. Gabel’s] findings and recommendations.’ The Commission then issued its “Notice of Intent to vary the Local Access Charge” on 8th November 2005, and that gave 14 days for all concerned to submit written reasons why the direction should not be finalized, as required by section 24(6) of the Act.14 There were then further exchanges between BTC and the Commission, and the Direction to vary was not finally issued until 10th February 2006, at which time it was accompanied by reasons for their decision dated 25th January 2006. I accept that BTC did not see those reasons until they were sent to it by fax on 10th February 2006, as confirmed by the fax imprint.

20. Given that history, I do not think that BTC can be said to have been taken unawares by the rejection of their SME data. It was, in my judgment, for BTC to put forward their 12 This was his oral evidence on 18th August 2008, but he also referred to paragraph 17 of his second affidavit of 24th January 2007, which refers to a meeting Kevin Hadall of BTC on 27th April 2005. 13 See the first affidavit of Leslie Rans, at paragraph [13] 14 Section 24(6) of the Act provides – “(6) Before the commission gives a direction under subsection (5) the Commission shall notify the Carrier of its intention and shall allow the Carrier fourteen days or such longer period as the Commission may specify to submit reasons in writing why the direction should not be given.” 9

best case from the outset. They could have been under no illusions about the nature of the process, and the risk in which they stood, particularly given the earlier decision of the Commission to order a reduction, which had been overturned by the Minister. The supposed ‘policy change’ is no more than an indication of what the Commission would regard as acceptable evidence. They could in fact have rejected the SME evidence without elevating it to the status of a ‘policy change’ simply on the grounds of its unreliability. Moreover, and in any event, BTC had ample notice from Dr. Gabel that he regarded the SME evidence as unacceptable, and it could hardly have been surprised when his report rejected it. It had a further opportunity to provide more information after the publication of Dr. Gabel’s report, and indeed did so. In the event Dr. Gabel and the Commission were unpersuaded by BTC’s further submissions, but that has little bearing on the allegation of procedural unfairness, which I do not think is made out. (ii) Did the commission act fairly in relation to using comparator data?

1. did the Commission give BTC a proper opportunity to comment on the use of Colorado data

2. Was Colorado a reasonable choice of comparator

3. Was the Colorado data of a higher quality (in terms of documentation/support) than the excluded data from BTC?

21. Having rejected the opinion of BTC’s SMEs on the appropriate allocation of costs between voice and data, Dr. Gabel produced two alternative allocations, based on different methodologies, which gave respectively 46% and 67% to voice, as opposed to BTC’s allocation of 85% to voice. The methodology which Dr. Gabel preferred (‘loops weighted by cost of service’), and which gave the lower allocation of 46% to voice, took into account that providing data services costs more, because special line conditioning and electronics are often required. In assigning these costs he used ratios derived from a study done for Colorado. The appellants strongly object to the use of that data.

22. The opportunity for BTC to comment on this was both before and after the publication of Dr. Gabel’s final report, and they availed themselves of that opportunity on 10

both occasions. During that period BTC supplied Dr. Gabel with information from their West Exchange which showed that for that Exchange only 1.65% of the copper loops were used to provide data. On 20th and 26th September 2005 BTC provided comments on Dr. Gabel’s draft report, which was issued on 9th September 2005, and on 21st October they submitted detailed submissions on his final report, which had been issued on 5th October 2006. In those submissions BTC argued that the Colorado data was inapplicable and led to an unreasonably high allocation to data. They pointed out that had South Carolina been chosen, a much lower ratio would have resulted. In the face of that it is hard to say that they were not given an opportunity to comment on the use of the Colorado data. Indeed, in their submissions on the appeal to the Minister, BTC stated “38. It is to be noted that the irrationality of the Colorado approach was pointed out to the Commission and the expert time and again during the process, as appears from the current Rans affidavit, in particular at paragraphs 9 (bullet four), 19 (bullet 3), 22, 28 to 30 and 51 (bullets one and two). The expert and the Commission simply ignored these concerns.” It seems to me to follow from that that BTC did have a proper opportunity to comment. Again, the fact that their arguments were not accepted has little direct bearing on the question of procedural fairness.

23. As to whether it was reasonable to use Colorado, as opposed to some other state, that was primarily a matter for the Commission. In its reasons, the Commission explained that it had chosen Colorado because its rates “had recently been reviewed and approved by the FCC”, and thus “reflect recent improvements and efficiencies in modeling techniques” and so “provides the commission with a reasonable and publicly available benchmark”. They said that they had had to do this as “BTC failed to adequately control for the higher costs associated with its data services”. There are numerous factors either way, which have been set out at length in the contending affidavits, but at the end of the day this was a matter for the Commission, and I do not think it can be said that they acted irrationally in their choice of comparator. 11

24. The final issue posited here (whether or not the data was of a higher quality that the SME data) is to compare apples and pears: it is not a question of the quality of the Colorado data, but whether it provided a fair analog for Bermuda in the absence of local data which the Commission found acceptable. It is, therefore, just another way of asking whether it was reasonable to use the Colorado data. Again, there are numerous and complex arguments either way. Again, this was a matter for the Commission, and was essentially a matter of fact: it was but one step in their assessment of BTC’s costs of providing the service. (iii) Did the Colorado data have a de minimis impact

25. On appeal the Minister said that “the choice of Colorado . . . had a de minimis impact on the estimated cost of providing access to the Class A carriers.” The ultimate source of this assertion was Dr. Gabel, as he made clear in his evidence. According to the appellants’ calculations, based on Dr. Gabel’s evidence filed after the Minister’s decision, using the SME data rather than Colorado would have increased the LAC by 0.09¢, which equates to some $90,000 per annum. Given that Dr. Gabel calculated the incremental cost of service at 2.70¢ per m.o.u., and that the Commission then appears to have rounded that up to 3.00 cents, 0.09 cents is, in my view, de minimis.

26. In any event, in my judgment the proper way to deal with this is for BTC to stop spending considerable amounts of time and money in arguing about the Colorado data, and to produce hard data of their own which will stand up to critical scrutiny. That leads to the next issue. (iv) Did the Commission act properly in making a final decision, rather than an interim measure, based upon comparator data

27. It is not apparent whether the Commission were ever invited to make an interim decision, but in any event the appellant’s case is that, where resort has to be had to a comparator to supply missing data, the rate which is fixed on such a basis should be interim not final, and it should include appropriate adjustment mechanisms in case the 12

final rate is higher. The appellant relies upon a ruling of the US FCC of 9th June 199715 (the ‘Expanded Interconnection’ decision), where the FCC did just that, and they complain that while the Bermuda Commission relied upon that case to support the use of surrogate data, it did not follow the practice established by it.

28. However it was and is open to BTC to supply new and better data to the Commission in order to justify a higher rate. Indeed, the Commission, in the Executive Summary (supra), expressly invited BTC to do just that: “4. The Commission finds that the data proffered by BTC failed to adequately disclose the relative cost of providing narrowband versus broadband services. • The Commission encourages BTC to submit other cost data in future proceedings, such as a long-run incremental or historical cost study of its operations, which identifies the relative cost differences between narrowband and broadband services. • Until such time as BTC addresses these concerns, the Commission finds that the UNE cost data from Colorado provides a reasonable and publicly available benchmark to evaluate the relative cost of providing narrowband versus broadband facilities.” It seems, therefore, that the decision is, in a sense, interim in that it is open to reconsideration on the production of better data, and there is, therefore, no finality as such about the rate.

29. To the extent that this issue is about the absence of any compensatory mechanism should BTC succeed in producing data to persuade the Commission at some future date that the rate was set too low, it seems to me that it would still be open to the Commission to deal with this. In fixing a new rate it would be open to the Commission, if it thought it appropriate, to include a compensatory adjustment. If they declined to do that, the matter could be addressed on a prompt and focused appeal. The first step in such a process is obviously for BTC to go back to the Commission with hard data, if it has it. 15 FCC 97-208, CC Docket No. 93-162. “In the matter of Local Exchange Carriers’ Rates, Terms, and Conditions for Expanded Interconnection Through Physical Collocation for Special Access and Switched Transport.” 13

(v) Was it appropriate for the Commission to take into account submissions by BTC’s competitors, which BTC never saw and has never seen?

1. Did such submissions have a de minimis influence?

30. I understood this to be abandoned once the submissions were produced.

APPROACH TO NON-TRAFFIC SENSITIVE (‘NTS’) COSTS

3. Was it right that, as of January 2006, Class A carriers should pay only 15% of the NTS costs of providing them with local access a. Was the Commission right that their decision was in line with FCC practice b. Did the Commission assume that the residential monthly charge was calculated16 to cover the NTS costs of the local network i. Was this a reasonable assumption ii. Was BTC given any or any reasonable opportunity to rebut this assumption

31. By way of introduction to this issue, some part of the cost of providing access to BTC’s system varies according to use, and so is ‘traffic sensitive’. The cost of other parts, and in particular the cost of the actual lines (or ‘loops’), is fixed and does not vary according to use. It is accepted by all sides that the long distance carriers should pay their share of the traffic sensitive costs. The debate is about whether they should pay for using the fixed system, the cost of which is not traffic sensitive and which is there anyway. BTC and its experts argue strongly that they should. However, it appears that the FCC in the US has been moving away from this for some time. This is explained in the overview given by Mr. Lipman (who was one of BTC’s expert witnesses) in his affidavit in the prior appeal. At one end of the spectrum are ‘fully allocated costs’, which is the system 16 The insertion of the word “calculated” in the statement of this issue may be misleading, for the Commission did not assume that the monthly charge was ‘calculated’ or intended to cover the NTS costs of the loop, but simply that it in fact did so. 14

used and preferred by BTC. At the other end are ‘incremental costs’, which represent the approach favoured by the Commission. Mr. Lipman explained it as follows17: “17. In its access charge proceedings, the FCC has long expressed a goal of seeking rates that reflect “cost causation,” that is, it seeks a rate design in which rate elements are closely related to the underlying costs. Historically, access charges in the United States were based upon fully-allocated costs, and recovered through per-minute access charges (as they are today in Bermuda). The FCC realized in the early 1980’s that this rate structure did not reflect cost causation, both because fully-allocated costs exceed (by a considerable margin) the economically-relevant incremental costs of providing service, and because per- minute charges should not be used to recover costs that do not vary directly with volume of usage (i.e., so-called “non-traffic-sensitive” or “NTS” costs). However, the FCC also recognized a competing policy goal of avoiding rate shock to consumers, and therefore did not seek to force rates immediately to levels based solely on incremental cost. Indeed, even today, more than 20 years after the FCC first articulated these principles, U.S. access charges still have not entirely achieved the cost-causation goal.”

32. In his report to the Bermuda Commission, Dr. Gabel gave them a choice between these two approaches. In fact he posed six choices, or ‘scenarios’, but for the purposes of this issue the choice was between Scenarios A and C, which were as follows: “A. Scenario A Position 1 – Worldwide, regulatory commissions have been pushing down access fees to the level of traffic sensitive costs, and thus, requiring international long distance carriers to pay no more (or less) than the incremental cost of service. This practice has been characterized by the FCC and other regulatory commissions as efficient. The assumptions used to allocate BTC’s loop, switching and transport costs to the LAC for Scenario A are consistent with this practice and appear below [Figures Omitted]. . . . C. Scenario C Position 2 – The second distinct position for the commission to consider is one with fully-allocated costs. This scenario reflects the opinion that international long distance carriers use BTC’s complete telephone platform, not just the traffic sensitive portions, and thus, it is reasonable to ask the international carriers to make a contribution towards the cost of the entire platform used to originate or terminate an international call. Scenario C reflects this position. In practice, Scenario C differs from Scenario A only in that all voice related loop and switching costs are assumed to be traffic sensitive. [Figures Omitted]” 17 See his affidavit of 27th January 2005 at paragraph 17. 15

33. Thus Scenario A represented the FCC’s preferred (but unattained) approach of fixing the LAC by reference only to incremental costs, while Scenario C represented fully allocated costs, the approach favoured and used by BTC. However, as Mr. Lipman said, “fully-allocated costs exceed (by a considerable margin) the economically-relevant incremental costs of providing service.” The Commission chose Scenario A (with a modification in BTC’s favour which I will come to shortly), and gave these reasons: “As noted in the Report, the price floor for the LAC is arguable no lower than the costs caused by toll traffic on BTC’s network. Some regulatory commissions, such as the FCC, contend that “. . . non-traffic sensitive costs – costs that do not vary with the amount of traffic carried over the facilities – should be recovered through fixed flat charges, and the traffic sensitive cost should be recovered through per minute charges. “The Commission’s pricing decision, which is founded on the cost analysis provided in the Report, is consistent with this principle. Regulatory commissions that have established local access charges that are above cost, have done so in order to cover the cost of the loop that is not recovered in the price of exchange traffic. However, since the cost data available to the Commission suggests that BTC’s current residential subscription charge covers 100% of the cost of the loop, it is therefore not apparent why the Class A carriers should be required to make a large contribution to the non-traffic sensitive costs as they are already recovered in full from the subscribers to basic voice services.” Given Mr. Lipman’s explanation, it is hard to see how BTC can challenge this in principle.

34. If the principle behind Scenario A is sound, then BTC are driven to hang their challenge on the last two sentences in Mr. Lipman’s statement quoted above, which notes that, more than 20 years after the FCC first articulated these principles, U.S. access charges have still not entirely achieved the cost-causation goal, because of the need to avoid ‘rate shock’ to consumers. However, it is also very clear that the Bermuda environment, and in particular its rate system, are very different from the US, and it may well be that the historical factors impeding the achievement of the goal of setting the LAC by reference to incremental costs alone in that country do not apply here. In 16

particular, if it is indeed the case that the local subscription charge already covers the cost of the NTS portion of the system as the Reasons state (supra), then it is not clear why there should be any ‘rate shock’ at all.

35. Much of this turns, therefore, on whether the current residential charge covers the NTS costs of the system. It is BTC’s case that this was not in fact true. Mrs. Rans (the CFO of BTC) gave evidence that various important items, which had been in their original submission, had been left out of the calculation of the cost of the loop. These included maintenance, billing and customer service. She calculated that they added a further $16M, which had not been included in Dr. Gabel’s calculation, and that when these were included the $26 monthly charge did not cover the cost of the loop. Dr. Gabel said that he had calculated the cost of the loop at $12 - $14 per month, and that included maintenance, and that the remaining items did not take the cost above $26. I do not think that I can or should attempt to resolve this dispute on this appeal – it is a pure question of fact. I am left, therefore, with the Commission’s finding that the residential subscription charge covers the cost of the loop. On that basis, all the arguments based on rate shock fall away, and there is no compelling reason why the Commission should not have adopted Scenario A.

4. Did the Commission give adequate reasons for its decision that Class A carriers should pay only 15% towards the NTS costs of providing local access

36. As I understand it, this refers to the difference between the rate in Scenario A ($0.027) and the rate actually adopted by the Commission ($0.03), which the Commission explained18 – “This rate is based on the cost estimate provided as Scenario A in the Report plus a markup for an additional contribution to common costs. The commission is of the opinion that Scenario A is the appropriate cost estimate because this rate compensates BTC for its direct costs, including a return on investments, and makes a reasonable contribution to shared and common costs.” 18 See the Commission’s Reasons dated 25th January 2006, at p. 16 (core bundle p. 114 at 129) 17

There is, therefore, an element of contribution towards common costs. It is not clear how it was fixed, but that does not assist the appellant, because the arguments adopted by the Commission would have justified accepting the $0.027 rate, and the fact that they have allowed a minor uplift from that is in the appellants’ favour.

5. Did the commission give any or any adequate reasons in deciding that Class A carriers should pay 0.3 cents towards the common costs of providing local access?

37. For the reasons given above, I consider that the Commission gave adequate reasons for choosing Scenario A. The additional 0.3¢ uplift was in BTC’s favour and called for no further explanation.

6. In deciding that Class A carriers should pay only 15% towards the NTS costs of providing them with local access, did the Commission have regard to the statutory factors listed in section 24: i. Did it have any proper or any regard to BTC’s actual costs of providing local access to the Class A carriers;

1. Are NTS costs part of the direct costs of local access; ii. Did it have proper or any regard to ensuring a reasonable return for BTC’s investors.

1. Was it reasonable for the Commission to rely upon its proposal that BTC could rebalance its rates to make up any revenue short fall.

2. is the Commission entitled under section [24] to take into account a carrier’s income from other services (in this case, the monthly charges paid by BTC residential customers) iii. Did it have proper or any regard to the public interest 18

1. Can it be in the public interest for residential customers to pay for local access by international carriers; a. Are the majority of the Class A carrier customers business customers

38. The Act sets out certain statutory factors to which the Commission is to have regard in fixing a rate: “(2) In the exercise of their discretion under subsection (1), or subsection (5) the Commission shall have regard inter alia to — (i) the cost to the Carrier of the service in question; (ii) the desirability and need to subsidize other non-profitable services offered by the Carrier in the public interest; (iii) the needs of the Carrier for adequate working capital and to establish reasonable reserves; (iv) international accounting standards where applicable and modifications thereto; (v) technological advances; (vi) market conditions in Bermuda and overseas including the likely impact of a new service on other Carriers; (vii) regulatory changes where applicable; (viii) investment; the need to afford investors a reasonable rate of return on their (ix) the question whether in the light of the foregoing, any proposed tariff is just and reasonable and any rates and charges contained therein are applied equally to all persons in substantially similar circumstances and conditions; (x) the question whether the Carrier, in respect of the application of rates and charges, the provision of services and the use of its facilities— (a) gives any preference or advantage to any person or to any particular description of telecommunication: (b) subjects any person or any particular description of telecommunication to any disadvantage; 19

(xi) the public interest.”

39. I have already dealt with the question of the cost of the service. The commission plainly took that into account in the ways set out above, I cannot say that they were wrong in principle in their approach, for the reasons already given.

40. As to the investor’s rate of return, Dr. Gabel based his calculations upon an allowable rate of return of 9%19. On its face that suggests that he, in making his recommendations, and the Commission, in adopting his Scenario A, considered the question of rate of return and took it into account. There is nothing to suggest that 9% is an inappropriate or inadequate rate20 .

41. As to rebalancing, that is the upward adjustment of other rates to cover any loss on the LAC. In its Reasons21, the Commission offered this, although noting at the same time that it did not believe it necessary: “Although the record indicates that the LAC adopted in this order is compensatory and BTC’s monthly connection charge covers 100% of its loop costs the Commission will nonetheless provide BTC the opportunity to submit a rebalancing proposal within 2 weeks of the publication of this order if it so chooses. We note that BTC’s rebalancing proposal need not be limited to increasing the price of basic local voice services.” BTC has not done that to date, arguing that it is wrong in principle. I do not see why that should necessarily be so, but it is impossible to say without seeing what would be entailed. But in any event, I do not think that BTC can challenge the Commission’s factual finding that the monthly subscription charge covers the cost of the NTS part of the system, and on that basis no question of rate rebalancing arises. 19 See page 28 of his Report” “All six scenarios . . . assume . . . a rate of return of 9%.” This is justified by the arguments in Section IV of his Report, pp. 27 – 28, under the heading ‘The Cost-of-Money’. According to that, BTC’s suggested rate of return was 8.65%, but Dr. Gabel argued for a higher rate. 20 BTC’s actual rate of return on its whole operation has been consistently higher than this since the reduction: Mrs. Rans accepted in cross examination that in y.e. 2008 BTC made profits of $12.6M on assets of $89.2M, and in y.e. 2007, profits of $10.8M on assets of $83.4M, giving an overall rate of return of 14% and 13% respectively. 21 See the Reasons of 25th January 2006, at p. 18. 20

42. The appellants, in stating the issue, also seek to argue that it is not appropriate to take into account the monthly subscription charge. To illustrate this, Mr. Elkinson advances an analogy based on the cost of a bus service. His argument is that, if the fares paid by local passengers on Bermuda’s bus service covered the cost of service, that would be no justification for letting tourists ride for free. The flaw in the analogy is that, if the NTS cost of the telephone service is already being paid for by the subscriber, then including an element in respect of the NTS costs in the LAC means that the subscriber is paying twice, because in the end the LAC gets passed back to the subscriber in the costs of the long distance calls which he originates and has to pay for. I think, therefore, that it is perfectly proper for the Commission to take into account the monthly charges paid by BTC’s subscribers.

43. As to the public interest generally, in argument and in the evidence, various points were made about the lay-offs and other cut-backs occasioned by the reduction in the LAC. If these are still live issues, then the answer is that the statute entrusts the decision of what is in the pubic interest to the Commission. It is for them to balance the benefits of a reduction in the rate against such effects. In doing so, one factor may well be a need to compel business efficiency. It is not for the Court to substitute its own judgment on this for that of the Commission, and there is nothing to show that in approaching its task the Commission got it patently wrong.

44. However, on the face of the statement of issues, the public interest point has come down to a misconception of the arguments in support of the Commission’s decision. The decision was based in part on a finding of fact that the fixed costs of the loop are already covered by the residential rental charge. That does not mean that residential customers in some way pay for international calls made by business customers, as the statement of the issue seems to imply. The point made was that business rates are higher than residential rates, but the residential rate alone covers the cost of the loop. Basing the calculation on residential rates was therefore conservative. 21

2. Was it reasonable to conclude that reducing the LAC was in the public interest since reductions would filter down to customers through reductions in Class A carrier rates; a. Did the Class A carriers reduce their rates.

45. The Minister and the Commission proceeded on the basis that price rivalry between the long distance carriers and the increasing impact of voice over internet providers, would combine to compel the long distance carriers to pass on the price reduction to their end-users22. However, the evidence is that that did not happen. According to Mrs. Rans, on the basis of her research into the publicly available figures, the Class A carriers did not reduce their rates23. Cable & Wireless Bermuda Ltd., one of the long distance carriers, filed a short affidavit but did not appear or participate in the proceedings. However, in its affidavit it said nothing about this issue, instead dealing with the amount of any repayment which BTC might demand, and the fact that it had negotiated interconnection rates with other carriers on the basis of 3¢ cents, and would be disadvantaged if the price now went up to 5¢.

46. In strict logic, the fact that the reduction was not passed on does not necessarily make the assumption that it would be an unreasonable one. Even reasonable assumptions can sometimes prove to be false. It is, however, a fairly compelling indication that the Commission and the Minister proceeded upon an erroneous basis in this respect. That does not, however, mean that the reduction in the rate was wrong if it was otherwise sustainable in principle. Nor should the point be considered in a vacuum. As the affidavit from Cable & Wireless indicates, the impact of the LAC on the Class A carriers may be more complex than simply providing them with a reduction which they should not be allowed to keep. The issue needs proper consideration in the round, and the appropriate body to do that is the Commission. There are various ways of achieving that: the appellant can take this new fact back to the Commission and ask them to reconsider the 22 See paragraph (1) a. iii. Of the Minister’s Decision. 23 See her supplementary witness statement of 12 August 2008. 22

LAC in the light of it; they can make a complaint to the Commission about the Class A carriers’ rates; or the Commission can consider those rates of its own motion on the basis of this new evidence. Given the lapse of time (which I address further below), I consider that some such course is preferable to my interfering with the Commission’s decision on this basis alone.

REMEDY

7. If the appeal is allowed, what remedy is available to BTC?

47. In view of the outcome, this does not arise. However, lest I am wrong, I indicate that, had I had to consider the question of remedy, it is likely that I would have refused BTC any retrospective remedy, given the delay. Any decision on this appeal affects third party rights, being those of the long distance carriers, and they were not given notice of the proceedings until very recently. In my view they were properly parties to this appeal, being directly affected by it, and should have been given notice at the outset. Indeed, Ord. 55, r. 4(1)(a) requires service of the notice of motion by which an appeal is brought to be served on “every party to the proceedings (other than the appellant) in which the decision appealed against was given.” In this case that must mean everyone who made representations to the Commission. This was expressly accepted by BTC’s attorneys in their appeal letter of 21st April to the Minister: “17. Here of course, as already pointed out, there are no adverse “parties” as there would be in court proceedings, or in regulatory proceedings where, for example, one party has made a statutory complaint against another . . . However, given all the circumstances, it is submitted, one might perhaps say that the “parties” here are BTC . . . ; the Long Distance Carriers, which stand to benefit hugely from this order; and the public . . . ”

48. The failure to serve the long distance carriers meant that, not having notice of the proceedings, they could not intervene to correct the delay or the procedural irregularities which contributed to it. In those circumstances I would consider it wrong to make any change which would adversely affect their position in the intervening years, and I would, therefore, only have considered a prospective order. 23

SUMMARY

49. For the reasons given above, I do not consider that there was any procedural unfairness in the way the Commission handled this matter, and I cannot say that its decision, or that of the Minister on appeal, was plainly wrong. I therefore dismiss the appeal. I will hear the parties on costs. Dated this 6th day of October 2008. ______________ Richard Ground Chief Justice 24