Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats by contacting us.
Volume 2, 12 July 2011
TRANSCRIPTION OF PROCEEDINGS BEFORE THE CANADIAN RADIO-TELEVISION AND TELECOMMUNICATIONS COMMISSION
Review of billing practices for wholesale residential high speed access services. Notice of Consultation CRTC 2011-77, 2011-77-1 and 2011-77-2
140 Promenade du Portage
12 July 2011
In order to meet the requirements of the Official Languages Act, transcripts of proceedings before the Commission will be bilingual as to their covers, the listing of the CRTC members and staff attending the public hearings, and the Table of Contents.
However, the aforementioned publication is the recorded verbatim transcript and, as such, is taped and transcribed in either of the official languages, depending on the language spoken by the participant at the public hearing.
Canadian Radio-television and Telecommunications Commission
Review of billing practices for wholesale residential high speed access services. Notice of Consultation CRTC 2011-77, 2011-77-1 and 2011-77-2
Konrad von FinckensteinChairperson
Crystal HulleyLegal Counsel
James WilsonLegal Counsel
Tom VilmansenHearing Manager and Manager, Costing Methods and Tariffs
140 Promenade du Portage
12 July 2011
- iv -
TABLE OF CONTENTS
PAGE / PARA
2. CIPPIC-OpenMedia 250 / 1482
9. Cogeco Cable Inc., Quebecor Media Inc. on behalf of its affiliate Videotron and Rogers Communications Partnership (collectively, the Cable Carriers) 315 / 1855
10. CNOC-CORC (Canadian Network Operators Consortium Inc.) 400 / 2363
CONSUMER PANEL (WESTERN PROVINCES)
13. Daniel Finnis 493 / 2977
14. Jesse A. Whitnack 494 / 2988
- v -
PAGE / PARA
Undertaking 296 / 1713
Undertaking 306 / 1788
Undertaking 308 / 1800
Undertaking 310 / 1818
Undertaking 396 / 2334
Undertaking 428 / 2520
Undertaking 437 / 2586
--- Upon resuming on Tuesday, July 12, 2011 at 0903
1478 LE PRÉSIDENT : Madame la Secrétaire, commençons.
1479 THE SECRETARY: Good morning, Mr. Chairman. Good morning, everyone.
1480 We will start today with the presentation from CIPPIC (Samuelson-Glushko Canadian Internet Policy & Public Interest Clinic) and OpenMedia.
1481 Please introduce yourselves for the record, after which you have 20 minutes to make your presentation.
1482 MR. ISRAEL: Thank you, Madam Secretary.
1483 Good morning. Mr. Chairman, Commissioners, my name is Tamir Israel and I am staff lawyer with CIPPIC, the Samuelson-Glushko Canadian Internet Policy & Public Interest Clinic.
1484 Here with me today is Steve Anderson who is the Executive Director of OpenMedia.ca.
1485 I just wanted to thank you for giving us the opportunity to come here today to participate in this proceeding. It touches on issues that are very important to OpenMedia.ca and to CIPPIC, so we are grateful for this opportunity.
1486 To start off this morning, Steve is going to talk a bit about some of our concerns that are raised by this proceeding, and I will address some of the questions you have put to us in your Commission letter last week in more detail.
1487 MR. ANDERSON: Thank you for having me and for holding this important hearing.
1488 I am here in the interest of perhaps the largest lobby in the country, the Canadian public. Over the last few months, 491,000 Canadians have signed OpenMedia's Stop The Meter Petition. Of those, nearly 100,000 followed up and sent submissions in for this hearing.
1489 As you know, that is an unprecedented level of citizen engagement for a telecom issue. The public cares, the public is watching, and they're depending on you to make the right choices.
1490 Acting on behalf of Canadians, OpenMedia and CIPPIC have put forward a submission that is deeply fact-based but also informed by these hundreds of thousands of Canadians, from policy experts, to small business owners, to independent ISPs, to people with disabilities just trying to scrape by. Their concerns are valid and there is clear evidence to back them up.
1491 Based on research and clear evidence, the outcome of this hearing should be a wholesale pricing regime that is cost-based, transparency-based and fact-based.
1492 Firstly, cost-based so that Canadians have a real choice for Internet access, wherein independent ISPs, with pricing autonomy, offer competing business models serving as alternatives to incumbent telecoms that currently dominate. Cost-based to enable competition and thus lower prices. Cost-based because we cannot afford to allow a few companies to stifle competition and impose a stranglehold on the critical Internet access market.
1493 Secondly, the outcome of this hearing must be transparency-based because it's crucial we have a market based on actual costs and not inflated numbers. We've seen several pricing schemes floated by Bell, the most recent being 19 cents per gigabyte, down from the previous ask of up to $2.50 per gigabyte. We clearly need to have some kind of mechanism in place that shows us the actual costs for telecom companies, whether it be through public disclosure and public scrutiny or independent costing audits.
1494 And lastly, the outcome of this hearing must be fact-based. If we are going to move forward in a positive direction, and I really think we can, we need to operate based on the actual reality before us. We can no longer allow decisions to be clouded by mischaracterizations of industry realities.
1495 FACT 1: We are simply not facing a bandwidth crisis or an explosion or anything like that. Growth rates have actually decreased in recent years, slowing to about 40 percent per year.
1496 FACT 2: Big telecom's investment in network infrastructure in Canada as a percentage of revenue has failed to keep pace with our OECD counterparts.
1497 FACT 3: ISPs can deal with demand through network investment. As we saw yesterday by the testimony by TELUS, they seem to be doing fine with network investment. We don't need metered pricing to deter customers from using the Internet.
1498 FACT 4: Cutting-edge business models -- from cloud computing to real time interactive media to peer-to-peer applications -- require more Internet usage, not less. We should not be deterring Canadians from using more Internet.
1499 As the recent OpenMedia.ca report "Casting An Open Net" makes clear, Canada is falling behind on several key Internet metrics when compared with our OECD counterparts, most strikingly in price. The thing Canada does lead on is capping Internet usage, and that's something none of us should be proud of.
1500 In countries like Japan, the U.K., Sweden and Chile, open access policies, including those that provide independent ISP pricing model autonomy, play a critical role. Open access policies, market choice and mechanisms to ensure ISP transparency are pushing these countries ahead of Canada in telecom performance.
1501 Functional marketplaces meet demand by increasing supply, not by squashing demand and seeing who can gouge Canadians the most. It is not okay for big telecom companies to invest less of their revenue than their global peers and then claim they need to gouge Canadians to make up for this waning investment. To do so would be to reward failure. The impact of this approach on Canadians is real and sometimes severe.
1502 For example, listen to the submission from Vince Dwyer, and I quote:
"I am a disabled father of 7 and [a] metered internet is unfair for parents of multiple children of school age. They all use the internet as a learning tool and it is mandatory for some of the classes. I have 7 children and get a disability cheque of 1300 hundred dollars a month. I'm barely scraping by as it is. Please stop the metered internet from happening as lower income families will suffer and our children will suffer."
1503 It is the stories of real Canadians, not the narrow, self-serving narrative produced by big telecom, that the Commission should listen to. I've heard from photographers, graphic designers and video makers who are being priced out of their craft. I've spoken to businesses that are unable to invest in online services.
1504 For example, Andrew Moore Crispin wrote to you:
"I work at butterscotch.com, a division of Tucows. ... Our online video network that focuses on helping people do more with the technology they own with video tutorials and helpful how-tos would not be viable in a climate where Internet users are worried about exceeding draconian and artificially low bandwidth caps."
1505 There's also Aaron Sinclair who wrote:
"Dear CRTC, I am an independent music producer and I depend on the internet for my livelihood. The arts, media and technology is one of the new ways for Canadian culture to thrive. ... The focus should not be on finding ways to make the internet cost more but on finding ways to make it faster and more accessible. ... I believe the CRTC should reverse its previous UBB rulings and allow independent ISPs to choose their own customer billing solutions."
1506 It's time to listen to Canadians. And Canadians are not just vocal, they are extremely well informed, as I am sure you have realized through the consumer testimony already.
1507 I invite you all to take a look at the OpenMedia's Facebook page. We have over 60,000 active contributors. With this page and social media in general, the Commission has a new and unprecedented opportunity to listen to Canadians who are remarkably well informed about this issue.
1508 What ties everything I've heard from Canadians on this issue together is what accounts for the largest citizens' movement in the country right now. That's the value of opportunity. The Internet represents opportunity: opportunity for a meaningful and rewarding job, opportunity for independence and autonomy, opportunity for a dynamic economy.
1509 The pro-Internet community and Canadians, writ large, are calling on this Commission to throw open the gates of opportunity that the Internet can provide on both a personal and a national level.
1510 You have heard some facts from me. You will hear more from my colleague here. The unprecedented 100,000 Canadians that asked you to enable pricing choice and the 491,000 that asked you to stop the meter await your decision based on these clear facts. I trust the Commission will make the right choice. Thank you.
1511 MR. ISRAEL: So, as you may have gathered from Steve's presentation, we are really concerned with the impact that UBB is having on consumers and also on online innovation.
1512 Of greatest concern for us is the rapid rate at which monthly caps have been dropping in the past year and a half or so on an industry-wide basis. This, to us, is indicative of a troubling trend and it comes at a time when annual wireline traffic growth rates are extremely modest and should pose no serious challenge to incumbents' investments.
1513 We have put a fair amount of evidence on the record documenting our concerns and I am not going to revisit it here in detail.
1514 In brief, we are concerned that at a time when annual traffic growth rates are modest, monthly usage allowances are not only failing to keep pace with that modest growth rate in the sense that they should reflect -- they should be rising to reflect the general trends in usage, but they are actually falling. In particular, the last year and a half or so has seen an industry-wide drop.
1515 These monthly caps now affect all levels of usage, and customers do not seem to like them. This is evident from the high level of engagement that this issue has attracted.
1516 The reason customers do not like these pricing tiers is because they are non-transparent, non-predictable and put customers at a real disadvantage in forcing them to guess how much they are going to be using ahead of time.
1517 They also burden emerging innovative online services, ranging from online delivery of video game purchases to online video streaming sites, to cloud computing/storage networks, with added costs.
1518 To foster innovation and customer choice, competitive models that encompass different pricing approaches are necessary. It's with this in mind that we turn to an assessment of the tariffs that are before you in this proceeding.
1519 At the outset, we do not think there is justification for the application of an economic ITMP with respect to wholesale. Given that GAS and TPIA tariffs have traditionally included cost-based components that are intended to reflect usage, this is, in our view, the best mechanism for ensuring incumbents are compensated for their wholesale services and not an economic ITMP. I will elaborate a bit in a bit.
1520 As noted above, there is no current phenomenal growth rates. For the last several years, wireline traffic has grown at a steady but modest 35-50 percent annual rate. This is far less then the 100 percent growth rate that has characterized much of the history of the Internet and falls far short of historical projections for this period.
1521 At the same time, the cost of network investment has been dropping. Falling annual traffic growth rates combined with falling equipment prices do not suggest out of control costs that require new pricing models.
1522 In fact, taking Bell as an example, its annual investment in wireline has remained steady since 2006, rising at about 1 percent compound aggregated growth rate from 2006 to 2010. With this constant level of investment, it was able to meet the roughly 45 percent traffic growth rates on its network while deploying new high-speed services such as fibre to the node.
1523 There has been no evidence of unmanageable network costs and we do not see justification for incentives to flatten usage across wholesale competitors based on that.
1524 As Steve noted, we set out our criteria for wholesale tariffs to be cost-based, fact-based and transparent. To the extent possible, we assess some of the tariff proposals before the Commission in light of their ability to meet these factors.
1525 A wholesale tariff should aspire to best emulate the actual costs incurred by the incumbent in provision of the service plus a reasonable mark-up. This includes any usage-based component included in such a tariff. Once a pricing model of this kind is in place, there is no longer any need to worry about incentives to control usage as the incumbent will be reimbursed for all usage.
1526 While we are not convinced that a variable usage-based wholesale component is required at this point, of the proposals currently before the Commission, CNOC's appears to most closely approximate actual incumbent costs of service provision in a number of ways in comparison to the AVP proposal, which is the other major proposal for variable rates.
1527 First, the AVP proposal, while a vast improvement over its predecessor -- which was an end-user-based economic ITMP -- is not designed to be cost-based but rather to put in place incentives to ensure usage control. We see that as a key distinction between the two.
1528 Second, CNOC's proposal avoids the double-dipping problem inherent in the current AVP proposal, which includes capacity costs in both the fixed and the variable usage components of its tariff. CNOC's proposal departs from this by disaggregating each fixed component -- port cost, customer connection cost, variable usage-driven costs -- and in this respect more closely aspires to our cost-based principle. This model will more closely replicate actual incumbent usage-based costs in that it will not be dependent on the end-user connection speed but rather on the degree to which that connection is utilized within the network if properly implemented.
1529 Third, the CNOC proposal is superior in that the risk allotment it places on competitors emulates that experienced by the incumbents. The competitors will provide the incumbents with their own usage projections and hence will not be forced to purchase in excess of their monthly usage or face stringent penalties. Instead, both CLECs and ILECs will be operating from the same types of usage projections that ILECs use in making provisioning decisions, and this is explained by CNOC in their proposal. I am just going to quote:
"The AVP proposal is unfair to wholesale customers on an individual basis. Each wholesale customer bears the risk of paying for usage it never uses if it overbuys pre-purchased usage blocks, and it bears the risk of paying more than necessary for post-paid usage if it underbuys pre-purchased blocks. A much better approach would involve having GAS customers provide rolling semi-annual usage forecasts to the Bell Companies on a monthly basis. Such forecasts in the aggregate would be more accurate than individual company forecasts and would therefore provide the Bell Companies the information they need for network planning purposes. This approach is consistent with the approach that local exchange carriers employ for network planning purposes. Unlike the AVP proposal, the use of forecasts only would not penalize wholesale customers based on whether they underbuy or overbuy pre-purchased blocks of usage."
1530 While competitors are better placed to project monthly usage than ordinary customers, the CNOC proposal is more fair on this point than Bell's in that it will provide the incumbents with usage projections and burdens both ILECs and CLECs with a somewhat equal level of risk.
1531 Finally, as explained in a little bit more detail below, the CNOC proposal is peak-period-based as opposed to monthly-usage-based and we find that this is more reflective of costs. However, we are not convinced that either of these proposals is required or preferable to the traditional mechanism for including provisioning costs in the GAS tariff at a fixed rate which is based on estimates of peak usage.
1532 Our second criteria is that the tariff be fact-based.
1533 It is critical that rates, and particularly variable-usage-based rates, be based on proven incumbent incremental costs. Currently, this data is not on the public record in that it is treated as confidential.
1534 Any per-gigabyte and particularly per-megabit-per-second rate that forms the basis of a tariff should be proven on the record and the underlying assumptions for such a rate made clear so that they can be challenged.
1535 There is reason to believe that the current proposed $0.195/GB rate put forward by Bell in its latest AVP proposal is significantly higher than Bell's actual costs, let alone the -- there's also a penalty component.
1536 There is evidence on the record that Bell's costs are as low as $0.08 or even $0.01/GB, and a mark-up of this calibre is unjustified in a cost-based tariff, and even, I would say, in an incentive-based tariff, I think it goes beyond. Indeed, as MTS points out, the $0.195/GB and $0.295/GB overage penalty are in fact higher than some of Bell's own retail service packages.
1537 A clearer on-the-record understanding of incumbent pricing assumptions is essential to ensure that the tariffs put forward are, in fact, cost-based. If it is impossible for confidentiality reasons to put these assumptions and costs on the public record, in our view, it becomes incumbent on the Commission to develop an in-house investigatory capacity so that it can better ensure this accuracy.
1538 This accuracy is even more critical if CNOC's proposal is adopted. As CNOC proposes to roll all usage-based costs into a variable per Mbps rate, inaccuracies in this rate may very seriously impact on competitors' ability to compete.
1539 Finally, a fact-based approach should include built-in reviews for any rates and particularly for any variable usage-based rates it sets. Variable usage rates are continually dropping at annual rates. If a tariff is locked in at one rate while incumbent rates decrease due to technological developments, the long-term prospects for competition will not be good.
1540 With respect to transparency, transparency in measurement is additionally important to a successful tariff.
1541 Network usage is not easily measured. In the United States, an examination of mobile usage measurements demonstrated overmeasurements of customer usage ranging from 7-14 percent on average, with specific examples ranging as high as 300 percent in overage measurement. So this is not like a billing mistake. This is equipment measurement.
1542 There have been examples of measurement issues in existing variable-usage-based schemes. CAIP, for example, noted on the record of Tariff Notice 242&338/7181 that:
"In late 2008, Bell made sample UBB data available to its wholesale customers. To say the data is flawed is an understatement. ISPs are reporting variances between Bell's data and the ISP's own data of as much as 800 percent. In one case, Bell has provided data indicating a residential end user consumed 1 TB (terabyte) of data over 20 hours of use."
1543 So if a variable message of usage-based billing is to be applied to wholesale pricing, moving the point of measurement to the competitor's point of interconnection, as proposed by CNOC, whether it's the POI or the A-HSSPI, it's likely to minimize billing discrepancies as incumbents and competitors will be working from the same measuring point.
1544 With respect to peak period versus monthly usage measurements, as expressed in far greater detail in our initial and reply comments to this proceeding -- and I will elaborate if there are questions -- network costs are a product of peak period usage and this remains the case.
1545 Peak period aggregate pricing more closely emulates actual incumbent costs. A gigabyte of usage over the course of one day can generate more provisioning costs than 10 gigabytes usage over the course of a month.
1546 Therefore, per terabyte monthly blocks, while superior to per end-user monthly caps are still not a tailored measurement of actual incumbent usage and costs.
1547 There is evidence on the record that while peak period usage remains the primary driver of network investment costs, only approximately 25 to 30 percent of all daily traffic occurs during such periods. This means that as much as 70 to 75 percent of daily usage is not likely to lead to substantial network costs, making monthly usage measurements of blunt and inaccurate statistic for determining a cost-based tariff.
1548 More data is needed to assess the level of correlation between CNOC's 95 percent measurement proposal and the actual network costs, but it appears more carefully tailored to peak period usage and, therefore, more closely cost based.
1549 With respect to incentives to build networks and manage traffic, in our view incentives to build new networks do not come inherently from any usage-based wholesale pricing scenario, whether fixed or variable. Instead, they come from increased competition. As long as incumbents are able to recoup any provisioning costs imposed on them by their wholesale services, along with a reasonable mark-up, a tariff of this nature will not deter incumbents from building their networks in any way.
1550 As noted above, once this cost recovery plus is assured, there is also no longer any need for traffic management incentives with respect to wholesale services because the costs incurred by the incumbent are covered.
1551 With respect to mark-ups, we do not make detailed submissions, but except to note that all elements of the proposed tariff should be cost-based with the same mark-up, depending on the appropriate category, essential/non-essential, that the tariff ends up in.
1552 In conclusion, with respect to application of the tariff legacy services, we note that the adopted tariff, whether it be Bell's AVP proposal or CNOC's or another alternative, should be applied to legacy services as well. We are concerned that Bell's proposal, which provides competitors with a discount -- Bell's proposal with respect to legacy services which provides competitors with a discount equal to Bell's projected accepted monthly average per customer usage rate will provide incentives to keep customers below the predetermined monthly usage caps because those are already in place. This will not be productive.
1553 Further, the existing legacy tariff is not consistent with CNOC's tariff structure, in which case it will need to be -- it won't be easily transposed upon that. We think that is a better proposal.
1554 So for this reason all services at issue should be migrated to a tariff based on the CNOC model.
1555 Thank you for your time and we will be glad to answer any questions.
1556 THE CHAIRPERSON: Thank you.
1557 First of all, let me remind you of the scope of this hearing. We are talking about residential wholesale internet, that's it.
1558 MR. ISRAEL: Sure.
1559 THE CHAIRPERSON: OpenMedia, you mentioned all sorts of very interesting topics such as accessibility, retail caps, business access, et cetera. I don't deny at all the importance of those subjects, but that is not the subject of this hearing. This hearing is devoted to residential wholesale services, so let's stick to those.
1560 You are using the words metered internet. Is that a simile for UBB or what do you mean by it? You used that expression several times throughout your --
1561 MR. ISRAEL: Just without getting into the retail markets in too much depth, we do analyze this tariff to some extent in light of the impact that it will have on the retail market in the sense that we believe that when we are talking about a metered internet I think we are talking more broadly the impact on consumers and the types of pricing models that they are subjected to --
1562 THE CHAIRPERSON: Okay.
1563 MR. ISRAEL: -- and we are hoping to instill a competitive check on --
1564 THE CHAIRPERSON: Okay. I'm just trying to understand your terminology. Let's say for argument's sake we adopt Bell's proposal of AVP, is that in your mind a metered internet system?
1565 MR. ISRAEL: I think that one is -- I think the key difference for us between Bell's AVP and CNOC is that Bell's AVP will make it more difficult for competitors to put competitive pressure on the retail market.
1566 THE CHAIRPERSON: I understand. I'm asking you a very precise question. I'm trying to understand your use of the words "metered internet".
1567 If we adopted AVP, would we be entering into the realm of what you call "metered internet"?
1568 MR. ISRAEL: I believe we are already in that realm and we are hoping to get out of it by putting in place a tariff that will allow the competitors to put higher competitive pressures on the incumbents to maybe reduce the metered internet that is available right now and that is present right now and prevalent right now in the retail market.
1569 THE CHAIRPERSON: I'm sorry, I cannot talk with you unless I understand the meaning of the terms you are using. I'm trying to understand what you mean by "metered internet".
1570 MR. ISRAEL: When we are talking about metered internet we are talking about the retail market and the fact that there is a situation.
1571 So right now there are a few unmetered plans.
1572 THE CHAIRPERSON: But just stay with the example. AVP, where it does not require any caps for the ISPs, they can do it, all they have to do, according to their proposal, is forecast what they need and buy accordingly. They can do whatever they want with their retail customer.
1573 Is that, in your view, metered internet or not?
1574 MR. ISRAEL: I believe that it's more likely to lead to one because it will be more difficult for competitors to stay away from those metered price -- for those metered prices in the retail market that they offer their customers, because their costs will be more tied to usage and less tied to costs and with deterrents to actually deter usage --
1575 THE CHAIRPERSON: You want to stay away from caps, I understand that, and basically metered internet, when you use it, leads to caps and that's what you want to avoid.
1576 Do I understand you correctly?
1577 MR. ISRAEL: Yes, Mr. Chairman.
1578 THE CHAIRPERSON: Okay. Now, you didn't mention at all cable. You heard yesterday from Bell that 70 percent of their end users are ISP customers and the use of 30 percent of the traffic, or something along that line, which is way more disproportionate to what is being used on cable, and cable actually has far more internet customers I understand at present than Bell.
1579 What is your view on this? Should the Commission try to make sure that the residential wholesale is carried as much by the cable as by the ILECs -- actually it's predominantly the Bell ILECs that is carrying the load.
1580 Is this an issue in which you have no view?
1581 MR. ISRAEL: Well, with respect to should we be encouraging cable as well, I think we should be. I think the Commission has taken steps to open up the TPIA tariff, which is really positive and that will help.
1582 With respect to the disproportionate usage by the whole -- if I'm understanding your question correctly, the disproportionate usage is that made by some wholesalers, we actually think that if there is cost recovery on that, that's not actually going to be a problem, as long as there is reasonable cost recovery.
1583 THE CHAIRPERSON: You see, there will be a shift if we adopt a full cost recovery model, like you suggest.
1584 MR. ISRAEL: Yes. Then there shouldn't be a problem. That's actually a positive, if customers want to use more and can find a place to do that and everybody is getting their money back with a reasonable profit, then that's good actually. That is a good thing.
1585 THE CHAIRPERSON: Good.
1586 Len, I believe you have some questions.
1587 COMMISSIONER KATZ: Thank you and good morning.
1588 MR. ISRAEL: Thank you.
1589 COMMISSIONER KATZ: I have a group of questions here, but I thought I would start with a general question.
1590 Welcome to the proceeding. This is the first time we are getting a number of interest groups and public participation, which I think is very valuable.
1591 The one piece that's missing from my mind is who you actually represent, because when we decide to sort of make a decision it's important to map against that decision where the interests lie and where they may not lie as well.
1592 I have seen your opening comments where you say you are a non-profit, non-partisan, national "organization engaging Canadians in governing and safeguarding the internet". I have a copy of the internet home page here where you list a number of members.
1593 How do we define who you actually represent at the end of the day, because I see on here some business, I see on here some industry, I see on here some unions. Give me an idea of your mandates, where you get your mandate, who gives you that governance authority that you are here before us representing Canadians.
1594 MR. ANDERSON: Sure. We are registered non-profit, our mandate is to push for an open and affordable internet and we are probably amongst the most grassroots-based organizations in the country, easily, and we represent and work in favour of the Canadian public in the public interest. I think that's pretty clear.
1595 I think the amount of engagement that we have been able to encourage in this hearing, and with our Stop the Meter Petition, which I believe has had the most signatures and citizen engagement of any on-line petition in Canadian history. It shows that we are an organization that is citizen-based and that's who we work for.
1596 We have a number of supporters and basically anyone who supports the open internet, the open and affordable internet and wants to support us in working on that, we are happy to work with. But we are primarily supported by the Canadian people and that's whose interests we work for.
1597 COMMISSIONER KATZ: How many individual members do you actually have that contribute to -- I don't know if it's a card or a membership or whatever?
1598 MR. ANDERSON: Well, we consider members people who take part in one of our campaigns. In that case this year over 520,000. Then if you are talking about financial support, over 10,000.
1599 COMMISSIONER KATZ: So over 10,000 individual Canadians have funded you?
1600 MR. ANDERSON: In that last year, yes.
1601 COMMISSIONER KATZ: As well as the members that are on here as well?
1602 MR. ANDERSON: Most of them. Some of them just have signed -- again, like we define membership as people who engage with our campaign. So some of those groups will provide us with support and other groups are just helping us spread the word and engage Canadians. It's pretty open-ended.
1603 COMMISSIONER KATZ: So groups like CACTUS, Canadian Association of Community Television and Users and Stations; groups like CIPPIC, obviously; CUPE, which is half a million members of various industries, all pay to support your representations?
1604 MR. ANDERSON: No. Like I said, it's very open-ended. So some groups and some individuals support us financially, other groups just help us reach Canadians. For example, CACTUS helps us do outreach; CIPPIC helps us do policy work and education work; other groups support us financially. So it's a mix.
1605 It's a network-based organization so I understand that is maybe a different kind of organization than you are used to, but I feel like it's reflective of the digital age.
1606 COMMISSIONER KATZ: Okay. I'm going to go through your opening comments here with some questions and then I will work backwards onto your previous submissions.
1607 In paragraph 18 you state:
"Functional marketplaces meet demand by increasing supply..."
1608 Demand comes from two sources, one source is more users who pay, the other is the same number of users who use more. In one case there is revenue to support that increased capacity and supply, in the other case, in your model, in your mind, there should be an unlimited usage.
1609 If that is the case, how is supply manufactured, if I can call it that, and where is the investment coming from?
1610 MR. ANDERSON: Sure. I think that we want to create -- there has been groups talking about incentive-based models and I think incentive-based should actually be incentive for investments, incentives for innovation and use. We want more use, not less, in Canada. That's what other countries are doing. We need to have that approach to keep pace.
1611 To answer your question, we are talking -- like what I would like to see if there to be flexible pricing models on the retail side that is enabled by a cost-based solution. So I think you can have unlimited options if we have an appropriate cost-based wholesale regime in place.
1612 So I think that if we appropriately support the costs of provisioning, then we can support the use and we can support the demand on the network. I think that will then -- if we give enough autonomy and I think a cost-based solution does -- hopefully the independent ISPs and hopefully down the line under competitive pressure the incumbents will provide options for unlimited use.
1613 COMMISSIONER KATZ: So you do recognize there is a cost associated with increased capacity?
1614 MR. ANDERSON: Yes. We are calling for a cost-based approach.
1615 COMMISSIONER KATZ: Okay.
1616 In paragraph 31 of your remarks you state that:
"Given that GAS and TPIA tariffs have always included cost-based components that are intended to reflect usage, this, in our view, is the best mechanism..."
1617 Do you know how much usage is imputed in those tariffs today?
1618 MR. ISRAEL: I can give you -- no, I can't --
1619 COMMISSIONER KATZ: I'm not looking for a number. You do understand there is a usage component in there?
1620 MR. ISRAEL: Yes. Yes. And it's reflected in the provisioning costs; right.
1621 COMMISSIONER KATZ: Right.
1622 MR. ISRAEL: So there is a provisioning component that says, you know, this tariff at this rate has an accessibility connection fee, a service upgrade, a customer service component, and then it has a provisioning upgrade component and that is built into the tariff to reflect usage.
1623 COMMISSIONER KATZ: Okay.
1624 MR. ISRAEL: It's factored into the Phase II costs, and I'm not an expert on Phase II costs so I can't analyze whether it's accurate or the degree of its accuracy, but it's definitely in there and it is there to reflect projections on how much the use of the tariff is going to cost in provisioning costs for the incumbent.
1625 COMMISSIONER KATZ: Okay. So let's for argument agree -- whether you do or not -- that there is 40 gigs imputed in a monthly rate for a GAS service today.
1626 MR. ISRAEL: Yes.
1627 COMMISSIONER KATZ: Everybody is paying the same amount of money, at the retail level in this case, for the same amount of usage. Some users are using an awful lot less, some are using an awful lot more, do you believe that is equitable?
1628 MR. ISRAEL: Are we talking about the retail market right now?
1629 COMMISSIONER KATZ: Yes.
1630 MR. ISRAEL: Okay. I did try to stay a little away from the retail market in our -- I can address that, though, if you would like.
1631 COMMISSIONER KATZ: Well, take it back to the wholesale market.
1632 MR. ISRAEL: Sure.
1633 COMMISSIONER KATZ: Do you believe it's equitable to impute a certain amount in the usage at the wholesale level where the value-added provider -- I won't call him a wholesaler -- the independent value-added provider will then turn around and repackage that in there when in fact there is a number, 40 gigs in this case, that is being used unilaterally in the costing exercise?
1634 MR. ISRAEL: Let me address wholesale first and then if you want I can address retail as well, because I do have an answer to that.
1635 For wholesale I think it is okay, as long as the -- because my vision for wholesale is that as long as the -- I am calling it wholesale, but the value-added -- if you prefer, the value-added ISP, as long as they are compensating the incumbent for whatever costs they are incurring, plus a reasonable mark-up, my vision is that if they want to take less of a profit off that with their customers to encourage more usage, we actually want that. That is the type of competition that we want in the general market because that puts the competitive pressure on the incumbent to then maybe meet that, if that is what customers actually want.
1636 That is where the market dynamics come in place, but I think I see that as fair as long as the incumbent is getting compensated.
1637 COMMISSIONER KATZ: Okay, but --
1638 MR. ISRAEL: As long as you can replicate the actual costs to the incumbent plus a reasonable mark-up I don't see a problem that the customers of the wholesaler are generating more traffic.
1639 COMMISSIONER KATZ: But I guess maybe I'm missing something.
1640 I was under the impression one of the big concerns six months ago was that there is a fixed amount of usage going into the wholesale tariff and as a result of that there is more flexibility available to the value-added independent ISP because they now have to assume that that is the amount of capacity is there, regardless of whether a customer they are serving only wants a fraction of that or wants a multiple of that.
1641 MR. ISRAEL: You are talking about the end-user-based --
1642 COMMISSIONER KATZ: I'm just looking at the wholesale customer right now. The CNOC customer --
1643 MR. ISRAEL: Right.
1644 COMMISSIONER KATZ: -- who is buying services --
1645 MR. ISRAEL: Right.
1646 COMMISSIONER KATZ: -- he's buying a legacy service --
1647 MR. ISRAEL: Right.
1648 COMMISSIONER KATZ: -- is getting an imputed amount of capacity that's fixed.
1649 MR. ISRAEL: Oh, okay.
1650 The problem with that there is that there is no room for competition there on pricing models, right. The incumbent is setting the pricing model for the wholesaler. Our whole objective here is to try to set up the wholesale ISPs in a way that they actually emulate the incumbents in what they are paying and then they can provide whatever -- so they can compete on pricing models as well as price.
1651 So if the incumbent is saying every one of your customers that goes over this amount you have to pay us per customer, then that is not reflective of the costs that the incumbent has to pay per customer on their end and it doesn't give the competitor the value-added ISP the wholesale --
1652 COMMISSIONER KATZ: So you are in favour of unbundling, if I can call it that, the usage component from the fixed component?
1653 MR. ISRAEL: We think that's one of the things that has been -- I think it was unbundled in that first usage-based tariff in the sense that it was reflected in two places, it was reflected -- there was a usage-based fixed rate in the provisioning portion of the Phase II costs, and there was another one that was a per-customer, if any one of your customers goes over this usage then you get another -- there was another usage sort of component in there, but in previous GAS tariffs it was built into just the Phase II costs without the pre-customer end-user overage.
1654 COMMISSIONER KATZ: Okay. In paragraph 34 you say that:
"A wholesale tariff should aspire to best emulate actual costs incurred by the incumbent in provision of the service plus a reasonable mark-up."
1655 Do you believe business risk is a cost?
1656 MR. ISRAEL: I do believe it should be accounted for in some respect and I believe that right now it is through the mark-up -- that was decided in your high-speed matching decision. To account for investment in new services and the risk that a business takes on in investing in new high-speed services, they get an extra mark-up if it falls into that category.
1657 COMMISSIONER KATZ: My difficulty -- and I will ask the same question to all parties during this proceeding -- is that I'm getting the sense that the alternative models that are being proposed all reduce business risk dramatically and shift the business risk from the independent ISP back to the service provider, whether it be a TPIA, cable company or phone company. What they are looking for is facilities when they want it, how they want it delivered, at the time they want it, at cost-based rates.
1658 But when you are running a business there is inventory costs, there is obsolescence costs, there is deliver delays, there are all sorts of things that the way this model is being purported -- and there is nothing wrong with paying for costs, the question is what is included in there as well, because from my perspective there is a huge transfer of business risk going from the independent ISPs back to whoever delivers the service to them as well.
1659 So my question is: How do you capture that? How do you account for that?
1660 MR. ISRAEL: I think that's a very fair comment.
1661 I think there is a way to calculate that maybe. I don't have the solution to that, but maybe on an annual rate.
1662 But there are a couple of risks here, right. There is one risk that is really tied to the usage, to the costs incurred from usage, right. So I think that risk has to be allocated really clearly in the way that you calibrate the usage-based component of the tariff.
1663 And I feel that the AVP -- some of these proposals really shift the risk away from the incumbent, 100 percent almost, and put it on either the competitor or, in the retail market, on the customer. I find that problematic.
1664 I guess that some of what we are talking about here is, how do you account for that division of risk. The ideal would be to allocate it in a way that the competitor and the ILEC are carrying it to equal degrees.
1665 With respect to those other risks, I think that if there is a way to calculate them, and they are reflective of actual costs, I think that is fair to build into a tariff. That is my view. Keeping a cost base is fair.
1666 COMMISSIONER KATZ: You will have time to think about it, because you are back next week, and maybe other people could think about it as well.
1667 In paragraph 41 of your remarks this morning, you are focusing on the CNOC proposal, which is a peak usage proposal, and you say, "As CNOC proposes to roll all usage based costs into a variable per Mbps rate...," and I think I read the same words in the PIAC submission as well.
1668 I am interpreting that to be an aggregate, which means that all independent ISP's traffic will be aggregated, and the peak model that is being proposed by CNOC will then apply to everybody equitably, so there won't be any distinction or delineation between members of CNOC who are buying services and running a business, some more effectively perhaps, and being able to manage peak, some less effectively.
1669 Is my understanding correct?
1670 MR. ISRAEL: It is a little bit difficult -- I should say this -- it is a bit difficult to assess the CNOC proposal in full at the moment, just because there is no price attached to that megabyte per second rate at the moment, and we don't know how that will be -- I don't think we understand how that will be calculated. I think that would have to come from the incumbent.
1671 But my understanding is that, yes, it is actually a line rate. So they are going to say, "We are generating traffic at this speed. That is the maximum that we are pushing traffic onto your aggregate network." Then, I think it falls to the incumbent to say --
1672 This is my understanding, because they will probably know what their costs are on their network --
1673 COMMISSIONER KATZ: Regardless of cost, if you have two players that are buying together, and one is able to manage its utilization by having a unique service -- I don't know, a security service -- that has limited utilization, but it's a value-added service, and the other one is actually selling service to people that are downloading videos or whatever, and that is driving the peak, as well, there is going to be one aggregated rate, and they are all going to pay the same rate.
1674 MR. ISRAEL: Each wholesale ISP will pay for the peak traffic that it pushes onto the network, regardless of whether it comes from their customers watching video or whatever. But I feel that that should be reflected in the depth of the peak, and maybe there is still some work to be done on how closely that is correlated to actual costs.
1675 But the idea is, if that is reflected in the amount of traffic that is being pushed through and the megabyte per second rate is priced properly to reflect the cost to the incumbent, then it shouldn't matter if it is people watching videos or...
1676 Am I missing the --
1677 COMMISSIONER KATZ: No, I think you have the point. You are telling me that it's a function of the cost at the end of the day --
1678 MR. ISRAEL: I think it is.
1679 COMMISSIONER KATZ: -- and I am saying, irrespective of what the cost is going to be, these two players are going to be treated the same way, and I hear you saying, I guess, that's reality.
1680 MR. ISRAEL: But I feel that that might be fair because -- you are talking about a business and --
1681 I feel that's fair because --
1682 COMMISSIONER KATZ: I am talking about wholesale and wholesale.
1683 MR. ISRAEL: Wholesale and wholesale.
1684 I feel that that's fair because the idea is to not start saying, "This is more legitimate." It's okay to --
1685 Businesses and video, those are all important, but the idea of the internet is that you don't start saying, "No, this is more --"
1686 The innovation that happens online, from our perspective -- and this is now getting into the net neutrality stuff a little -- is that you set up the elements, and then you let the innovation happen.
1687 You don't know what the next -- the video thing could turn out to be a business.
1688 You don't know what is going to happen on there, so you just want to set the conditions and then let it go.
1689 COMMISSIONER KATZ: You are a supporter of the CNOC model, which is charging for peak. Obviously, we heard from Bell yesterday, and they are a supporter of the usage model.
1690 I asked them the question; they didn't have the information. Have you run any regression analysis or models to decide whether there is a correlation between heavy usage and peak usage?
1691 MR. ISRAEL: We don't have the data to do that, but what we do have is -- we know how much traffic happens in off-peak periods and how much of it happens during peak.
1692 I have that in here. There is a submission on the record here, and it's based on Rogers' numbers. We don't have the actual numbers, but we have the conclusions based on them.
1693 This is at paragraph 45. There they are saying that 25 to 30 percent of daily traffic happens during the peak period, and that is the key driver of the cost.
1694 The rest of the daily traffic is happening at off-peak periods, but it is spread out in a way that doesn't incur a cost.
1695 But that doesn't mean that -- there probably is some -- the people who are using more --
1696 Like, I imagine that there is some correlation between the people who are using more in the 70 to 75 percent chunk and the 25 to 30 percent chunk.
1697 I think we would need a much clearer -- we would need someone to actually --
1698 We can't do that analysis because we don't get the numbers, but I think there is --
1699 Particularly if you are charging by the megabyte, by the gigabyte, I think you would want a much clearer correlation to what individuals are actually using, because it is not going to be a uniform -- you know, every gigabyte in that 70 to 75 percent chunk, it's not going to be a clear correlation because there is that disaggregation.
1700 COMMISSIONER KATZ: Yes, there never is a clear correlation. It's never 100 percent, but that's fine.
1701 My last question is one that I want you to think about. Rates, at the end of the day, are going to be driven by forecasts of usage, whether it's peak usage or heavy usage.
1702 I am sure you heard the cross-examination yesterday with Bell, where their submission said that there was huge growth in the industry, and yet their costing models have a much more refined amount -- a modest amount of growth.
1703 Can you think about what you would suggest would be a reasonable forecast of growth of demand over the next three to five years, and when you come back next week --
1704 MR. ISRAEL: I can, but I can speak to that a little bit right now, if --
1705 COMMISSIONER KATZ: If you would like to, sure.
1706 MR. ISRAEL: This is an issue that we have tracked for a couple of years now with our experts, who did some traffic measurement on this. I think it's really important to pay attention to the annual growth rates, because we are actually not in --
1707 You hear about video and all this other stuff, and it sounds like we really are in that period of phenomenal growth rates, because people are doing a lot more online. But what happens is, the internet grows at a steady rate every year, and we have gotten to the point where there is a lot of traffic already, so that video is adding actually less than --
1708 The annual growth rates right now are lower than they have been at most historical levels, and there are projections, and it is really hard to predict what will happen in the future. I imagine that they will go higher, because they are actually really low right now, compared to what they have been historically.
1709 Our line on this has been that you don't want to base policy -- because it's a little bit unpredictable, you don't want to base --
1710 You can say, "Oh, we are going to expect traffic to quadruple next year," but we have heard a lot about that over the years. I don't think we want to base policy decisions on the assumption that that is going to happen, because nine times out of ten it doesn't.
1711 I think that you do want to build in some contingency, but you want to look at what the annual growth rates are now, and they are 35, 40, 45 percent, and what they have been historically, which is around 100 percent, and say that maybe something within that range is what you could expect, which is not unreasonable.
1712 COMMISSIONER KATZ: Think about being more specific when you come back next week.
1713 MR. ISRAEL: Sure.
1714 COMMISSIONER KATZ: Those are my questions.
1715 THE CHAIRPERSON: Tom...
1716 COMMISSIONER PENTEFOUNTAS: Thank you, Mr. Chairman.
1717 You guys raise some interesting points. That being said, you mention in paragraph 20 that you are here representing true Canadians and their true stories, and not the narrow, self-serving narrative produced by big telecom.
1718 Your narrative is just as self-serving as the narrative of big telecom, is it not, sir?
1719 MR. ANDERSON: No, I don't think so, and one reason I say that is because you had a representative from Bell here yesterday who actually sat before you and insinuated that Canadians and their customers actually are happy with usage-based billing.
1720 That is clearly not the case, and it's disingenuous.
1721 But in terms of our interests, we are supported, fundamentally, by the Canadian public, and our interests are in making the internet more open and affordable.
1722 If you can call that a self-interest, sure. It's in the interest of consumers and Canadians in general to have a more open and affordable internet, so we can encourage innovation, so we can encourage business investment and consumer choice.
1723 I think that's a public interest. I think it's one that we collectively have as a country, and I think that is very different from Bell or the incumbents' narrow commercial interest, which is to make the most profit possible.
1724 Certainly, they are welcome to have that interest, that's their job, but I think it's important to note the distinction.
1725 COMMISSIONER PENTEFOUNTAS: Getting back to profits, you mentioned that the focus of the telecom companies should be on finding ways to make the internet faster and more accessible, and not increase costs.
1726 Would you be in agreement with me in saying that profits motivate investment and would contribute to making the internet faster and more accessible?
1727 MR. ANDERSON: Yes, I think they do, but I think it's -- do we create incentives for --
1728 Because there are different ways to profit. You can profit from gouging consumers, or you can profit by creating a better product and competing on those terms.
1729 You can create incentives for bandwidth scarcity, to allow them to gouge consumers, rather than invest, or you can create incentives based on a cost-based model that will encourage the incumbents to invest.
1730 What it seems to be, from what I can see, is that there haven't been enough incentives for them to invest in the network and to increase profits by creating a better product, and there has been too much of an incentive -- or there has been too much allowance for there to be an incentive for them to gouge consumers.
1731 COMMISSIONER PENTEFOUNTAS: Don't the independent ISPs serve that purpose, in large part? Don't they contribute to creating motivation?
1732 MR. ANDERSON: They do, and that is sort of the whole point. We need to encourage them to have autonomy, so that we can have competitive pressure.
1733 COMMISSIONER PENTEFOUNTAS: You also mention at paragraph 25 that you really find it is irresponsible and undemocratic to have a few companies controlling the internet.
1734 What impedes independent ISPs or anyone else from investing in networks?
1735 MR. ANDERSON: I think that the incumbents' networks are legacy networks that they have built up over many, many years, and there are public rights-of-way there. They have that historically to build on.
1736 I would say that the independents can invest in networks, and I think they do invest already --
1737 COMMISSIONER PENTEFOUNTAS: Do you have figures for that? Do you know how much they have invested in internet networks over the last year?
1738 MR. ISRAEL: We do, actually. If you go to Figure D on page 11 there is --
1739 I'm sorry, it's not very well reproduced because it's a colour graph.
1740 I think the curved upward line is the traffic growth that we have right now, at about 45 percent. It's probably a bit lower -- compound aggregate growth.
1741 The middle line across, I think, is Bell Canada's investment, based on information that they put on the record.
1742 The top line, which is hardly visible, is, I believe, incumbent TSP investment in Canada, and that is from the --
1743 COMMISSIONER PENTEFOUNTAS: I understand incumbents, but what about independent ISPs? That's what I am asking about.
1744 MR. ISRAEL: That is on the record, but I don't have it here. It's lower, but I believe that their --
1745 COMMISSIONER PENTEFOUNTAS: Almost non-existent?
1746 MR. ISRAEL: I believe that --
1747 COMMISSIONER PENTEFOUNTAS: You are shaking your head. Does that mean yes?
1748 MR. ISRAEL: I don't have the exact figures, but I believe that part of their issue is that they can't get to the last mile. So there are limits on --
1749 COMMISSIONER PENTEFOUNTAS: Isn't part of the issue that the deal is so good and they can piggyback on the incumbent, as you say, internet providers, and still be tremendously profitable, and there is no need for them to invest in networks on their own?
1750 MR. ISRAEL: In a prior proceeding, there was a request for a tariff that would let the wholesalers connect closer to the final -- at the home office -- the customer office -- but I think it wasn't picked up -- it wasn't put in place by the Commission.
1751 That would have been a higher priced tariff, and they would have had to build closer that way.
1752 But I think they are facing -- I think there is a bit of an inefficiency there, as well, just looking at it from the broader public interest level, in the sense that we don't want, necessarily, every single ISP building up that last copper wire. We don't want eight wires going into everybody's home, right?
1753 So there is an efficiency to splitting -- sort of a broader economic efficiency to splitting that out, I would say, to some extent.
1754 But you still need to maintain competition, which is why I feel that this is --
1755 COMMISSIONER PENTEFOUNTAS: But there is no real incentive for independent ISPs to build their own networks, just getting back to that.
1756 MR. ISRAEL: I think there are incentives for them to build closer to the last mile, but I don't think we want them to build to the last mile, necessarily, in all places.
1757 COMMISSIONER PENTEFOUNTAS: I am trying to find out what is undemocratic about the system we have right now: "...allowing a few companies to control the internet access market would be irresponsible and undemocratic."
1758 MR. ANDERSON: I think what is undemocratic is that Canadians have been pretty clearly -- have been pretty vociferous about what they want, which is a competitive market.
1759 And if we look at other countries, like in the U.K., for example, the way to enable a competitive market is open access policies, it is encouraging pricing autonomy and encouraging the independent ISPs to have control over their own businesses.
1760 So I think that it is undemocratic for us to continue to allow the big telecom companies to dominate the market.
1761 COMMISSIONER PENTEFOUNTAS: How are we allowing that?
1762 I am getting back to the same point. Why aren't independent ISPs building their own networks, ergo, making themselves more independent?
1763 MR. ANDERSON: I think, again, like Tamir said, we don't necessarily want them all to build their own networks. That is not what they are doing in the U.K. -- and I encourage the Commission to take a look at what they are doing there, because they have a very competitive market.
1764 COMMISSIONER PENTEFOUNTAS: How many internet providers are there in the U.K.?
1765 MR. ANDERSON: Do you know the number?
1766 MR. ISRAEL: Not offhand, but there is one big one, right? There is one big one, British Telecom, and there are a lot of smaller ISPs, so they have structural separation there.
1767 I think, Commissioner, that you are right to a certain extent, that we do want to incent the competitors to build, maybe not an entire network, but definitely more. I think that would be a good thing.
1768 MR. ANDERSON: I actually think that you should ask them, in particular, what would encourage them to invest more. I think that's a really important question.
1769 But I would think that having a more cost-based approach and enabling them to have more business autonomy would encourage them to build out their business, because right now there is sort of a logjam in terms of them -- or there is a proposed logjam in terms of them controlling their own services.
1770 COMMISSIONER PENTEFOUNTAS: Also, at paragraph 40, you mention that there is evidence on the record that Bell's costs are as low as .08 cents, or even .01 cent per gigabyte.
1771 I don't see a footnote following that statement. Do you have that somewhere?
1772 MR. ISRAEL: I do. It is referenced in our submission.
1773 I'm sorry, there should be one, you're right.
1774 COMMISSIONER PENTEFOUNTAS: What is the basis for that?
1775 MR. ISRAEL: There are two studies that were put in that are cost studies. One is based on -- for Bell, in particular. They put in a cost -- this is part of a different proceeding, to build out to -- with respect to how the deferral funds that were put aside in a former proceeding would be used to potentially build out to rural areas.
1776 They put in a lot of costing information that said, you know: Here is how much this equipment costs. Here is how much it would take to have a network that manages this many customers, at these service levels.
1777 That information was taken -- not by us, but that information was taken and analyzed and a range of prices was put together.
1778 I am not saying that these are the actual rates, I am just saying that --
1779 COMMISSIONER PENTEFOUNTAS: Do you want to take that sentence back, or come back to us?
1780 MR. ISRAEL: I am not taking it back. Or, sorry --
1781 COMMISSIONER PENTEFOUNTAS: Are you comfortable with that?
1782 MR. ISRAEL: I am comfortable with it, what I am saying --
1783 Maybe I should clarify what I am saying here. I am saying that there is evidence on the record that their costs could be as low as that, but what I am really getting at here is that we should be -- there should be a way to challenge --
1784 I feel that there are many different ways to put these costing apparatuses together, and I feel that when we are talking about a tariff, there should be a way to analyze the assumptions that, you know, a per gigabyte is based on.
1785 COMMISSIONER PENTEFOUNTAS: And I am telling you that I would like to see evidence on the record to justify these numbers.
1786 MR. ISRAEL: It is on the record, but I can -- in rebuttal, or whatever, I can bring those in.
1787 THE CHAIRPERSON: Maybe you could bring it next week. We are running out of time. Okay?
1788 MR. ISRAEL: Yes, I can do that. That's no problem.
1789 COMMISSIONER PENTEFOUNTAS: I gather that we are out of time. Thank you.
1790 THE CHAIRPERSON: Tim...
1791 COMMISSIONER DENTON: Hi. Just a few moments of factual information, which you can supply now or later.
"FACT 2: Big telecom's investment in network infrastructure in Canada as a percentage of revenue has failed to keep pace with our OECD counterparts."
1792 That is at page 2 of your evidence, paragraph 15, Fact 2.
1793 MR. ANDERSON: Yes, those are based on OECD numbers, and we can --
1794 Do you want it in rebuttal or do you want it now?
1795 COMMISSIONER DENTON: You don't need to read it now, I just need to see it.
1796 MR. ISRAEL: I think the source is in there, but let me list those.
1797 COMMISSIONER DENTON: All right.
1798 Paragraph 17:
"In countries like Japan, the UK, Sweden, and Chile, open access policies, including those that provide independent ISP pricing model autonomy, play a critical role."
1799 That might be something to unpack, but why don't we start orally now, and you can pursue it in argument, should you so choose?
1800 MR. ANDERSON: Sure. I think we can provide more evidence in rebuttal for that, but OpenMedia recently published a report called "Casting an Open Internet". It's 150 pages. There is --
1801 COMMISSIONER DENTON: I assure you that I am not going to read it before --
1802 MR. ANDERSON: Yes. The international section, I think, may be more digestible. There are international comparisons, where we look into what Japan has done, and we look into what the U.K. has done, and we compare the --
1803 COMMISSIONER DENTON: In the U.K., as I understand it, they took British Telecom and broke it into two parts, an infrastructure part and a services part.
1804 Are you saying that Japan, Sweden and Chile have followed that example?
1805 MR. ANDERSON: No, not exactly. We are saying that if you look at them together, there are some principles that carry over. Open access policy, transparency, and a very active, bold and engaged regulator are things that we see that cross over those and are general themes.
1806 We can attach that part of the report to our rebuttal.
1807 COMMISSIONER DENTON: I think that might be interesting, but you have to remember the advice of Sir Humphrey Appleby in Yes Minister that a bold proposal is one that is bound to lose a general election.
1808 So, be careful.
1809 MR. ISRAEL: We treat you guys as impartial commissioners, not as bureaucrats. Does that help?
1810 COMMISSIONER DENTON: Yes.
1811 The other thing is that it says, "provide independent ISP pricing model autonomy." What on earth do you mean by that?
1812 MR. ANDERSON: It's funny that you picked that out of all the acronyms. I think that's probably the most clear to Canadians than what is being spoken of here.
1813 What I mean by that is that independent ISPs should be able to decide their own business model. They shouldn't have it dictated by the big telecom companies.
1814 COMMISSIONER DENTON: Is that not what Bell's AVP proposal achieves?
1815 MR. ANDERSON: It's a spectrum, right? It's a spectrum.
1816 Having it more cost-based and allowing them more autonomy is what we are arguing, and we think that there are some proposals that enable autonomy more than others, and AVP, I would say, would enable a little bit less autonomy.
1817 COMMISSIONER DENTON: I would be interested to see any kind of international, or whatever comparisons you choose to provide, that show what that spectrum is.
1818 MR. ANDERSON: Okay.
1819 COMMISSIONER DENTON: Because I think that is something that may be in your head that needs to be unpacked, so that we see what you are talking about.
1820 MR. ANDERSON: Sure.
1821 COMMISSIONER DENTON: Paragraph 27: "...we are really concerned with the impact that UBB is having on consumers, and also online innovation. Of greatest concern for us is the rapid rate at which monthly caps have been dropping in the past year and a half or so industry wide."
1822 You say that wireline traffic growth rates are extremely modest and should pose no serious challenge to incumbents.
1823 The source and meaning of that? Forty percent growth a year is not 100 percent, but it's still huge growth.
1824 MR. ISRAEL: It is, but there are a couple of things.
1825 If you turn to page 11, Figure D, and also Figure E on page 12, but I will start with Figure D --
1826 The two flat lines across the top reflect incumbent wireline investment since 2006.
1827 One is 2006 to 2009, and that is TSPs in general, and the other one is Bell Canada.
1828 Those are pretty flat lines. So there hasn't been an increase in investment over that period, and that has managed to mean 45 to 50 percent traffic growth, plus investment in new wireline, plus whatever else that goes to.
1829 I guess what we are saying is, we don't see anything to justify a departure from that model.
1830 And since then, from what I understand, the annual traffic growth rates have actually dropped a bit. Now we are below 40. It was 43 percent when we had our last proceeding, the net neutrality proceeding, and I think now we are at 35 or something like that.
1831 So we don't see any dramatic, phenomenal growth rates that we feel would justify a departure from the traditional sort of non-usage-based --
1832 COMMISSIONER DENTON: In other words, am I correct in understanding that, of the three possible outcomes -- no usage-based billing, AVP, or the CNOC proposal -- you would prefer no usage-based billing whatsoever?
1833 Is that correct?
1834 MR. ISRAEL: I'm sorry, I should clarify. From a framing perspective, we do address retail a little bit here, and I think that's more what we are getting it. It's just to show that --
1835 If I could take you to Figure E for a second, on page 12 --
1836 COMMISSIONER DENTON: Yes.
1837 MR. ISRAEL: The darker line is 45 percent annual traffic growth, and the lighter line is where the caps have been going.
1838 COMMISSIONER DENTON: Yes. I think that's quite persuasive and important evidence.
1839 MR. ISRAEL: Right. On the retail side, that's what we are concerned about, and that's what we are talking about when we are talking about that. So that's not --
1840 COMMISSIONER DENTON: So you are trying to say that caps are declining and growth is significant.
1841 MR. ISRAEL: We are saying that growth is actually not -- it is growing at a steady pace, but a modest one; not at an unprecedented level that is difficult to match.
1842 COMMISSIONER DENTON: I get it.
1843 MR. ISRAEL: But, at the same time, the caps are going down.
1844 So we are saying that on the retail market that is what is happening, and we want to, maybe, slow that down a little. That is something to definitely keep an eye on --
1845 COMMISSIONER DENTON: What is the "it" to which you are referring that you want to slow down?
1846 MR. ISRAEL: I suppose I should say that what we want to do is make sure that that trend doesn't continue, and one way to do that would be to give the competitors as much flexibility in pricing models as possible.
1847 COMMISSIONER DENTON: Thank you. Those are my questions.
1848 THE CHAIRPERSON: Thank you very much. Those are our questions for you.
1849 We will take a ten-minute break before we continue.
--- Recessed at 1016
--- Resumed at 1028
1850 THE SECRETARY: Order, please. A l'ordre, s'il vous plaît.
1851 THE CHAIRPERSON: Okay, commençons.
1852 THE SECRETARY: Mr. Chairman, we will now proceed with the presentation from Cogeco Cable Inc., Quebecor Media Inc., on behalf of its affiliate Videotron and Rogers Communications Partnerships, collectively the Cable Carriers, who will make a joint presentation.
1853 Appearing for the Cable Carriers is Mr. Dennis Béland.
1854 Mr. Béland, please introduce your panel for the record, after which you have 45 minutes.
1855 MR. BÉLAND: Good morning, Mr. Chairman, Commissioners, Commission staff.
1856 My name is Dennis Béland and I am Senior Director, Regulatory Affairs, Telecommunications for Quebecor Media.
1857 With me today is Pierre Roy Porretta, Vice President, Engineering, Research and Development and in the second row we also have Mr. Patrick Simoneau, Francis Claprood and Sylvain Archambault from Videotron.
1858 Representing the other Cable Carriers and appearing with us on the panel today are:
1859 - From Rogers Communications, Ken Engelhart, Senior Vice President, Regulatory; Dermot O'Carroll, Senior Vice President, Engineering, Access Networks; David White, Director, Engineering IP Networks, David Watt, Vice President, Regulatory Economics. We also have Ddin Morbi from Rogers.
1860 - From Cogeco Cable we have Philippe Jetté, Vice President and Chief Technology Officer, Michel Messier, Director, Regulatory Affairs and in the second row we have Michael Coltart and Lucie Adam.
1861 - Also appearing with us is Suzanne Blackwell, President of Giganomics Consulting Inc.
1862 In our opening remarks today, we will address the six issues outlined in the Commission's letter on the organization and conduct for this hearing.
1863 The Cable Carriers submit that the aggregated usage model, proposed in our March comments, is the appropriate billing model for the provision of residential high-speed internet services to wholesalers.
1864 Under this model, the monthly charge per end-user bundles the access service with approximately 25 gigabytes of usage (for our most popular speed tier). In addition, the tariff will require the payment of an additional charge for usage over the monthly usage allowance.
1865 However, the wholesaler will be allowed to aggregate the usage charges for its end-users. So if one end-user was away on vacation and used no gigabytes that month and another user used 50 gigabytes, the wholesaler would not have to pay the overage charge for the customer that used more than the 25 gigabytes allowance.
1866 The monthly usage allowance will be based on each cable carrier's average monthly usage and will be adjusted each year as traffic per end-user increases.
1867 The monthly charge as well as any additional usage charge will be calculated at cost-based rates. The usage charge will be much lower than under the current tariff.
1868 The price per gigabyte of additional traffic is based on our cost studies where the traffic sensitive costs are divided by the forecast traffic volume. The volumes are based on the "all carrier demand" principle so that wholesalers gain the benefit of our economies of scale.
1869 We believe that our proposal strikes an appropriate balance and will result in a set of wholesale tariffs that will provide for a vibrant TPIA service and which will be in the public interest. This model is similar to Bell's proposed approach for existing GAS wholesale access services. The aggregated usage model adopts the proposal outlined in Commissioner Molnar's dissent to Telecom Decision 2010-255.
1870 The benefits of our proposal are as follows:
1871 i. It reflects the networks we are building which provide for both access to the internet and capacity to use that connection.
1872 ii. It facilitates more effective and cost-efficient network planning.
1873 iii. It ensures that the TPIA tariffs foster more usage of the internet rather than less.
1874 iv. The aggregated usage model provides wholesalers with significant additional flexibility in the retail market because it effectively breaks the link between wholesale and retail rate structures.
1875 v. It lowers costs for wholesalers because wholesalers that serve a mix of lower and higher volume users will receive an allowance for lower volume end-users that can be directed to offset higher volume use by other end-users.
1876 vi. By greatly lowering the usage charge, it is easier for our wholesale customers to offer unlimited packages. This rate will decrease over time as traffic volumes rise and costs decrease.
1877 vii. Unlike the Bell aggregated volume plan, the Cable Carriers are not proposing that wholesalers pre-purchase terabyte blocks of usage. There is no premium to be paid for usage not purchased in advance.
1878 viii. By recovering short term variable costs through usage-based charges, the tariff sends the appropriate economic signals to wholesalers regarding usage.
1879 The monthly access rates per end-user are unchanged from the proposed TPIA tariffs that were filed pursuant to Telecom Regulatory Policy 2010-632. The monthly usage allowances are simply those included in our cost studies, as indicated in our responses to interrogatories.
1880 The rates for usage above the allowance are provided in Schedule 1 to our remarks. These rates are based on the cost studies and assumptions already filed on the record of this proceeding.
1881 These rates take account of the point to multi-point distributed architecture of our cable networks. These networks are very different from the point-to-point transit networks that some parties suggest cost mere pennies per gigabyte.
1882 The two types of networks are fundamentally different and so are the costs. Comparing our shared access network to a dedicated transit network is not an "apples to apples" comparison.
1883 The Cable Carriers intend to make annual adjustments to the price per gigabyte of usage and the monthly usage allowance per end-user. The adjustment could be based on updated forecasts for the forthcoming year or the previous year's actual volumes. Annual adjustments will provide the wholesalers with the benefit of cost savings from increased traffic volumes.
1884 We have included in Schedule 1 a chart that indicates the expected trends in the rate per gigabyte per year based on our current 10-year forecast of traffic volumes and usage sensitive costs.
1885 The Cable Carriers consider the forecasts for latter years to be highly speculative and subject to change. Therefore, we are not proposing rates for these additional years at this time.
1886 We believe that the proposed aggregated usage model fulfils the objectives of this proceeding to provide wholesalers with the flexibility to compete, to be fair to consumers with varying usage profiles and to achieve reasonable economic internet traffic management practices.
1887 However, if the Commission were to adopt another model we would like to comment on how this should be structured. The various models put forward all have in common two main components: a fixed access rate that reflects the non-usage sensitive costs and a variable usage rate that recovers short term usage-sensitive costs.
1888 Whichever model the Commission adopts, it is crucial that the dividing line between fixed costs and usage-sensitive costs be drawn in the right place. Getting this division wrong will send false economic incentives to wholesalers and will threaten our ability to properly manage our access network to the benefit of all end-users, ours and theirs.
1889 In June the Commission asked a series of interrogatories which directed us to develop various sets of rates assuming a specific delineation between fixed costs and usage-based costs. Where the Commission drew the line between these costs, we submit, was inappropriate for our networks.
1890 Specifically, the Commission's interrogatories directed us to assume that the monthly access rate component of our tariffs, that is to say the fixed rate component, recover all costs associated with the equipment and facilities between the customer premises and the cable modem terminating system or CMTS. These fixed costs would include all node segmentation costs and all costs associated with the CMTSs themselves.
1891 On the flip side, the Commission's interrogatories directed us to assume that the usage rate component of our tariffs, recover all other costs, including transport and the aggregated point of interconnection.
1892 This cost delineation is both incorrect and improper. It effectively treats the Cable Carrier's "access" or last mile facilities that connect the end-user to the CMTS as if they were no different from the telephone company's last mile facilities. However, a telephone company's last mile facilities are far less usage sensitive than those of a cable company. This telco-centric allocation results in an understatement of the Cable Carriers' usage-sensitive costs and an overstatement of our non-usage sensitive costs.
1893 To help explain the true nature of our networks, we have included in Schedule 2 to our remarks a simplified diagram of a typical cable hybrid-fibre coax, or HFC, network. As you can see, a large number of homes rely on a shared -- I will give you a moment to turn to that.
1894 As you can see, a large number of homes rely on a shared coaxial cable pipe to a node which is connected via fibre to a CMTS in a hub. Another fibre connection links our geographically distributed head-ends to the public internet cloud. Our CMTSs reside at our head ends.
1895 When traffic volumes increase in a given node, this puts more demand on the CMTS and the fibre facilities that serve that node. Eventually, we have to split the node, which means adding capacity to the CMTS and lighting more fibre. These costs are usage-sensitive costs and must be treated as such.
1896 To be more specific, the first cost element that that must be recovered through usage-sensitive charges is the line cards in the CMTS, including the associated equipment and maintenance of the cards. The CMTS is made up of a chassis and associated central processing components, as well as a number of line cards and provisioning ports within it. While the chassis is a somewhat fixed cost, the line cards and ports needed to support growing traffic volumes are not.
1897 The second cost element that must be recovered through usage-sensitive charges is the optical-electronic equipment and associated power equipment required to light additional fibre in the newly split node. While unlit fibre is a somewhat fixed cost, the incremental optical-electronic equipment required to transmit growing traffic volumes is not.
1898 We have reviewed this cost delineation issue at some length and we hope that our description has been of value. Whatever model you ultimately adopt, the first step is to put the right costs in the right place. If you get this wrong, your economic signals will be wrong.
1899 I will now turn to Ken who will outline our views on how best to recover those costs that have properly been placed into the usage-sensitive category.
1900 MR. ENGELHART: We now turn to the Commission's second issue, the methodology that should be applied to the usage rate component, volume or peak traffic.
1901 The Cable Carriers believe that a volume based form of usage billing is more appropriate than peak traffic. That is, we believe that there should be a charge assessed in cents per gigabyte of traffic.
1902 CNOC proposes measuring the peak traffic at the POI between the cable carrier and the wholesalers. They propose measuring the peak downstream traffic using a speed measurement and defining the peak as the busiest five minute period in a month after discarding the top 5 percent of the intervals.
1903 While this method is common for transit or backhaul services, TPIA is not a transit service. It is a point-to-multi-point distributed last mile or access service. Our last-mile network is the mirror opposite of a transit network.
1904 There are four concerns with the peak traffic proposal as put forward by CNOC in the form of the 95th percentile speed measurement:
1905 First, in the cable network the usage that drives costs are not at the POI with the wholesaler. They are at the neighbourhood nodes. Not only is POI peak traffic at the wrong place, it is at the wrong time. That is because each cable node has its own peak at a different time.
1906 For example, in the case of Rogers, only 13 percent of the nodes experience their peak traffic during the overall busiest traffic hour, 10 to 11 p.m., let alone during a five minute time slot as proposed by CNOC.
1907 The peaks of nodes occur over a wide range of times. Some have their peak at 4 a.m and some have their peak at 4 p.m.
1908 Furthermore, these peaks are not stable. They change from month to month and even from day to day. We provided evidence of this in our interrogatory response.
1909 If the CNOC proposal is adopted, the wholesalers' traffic that is causing a cable carrier to augment facilities at a node, will likely not be incurring any incremental usage based charges.
1910 Second, it creates incentives for wholesalers to design retail services which "game the system." They could for example sell a service to business customers that sends a steady stream of traffic through the cable network from 9 in the morning to 5 in the afternoon. The presence of this traffic would cause cable carriers to augment facilities at some nodes at considerable cost. However, the wholesalers will pay no incremental charges for this traffic, because it is not sent during their peak.
1911 Third, it is important to remember that in the cable network, peaks are not as important as in the telco network. This is because the cable network is shared in the neighbourhood. With a shared network all traffic impacts all other traffic, at the peak or otherwise.
1912 Think of the road system. A car parked in the right lane, in front of a coffee shop is going to affect your drive down that road. It will affect you more at rush hour, the peak, but it will affect you at the other times of the day as well. However, these localized traffic jams would not be reflected in the 95th percentile measurement and, therefore, have no cost impact on the wholesaler.
1913 Fourth, the CNOC proposal also ignores all upstream traffic. The cable carrier networks are severely bandwidth constrained in the upstream. They are asymmetric. The downstream pipe is much bigger than the upstream pipe. Under the CNOC proposal, all of their upstream traffic does not incur any incremental costs, even though it may be causing us to augment facilities.
1914 While volume and speed are both proxies for the underlying cost drivers, the Cable Carriers' proposal to use volume represents a better proxy because it is more closely aligned with the true cost drivers.
1915 A volume-based usage billing model would not impair the wholesalers' ability to compete with the Cable Carriers' retail services any more than one based on speed. The real issue is how the usage billing model would impact the network. For the reasons noted above, the risks of the peak model on network quality are significant.
1916 There are also implementation concerns with the 95th percentile model. It would cause us to invest a substantial amount of time and money to develop billing on this basis.
1917 A volume based pricing model is straightforward since it mirrors what we do in the retail market and it is used around the world. Conversely, a peak speed based model would be difficult. It would also be costly.
1918 We would also like to comment on the proposal by CNOC and MTS Allstream to charge the same fixed rates for all speeds. This makes no sense.
1919 The fixed costs include fibre, cable channels and the CMTS chassis. Higher speed services need more fibre capacity, more cable channels and more CMTS than lower speed services. So of course, a proper costing model will allocate more of these costs to a higher speed service.
1920 The request for a uniform fixed monthly charge for each speed tier is a transparent attempt to gain a regulatory arbitrage opportunity.
1922 M. MESSIER : Madame et Messieurs les Conseillers, je vais poursuivre en français.
1923 Les services Internet haute vitesse ont atteint un taux de pénétration très élevé dans les foyers canadiens. Les réseaux des câblodistributeurs sont parmi les plus rapides en Amérique du Nord. Cela a été confirmé par une étude récente de Netflix.
1924 Le Canada se doit de garder en place des mesures incitatives favorisant l'investissement par les fournisseurs de réseau actuels s'il veut maintenir ses avancées en matière de commercialisation de services à large bande de haute vitesse et de haute qualité.
1925 Le modèle de facturation suggéré par les câblodistributeurs est celui qui enverra les signaux économiques appropriés pour gérer le trafic.
1926 Toute obligation d'accès de gros augmente les risques reliés à l'investissement. Ces risques sont décrits en détail dans un rapport déposé par les câblodistributeurs, intitulé " Mandated Access in Canada ", écrit conjointement par Leonard Waverman et Kalyan Dasgupta.
1927 Les auteurs notent que le Canada est le seul pays où il existe une concurrence fondée sur les infrastructures entre deux réseaux largement déployés -- soit le câble et la téléphonie -- et que, malgré cela, ces deux réseaux sont également soumis à des obligations d'accès de gros.
1928 La situation unique au Canada soulève un certain nombre de préoccupations quand il s'agit de réglementer l'accès de gros. L'extrait suivant traite de ce point.
"In the Canadian context, however, the basis for applying regulation based upon typical notions of natural monopoly (where there is one platform owner who possesses significant and enduring market power) looks somewhat shaky in light of the fact that an alternative "pipe" reaches just as many homes as the incumbent network. Further, the disincentive effects of regulated access upon investment in new networks still apply, but there is some evidence also that these disincentive effects apply to all facilities-based investments, not just incumbent networks. In other words, the practical application of the "ladder of investment" has served to reduce the intensity and the beneficial effects (especially, but not exclusively, the dynamic long-term effects) of inter-platform competition."
1929 Les auteurs font une distinction entre la situation au Canada, où il existe une concurrence inter-plateforme, et celle prévalant au sein d'autres pays, notamment en Europe, où l'on a grandement misé sur l'accès de gros afin de permettre la concurrence à l'unique opérateur de réseau en place.
1930 Dans ces pays, lorsqu'un client passe à un concurrent, le fournisseur de réseau délaissé continue de percevoir une certaine forme de revenus en contrepartie de ses investissements.
1931 THE CHAIRPERSON: Excuse me.
1932 M. MESSIER : Oui, Monsieur.
1933 THE CHAIRPERSON: Apparently we have lost the translation.
1934 COMMISSIONER MOLNAR: Can you just start again at paragraph 32, please?
1935 M. MESSIER : O.K.
1936 Les auteurs font une distinction... Is it okay?
1937 THE CHAIRPERSON: Okay.
1938 M. MESSIER : Merci.
1939 Les auteurs font une distinction entre la situation au Canada, où il existe une concurrence inter-plateforme, et celle prévalant au sein d'autres pays, notamment en Europe, où l'on a grandement misé sur l'accès de gros afin de permettre la concurrence à l'unique opérateur de réseau en place.
1940 Dans ces pays, lorsqu'un client passe à un concurrent, le fournisseur de réseau délaissé continue de percevoir une certaine forme de revenus en contrepartie de ses investissements. Les fournisseurs de réseaux canadiens ne bénéficient pas d'une telle garantie.
1941 Dans les faits, contrairement à ce qui se passe ailleurs, les compagnies de téléphone et les câblodistributeurs canadiens mettent en place des infrastructures permettant le branchement de tous les foyers, tout en sachant pertinemment qu'ils n'ont aucune garantie quant aux revenus qu'ils pourront percevoir en retour.
1942 L'assiette des revenus est divisée. Superposé à cet environnement extrêmement concurrentiel, un régime d'accès de gros augmente considérablement les risques liés aux nouveaux investissements.
1943 Ce sont les investissements effectués par les câblodistributeurs dans la modernisation et l'expansion de leurs réseaux qui ont permis aux grossistes d'offrir au détail des services d'accès Internet à haut débit concurrentiels, et ce, à des vitesses de plus en plus élevées.
1944 Le Conseil ne devrait pas perdre de vue le fait qu'une politique publique misant sur des mesures favorisant l'investissement présente des bénéfices substantiels, et que l'imposition d'ententes de gros trop agressives risque de rendre impossible le recours à de telles mesures.
1945 Nous sommes particulièrement préoccupés par l'impact qu'aurait un régime d'accès de gros basé sur une délimitation incorrecte des coûts entre l'accès et l'utilisation ou encore sur une méthode de recouvrement des coûts d'utilisation déficiente.
1946 De telles erreurs rendraient les investissements davantage risqués, saperaient les mesures favorisant l'investissement et entraveraient notre capacité à gérer le trafic. En bout de ligne, notre capacité à fournir des services de pointe de haute qualité à tous les consommateurs d'Internet s'en trouverait affectée.
1947 Par exemple, si la répartition des coûts était trop axée sur le volet " accès ", seule une très petite partie des coûts reliés à l'usage pourrait être récupérée. Cela favoriserait des volumes extraordinairement élevés qui exigeraient que nous devancions et peut-être même que nous augmentions les investissements dans nos réseaux. Parallèlement, les tarifs d'utilisation très faibles résulteraient en des revenus minimes ne permettant pas de compenser les coûts d'investissement.
1948 De façon concrète, cela aurait pour effet de dissocier la cause de l'augmentation des coûts et la possibilité de récupérer ces coûts. Cela nuirait à l'incitation à l'investissement. De plus, cela transférerait des grossistes et leurs clients le risque associé à l'augmentation des coûts vers les câblodistributeurs.
1949 Et, bien sûr, nous devons tous garder à l'esprit que seuls les câblodistributeurs investissent dans les réseaux d'accès. Les grossistes n'investissent pas dans les réseaux d'accès et n'ont pas à supporter les risques reliés à des investissements substantiels.
1950 L'autre conséquence entraînée par un taux d'utilisation très bas porte sur la capacité des câblodistributeurs de gérer le trafic en utilisant les pratiques de gestion de trafic Internet de nature économique. Des frais d'utilisation exceptionnellement bas enverraient un mauvais signal économique.
1951 Choisir le mauvais modèle de facturation basé sur l'utilisation peut également avoir un effet négatif sur les mesures incitatives favorisant l'investissement et la gestion du trafic.
1952 La proposition du Consortium des opérateurs de réseaux canadiens à ce qu'une compensation soit accordée pour l'utilisation basée sur la période d'utilisation maximale agrégée ne tient pas compte du fait que nos réseaux sont sujets à des périodes d'utilisation maximale à divers moments au cours d'une journée donnée et à différents endroits au sein du réseau. Bien que ces périodes d'utilisation maximale aient un effet sur la planification et l'approvisionnement de notre réseau, elles peuvent devenir invisibles une fois le trafic regroupé vers un point d'interconnexion centralisé.
1953 La proposition du Consortium de ne payer que sur la base des périodes d'utilisation maximale observées, par analogie, " à la bretelle d'accès pour l'autoroute " repose donc sur un vice fondamental puisque ces périodes devraient plutôt être observées " aux intersections " à l'intérieur de la ville.
1954 De plus, en vertu de la proposition du Consortium, seul le trafic en aval ferait l'objet d'une compensation. Cela ne tient pas compte du fait que la capacité en amont du réseau est très limitée, ce qui le rend plus sensible à la congestion.
1955 Le modèle mis de l'avant par le Consortium dissocie les facteurs qui causent l'augmentation de nos coûts de l'obligation d'assumer ces coûts. Le résultat s'avère tout aussi dommageable sur nos capacités de gestion de trafic et notre disposition à investir dans le cas de l'établissement de frais d'utilisation trop bas.
1957 MR. BÉLAND: The Cable Carriers support symmetry in applying the mark-up on costs among all network providers.
1958 The Commission's approach to setting the mark-up for non-essential services such as TPIA has been not to constrain this to a specific or uniform mark-up such as the 15-percent limit imposed on essential services. While parties have argued before, and continue to argue in this proceeding, that a 15-percent limit should be imposed even on non-essential services, the Commission has rejected this approach and should continue to do so.
1959 It is important to recall that the purpose of a mark-up is to provide a contribution to the Cable Carriers' fixed common costs and a reasonable opportunity to recover variable common costs that exist in hundreds of activities that cannot be individually costed and explicitly included on a discrete item basis in the costing studies. The mark-up does not create a profit for the Cable Carriers; it allows the Cable Carriers to recover actual costs that cannot otherwise be captured in the cost studies.
1960 The fact that the mark-up for non-essential services is not fixed at 15 percent does not mean that the rates for TPIA services are not cost-based. The rates that the Cable Carriers have filed in this proceeding are based on cost studies with the same mark-up that the Commission has approved for TPIA services over the past decade.
1961 There is an asymmetry in the current wholesale access regime and it favours the ILECs. The Commission granted the ILECs a higher mark-up for their higher speed wholesale access services because these are based on FTTN networks. The ILECs were permitted to increase the mark-up for these services by an additional 10 percent over and above the mark-up that they had been applying to their legacy wholesale access services.
1962 The Cable Carriers' TPIA services are also based on FTTN networks. If there is to be symmetry, then the Cable Carriers should be permitted a comparable mark-up for our TPIA services.
1963 We would also note that Cable Carriers have always been required to do 10-year cost studies, while the phone companies have been permitted to do shorter cost studies of five years, or in the case of TELUS, only three years. The effect of course is likely to give the telcos higher costs.
1964 In the current proceeding, the Commission asked the phone companies to do 10-year studies and they flatly refused. We also believe that 10-year studies are not appropriate for Cable Carriers either.
1965 The same mark-up should apply to both access and usage, whether the access and usage components of the service are billed jointly or separately. A TPIA customer would need to subscribe to the access component, charged on a fixed amount per end user, and the usage component, charged on a volume of usage aggregated over all end users served by the TPIA customer. Neither component on its own would allow a TPIA customer to provide a functional retail high-speed Internet access service.
1966 TPIA service has been treated as non-essential since the service's inception. The Commission's reevaluation of the service under its essential services framework in Telecom Decision 2008-17 verified that this is the appropriate classification for the service, taking into consideration the access and transport functionalities involved. The non-essential status was most recently confirmed in TRP 2010-632.
1967 Regardless of whether there is some usage allowance included in access or it is billed for separately, the nature of TPIA service remains the same: both last mile and transport functionalities are intrinsic to the service. There is no rationale for applying a different mark-up to different components of the same service.
1968 The Commission has already established uniform mark-ups for other parts of the TPIA service, such as the Point of Interconnection and associated service charges. The same reasoning applies to setting the mark-up for the access and usage components.
1969 Les services d'accès Internet de gros et de détail des câblodistributeurs sont basés sur des réseaux à forte densité de fibre. La fibre est déployée de plus en plus près des foyers de nos clients sur une base continuelle, et ce, dans le but d'augmenter la capacité, d'éviter la congestion et de satisfaire aux exigences des consommateurs en matière de nouveaux services avancés de haute qualité.
1970 Les réseaux des câblodistributeurs devraient être considérés comme étant des réseaux de nouvelle génération, tout comme les réseaux dits de " fibre jusqu'au répartiteur " des compagnies de téléphone. En réalité, vu sous cet angle, aucune des composantes des réseaux des câblodistributeurs ne peut être considérée comme étant de nature " traditionnelle. "
1971 Il est également faux de prétendre que le réseau câblé ne nécessite que quelques investissements mineurs additionnels afin d'être en mesure de répondre aux exigences du marché pour des nouveaux services avancés à haute vitesse. L'investissement des câblodistributeurs dans nos réseaux est substantiel et continuera de l'être tandis que la fibre continue d'être déployée de plus en plus près des foyers de nos clients.
1972 L'approvisionnement des réseaux DOCSIS 3.0 nécessite également de nouveaux investissements considérables, mais cela n'élimine pas d'autant le besoin d'accroître sans cesse les installations de fibre optique. Bien au contraire, puisque le déploiement de la technologie DOCSIS 3.0 fait en sorte que la capacité de nos réseaux est de plus en plus sollicitée.
1973 Nous serions très inquiets si le Conseil en venait à considérer les services actuels offerts par les câblodistributeurs comme étant des services " traditionnels " simplement parce que ces services sont déjà largement disponibles dans les marchés que nous desservons.
1974 Toute ligne de démarcation basée sur un point arbitraire dans le temps quant au moment où un service a été lancé violerait le mandat confié au Conseil par les Instructions de voir à ce que toute réglementation soit technologiquement neutre et de nature symétrique.
1975 In conclusion, if Canada is to build upon its past success of providing Canadians with access to high-speed, high-quality broadband services, the Commission's decision in this proceeding must uphold our incentives to invest in networks. This requires a model that relies on the right costing and a rate structure that sends the right economic signals.
1976 The Cable Carriers believe that the model we have proposed will achieve these goals. In summary, our positions on the Commission's six issues are as follows:
1977 1. Our aggregated usage model as proposed in March remains our preferred model.
1978 2. There should be a usage component based on total volume, not peak traffic measured at the 95th percentile.
1979 3. All models of wholesale access increase the riskiness of investment, particularly in Canada where we have the advantage of facilities-based, inter-platform competition. However, our model will be less likely to interfere with these incentives or our ability to manage traffic.
1980 4. We support symmetry in the mark-up on costs.
1981 5. The same mark-up should be applied to all wholesale rates.
1982 6. Special treatment of legacy services is not justified.
1983 We would now be pleased to respond to any questions you may have.
1984 THE CHAIRPERSON: Thank you very much for your presentation. It was very clear and you followed the outline that we asked and I appreciate that.
1985 In your first point you say on page -- I guess it's page 2, point 1:
"The monthly usage allowance will be based on each Cable Carrier's average monthly usage and will be adjusted each year as traffic per end-user increases. "
1986 And then at some other point you say:
"The adjustment could be based on updated forecasts for the forthcoming year or the previous year's actual volumes."
1987 Which one are you actually going to use? Are you going to use forecast or previous years?
1988 MR. BÉLAND: Frankly, we are open to both. We've heard in the proceeding to date differing points of views from different parties -- or I should say differing concerns from different parties.
1989 The value of fixing the amounts now based on the presumptions that are currently in our cost studies is a value of certainty. The value of fixing the amounts each year based on the real experience in the previous year may be a value of accuracy and we're open to both.
1990 THE CHAIRPERSON: What happens if an ISP -- for argument's sake, let's say -- you say you're open to both -- we adopt the forecast and an ISP finds out it forecasted wrong and its traffic is actually increasing much faster than it expected, is there some sort of remedial action it can take?
1991 MR. BÉLAND: Well, to be clear, the average consumption that would be included in our tariff, the allowance that would be included, is an all-company average. So that will be the average of the entire end-user base of each of our companies at each speed.
1992 So an ISP wholesaler who finds itself with an average that is higher than an assumed average, they are now into the realm of the usage-based charges and what we've now structured for them is a consolidated regime where they will pay the cost-based rate but pay on a consolidated basis.
1993 So, for example, a very-low-volume user would be able to offset a very-high-volume user and they end up really paying on only the increment that they have above the network-wide average.
1994 THE CHAIRPERSON: Can you say that again? I am not sure I quite understood that.
1995 MR. BÉLAND: Sure.
1996 THE CHAIRPERSON: So, let's take a specific -- an ISP buys from you and finds out it mis-estimated and the usage is actually 50 percent higher than they thought for that year. What do they do? What's the impact for them?
1997 MR. BÉLAND: Well, first thing first, there's no requirement for them to estimate in advance. We have no prepaid price and postdate price. So we're not asking them for an estimate. That's the first point.
1998 What will happen is month by month, based on their real usage, they will pay us according to our tariff.
1999 If their real usage for a given month on average is higher than the allowance that is included in the tariff, then they will pay for those gigabytes at the cost-based gigabyte rate.
2000 THE CHAIRPERSON: Okay. That I understand, but I thought in my earlier question you said you were open to whether you do the adjustment on the basis of forecasts or on the basis of previous years.
2001 MR. BÉLAND: Yes. That's purely an issue of tariff-setting mechanics.
2002 THE CHAIRPERSON: Oh, just for the setting, but not on a monthly basis?
2003 MR. BÉLAND: That's right. That's right.
2004 THE CHAIRPERSON: I see. Okay.
2005 And then you went through quite some detail to point out to us that the cable networks are different than the ILEC networks and that our interrogatories were wrong and that what should be considered fixed costs and what should be incremental costs are different for you.
2006 Now, if I accept this, does that mean at the end of day we have a different model for the cable companies and for the telco companies or is it just that we have to do different cost allocations?
2007 MR. BÉLAND: We are not proposing a different model.
2008 THE CHAIRPERSON: Yes.
2009 MR. BÉLAND: The only issue you need to consider is that when you're making the delineation between the fixed and usage costs for each type of network, you make the appropriate delineation for that network.
2010 THE CHAIRPERSON: Use one set of delineation for cable and another one for telcos? That's really what you're --
2011 MR. BÉLAND: And to the extent that that truly reflects the nature of the networks of each category of networks, that's entirely consistent with the symmetrical regulatory principles.
2012 THE CHAIRPERSON: And what is the advantage of your system over the AVP that Bell proposed yesterday?
2013 MR. BÉLAND: Several advantages.
2014 One is we wipe away this distinction between pre-purchasing and post-purchasing which seems to cause some consternation, a single cost-based rate.
2015 We also think that our proposal reflects very well the true cost structure of a network and also provides maximum flexibility for the wholesaler to design its own retail plans.
2016 THE CHAIRPERSON: How does it do this more than the Bell model?
2017 MR. BÉLAND: In part by wiping away that distinction between the prepaid and post-paid.
2018 THE CHAIRPERSON: That's really the key distinction. The disincentive or the disciplining effect of the two-tier rates from Bell, that's really the main differentiation, if I understand it.
2019 MR. BÉLAND: Another important distinction relates to the fact that you remember that Bell has proposed one model for legacy and one model for --
2020 THE CHAIRPERSON: Yes, right.
2021 MR. BÉLAND: -- non-legacy let's call it. Ours is a single model which bundles some usage in with the base monthly rate and we think that's a rational model that reflects how networks are built in the real world. In the real world networks are built to accommodate a certain level of usage, they are not built to accommodate no usage.
2022 THE CHAIRPERSON: So you don't accept Bell's argument that you are causing an awful lot of dislocation and confusion by applying the same model to legacy as well as FTTN.
2023 Bell's point yesterday was basically the legacy costs, everybody knows they work, people know how to live with them, et cetera, don't touch them, let's just concentrate on the FTTN. Logically applying the same system to both makes sense; in reality the pragmatic effect of changeover, et cetera, you are just going to cause a lot of uncertainty and dislocation.
2024 MR. BÉLAND: To be frank, we heard that the whole discussion is a very Bell-specific discussion. We have put forward what we think is the rational model with some bundling of usage, it's consistent with what we do today, it will increase flexibility for wholesalers relative to what we do today, that makes sense to us.
2025 MR. ENGELHART: Mr. Chair, if I could just --
2026 THE CHAIRPERSON: Mr. Béland, we accept what you are suggesting for the cable companies, but for the telcos let's leave legacy as it is and only apply your system to the FTTN.
2027 Would that cause you any problem?
2028 MR. ENGELHART: Actually, our system is quite similar, Mr. Chair, to the Bell legacy proposal, so what we are really saying is that is the one that we think makes the most sense.
2029 THE CHAIRPERSON: Tom, you have some questions.
2030 COMMISSIONER PENTEFOUNTAS: Yes.
2031 The learned Chairman went through some of the questions I may have had, but yesterday Bell mentioned the fact that the cablecos are the big boys in this field, in the field of internet service. Globally Rogers -- cable companies in general have about 55 percent market share, Bell I would say is around 39 and the independent ISPs are about 6, but as a percentage of the independent ISPs, Bell is basically the only show in town.
2032 Why is that, that Bell dominates offering services to these wholesale companies?
2033 MR. BÉLAND: It partly reflects historical realities. At the very beginning when independent ISPs were getting into this business, deciding I would like to resell someone else's network, or have access to someone else's network in order to provide services myself, they were focused to a large extent on the business market, residential and business.
2034 If you look at just the footprint of cable networks versus telco networks, telco networks are ubiquitous, they cover effectively all homes and all businesses in the country. Ours are much more residential-focused, we have some business presence, it's growing.
2035 So at the time that independent ISPs got into this game they were looking at two possibilities and they saw one that gave access to both residential and business and one that gave access to only residential. I think that's a big factor, probably not the only factor, but that is a big factor of why they went that direction.
2036 CONSEILLER PENTEFOUNTAS : Avez-vous essayé de cultiver cette clientèle de grossistes? Est-ce que ça fait partie de votre plan d'affaires?
2037 Peut-être monsieur Messier peut répondre à ça, ou monsieur Béland.
2038 Avez-vous intérêt à cultiver cette clientèle-là?
2039 M. MESSIER : Bien, écoutez, on offre ce service-là et on répond à chaque demande. On a un certain nombre de points d'interconnexion qui ont été développés dans le passé. On a un certain nombre de clients qui sont là. Donc, on répond adéquatement à chaque demande qui nous est faite.
2040 Maintenant, si vous me demandez si on a intérêt, je pense que notre position est assez claire sur le dossier. On ne croit pas que c'est une obligation qui devrait s'imposer. Maintenant, c'est une obligation que vous nous avez transmise, c'est une obligation avec laquelle on vit, et puis on essaie de servir nos clients wholesale du mieux que l'on peut.
2041 Si vous me demandiez, si on était dans un contexte complètement déréglementé, est-ce qu'on en ferait? Une question d'affaires...
2042 CONSEILLER PENTEFOUNTAS : Ça ne vous intéresserait pas du tout?
2043 M. MESSIER : Bien, ce serait à examiner, effectivement, mais les conditions seraient totalement changées puisqu'on serait dans une condition où est-ce qu'on aurait à négocier directement avec eux et à revoir notre modèle d'affaires, à revoir...
2044 Donc, présentement, on fonctionne dans le cadre des conditions qui ont été approuvées par le Conseil.
2045 CONSEILLER PENTEFOUNTAS : Mais les conditions seraient changées dans quel sens si on fait abstraction du fait que c'est réglementé présentement?
2046 M. MESSIER : Bien, lorsque nous avons eu l'instance sur les services essentiels, il y a environ sept ans, nous avions -- six ans maintenant -- nous avions examiné un peu la possibilité -- et puis, je pense que Bell avait expérimenté -- d'aller directement dans un régime de négociation entre les parties.
2047 Est-ce que cela... Est-ce qu'il y aurait un point de rencontre entre nos intérêts, le modèle d'affaires que nous pourrions développer, et aussi la demande que les fournisseurs pourraient avoir? Je pense que c'est une question qui est ouverte maintenant.
2048 Pour être franc, nous n'avons pas développé quelles seraient les conditions dans un tel contexte de déréglementation. Nous vivons avec le régime que nous avons débattu ici, il y a environ... les deux dernières années, où est-ce qu'on a discuté, entre autres, des questions d'établissement du -- des différentes conditions. Alors, c'est dans ce régime-là, dans cet univers-là que nous fournissons des services actuellement.
2049 CONSEILLER PENTEFOUNTAS : C'est bien compris.
2050 Just briefly, to get to some general issues before we get into some more substantive issues, we heard a lot about investment in internet services and we understand the importance of it, we heard some comparisons with respect to OECD figures.
2051 Have you looked at those figures and done some kind of comparative and are you comfortable with where we are in Canada in terms of investing in our internet network as it regards our OECD partners?
2052 MR. ENGELHART: Yes. First of all, household penetration in Canada is quite high. We have penetration, I believe in the top 10 percent of the OECD countries.
2053 Second, speeds in Canada are quite high and you have to be a bit careful when you look at the OECD numbers because they look at advertised speed and those are notoriously unreliable.
2054 If you look at a measurement like Akamai, which is a measure of real speed, we are again in the top 10. So we have very broad penetration of our broadband and we have very high speeds compared to other countries.
2055 COMMISSIONER PENTEFOUNTAS: Okay.
2056 Is there a bandwidth crisis, simply put?
2057 MR. ENGELHART: Well, thank you very much for that question because we don't think so. One of the things that disturbs us about this whole debate is this notion that this is all about congestion.
2058 You only saw the word "congestion" in our opening remarks once and that was that we make the investments to avoid congestion. So we work very hard so that there is no congestion.
2059 This is not about a bandwidth crisis, this is not about a crisis at all. There are some costs that are usage sensitive. When costs are usage sensitive, that means that the more you use, the more you spend and it's appropriate to send the right economic signals, it's appropriate to charge for usage-based costs on a usage basis and if people want to pay that price and use tons and tons of bandwidth, we are happy to supply it.
2060 So we don't see any crisis, we just think that pricing should involve the correct signals.
2061 COMMISSIONER PENTEFOUNTAS: Is it possible to calculate cost -- and we are talking about cost-based rates -- given the fact that costs are constantly decreasing, given technological developments.
2062 Would you be in agreement with that statement?
2063 MR. ENGELHART: Yes. I mean, Moore's Law does reduce the cost of chips so that they get more powerful as time goes by.
2064 I would caution, though, that there are two caveats I would put on that.
2065 One is, digging holes in the street is not subject to Moore's Law. Digging holes in the street is getting more expensive not less expensive as time goes by.
2066 The other thing is, if you buy a new computer every two years, yes, it's way more powerful than your old computer, but you still had to buy a new one every two years. So it's not like your family computer costs are going down, you just have a more powerful computer.
2067 COMMISSIONER PENTEFOUNTAS: Don't talk to me about it, my computer costs are going up.
2068 There was also a lot that was mentioned with regard to measurement and accuracies and I would like you to perhaps answer that question.
2069 Some people may not be confident that you can accurately measure consumption. Would you address that question? Is that a legitimate issue?
2070 MR. BÉLAND: I will address that question from Vidéotron's perspective. We are the cable carrier currently that does apply usage-based billing charges at both retail and wholesale.
2071 There is simply no issue with regards to the accuracy of our billing charges at retail or at wholesale. Every once in a while there is someone who comes along and questions whether they really use that many gigabytes in the month, just like in the telephone network there is someone who questions, "Did I really call Alberta that many times in that month?", but those are isolated and anecdotal.
2072 Our measurements of gigabyte consumption are extremely, extremely reliable.
2073 COMMISSIONER PENTEFOUNTAS: Do you have a problem with where we measure consumption? It has been raised that there are different ways of measuring it and there are different places to measure it.
2074 MR. BÉLAND: Well, in fact that --
2075 COMMISSIONER PENTEFOUNTAS: Does that raise a concern for you?
2076 MR. BÉLAND: That has been raised as a concern with regard to the peak traffic sorts of measurements. That in fact is one of the advantages of the gigabyte volume sort of measurement, is that it is very reliable. You add up the volume being used by all of the modems and that's your total volume figure.
2077 COMMISSIONER PENTEFOUNTAS: You mentioned peak traffic and my question basically would be: Doesn't peak period aggregate pricing more closely emulate actual incumbent costs?
2078 MR. ENGELHART: I think that's a fair statement if you were measuring the peak at the right place. As I mentioned in our opening remarks, measuring it at the point of interconnection between the cable network and the wholesaler network is completely the wrong place.
2079 We have a bunch of little peaks, 20,000 of them, spread throughout our network and to look at the point of interconnection between our network and the wholesale network, the peak that you end up with is an average of a whole lot of small peaks.
2080 So it's sort of like the old joke about averages, you have one foot by the fire and one foot in a bucket of ice water and on average you are comfortable. You know, it doesn't really reflect what's going on in those individual nodes.
2081 So you could have an ISP that is driving us to augment facilities in an individual node and it would have very little impact on the peak. So that's why if you are talking about peak pricing you have to be down at our neighbourhood node.
2082 The second thing is, the whole idea in telephony of looking at the peak was really built on the fact that the network, the telephony network was a statistical average of a lot of different customers making phone calls and because you were averaging a lot you had kind of a stable peak. The legend, whether it's true or not, was it was on Mother's Day. That's when everybody used the phone network.
2083 But when you are down in the cable node you are talking about 300-400 homes. That is not a big enough group to have a stable peak. So our peak changes from day to day, month to month, week to week in those nodes and it's not that important because the thing that keeps my friend Dermot here working is not the peak, it's the annual growth, that 40 percent growth that CIPPIC talked about.
2084 It follows as night follows day that Dermot has to augment those nodes every two years. It doesn't matter if the peak is at 2:00 in the afternoon or 2:00 in the morning, that node has to be augmented. So the peak isn't that important.
2085 The last point I would mention is that the cable network down in those neighbourhood nodes is a shared network. Everybody is riding on the same pipe. So again, the peak is not as important because every gigabyte affects the other gigabytes.
2086 COMMISSIONER PENTEFOUNTAS: Now, you are talking about nodes and talking about node splitting at one point, is that usage-based or is that part of a fixed cost?
2087 You understand my question? Go ahead, Mr. O'Carroll --
2088 MR. O'CARROLL: I understand your question, Commissioner.
2089 COMMISSIONER PENTEFOUNTAS: -- to save some time.
2090 MR. O'CARROLL: It's very much usage-based. As Ken mentioned, we have 20,000 ports in our network that --
2091 COMMISSIONER PENTEFOUNTAS: I'm just going to interrupt you, I understand it's usage-based, but it's not usage-based specifically as it pertains to the independent ISP usage.
2092 MR. O'CARROLL: Well, it's usage-based as it pertains to the customer's behaviour.
2093 COMMISSIONER PENTEFOUNTAS: The end user absolutely, yes.
2094 MR. O'CARROLL: The end user. As Ken mentioned, in any particular node there is a relatively small number of customers and we monitor those individual nodes on a continuous basis and every single night, 365 days a year, we are adding capacity to some of our nodes. We have 20,000 ports, all of which are exhausting based on customer usage.
2095 COMMISSIONER PENTEFOUNTAS: I understand that, but it's end customer usage and it's not necessarily independent ISP end customer usage. That's the only point I really want to make, unless you can tell me otherwise.
2096 MR. O'CARROLL: A wholesaler's --
2097 COMMISSIONER PENTEFOUNTAS: Yes.
2098 MR. O'CARROLL: -- third party customer can dramatically impact the exhaust on any particular node because, depending on the behaviour of the customer, as few as 10 customers can exhaust a node. So individual customers can have significant impact on that exhaust.
2099 COMMISSIONER PENTEFOUNTAS: Okay.
2100 Back to your model, we have an aggregate model whereby there is a certain percent -- we would take all the customers, your customers and the wholesalers' customers and we get an average, right, and we have a rate for that. Anything above and beyond that, there is a surcharge. The costs on the additional usage are no different than the costs of the aggregate usage.
2101 So how would you justify an increased surcharge on the additional usage?
2102 MR. BÉLAND: You are correct in saying that the costs on the additional usage are no different than the costs on the amount of usage that has been bundled in with the access --
2103 COMMISSIONER PENTEFOUNTAS: Okay.
2104 MR. BÉLAND: -- but all of that is usage. You use more, it costs more. So someone who is using more than the average is imposing --
2105 COMMISSIONER PENTEFOUNTAS: so there is no cost-based justification --
2106 MR. BÉLAND: -- is imposing greater costs on our network.
2107 COMMISSIONER PENTEFOUNTAS: In other words, there is no cost-based justification for that surcharge?
2108 MR. BÉLAND: No, in fact the opposite, it's extremely cost-based. It's as usage goes up the costs imposed on our network goes up. Part of that usage will be rolled in with the monthly access, but everything above and beyond that allowance, there needs to be compensation for that.
2109 COMMISSIONER PENTEFOUNTAS: But there is no additional cost. So many terabytes are being used, you have a cost. The additional terabytes that may be used, there is no additional fixed costs on that excess usage.
2110 MR. BÉLAND: No. In fact there is, because those additional terabytes that are being used are potentially --
2111 COMMISSIONER PENTEFOUNTAS: That's what causes you to start node splitting --
2112 MR. BÉLAND: -- potentially blowing a couple of our nodes, yes.
2113 COMMISSIONER PENTEFOUNTAS: -- and fibre adding.
2114 MR. BÉLAND: We are expanding, we are spending money to accommodate that usage.
2115 COMMISSIONER PENTEFOUNTAS: What would be the problem with Bell's model? It provides additional risk to the wholesaler, because they have to predict what they are going to use. That would help you also know what potential consumption there may be on your network.
2116 Where is the problem with their model?
2117 MR. BÉLAND: Again, there is not a big difference between Bell's legacy model and our model. The biggest difference is probably this issue of prepaid and post-paid whereas we --
2118 COMMISSIONER PENTEFOUNTAS: Setting aside the prepaid/post-paid issue.
2119 MR. BÉLAND: There is not a huge difference between the two models. They are quite close.
2120 COMMISSIONER PENTEFOUNTAS: But you are not asking your clients to predict their usage to the wholesale client; Bell would be.
2121 MR. BÉLAND: That's correct, we are not asking.
2122 COMMISSIONER PENTEFOUNTAS: You wouldn't see an advantage for you to know and to have them predict their usage in planning for your network capacity?
2123 MR. ENGELHART: Well, that's one reason why our model bundles some usage in with the access. So when they order 1,000 lines, we know they are at least prepaying for 1,000 bundles of usage. So it provides some stability that way.
2124 COMMISSIONER PENTEFOUNTAS: I think it was pretty clear -- I had a question on the CNOC model and the fact that there were technical difficulties. I think you were quite clear on that, that technically it would be a nightmare.
2125 Je pense que monsieur Messier a touché... ou c'est monsieur Béland qui lisait cette partie-là de votre présentation. Je pense que vous n'avez rien à ajouter là. Ça serait vraiment difficile au niveau technique et technologique de faire face au modèle qui a été présenté par CNOC; c'est exact, Monsieur Messier ou Monsieur Béland?
2126 M. BÉLAND : Fondamentalement, c'est non seulement difficile, mais ce n'est pas exact, parce que ce qui se passe au point agrégé en termes de peak, non-peak, ça n'a aucun rapport avec ce qui se passe dans le réseau, au noud distribué dans le réseau.
2127 COMMISSIONER PENTEFOUNTAS: There was also a point made, I think it was paragraph 4(iii):
"... ensures that the TPIA tariffs foster more usage of the internet ..." and not the contrary.
2128 How would that be the case? How would your model foster more usage?
2129 MR. BÉLAND: It fosters more usage by setting an accurate usage-based price for the variable component. So there is no punitive element, let me put it that way. The cost is the cost, you want to use it more, you use it more, just pay us the tariff rate.
2130 COMMISSIONER PENTEFOUNTAS: You don't mind them using it as long as they pay for it.
2131 MR. BÉLAND: Exactly.
2132 COMMISSIONER PENTEFOUNTAS: Okay. You are not trying to manage consumption in that sense? Managing consumption doesn't interest you, you are just interested in getting paid for the service you provide in other words?
2133 MR. BÉLAND: If the tariff is properly structured --
2134 COMMISSIONER PENTEFOUNTAS: Yes.
2135 MR. BÉLAND: -- and fixed costs are where they belong and usage costs are where they belong, go for it.
2136 COMMISSIONER PENTEFOUNTAS: Internet traffic management is not your concern. You don't stay up at night worry about that?
2137 MR. BÉLAND: No. In fact, a properly structured tariff is a key component of internet traffic management.
2138 COMMISSIONER PENTEFOUNTAS: To managing, okay.
2139 I think it's the last question. Juste pour retourner sur quelque chose que monsieur Messier a lu, et j'essaie de comprendre ça.
2140 Votre investissement pour avoir la fibre jusqu'au répartiteur, est-ce différent pour les câblos que pour Bell, par exemple, ce dernier mille-là? Y a-t-il une véritable différence dans vos coûts par rapport à leurs coûts?
2141 M. MESSIER : Je ne pourrais par vraiment... Je n'ai pas comparé les deux modèles de ce côté-là. Mais de notre côté, l'ensemble des coûts qui sont facturés pour la partie accès...
2142 CONSEILLER PENTEFOUNTAS : Il va falloir que vous vous rapprochiez du micro.
2143 M. MESSIER : D'accord.
2144 CONSEILLER PENTEFOUNTAS : A mon âge avancé, c'est la première chose qui nous laisse.
2146 M. MESSIER : Alors, comme j'ai dit, on n'a pas comparé les deux modèles. Donc, je ne peux pas me prononcer là-dessus, tout simplement.
2147 CONSEILLER PENTEFOUNTAS : O.K. Ça va. Merci.
2148 THE CHAIRPERSON: Thank you.
2149 Before I give you to the tender mercies of Mr. Katz, I asked Bell what does the model do for consumers, let me ask you the same question.
2150 MR. ENGELHART: Well, as you know we are not huge proponents of the need for a wholesale tariff, but we are not fighting that today. We are here to be as helpful as possible.
2151 There are certainly those who believe, as CIPPIC mentioned, that the wholesale tariff creates competitors that provide competition, so it will give consumers a competitive outlet where wholesalers can design services that meet consumer needs.
2152 THE CHAIRPERSON: So all the benefits are indirect. It's not a direct benefit to consumers, it's indirect by providing more competition and consumers getting the benefit of competition?
2153 MR. ENGELHART: And by allocating the costs properly, we believe that it will continue to incent us to build networks that are fast and reliable.
2154 THE CHAIRPERSON: As a Commission, we are quite concerned about the disproportionate amount of third party traffic that Bell carries, as opposed to the cable companies. If we adopt this model, do you see that more of the ISPs will switch to you as their provider, rather than the telcos?
2155 MR. ENGELHART: First, I think the gap is between Bell and everybody else, not really between cable and telco.
2156 THE CHAIRPERSON: Okay, let's say between Bell and the cable companies.
2157 MR. ENGELHART: TELUS mentioned yesterday that they have about 4 percent end users. I believe that Rogers has 2 percent.
2158 And even without the changes that the Commission has mandated, that number is growing fairly rapidly for Rogers.
2159 I think the fundamental problem, as Dennis said earlier, is the difference in our footprint. You heard the B.C. Broadband Association say that yesterday. They said that they make most of their margin on business customers, not residential. Our networks are mostly residential. So I do think that gap will continue.
2160 THE CHAIRPERSON: Okay. Len...
2161 COMMISSIONER KATZ: Thank you, Mr. Chairman.
2162 Good morning. I want to continue with the discussion you had with the Vice-Chairman regarding your model. You are saying that your model is very, very similar to the Bell legacy model. That model has an imputation of some usage in there, which is what your model has, as well.
2163 Why are you so insistent in your model that there be some component of usage bundled into your access charge, as opposed to having them totally separate?
2164 MR. BÉLAND: Fundamentally, and very briefly, Mr. Vice-Chair, that is because that reflects how networks are built in the real world. No one goes out and builds a network on the basis that no one will use it, and then bills usage afterwards. You build a network because you expect people to use it.
2165 COMMISSIONER KATZ: But it's not a matter of billing them afterwards, it's a matter of billing them for consumption, distinct from the network component, the fixed component.
2166 When you get your phone bill, you get the front page, which is your flat rate monthly charge, and the back page is your long distance. You don't bundle a certain amount of long distance -- maybe you do today -- into local.
2167 MR. BÉLAND: I will hand it over to Dermot in a moment, but, again, what is artificial about the CNOC proposal and the suggestion that you are able to delineate costs on that basis -- what they are basically saying is that cablecos, or telcos for that matter, go out and build networks in two stages. You build something that is called access, and everybody has a pipe. It is ready to go. Then you build capacity that is going to handle what people are going to do with that pipe. That is not how networks are built in the real world.
2168 COMMISSIONER KATZ: Yes, but the distinction I am trying to make is not your model to CNOC's, it is yours to Bell's, which we heard yesterday. All I am saying is, why can't there be a separation between usage -- pay as you go, you pay for what you use, subject to what is included in costs -- and a flat rate for that which is fixed, basically?
2169 MR. ENGELHART: Vice-Chairman Katz, of course it could be done that way, and given that most -- I think all TPIA wholesale customers have a higher average monthly usage than our own customers' average monthly usage, mechanically it would work out to be the same.
2170 I guess the only disadvantage to our proposal for TPIA wholesale customers is if they want to provide very tiny packages to customers. We don't see any evidence that they want to, and even if they did, our aggregation model would allow them to give some small packages and some big packages.
2171 So we think it is the right way to do it, but mechanically you are right, it could be done either way.
2172 COMMISSIONER KATZ: That leads me to the next question. You probably heard my discussion with Bell yesterday with regard to the additional cost that their model includes if the independent ISP misses his purchases.
2173 What I was talking to him about was not giving them extraordinary or super-normal profits, because they would benefit from that, but blending it in the rate in some way, as well.
2174 But what I was trying to get at here was to try, to the extent possible, to put the business risk where it belongs. Let the independent ISP assume some degree of business risk for his business, rather than passing all the business risk to the cablecos or the ILECs, while, at the same time, making sure that there are no additional profits that are being borne by the ILECs and the cablecos.
2175 Your model doesn't do that. You are accepting that business risk. You are saying: We will serve them at one price. We will give them whatever they want, when they want it.
2176 In my days working for Ted Rogers, he always liked his competitors to have as much business risk as possible, and here you are saying: I'll take it.
2177 MR. ENGELHART: I certainly sympathize with what you are trying to accomplish, but I don't think it is possible when you have a wholesale tariff regime. The business risk sort of occurs when that investment decision is made.
2178 If I drew an analogy to a hotel, I am building a big, fancy hotel, I am spending the $100 million to build the hotel, and then I hope that people will rent rooms to pay for my investment.
2179 The wholesale tariff is really saying that the wholesalers have the right to rent rooms, anytime they want, at 50 percent off. They don't have to rent any rooms if they don't want to, or they can rent as many as they want at 50 percent off.
2180 So the idea that, "Well, maybe if they paid 55 percent or 60 percent, that compensates me for my risk," no, it doesn't. The only thing that compensates me for my risk is if you are a partner when I am building that hotel.
2181 So the very nature of a wholesale tariff means that the person who is investing in the network bears the risk, and higher rates would certainly be beneficial to the network builder, but even a higher rate doesn't compensate us for the risk that we incur.
2182 COMMISSIONER KATZ: I put this question to other parties, as well, so I will pose it to you. Do you know of any correlation between peak and total usage?
2183 Have you done anything? Have any of your companies taken a look at whether increased usage across the board actually increases peak?
2184 MR. BÉLAND: On an individual end user basis, it is largely self-evident. An end user that uses a lot more than another end user will tend to have a peak that is a lot higher than another end user. So that correlation is there.
2185 Where it breaks down is on a network-wide basis, when you try to bring that all out to an aggregation point somewhere and say: That peak over there is correlated with all the usage and peaks of usage throughout the network.
2186 That's where it breaks down.
2187 COMMISSIONER KATZ: This is my last question. I am trying to help you here, Mr. Béland, get more business, even though, I guess, you may not want it.
2188 When I look at your Schedule 1 and I look at your rate -- and I respect the fact that it is coming down based on some forecast of demand, but in 2011 you are looking at a 48 cent rate, and I am assuming that it is cost-based, based on your costs.
2189 Yesterday, Bell came in with a usage-based rate at 19.5 cents.
2190 You are uncompetitive. Why would someone want to buy your service?
2191 And, yet, the cablecos have the lion's share of the business.
2192 MR. BÉLAND: I will be frank with you, and I will reflect the comments of Mr. Messier earlier. I am not here to market a service, I am here to make sure that the tariff structure for that service accurately reflects our costs and sends the right signals.
2193 That's my measure of success.
2194 COMMISSIONER KATZ: Rogers, you are at 29 cents versus 19.5 cents.
2195 MR. ENGELHART: The whole reason we discussed all of the node structure that is inherent in our network was to convey the idea that we have a more usage-sensitive network than the phone company does. Their network is more fixed, ours is more variable with usage. So it would follow that we would have a higher usage-based rate.
2196 COMMISSIONER KATZ: Would it follow that your fixed rate is lower than the ILEC fixed rate, if you have a lower fixed and a higher variable?
2197 MR. ENGELHART: I would have to look at it, but, I guess, the two networks are not completely comparable, in that ours is faster.
2198 The old DSL networks, before they brought in their five services, just couldn't provide anywhere near the speed and reliability that cable can.
2199 So I wouldn't expect the total number to be the same.
2200 COMMISSIONER KATZ: When I look at the cablecos' rates relative to the phone companies' rates, at the same 16 megabyte speed or 25 megabyte speed, what I see is that the cablecos have a higher fixed and a higher variable.
2201 That is the dilemma that I have, that either, from a cost perspective, your infrastructure is that much higher -- and I don't want to get into your business, but, at the same time, you are getting the lion's share of the business and, for some reason, all of these costs seem to be much higher.
2202 Maybe it's based on forecast of demand and your demand is lower, so your costs are higher relative to the ILECs, but there is an incongruency here that I am trying to understand.
2203 MR. ENGELHART: It is because our 10 megabyte per second service actually delivers 10 megabytes per second, and theirs generally doesn't, depending on where you are relative to the Central Office.
2204 COMMISSIONER KATZ: Okay. Those are my questions.
2205 THE CHAIRPERSON: Thank you very much.
2206 Candace, I believe you have some questions.
2207 COMMISSIONER MOLNAR: I do have some questions, but I want to begin by clarifying a point that I see in paragraph 3, where you state that your usage model adopts the proposal outlined by Commissioner Molnar.
2208 Just for clarity, I said "an aggregated model, if appropriately designed."
2209 The jury is still out on whether or not this is properly designed, and I think that some of the key criteria is whether it provides appropriate economic signals to all providers in the system, and provides flexibility for independent ISPs to innovate.
2210 I want to look at some of the statements you made about your model based on that.
2211 I would like you to expand, if you would, on a couple of the points you made here. Point 1 says that it facilitates more effective and cost-efficient network planning.
2212 What is it exactly that your model does relative to, for example, the AVP or CNOC that it facilitates this efficient network planning?
2213 MR. BÉLAND: This comes right back, Commissioner Molnar, to the proper delineation and measurement of costs.
2214 COMMISSIONER MOLNAR: So you are talking about where the tariff would break between fixed and usage.
2215 MR. BÉLAND: That's right. If you break it at the right place, and the costs are of the right magnitude, and the markup is appropriate, we believe that facilitates effective and cost-efficient network planning.
2216 COMMISSIONER MOLNAR: Thank you.
2217 MR. ENGELHART: It is also the bundling of the usage with the access in our model, as with the Bell legacy model.
2218 As I was saying to the Chair, we know that if an ISP is buying 1,000 TPIA accesses from us next year, they are buying a minimum of 25,000 gigabytes. So that assists with the planning process.
2219 COMMISSIONER MOLNAR: Okay. Fair enough.
2220 The No. 2 point that you make here I find really quite confusing, to say that the tariff fosters more usage rather than less.
2221 How is it that a tariff that charges based on volume of usage could encourage or foster usage of the internet?
2222 MR. BÉLAND: We come back to, I guess, stepping back, looking at why we are here, and contrasting our bundled model with some of the other models, notably the zero usage, fixed access model.
2223 We are not here because Canadians were out in the street saying: I want internet access. It's my right, but I don't want to use it.
2224 The reason we are here is, people were concerned about what they believed to be disincentives for usage of the internet.
2225 So what we have put forward is a bundled model that includes the amount of usage that corresponds to the general network average, and then says, beyond that, here is a cost-based rate, and go for it.
2226 COMMISSIONER MOLNAR: But it still charges for accumulated usage through the month, so how could that possibly be an incentive that fosters usage?
2227 MR. ENGELHART: As Dennis said, the difference between our model and some of the others is, by bundling some usage in with access, it gives the ISPs an incentive to have an average usage that is, at least, equal to our average usage, because that is how much they are buying at a minimum every month. It doesn't incent them to provide very small packages to end users, although they can with the aggregated model. But the incentive is to provide larger packages.
2228 COMMISSIONER MOLNAR: Would it be true that the incentive would be to offer packages quite similar to the packages you offer yourself, since they are paying based upon what is, in essence, your packaging?
2229 MR. ENGELHART: No, because the usage --
2230 COMMISSIONER MOLNAR: I'm sorry, I am going to throw in my second question, because I may need some clarification here.
2231 If I understood your tariff, you actually were proposing these caps by speed tier. Is that true?
2232 MR. ENGELHART: Yes.
2233 COMMISSIONER MOLNAR: So, actually, by speed tier, by cap, it's linked directly to your retail.
2234 MR. ENGELHART: Yes, but we have found that the TPIA customers that we have today, on average, use more than our customers use for all speed tiers. So it's not like this is going to constrain them. They are generally providing more.
2235 Under the proposal that the Commission suspended, the ISPs were paying $1.70 per gigabyte. Now, as you see from Schedule 1, in the case of Rogers, it's 29 cents. Obviously that makes it much easier for them to provide big packages or unlimited packages.
2236 COMMISSIONER MOLNAR: I understand the point you are seeking to make.
2237 I just want, once again, a point of clarification. You have made the argument that your network is driven -- an augmentation of your network is driven by usage. But can I just be clear? It is still driven -- augmentation of your network is still driven by congestion and not usage.
2238 Is that true?
2239 It's not the volume, but it is the usage of a particular node that would drive augmentation.
2240 MR. ENGELHART: You are correct that Mr. O'Carroll looks at the percentile usage of a network at a peak point when he makes that augmentation decision, and that is true of all telecom network augmentation, I believe.
2241 So, yes, when a peak reaches a certain point, Mr. O'Carroll orders that augmentation to take place.
2242 The point we were making is that, when you look down at the smaller nodes, and when you look at internet versus other telecommunications services, like telephony, peak/non-peak is not really the story. The story is the inexorable 40 percent growth, year after year after year, that drives augmentation. That is the kind of growth for which total volume is a pretty good proxy.
2243 I take your point, it's not a perfect proxy. And it is a proxy, but we are saying that it is a pretty good proxy, and it is a way better proxy than looking at peak usage at the POI interface.
2244 COMMISSIONER MOLNAR: You know TELUS. I left yesterday remembering one of the statements they made, that they are all imperfect solutions. That, unfortunately, is the case when you are trying to put in place a system that addresses congestion within neighbourhoods and aggregate it.
2245 I think the challenge is, you know, they are all imperfect, in that they can't be scientific and surgical. What can we do that is reasonable, flexible, and provides proper incentives?
2246 There are some concerns here, where you are linking it in each speed tier to a cap of usage for your own customers. That seems somewhat concerning.
2247 And, then, driving additional charges based on cumulative volume over a month, versus what drives augmentation.
2248 They may all be imperfect, but some seem more concerning -- seem less perfect than others.
2249 MR. ENGELHART: It is not really a cap. The 29 cent per gigabyte rate that Rogers is charging, you are paying that for the 1st gigabyte as well as the 26th gigabyte, it's just that we are bundling the first 25 gigabytes of usage in the monthly tariff.
2250 It's not like you pay a different rate for Gigabyte 1 than you pay for Gigabyte 26, it is just that one is à la carte and one is bundled with access.
2251 The second thing is, getting back to the issue of what is a good proxy and what is not a good proxy, you have to realize that the growth in our network causes a sort of irregular augmentation. So, really, what difference does it make whether one gigabyte is the gigabyte that pushes you into augmentation or not?
2252 Say that we have segmented in your neighbourhood. We don't have to segment for another two years. Does that mean we shouldn't charge you for extra gigabytes and we should charge the person who lives a few blocks away, whose node is about due for augmentation? Of course not.
2253 The fact is that the constant growth in traffic causes us to segment and to augment. There is nothing bad or scary or difficult about that. It's our job, it's what we do, but those are costs that are driven by usage.
2254 So charging on a per gigabyte basis seems to us to be a pretty good proxy to measure what is going on.
2255 Indeed, when we brought in retail UBB, we found that there was less augmentation and that some users did use less than they had before. That, to me, indicates that it's working.
2256 COMMISSIONER MOLNAR: I have one final question, and it's a costing question.
2257 The tariffs you filed, if we understand appropriately, you have filed an access rate that includes both the fixed charges and the usage. Correct?
2258 MR. ENGELHART: Correct.
2259 COMMISSIONER MOLNAR: And that usage would have included both heavy users and light users in creating an average usage per customer?
2260 MR. ENGELHART: Correct.
2261 COMMISSIONER MOLNAR: What cost, if any, is there to additional usage?
2262 If you have already incorporated both heavy users and light users into the single access charge, why is there an additional charge, and what is that representing?
2263 MR. ENGELHART: It is a cost model based on a network that can deliver 25 gigabytes per end user. A network that delivers more than 25 gigabytes per end user costs more.
2264 We have costed out what it would cost to have a network that uses 25 gigabytes, so we are charging an extra amount, a cost-based amount, for the additional usage.
2265 If you will permit me, Commissioner, let me just -- I hope it's not a digression, but you mentioned heavy users and light users in your question.
2266 There is this sort of concern that heavy users are a bad thing and light users are a good thing. It's not always the case. Our network is really designed for people who use a lot in short bursts. That's great. Short bursts use a lot. It's wonderful.
2267 Some of the light users can actually impose more problems for our network management if they are using it 24/7. So it's like my analogy of the car parked in front of the coffee shop. They are blocking the whole lane and that affects the road.
2268 So it's really -- it's even difficult to talk about heavy and light because in our case a lot of heavy users who are very spiky or peaky are easy for us to accommodate.
2269 THE CHAIRPERSON: But, Mr. Engelhart, that means that one of the questions which we set out as principles that the average users or light users should not subsidize the heavy users, you more or less ignore that, if I understand you correctly.
2270 MR. ENGELHART: Well, no. You know, back to my point with Commissioner Molnar, we have to have a proxy. And so the proxy is you know, the average usage per user.
2271 But I do think that our proposal does accommodate the Commission's requirement because it does have that usage-based rate for the additional usage. So I think we have complied with your instructions.
2272 THE CHAIRPERSON: Okay. I see.
2274 COMMISSIONER PATRONE: Thank you, Mr. Chairman, and good morning.
2275 Just on that point, Mr. Engelhart, if it's easier to accommodate spiky users why the need to monetize peaks?
2276 MR. ENGELHART: Well, I guess you sort of have -- it gets back to the nature of statistical events in a telecommunications network. So the decision to augment happens when you reach -- you know x percent of the network is full. In our case and in the case of my colleagues, we do that long before there is any congestion because we want our customers to have a fast, consistent experience.
2277 But if a whole bunch of people are spiking their usage at sort of random times, that's fairly easy for telecommunications networks to accommodate. It's actually the 24/7 usage that builds you up towards that peak more quickly.
2278 COMMISSIONER PATRONE: And yet the whole concept of charging people more or having light users not having to subsidize the heavy ones, you know, that jumps out at me when I hear you talk about the fact that the spiky users are easier to deal with than the people who are using a consistently moderate rate throughout the day.
2279 MR. O'CARROLL: If I could try and answer this, Commissioner?
2280 I think they shouldn't necessarily be characterized as light users. When you start billing based on speed that is where the issue is because the issue is low speed users can have heavy usage. High speed users can have very low usage.
2281 So when we look at a node to relieve it, we are looking at the usage on that node. What contributes to that usage is a combination of the speed, but also the duration for which that speed is contained.
2282 So a typical node in our network with typical usage can handle well over a 100, 10 megabits per second customers. But if I have 10 users doing a steady continuous stream at 3 megabits they will exhaust that node and I have to relieve it. As the 10 megabits users use more and their spikes become broader, they contribute more to the exhaust.
2283 So from our point of view, usage is a very, very good proxy for what actually causes the exhaust of any node in our network.
2284 COMMISSIONER PATRONE: Does augmentation actually mean you incur more in the way of costs or is it pure gravy from your --
2285 MR. O'CARROLL: Oh, no, absolutely.
2286 COMMISSIONER PATRONE: It means more users. As you were suggesting earlier, the more users the better, right?
2287 MR. O'CARROLL: Yeah.
2288 COMMISSIONER PATRONE: I mean, you are built --
2289 MR. O'CARROLL: Augmentation absolutely results in significant cost, and depending on the specifics of the node, the costs can vary from thousands of dollars to hundreds of thousands of dollars.
2290 So, for example, if I'm in a node where I have, you know, 200 customers and I have multiple fibre strands available to that node then I may have to add line cards to the CMTF, as we talked about. You know they can cost tens of thousands of dollars.
2291 If, however, I need to extend my fibre closer to the home in order to split that node, I may have to spend hundreds of thousands of dollars in that specific case.
2292 So the individual costs per node vary, very much, on the conditions and the situations in the node but it's never free.
2293 COMMISSIONER PATRONE: Okay.
2294 MR. BÉLAND: If I might just add, Commissioner Patrone?
2295 COMMISSIONER PATRONE: Sure.
2296 MR. BÉLAND: I think it's also important to remember Vice-Chairman Katz's point from this morning, I believe. There are two things that might cause demand increase.
2297 One is new end-users which you are correct, that is new revenue. The other is existing end-users using the service more. And from there flows the importance of these usage-based tariff rates.
2298 COMMISSIONER PATRONE: Can I take you to page 8 of 18 on your oral, the second bullet where you talk about -- when you are comparing your model to the CNOC model and you are referring to "gaming the system"?
2299 I believe you are saying -- suggesting that the CNOC model allows for gaming, right, that they are going to somehow get away with something or by -- as you say here, they could for example sell a service to business customers that sends a steady stream of traffic through the cable network from 9 to 5 in the afternoon.
2300 And the presence of this traffic would cause cable carriers to augment facilities, as you were talking about, Mr. O'Carroll, and the wholesalers would pay no more incremental charges.
2301 As we are dealing with wholesale residential, do you see any potential for gaming on the residential side under the CNOC model?
2302 MR. ENGELHART: Unfortunately, the TPI tariff doesn't distinguish between business and residential and the Commission has required us to make TPI available for all customers, business or residential.
2303 COMMISSIONER PATRONE: But as you were saying, the vast majority of your customers are residential.
2304 MR. ENGELHART: Yes. But if you have a business at your home or if it's a business location and there is cable going there, you are entitled to TPIA.
2305 COMMISSIONER PATRONE: So you are saying that the gaming argument, even though the example cited applies to business customers --
2306 MR. ENGELHART: It is something they could do in the real world. They could take our TPI tariff. They could provide the service to business customers.
2307 I said that our footprint didn't cover all business customers. It does cover some and it's an example of the kind of thing you can do.
2308 So, whereas our proposal, our proxy says each and every gigabyte that goes through that point of interconnection is charged for, their proposal says you look at all the five minute intervals in a month and you select one of those five minute intervals and you pay on that five minute interval and you pay on nothing for the rest of the month.
2309 It is possible for them to design services that would cause traffic to flow in periods other than that five minutes and, therefore, it would be free for them but it would still cause us augmentation costs.
2310 COMMISSIONER PATRONE: Right.
2311 MR. ENGELHART: Because our point of augmentation if not at the POI interface. It's down in the node.
2312 COMMISSIONER PATRONE: I get what you are saying, Mr. Engelhart, that it would cause you to augment and that augmentation could not be passed on to the wholesale customer. Is that correct?
2313 MR. ENGELHART: Correct.
2314 COMMISSIONER PATRONE: A couple of short snappers.
2315 I heard you say that competitors now account for 2 percent of your network and growing rapidly. Is that correct, Mr. Engelhart?
2316 MR. ENGELHART: Yes, sir.
2317 COMMISSIONER PATRONE: Are those new wholesale customers switching over from ILEC networks or are they new to the market?
2318 MR. ENGELHART: I think you probably get all three; people moving from Rogers to Rogers' TPIA, people moving from Bell to Rogers' TPIA and new customers.
2319 COMMISSIONER PATRONE: The final question. Would the CNOC model have any more appeal if the measurement of those peaks took place at multiple places on the cable network?
2320 MR. ENGELHART: You would have to take it down to all 20,000 line cards that we have in the CMTS which would be impossible. So I don't think their proposal will work.
2321 COMMISSIONER PATRONE: Thank you very much for your answers.
2322 LE PRÉSIDENT : Michel, je crois que tu as des questions.
2323 CONSEILLER MORIN : Oui, merci, Monsieur le Président.
2324 Madame la Secrétaire, je pense que vous avez un petit tableau à distribuer à nos invités.
2325 Hier, comme vous le savez, j'ai essayé de donner tous, mais alors tous les chiffres jusqu'à l'utilisateur, selon que le petit fournisseur d'Internet avait 100 ou 200 ou 1 000 clients, et je l'ai fait pour Bell en comparaison avec un modèle dont on a à peine parlé aujourd'hui et hier, celui de la compagnie manitobaine MTS.
2326 MTS propose un modèle où, finalement, on confie au petit fournisseur le soin de décider de la grosseur du tuyau. Autrement dit, on les responsabilise.
2327 Ce matin, on a évoqué le fait que vous êtes obligés de fournir l'accès avec des prix de gros aux fournisseurs. Vous savez, comme moi, que cette situation n'existe pas aux États-Unis. Les gens ne sont pas obligés, comme vous, de fournir des prix de gros. C'est une invention de notre système. Certains parlent de compétition artificielle. Quoiqu'il en soit, la moyenne des 10 dernières années, les petits fournisseurs d'Internet n'ont pas augmenté considérablement leur marché.
2328 Alors, donc, sur ces remarques, moi, j'aimerais éclaircir, parce qu'on a lancé toute sorte de prix depuis hier, j'aimerais éclaircir, d'ici lundi prochain, ces tarifs.
2329 Je les vois par rapport à MTS. Ils sont, grosso modo, 25 pour cent plus élevés ou vice versa, sauf peut-être Vidéotron qui est plus dispendieux. Et c'est les mêmes hypothèses que j'ai prises avec Bell, et c'est probablement les mêmes hypothèses que je prendrai avec d'autres fournisseurs.
2330 Donc, ma question est toute simple, et je ne vous demande pas d'y répondre ce matin : Est-ce que ce tableau-là est vrai?
2331 Et comme vous l'aurez sur le site Internet, comme celui de Bell hier, bien, je pense que vous pourriez corriger des prix si ces prix-là ne vous apparaissent pas valables, incorrects, ou ajouter.
2332 Pour les autres questions, pour l'instant, j'en reste à cette question-là que je trouve fondamentale. Il y a un fournisseur de service aux petits fournisseurs de service Internet qui a des prix qui m'apparaissent beaucoup moins élevés que les vôtres et que ceux de Bell hier.
2333 Alors, ce serait... Moi, j'aimerais avoir une réponse écrite. J'essaie d'être le plus transparent possible en vous donnant toutes les hypothèses et tout ce qui est entré en considération : le taux d'accès, les frais d'utilisation, le coût mensuel de l'interface. Et ça ne comprend pas, évidemment, les frais de service, qui sont exclus, et les charges.
2334 M. BÉLAND : Oui, Monsieur le Conseiller Morin, je peux prendre l'engagement au nom des trois compagnies qu'on va vous donner une réponse écrite par rapport à cette analyse pour lundi prochain.
2335 M. BÉLAND : Il faut dire, de façon très préliminaire, que ça va prendre l'analyse de notre part parce que le tarif de MTS est drôlement fait, de notre perspective, drôlement pas parce que c'est mauvais, mais parce que la structure est complètement différente de la nôtre. Ils semblent mettre beaucoup d'emphase sur le tuyau qui est présent au point d'interconnexion.
2336 Au niveau purement conceptuel, ce qui est important pour nous, ce n'est pas le tuyau qui part du point d'interconnexion, c'est le réseau qui est avant le point d'interconnexion, qui est notre réseau, qui est notre responsabilité, qu'on a essayé de "coster."
2337 CONSEILLER MORIN : Je comprends tout à fait votre argument.
2338 M. BÉLAND : Comme commentaire tout à fait préliminaire.
2339 CONSEILLER MORIN : Absolument. Je comprends tout à fait votre argument.
2340 Mais au fond, ce que disent les gens de MTS : Laissons aux gens, aux petits fournisseurs le choix de décider. Si les risques dont vous parliez avec le vice-président Katz ce matin, bien, les risques sont portés par les petits fournisseurs eux-mêmes, ce qui ne semble pas être beaucoup dans la discussion actuellement, et c'est dans ce sens-là que je présente ces chiffres, parce que je veux d'abord avoir votre avis là-dessus et toutes les considérations que vous pourriez me faire.
2341 M. BÉLAND : C'est parfait. On prend l'engagement de donner une réponse écrite.
2342 CONSEILLER MORIN : J'apprécie.
2343 C'est tout, Monsieur le Président.
2344 THE CHAIRPERSON: Let's hear from MTS first too before you do your response. They are on Thursday.
2345 M. BÉLAND : On va les écouter de près.
2346 THE CHAIRPERSON: Tim?
2347 COMMISSIONER DENTON: Mr. Engelhart, I believe I quote you correctly -- and correct me if I'm wrong -- you said at some point talking to Mr. Katz, "Even the higher rate doesn't compensate us for the risk we incur."
2348 Can you unpack that for me because it's sort of like -- is there no rate at which your risk is paid for? Just tell me more about that, please.
2349 MR. ENGELHART: Right. Well, the risk is incurred when you make the investment. You know, you make investments and sometimes it works out wonderfully and sometimes not so wonderfully. The very nature of a wholesale tariff is the wholesale customer comes along after the network has been built and then, as a free option, they can take the service or not take the service.
2350 So it's not like -- back to my hotel analogy if they said, "If you build this hotel I will commit to take 10 rooms a night for sure", well, that would be -- that would be diminishing my risk. But because they have a free option after that hotel is built and because they are paying a lower rate than retail by the very nature of wholesale, they can't really compensate you for the risk you have incurred.
2351 COMMISSIONER DENTON: Okay. So in your view there is no conceivable rate at which you could be compensated because, according to the logic you have just presented, they do not share the going in costs of establishing the hotel, the infrastructure, the investment?
2352 MR. ENGELHART: Right. Mandated wholesale increases the riskiness of the investments for facilities-based carriers no matter what the rate is, unless the rate was equal to retail and that obviously wouldn't work.
2353 So, yeah, you are always going to face increased risk as a result of mandated wholesale.
2354 COMMISSIONER DENTON: Okay, your meaning is clear. Thank you very much.
2355 THE CHAIRPERSON: Okay. Thank you very much for your presentation. Look forward to seeing you in the rebuttal.
2356 I think those are all our questions. We will adjourn now and we will resume at 1:30. Thank you.
--- Recessed at 1218
--- Resumed at 1334
2357 LE SECRÉTAIRE: Prêt à commencer.
2358 Order, please.
2360 LE PRÉSIDENT: Commençons.
2361 THE SECRETARY: So we will now proceed with the presentation of CNOC, the Canadian Network Operators Consortium Inc. And Mr. Chris Tacit represents CNOC.
2362 Mr. Tacit, please introduce your panel for the record, after which you will have 20 minutes for your presentation.
2363 MR. TACIT: Thank you.
2364 Mr. Chair and Commissioners, we are very pleased to appear before you today in this very important proceeding. My name is Chris Tacit and I represent Canadian Network Operators Consortium or CNOC.
2365 Seated to my right is Mr. Bill Sandiford, Chair of the Board and President of CNOC, as well as President of Telnet Communications. Seated to his right is Keith Stevens, Chair of CNOC's Regulatory Committee and Chair of Execulink Telecom. Paul Andersen, President of eGate Networks is seated to Mr. Stevens' right. Peter Rocca, Director of Start Communications is seated to my left, and Marc Gaudrault of TekSavvy Solutions, the CEO, is seated to his left.
2366 In this presentation we will respond to the questions posed by the Commission in its July 4, 2011 letter.
2367 I understand there is a bit of a procedural issue. We attempted to be helpful to the Commission by providing a short summary at the end of some issues that are not being covered, because we don't have time, some costing and tariffing issues. And we would ask that Chair's indulgence to leave those attached to the opening statement, although we clearly won't have time to read them.
2368 THE CHAIRPERSON: Are you talking about Annex B?
2369 MR. TACIT: Annexes A and B, yes.
2370 THE CHAIRPERSON: Okay. Since we are going to have a reply phase next week and people have a chance to look at it, I don't think there is going to be any procedural unfairness or bias. Make sure it gets tabled by Madame Secrétaire and we can put it on the web. Okay?
2371 MR. TACIT: Thank you.
2372 Bill is now going t set the stage for the rest of our presentation.
2373 MR. SANDIFORD: Thank you, Chris.
2374 The Commission has bee committed for many years to increasing competition in the provision of telecommunication services so that consumers could reap the corresponding rewards.
2375 Competition means that consumers end up with better, faster, and less expensive internet access and other broadband-enabled services. Competition increases consumer choice, innovation and creates incentives for investment in the network infrastructure.
2376 This proceeding is taking place at a time when it is clear that demand for internet services is increasing rapidly as streaming video becomes more popular. The heavy user of today is the average user of tomorrow. Consumers need choices and the role of ISPs should be to meet increasing demand and provide new service offerings that consumers want. ISPs should not stifle demand.
2377 The cost-plus approach that the Commission has taken to regulation of wholesale services provided by incumbents ensures that they are properly compensated for the use of their networks.
2378 Despite this, the incumbents want to charge excessively for carrying traffic across their network. Once a network is built and paid for, there is almost no cost to move internet and other broadband traffic across it.
2379 However, the incumbents want to charge significant amounts for all usage going through their networks at all times, even when moving that traffic costs them very little.
2380 If this practice is allowed incumbents will reap a windfall. At the same time, the price of internet access will increase significantly for all consumers and demand for internet services will be constrained. Competition will be reduced at the consumer level.
2381 ISPs are willing to pay for those things that cost the incumbents. Just as ISPs pay a fair amount to use the incumbents' networks, they are also prepared to pay for network usage when that usage actually imposes costs on the incumbents.
2382 It turns out that when it comes to usage the only type of usage that can lead to significant costs for the incumbent networks is peak usage. It is peak usage that can cause congestion and require the incumbents to increase their network capacity.
2383 In this proceeding we are proposing an approach to billing for wholesale high-speed services that compensates the incumbents for the peak usage generated by ISPs sharing the incumbents' networks.
2384 This approach is fair to the incumbents, while not interfering with competition at the consumer level and is already widely used in the telecommunication industry for measuring usage and engineering networks. It also disciplines the market power of incumbents and allows ISPs the ability to earn a reasonable return so that they can continue to operate and invest.
2385 Our approach focuses on identifying peak usage when it actually occurs and paying for that usage of network capacity, but it does not require payment for non-peak traffic that is virtually free to carry.
2386 There is a lot at stake in this proceeding. The proceeding is not simply about rate levels, but also about a fundamental restructuring of rates for the wholesale high-speed services of the incumbents without which ISPs cannot provide internet access and other broadband-enabled service to consumers.
2387 There are many ways that the rate restructuring can be carried out, but most of them will lead to a competitive dead end. If Canada reaches that dead end, then the market for the provision of internet and other services will be, at best, an entrenched duopoly between telephone and cable companies. In that case, consumers in the Canadian economy will pay the price.
2388 The Commission must be vigilant to ensure that speed-matching, point of interconnection aggregation, and the introduction of billing for usage do not end up making ILEC and cable carrier wholesale services economically unusable.
2389 In particular, the Commission should ensure that rates are cost-based and mark-ups are not based on value of service principles. Such principles have no place in a regulated wholesale services environment that should be focused on compensating incumbents fairly for network resources employed by incumbents while disciplining the incumbents' market power.
2390 We urge the Commission to be mindful of the fact that ISPs have many costs to bear in providing services to their customers. Burdening ISPs with additional unnecessary costs will not promote competition.
2391 Since the Commission really wants competition in the provision of broadband services to flourish, it will need to adopt a highly principled approach to the rate structure that it approves for incumbent wholesale services.
2392 In a competitive environment incumbents will lose some revenues, otherwise competition will not work. As the Commission's goal is to foster competition, then its objective cannot be to protect the incumbents and keep them revenue neutral or to blunt the economic impact of rate restructuring on them in an artificial manner when that is not warranted by the sound economic principles.
2393 Competition will not work until the incumbents actually lose significant market share. And I want to repeat that point, because I think it is very important. Competition is not working until the incumbents actually lose significant market share. Six per cent that the competitors have today is not significant market share.
2394 We urge the Commission only to allow the incumbents to recover the costs associated with providing high-speed services plus a reasonable mark-up, without otherwise tying the hands of the competitors.
2395 We will now turn to a discussion of the six questions posed by the Commission in its letter dated July 4th. Peter will now address the first two questions.
2396 MR. ROCCA: Thank you, Bill.
2397 First, I am going to address the issue of which of the billing models proposed by the interveners should be adopted.
2398 We believe that the model proposed by CNOC, which separates wholesale services into separate access, usage, and interface rate components is the appropriate one. Our model is focused on compensating the incumbents for costs they actually incur, but not providing them a windfall when they do not incur costs.
2399 Our model creates proper economic incentives. It prices network usage appropriately, leading to efficient use of incumbent network capacity and creating efficient incentives for investment in new capacity by incumbents.
2400 Models proposed by the other parties do not create these incentives, and instead create windfalls for the incumbents and increase costs for independent ISPs without any economic justification. Such unjustified anticompetitive cross-subsidies should be rejected. Competitors must not be handicapped and consumers should not be disadvantaged in this manner.
2401 Competitors should be encouraged to innovate and differentiate their services for the public good. Our model also has the following additional advantages. It applies the Commission's Phase II costing principles rigorously. It applies rational and economically justifiable mark-ups on Phase II costs. It eliminates economic distortions by more closely matching costs and rates for various wholesale service components, thereby reducing barriers to entry.
2402 I now want to turn to the Commission's second question regarding whether if a usage component is adopted, should it be measured via peak traffic or via volume.
2403 The incumbents argue that explicitly charging ISPs for data usage by volume for aggregated network use is the fairest way of ensuring that those who cause network congestion pay for it. This is not so.
2404 Unlike utilities such as gas or electricity, internet traffic is not consumed. Once a data network is in place, the cost of carrying data on it is negligible until more peak capacity has to be added.
2405 Data networks are like pipes, when the data is being flowed through a pipe it does not constrain its peak capacity (i.e. it does not fill up the pipe), carriage of that data does not impose any significant costs on the network. The cost is the pipe, not the content that flows through it.
2406 What drives the cost of a carrier that wants to avoid congestion is the peak demand on its network measured as the speed of traffic in megabits per second. In other words, the same amount of data transmitted through the pipe at a faster rate uses more of the capacity of that pipe than if data is transmitted at a slower rate.
2407 A charge that is based on volume is not reflective of costs. Such a charge will only lead to a windfall for the entity that levies it, which amounts to an anticompetitive cross-subsidy.
2408 Pricing usage on peak demand of a pipe based on speed is the correct approach, since it appropriate signals to a carrier when it needs to add more pipes or increase capacity of the network. This method also encourages those who use the pipe to mange their usage more efficiently.
2409 The beauty of pricing usage according to peak capacity is that there is a well-developed and simple method for doing this, that method is the 95th percentile method that is described in the submissions of CNOC in this proceeding. This method is how carriers, including the incumbents, sell transit and transport services as well as design and manage their own networks.
2410 The 95th percentile method does not artificially constrain network usage. The 95th percentile method does not create inappropriate cross-subsidies. The 95th percentile method does not interfere with competition among ISPs. All data network operators measure traffic according to the 95th percentile.
2411 By charging for usage separately in this manner, monthly flat-rate access and interface prices should also fall, as those rates will not be based on costs that are usage or speed-sensitive.
2412 This approach should facilitate a rate restructuring that will be even more competitively neutral by enabling competitors to obtain wire speed rather than matching speed access to incumbent networks. Such an approach will provide independent ISPs maximum flexibility to compete in the market and increase consumer choice.
2413 The 95th percentile method is easy to apply for the industry as a whole. For example, the method does not require incumbents to segregate volume measurements by speed tier. It also does not require measuring usage on hundreds of thousands of individual users, ports. One measurement at each ISP interface is sufficient and easy to audit.
2414 We do not believe that there is any significant implementation or ongoing challenge associated with the use of the 95th percentile method given its widespread use and acceptance in the industry.
2415 Marc is now going to address the third and fourth questions in the Commission's letter.
2416 MR. GAUDRAULT: Thank you, Peter.
2417 With regards to the third question, we have already explained that our model creates proper economic incentives. It leads to more efficient use in incumbent network capacity and more efficient incentives for investment in new capacity by incumbents.
2418 This issues is covered well and in much greater detail in the reply of CNOC dated April 29, 2011 in this proceeding.
2419 The fourth question asked whether there should be symmetry with respect to the mark-up on costs with all network providers. We are of the view that such symmetry is generally desirable.
2420 However, in TRP-2010-632 the Commission recognized that significant upfront investment is required to construct the facilities, ILECs, used to provision new high-speed wholesale service option.
2421 Therefore, the Commission stated that ILEC rates for those new high-speed wholesale service options will include, in addition to the mark-up on costs that would otherwise be used, a supplementary mark-up of 10 per cent.
2422 The cable carriers did not make a plea for similar relief in the proceeding leading to TRP-2010-632 based on any reluctance to invest if they are required to share facilities with competitors.
2423 Therefore, cable carrier mark-ups should not include this 10 per cent premium.
2424 Paul will now address the last two questions.
2425 MR. ANDERSEN: Thank you, Marc.
2426 When considering whether the same mark-up should apply to monthly access rates and usage rates, the critical issue is the principles that should apply to the determination of mark-ups.
2427 The Commission applies a 15 per cent mark-up for near-essential services over Phase II costs. This mark-up enables incumbents to recover fixed and common costs. The CNOC model splits out access and interface port from transport costs.
2428 Since the access network and interface ports are not duplicable, the mark-up on these elements should be treated as near essential and be no more than 15 per cent.
2429 In the case of transport, mark-ups should be based on those that would be expected to be present in a competitive marketplace. Existing transit and transport services that are available in the marketplace could be used to derive such mark-ups.
2430 Any notion of value-based pricing provided by incumbents is arbitrary and should be rejected. Higher mark-ups will simply create anticompetitive subsidies from competitors to incumbents that will interfere with the operation of competition in retail markets.
2431 The principles we propose should apply regardless of whether access rates and usage rates are billed jointly or separately. Although, some additional clarification is required on this point.
2432 In the case of DSL, it turns out that access and interface costs are non-usage sensitive, whereas transport costs are user-sensitive. Therefore, the access and interface flat rates should be based on a 15 per cent mark-up, whereas peak rates should be based on the mark-up applicable to the transport.
2433 The model also works for cable carriers. However, in that case some of the access and interface costs have been claimed to be usage-sensitive by them. We believe that the flat rates for access and interface should only reflect to the non-traffic sensitive costs, and the mark-up on those costs should be 15 per cent.
2434 The peak charge should recover all traffic-sensitive costs incurred. Therefore, a blended mark-up should be included in the peak rates. The mark-up on the traffic-sensitive access and interface costs should be 15 per cent and the mark-up on transport costs should be the competitive mark-up applicable to transport services in the marketplace.
2435 One important factor that needs to be addressed is to reflect the increasing efficiencies. All rates for aggregated wholesale services need to be reduced over time to take into account productivity improvements or else the prices of wholesale services will be higher than appropriate.
2436 If that occurs, the incumbents will end up with an unfair pricing advantage and competition at the retail level will be unduly lessened to the detriment of consumers.
2437 With regards to the last question, we are of the view that there is no reason, conceptually, that the principles, provided by CNOC cannot be applied to legacy services. In fact, we support the applications of these principals to all wholesale high-speed services.
2438 However, if mark-ups are excessive or some other method is applied for usage billing, the impact on legacy services would be devastating. CNOC opposes such an outcome.
2439 I now invite Keith, who will provide our concluding remarks with respect to the issues raised in the Commission letter.
2440 MR. STEVENS: Thank you, Paul.
2441 The Commission's July 4th letter dealt with the rate restructuring aspects of this proceeding, rather than the setting of rates for the new services being introduced by ILECs and the cable carriers. This is rightly so.
2442 If the Commission does not get the rate restructuring right nothing else will matter, as there will be no competitive markets.
2443 It is clear that under the current rate structure competitors are not making sufficient competitive inroads. Something needs to be changed. We have placed a principled and economically sound option before you in this proceeding that will enable competitors not to just survive, but also to innovate to a greater extent than has been possible up until now.
2444 The rate restructuring we have proposed is logical. It allows for the recovery of traffic-sensitive costs through traffic-sensitive charges and non-traffic sensitive costs through non-traffic-sensitive charges. Such a rate structuring facilitates the operation of competitive markets.
2445 Our approach also provides for the measurement and billing of usage according to industry standards, peak capacity, 95th percentile method.
2446 Finally, we are prepared to compensate the incumbents for costs that they actually incur, but we do not agree to cross-subsidize them when there are no costs to be recovered.
2447 For all these reasons, we ask you to approve the principles and approach advanced by CNOC in this proceeding.
2448 Chris will now wrap-up.
2449 MR. TACIT: Thank you, Keith.
2450 In addition to the matters discussed in this opening statement, there are a number of important matters relating to costing principles and issues, rate and non-rate tariffing issues and the treatment of interim rates on a final basis.
2451 CNOC will be dealing with these matters in its final written submission on the 29th of July.
2452 We would like to thank the Commission and Staff for their courtesies extended during this proceeding. We are now available for the Commission's questions.
2453 THE CHAIRPERSON: Thank you very much for your submission.
2454 Now, help me here. Did you not come and speak to what we call the FCM, the Full Commission Meeting on January 26th, 2011, just before we decided to review and -- the UBB decision?
2455 MR. TACIT: Correct.
2456 THE CHAIRPERSON: And at that point in time I thought your position was that UBB as such was not so bad, what was bad was at the point that we had implemented it, we should have followed Commissioner Molnar's advice and applied it as the aggregate level rather than at the final level.
2457 Was that not the subsequent --
2458 MR. TACIT: I am not sure we talked about a specific approach. I think --
2459 THE CHAIRPERSON: But you showed me a chart and said you went at the wrong level, you should have gone earlier. Was that not the substance of your presentation at that point in time?
2460 MR. TACIT: I think we did talk about that as being better than then end user-based approach. But at that point, frankly, we didn't have much hope. We were frankly swimming upstream against a very big tide and we --
2461 THE CHAIRPERSON: Your view has evolved since then?
2462 MR. TACIT: Well, our view has evolved, and partly because we have really had time to think about this, you know, as a group rather than reacting very quickly to a decision that, at the time, was actually very devastating for the members.
2463 Clearly, a lot of time has gone into and though by many members at CNOC into putting this proposal forward and trying -- we really tried to follow sound economic principles, and so that is the approach that we -- but I think --
2464 THE CHAIRPERSON: Then help me out here, because I don't understand. You are saying the 95th percentile is the right way to go and you are making a very logical convincing argument. But I have heard from Bell, I have heard from the cable companies and they say two things; (a) we don't measure it that way, we don't actually -- they have the chart, you saw that Bell says it's (indiscernible). Right here we only do -- we would have to do it at various points. So first of all, there is a cost of implementing this if you wanted to.
2465 And secondly, they say that there would be a time delay in doing it. And so while it is feasible, it is impractical. In fact, it is not a realistic solution.
2466 You say exactly the opposite. You say, yes, it can be done and do it tomorrow. How am I supposed to approach this? I hear two people who are basically telling me the opposite thing.
2467 MR. TACIT: Well, let me suggest what we know about the industry and then you can draw your own conclusions from that. And let me suggest a possible approach that will test this out as well.
2468 So in terms of what is known in the industry, as we know all the major providers, including the incumbents, actually use the 95th percentile method to bill when they provide transit and transport services as such, as opposed to aggregated DSL services. So to suggest that they don't know how to measure and bill this is just not --
2469 THE CHAIRPERSON: No, they didn't say that. They don't say that they don't know how to do it, they actually don't do it, they don't have the measurements in place.
2470 MR. TACIT: The tools are very simple and my colleagues here, you know, can also speak to that. There are tools that are available for download from the internet that you can use, if you have a router, to measure traffic this way.
2471 With regards to where you should measure, we think that is a red herring and we think so for the following reason. If you are paying for the capacity that you are using -- in other words, if you get the price correct for the 95th percentile measurement and if you apply the correct rate to that, we will be properly compensating the carrier for the peak demand that we cause.
2472 THE CHAIRPERSON: M'hmm.
2473 MR. TACIT: What they are saying is you are not helping us manage our network and figure out which link inside our network, you know, we have to upgrade at a specific point in time.
2474 But that is not our job, that is their job. They have to do that today. When they have congestion today they have to figure out today which of those links have to be upgraded presumably hopefully before they hit that point.
2475 It sounds like some ILECs are doing a better job of doing it beforehand than others. But the point is, that is their network management function. They are trying to delegate that to us and it muddies the water.
2476 THE CHAIRPERSON: So you think basically the measurement is feasible, it is doable, and it is not a major cost to implement it?
2477 MR. TACIT: That is right. And Bell --
2478 THE CHAIRPERSON: What about the second part, which they say that also means we have to totally regig our billing system?
2479 MR. TACIT: Well, of course there is going to be some adjustments. But they have had to make adjustments to bill us for volume as well.
2480 And there are only -- you know, we are talking about a few hundred ports where they have to bill here as opposed to, you know, tracking the usage from hundreds and thousands of end users of all the wholesale customers and blending that together.
2481 And then you get into all of the problems with whether the measurements are correct, how we as their customers audit that.
2482 Even today we have all sorts of issues where we have billing disputes because we're not sure how they audit, whether they've actually captured when we have a new installation correctly or when somebody has stopped their service correctly.
2483 And it's a lot more straightforward to say we're going to synchronize these measurements, you audit it if you want and we're going to audit it, starting at this time period we'll follow the peaks and we'll see. Everybody can do that with software that's available for free from the Internet.
2484 THE CHAIRPERSON: On Monday I specifically pushed Bell on this point and they said: We could do it, we have never looked at doing it and we don't know how long it would take us to implement it.
2485 You're basically telling me they could do this within what period of time?
2486 MR. TACIT: Well, I am not going to speak for how quickly they could do it, but I'm sure if you told them, we're going to strip usage out of your flat rates and approve those, and as soon as you start measuring usage we're going to approve your usage rates, I bet you they will do it pretty quickly.
2487 THE CHAIRPERSON: Sorry, you wanted to say something?
2488 MR. SANDIFORD: Yes. The notion that this is a difficult concept to implement or do is very foreign to us. I mean we have some of the smallest and largest ISPs in the country in our midst and I don't know of a single one of our members that hasn't been able to implement 95th percentile billing.
2489 Further, we know for a fact that in the case of the incumbents they do bill 95th percentile today. They send bills to our customers for our Internet transit services on a 95th percentile basis.
2490 It surely couldn't be difficult to use those same systems that they already have in place to look at our point of interconnection port as opposed to our Internet transit port and bill for it appropriately.
2491 THE CHAIRPERSON: What about the argument that you're opening the door to gaming the system because they can't distinguish between residential and business traffic?
2492 MR. TACIT: Well, I respond here with some significant degree of trepidation, but the fact is that CNOC would actually like to see this method applied to all wholesale traffic, and from our perspective it's a little awkward because we have two parallel proceedings.
2493 If the Commission were not to decide to go that route ultimately, there are technical solutions that could be applied. They're a little less elegant than the regulatory solution but the fact of the matter is that we would actually welcome application of this method because we believe it's economically correct.
2494 THE CHAIRPERSON: In your written submission regarding the business wholesale, you will put the same idea forward --
2495 MR. TACIT: Exactly.
2496 THE CHAIRPERSON: -- hoping we will implement both, so therefore a distinction between the two is irrelevant?
2497 MR. TACIT: Yes. And in fact before the proceeding was split out, I think we said that in our initial submissions back in March or April.
2498 THE CHAIRPERSON: Okay. Thank you.
2499 Candice, I believe you have some questions.
2500 COMMISSIONER MOLNAR: I do.
2501 I want to begin just by ensuring I'm understanding all that you have proposed with your own model.
2502 First of all, just to confirm, you had proposed that each ISP have a separate peak and not be the peak where they would be applied -- usage would be based on each individual ISPs peak traffic?
2503 MR. SANDIFORD: That is correct.
2504 MR. TACIT: Yes.
2505 COMMISSIONER MOLNAR: So clearly you heard, and I heard you speak with the Chair a minute ago, about the argument put forward by the ILECs that the individual peaks of any one ISP may not correlate to the actual peaks or where congestion is occurring in the network, and when you move to a per-ISP basis that just compounds that problem. Have you thought about that?
2506 MR. TACIT: Yes, we have.
2507 First of all, it's important to note that when you're talking about -- if you get the rate right, we're actually going to be paying for the full use of that pipe even though it only peaks a small amount of the time.
2508 In fact, that's how transit and transport providers make their money today, is they have the same issue that Bell and TELUS does, they have many customers that they have to interconnect with at various interface points and they charge each one based on peak, and it's through the statistical advantages that they actually get to over-recover in many respects for that capacity. So I think that's, you know, the first important point to mention here.
2509 I'm sorry, can you just repeat the second half of your question there?
2510 COMMISSIONER MOLNAR: I can't --
2511 MR. SANDIFORD: I think it's important, further to Chris' comment, to say what he just said to the Chair a moment ago, is that we think this notion of where to measure is a bit of a red herring.
2512 Our proposal that CNOC has put forward is a proposal that deals with what this proceeding has set out to deal with, which is the effective management of the billing of the wholesale services.
2513 We're not here to determine how the incumbents manage their 20,000 links downstream. We're here to determine how they bill us for it, and the only appropriate place to bill us for it as at the aggregate point. And they will still do all their network management downstream.
2514 MR. ROCCA: I would like to just follow up to your question about the individual ISPs having different peaks.
2515 It's actually to the incumbents' advantage. The worst-case scenario would be if every ISP peaked at the exact same time. That would drive the maximum amount of capacity that the incumbents would have to bill.
2516 By having ISPs with different peaks, they're still billing the aggregate peak of all of those ISPs individually, with actually not having to deliver the aggregate sum of all those peaks in their network.
2517 MR. SANDIFORD: I can undertake to put together a chart that will illustrate that for you for next week.
2518 MR. TACIT: We actually have an interrogatory response that shows that very thing actually.
2519 COMMISSIONER MOLNAR: Yes. Thank you. I would appreciate the reference.
2520 MR. TACIT: We will look it up.
2521 COMMISSIONER MOLNAR: Just before I leave, the cablecos were on before you and spoke about their constraints on the upstream traffic.
2522 So how does your method address constraints on upstream traffic?
2523 MR. TACIT: We recognize that cable has a specific issue and we would be fine with paying on both downstream and upstream separately.
2524 COMMISSIONER MOLNAR: So you would suggest you measure a peak based on upstream as well?
2525 MR. TACIT: Yes.
2526 COMMISSIONER MOLNAR: And is that on the record?
2527 MR. TACIT: Not so far. This has come up in response to what we heard today.
2528 COMMISSIONER MOLNAR: Okay. Fair.
2529 You had a discussion about the potential difficulty and timelines of the incumbents implementing a billing system based on your 95th percentile, and while I understand you question how difficult that might be or how long it might take, assuming that there is both time and cost associated with putting in that methodology that doesn't exist with a volume-based approach, are you prepared to incur those costs? I mean they're costs of the service.
2530 MR. TACIT: Well, again, I mean, Phase II -- we don't have to create special costing and pricing rules for this case. Phase II provides the Commission with all the tools.
2531 COMMISSIONER MOLNAR: Right. And it would be an incremental cost to the service?
2532 MR. TACIT: Right. And the fact is that it would be shared with their service as well because presumably they need to manage their network capacity as well.
2533 Whether they choose to build that way to their residential retail consumers is not relevant here. What's relevant is what the wholesale service is like and they should be delivering it to us the same way they deliver the wholesale piece to themselves.
2534 COMMISSIONER MOLNAR: Right. But we're only talking about a billing mechanism here. As it was pointed out, we're not talking about how they're managing their network, we're talking about a billing system designed for independent wholesale customers.
2535 And so to the extent they need to put in place a new billing system with new tools, those costs, you accept they would be costs of the service passed on to --
2536 MR. TACIT: Yes, but we wouldn't want to see their effort replicated. So in other words, to the extent that they already can do this and they're moving code from one module to another, we wouldn't want them taking the approach that they've rewritten this from scratch. So there would have to be some scrutiny as to how they come up with those costs.
2537 MR. SANDIFORD: We believe that this is such a simple exercise that the costs would be trivial nonetheless.
2538 COMMISSIONER MOLNAR: Okay.
2539 Can I --
2540 MR. TACIT: Peter has something to add.
2541 MR. ROCCA: Actually, as a customer of another ILEC service, a transit service from Bell, I already receive a 95th percentile bill every month from them.
2542 COMMISSIONER MOLNAR: Can I confirm as well you mentioned that you have a number of smaller ISPs part of your organization as well as larger because we did hear there were some issues with some of the small ISPs who, I think, are not riding yet on --
2543 MR. SANDIFORD: It was interesting when the comment came up yesterday, the one that you're referring to. My Messenger went off on my computer, with one of our members sending a message saying: Who do they think that I am? Do they think I have tin cans and strings for routers? Of course I can do 95th percentile.
2544 And that was one of the smallest ISPs that we have, perhaps one of the smallest ISPs in the country.
2545 COMMISSIONER MOLNAR: So you are confident that from the independent ISP world this is workable?
2546 MR. SANDIFORD: Absolutely.
2547 MR. TACIT: Yes.
2548 COMMISSIONER MOLNAR: Okay.
2549 In paragraph 54 of your comments you stated that:
"...a blended mark-up should be included in the peak rates."
2550 I'm sorry, I just need you to explain to me exactly what you're talking about there.
2551 MR. TACIT: Sure. So this is because of the -- and by the way, we agree with a lot of what the cable people said earlier today about how you need to distinguish between the usage- and non-usage-sensitive costs when it comes to access in the cable network. We actually agree with that and so what this tries to do is to account for that.
2552 So one of the basic economic principles that we try to start with is let's build for non-traffic-sensitive costs using flat rates, and for usage-sensitive costs or traffic-sensitive costs we have usage-sensitive rates, because from an economic efficiency and competitive neutrality perspective, that is the best way to do it.
2553 So having accepted that notion, we then say, okay, when you're talking about the access piece, we are proposing that since access is a near-essential service that the mark-up on the access part be 15 percent.
2554 But the flat rate portion, in the case of cable you really kind of have two access pieces, although you're not going to see two rates.
2555 You're going to see a rate for the -- a flat rate that recovers the non-traffic-sensitive rates, with a mark-up of 15 percent.
2556 Then you're going to have a usage component which is going to be packaged according to the 95th percentile method, the rate that will be applied using that methodology, but it's a traffic-sensitive rate and it is meant to recover the traffic-sensitive costs of the access piece, because cable has some of those, and that mark-up is going to be different. It's going to be some competitive mark-up that you determine.
2557 And then on the transport piece we're going to have the same thing apply as you apply to the -- but on the access portion that's traffic-sensitive, because it's still access the mark-up is 15 percent, whereas on the transport portion, which is also usage-sensitive, the mark-up is a competitive mark-up, not the 15 percent.
2558 So you end up with a blended mark-up for the usage-sensitive component but that's because the usage-sensitive component is really a sum of two -- or some sort of weighted average of two components, one that's access-related and one that's transport-related. And this is just because of the quirk of the cable world.
2559 Is that clear?
2560 COMMISSIONER MOLNAR: I do understand what you said.
2561 MR. TACIT: Okay.
2562 COMMISSIONER MOLNAR: Did you put on the record what you view to be a competitive mark-up?
2563 MR. TACIT: No, we did not, but again, the beauty of this is that there are rates out there that exist, that people charge using the 95th percentile method.
2564 And we recognize adjustments might need to be made because, you know, when you talk about incumbent networks, the traffic passes through multiple links and so on, so there are some differences.
2565 But as a starting point, there are rates out there in the world today that are charged on this basis that could be used as a starting point for that exercise by the Commission.
2566 We didn't presume to tell the Commission what the ultimate rate should be. Part of the problem we always have of course is that we don't know what costs the ILECs and the cable carriers file in confidence, nor do we know the mark-ups.
2567 COMMISSIONER MOLNAR: Right. But of course we're not talking here about what is the cost or what is the price. I was asking you about the mark-up.
2568 MR. TACIT: Right.
2569 COMMISSIONER MOLNAR: So knowing what a competitive price is may not help me know what the mark-up is.
2570 MR. TACIT: Right.
2571 COMMISSIONER MOLNAR: So how do we get the mark-up, the competitive mark-up that you're proposing?
2572 MR. TACIT: Well, I guess the way I would look at this is look at the specific transit and transport rates that are up there, account for the fact that there may be an average number of links that have to be transited within an incumbent network when they're providing an aggregated service, and figure it out, you know, on that basis, maybe not focus so much even on the mark-up but on the actual dollar amount at the end of the day because that's really what matters.
2573 So what we're really saying is the dollars are out there, the dollar rates are out there for these types of services. You can use those as starting points and then just logically go from there and see what adjustments, if any, you need to make.
2574 MR. SANDIFORD: Are you asking us to propose a rate of mark-up?
2575 COMMISSIONER MOLNAR: I am asking -- yes, if you haven't proposed, based on some evidence -- you have said here that there should be a competitive mark-up. You did not say there should be a competitive rate. You said there should be a competitive mark-up.
2576 MR. SANDIFORD: We will undertake to get a competitive mark-up on the record for you.
2577 COMMISSIONER MOLNAR: With the evidence as to how you determine that competitive mark-up?
2578 MR. SANDIFORD: Yes, of course.
2579 COMMISSIONER MOLNAR: Thank you.
2580 MR. TACIT: Or rate. We'll see what we can derive or do you have a --
2581 COMMISSIONER MOLNAR: I see those as two very different things. You've suggested mark-up. You wanted cost-based rates.
2582 MR. TACIT: Right.
2583 COMMISSIONER MOLNAR: So now going to competitive rates does not really align, in my view, with your evidence, would you agree, that says you want cost-based rates with competitive mark-ups?
2584 MR. TACIT: Well, I guess the thing is --
2585 COMMISSIONER MOLNAR: Anyways, sort it out and you can come back.
2586 MR. TACIT: Yes. All right.
2587 COMMISSIONER MOLNAR: Thank you.
2588 One more question on your proposal.
2589 You say at paragraph 62 that moving to this model would allow you "to innovate to a greater extent than has been possible up until now."
2590 Now, if I -- maybe you could just explain that. I'll just wait for you to explain it.
2591 MR. TACIT: Sure, I'll be happy to do that.
2592 So part of the problem, even under existing legacy rates and under, for example, the proposals that you heard from the cable carriers and from Bell for legacy AVP, is that there is a certain amount of presumed usage in the flat rates that are being charged to consumers.
2593 So in our mind, a wholesale service should be like a LEGO set. You provide us the building blocks to put together the service and if you don't glue the pieces of LEGO together before we get them, we have the freedom to create all sorts of amazing things with those LEGO blocks as we put them together and then take them apart and do it differently the next day.
2594 So if you strip out all of the usage and all of the other things and provide us with a flat rate and don't allow for selective mark-ups to be applied based on speed and other factors that are not really cost-based and not competitively neutral, then we can innovate because we will have the basic building blocks.
2595 We will compensate incumbents for the flat rate, we will compensate them for the usage, but we will then be able to take those building blocks and there won't be anything predetermined within them about how we must package our services based on the fact that there's a certain amount of assumed usage within a flat rate or because there's a different mark-up because it happens to be a 25-megabit-per-second service instead of a 7-megabit-per-second service.
2596 So that's what we're saying, is give us the actual building blocks and we can innovate. We can come up with all sorts of creative packages.
2597 MR. STEVENS: If I can give a very concrete example. We at Execulink have -- through our On-Net, using the wholesale services, we've introduced a couple of products that customers find very appealing.
2598 In the package you heard from Bell and from the cable companies they all include a fair bit of usage, 25 Megs or whatever it is, with the package. There's a lot of customers out there that don't use that much and don't want to pay for it.
2599 So we want to be able to offer those customers a package that may be very high-speed but a very small amount of prepaid usage so they can actually pay for what they need. This would allow us the flexibility of doing that.
2600 If we have to take, as the cable companies proposed this morning, their packages with their large amount of usage associated with those packages for the customers, we can't offer those packages of very small usage that customers may want.
2601 That's just one example. There's others that we could supply as well.
2602 COMMISSIONER MOLNAR: Do I understand from what you said that you do not support flat-rate wholesale billing?
2603 I mean there is, as you know, in this proceeding, on the telco side at least, a number who propose to maintain flat-rate billing for wholesale services.
2604 MR. TACIT: I think we would like to see -- if the costs are done right and if the economic principles are applied in a stringent way, without any kind of compromises for other reasons, we would love to see this method nationwide. That's how strongly we believe in it.
2605 COMMISSIONER MOLNAR: And you would agree that there is no requirement to link -- I mean there used to be this old equivalent treatment that caused some of this fear actually, the equivalent treatment rule that said you don't apply usage-based charges to wholesale unless you've applied them to your retail base.
2606 You would agree that link should be --
2607 MR. TACIT: Absolutely, it should be severed.
2608 COMMISSIONER MOLNAR: -- severed?
2609 MR. TACIT: We want the blocks. We don't want them glued together.
2610 COMMISSIONER MOLNAR: Yes. Thank you.
2611 I want to talk to you a little bit about one of the arguments that Bell made in its -- and I understand that you're not supportive of the AVP proposal.
2612 The one component of their proposal, as you know, was that you make commitments to certain amounts of traffic and then there would be essentially surcharges if you don't manage your traffic to your commitment.
2613 I don't anticipate you to support that notion, but underlying that notion is who incurs the risk, right? You make the investment on anticipated traffic and so speak to me about how yours, where you pay simply for what you use on a monthly basis, aligns with the notion that the risk of investment is up front and not on a month-to-month basis.
2614 MR. TACIT: Sure.
2615 So when it comes to investment, first of all, the Commission studied the whole investment risk issue very, very carefully and dealt with it by giving the ILECs an extra 10 percent because they were claiming that they needed this to deal -- they were claiming that they needed something if the Commission was going to impose matching speeds, FTTN, the Commission had to do something for them because of the extra risk. The Commission to some extent agreed with them and gave them that amount.
2616 So now it's another trip to the well for the incumbents. Well, I don't buy that.
2617 Now, the cable carriers at the time, very interestingly -- and it's important to remember this -- and also MTS Allstream said to you at the time: You know, what drives our investment is the need to compete with the ILECs and we're going to compete with them regardless of what you do here and whether you give us any other premium or whatever.
2618 And that is in fact why in the decision the Commission didn't give the cable carriers a risk premium, but it did impose it on the ILECs. The ILECs were the ones that were claiming this.
2619 So to some extent, I think that this issue has kind of been put to bed to begin with, the whole risk issue.
2620 But in addition to that, there are benefits also to the incumbents by having the wholesale business. We bring a whole ton of customers. They're on their network. They get incremental revenues. They have to do none of the marketing, none of the advertising. They have to do nothing but be available to take our business.
2621 And the proof -- you know, I heard today what was said by the cable carriers, that, you know, the situation is much worse in Canada because we have two networks. In other countries, you know, if you go from one competitor to another, it still says on the incumbent's network, so they still get some revenue.
2622 Well, if they were that worried about this risk, why the heck are they not competing for the wholesale business vigorously today? That is the real question, I think, that should be asked here, because if they were truly worried about that risk, they would be doing everything they can and falling over each other to make wholesale conditions as favourable as possible to get as much of that business.
2623 COMMISSIONER MOLNAR: Okay, my final question is really a follow-up to that and that's a question related to the cable infrastructure.
2624 Under your proposal would you see cable being a viable alternative to the ILEC infrastructure? I mean you are proposing essentially the same structure for both, yet when we look at the numbers as to where the wholesale business is today, of course it's residing primarily on the telco's network.
2625 Is this something that is designed to make both networks viable alternatives?
2626 MR. TACIT: Well, it will go a long way. It's not the only solution.
2627 One of the things that we in the industry need is to have access to multiple configurations.
2628 For example, as you know, to some extent the lack of previous POI aggregation has been one impediment that existed and it's being removed, but conversely in order to have competition we need a continuation of local POIs as well, which also encourage independent ISPs to invest in infrastructure.
2629 So the answer doesn't hinge strictly on the pricing, but the pricing has to be right to begin with. The reason we only have 6 percent market share today is because of distortions caused by the current pricing regime, in large part, but there are other pieces to this puzzle.
2630 The competitive options that are available to us over time will also be helpful. POI aggregation helps and retaining the existing kinds of POI or local POI arrangements is also a very important factor.
2631 I think Bill had something to add.
2632 MR. SANDIFORD: You said most of it for me.
2633 I mean the reason why there is a huge difference in the fact that the majority of the wholesale business lies with the telephone incumbents today over the cables is that they are offering from wholesale services going back to when these services started 10 years was workable, whereas for many years the solution that was put forth by the cable carriers was not workable. A business case could not be made from it.
2634 Now, with POI aggregation coming into play, some transport services become cheaper in the marketplace, we have started to see some carriers, including my colleague to the far end of the table to my left, making significant inroads into the cable market. But previously a lot of the economic factors that allowed that to occur didn't exist. That's why we have a huge disparity today.
2635 COMMISSIONER MOLNAR: Fair enough. While you may have answered, I just want you to answer one more time.
2636 Under what you have proposed, are those disparities significantly addressed? I did hear what you said, I mean POI aggregation is part of this.
2637 MR. SANDIFORD: We believe that what we have proposed is a suitable solution for both DSL and the cable regime.
2638 COMMISSIONER MOLNAR: And would cause you to look at cable as a viable alternative to grow it?
2639 MR. TACIT: Yes.
2640 MR. SANDIFORD: Absolutely.
2641 MR. TACIT: It would certainly do that.
2642 COMMISSIONER MOLNAR: Okay, thank you.
2643 Those are my questions.
2644 THE CHAIRPERSON: Thank you.
2645 Len, you have some questions?
2646 COMMISSIONER KATZ: Oh, yes, quite a few.
2647 COMMISSIONER KATZ: Good afternoon.
2648 MR. SANDIFORD: I knew that was coming.
2649 COMMISSIONER KATZ: I have to follow up on some of these answers, I'm sorry.
2650 You said that you don't know why the cable companies and the ILECs wouldn't be falling all over you folks to get your business. I asked Bell the question yesterday as to whether and why not and what I was told was I was delusionary -- I think that was the word they used -- to think that you folks and others would want to negotiate a deal when you can come to the CRTC and take your chances with us.
2651 Can you expand upon that first of all and tell us whether you really believe you can't negotiate a deal without the CRTC?
2652 MR. TACIT: Absolutely. Commissioner Katz, you were here as well as I was in the previous proceeding leading up to Regulatory Policy 632 when what we heard from Bell Canada is: We will negotiate when we feel like it, if we feel like it. We want to appoint our wholesalers under conditions that suit our particular business needs. That was the language that was used.
2653 They have had years to negotiate with us. Ever since the essential services decision there has been a negotiation option available. They are Type II arrangements that can be put into place that don't have to be publicized, there are all manner of ways if they had wanted to negotiate.
2654 To the extent that there is any renewed interest in negotiation, I think it's because they actually feel the heat that something might be coming that's less favourable.
2655 But we have tried for many, many years to negotiate with incumbents on all manner of things and we still do. There are other issues that aren't even rate issues. We have all sorts of operational issues and problems that you would think somebody who values a customer and the fact that they are bringing revenue into their pocket would try to address and they get very poorly addressed or take a long time to address or they never get addressed.
2656 So I don't think that regulation is the problem. We have regulation because without it we would have a duopoly. That's the reason we have it. So to say that we can just somehow get rid of it and they will suddenly negotiate with us, they are going to exercise duopolistic market power. It's not going to happen, not in any meaningful way.
2657 MR. SANDIFORD: To further Chris' answer, Commissioner Katz, Chris pointed out the example that Bell gave to you in the proceeding approximately a year ago; TELUS gave you the same answer when you were questioning him with regards to Allstream and why he hasn't negotiated. The answer you got was: Because I don't have to.
2658 We have tried to negotiate. CNOC as an organization actually has a Negotiations Committee. Our Regulatory Committee sits before you today, we have a Negotiations Committee who is trying to negotiate deals with the incumbents. We are working heavily with Bell right now, as Mr. Little told you yesterday. We are all doing so in good faith, but we are just not getting there.
2659 COMMISSIONER KATZ: I thought I heard Mr. Engelhart say what would be helpful -- and this is on the cable side obviously -- is if they were building a hotel if you bought 10 rooms and guaranteed 10 rooms -- are you prepared to make commitments when you negotiate with whoever? I don't want the specifics, but are you prepared to formally negotiate commitments?
2660 MR. TACIT: Well, if they make it attractive companies will negotiate commitments, but it's a business thing, it has to make business sense. Are we prepared to give carte blanche and say we will make commitments, no. But it's like any other business deal, if they want our business I'm sure that they could negotiate commitments if the terms were fair to both sides and not just themselves.
2661 COMMISSIONER KATZ: The fallback is that we will be required to make commitments on all of your behalf --
2662 MR. TACIT: Correct.
2663 COMMISSIONER KATZ: -- and if you would rather that then I guess that's going to be the byproduct.
2664 MR. SANDIFORD: If the terms are fair to all we are happy to make commitments.
2665 COMMISSIONER KATZ: Okay.
2666 We are on record as saying that there is a need to protect consumers and obviously the need to reinforce the services that you offer as an alternative to the ILECs and the cable companies is paramount right now, but I look in a crystal ball and, as you are aware, I worked in the wireless industry, satellite and wireless are fast becoming alternatives.
2667 When I look down south I see all this optioning going on for spectrum, for broadband services to provide the same levels of speeds and capability at the same price give or take. That's going to happen in Canada as well and when the non-vertically integrated companies -- not the Bells of the world and the TELUS who are buying spectrum, but those that are out there today, the Winds of the world, the Mobilicitys of the world, the Public Mobiles and the new ones to come on as well -- come in and form a service providing capability that will provide and protect consumers, what happens to you folks?
2668 MR. TACIT: Well, first of all, there is an awful lot of speculation there. In some ways I feel like I'm reliving the last proceeding we just had that ended in 2010, whereas this, in my mind, is an implementation proceeding in many ways.
2669 But the fact of the matter is that the laws of physics aren't going to change, wireless is always going to have some coverage problems and it's always going to be more expensive than wireline.
2670 If we get to some nirvana state where wireless is substantially competitive with wireline, so be it. We have an essential services review proceeding coming up in a couple of years, everybody can try and make their case that that is happening or not happening, but that's not where we are today.
2671 Today, in 2011, we are in a situation where the ISP industry is hurting very badly and needs to be given not a handout but proper terms under which they can compete fairly.
2672 That's all we are saying. This is an implementation proceeding and let's implement what's before us. If the Commission wants to have a proceeding about wireless substitutability, as I said, the essential services review is coming and there will be other opportunities to do that.
2673 COMMISSIONER KATZ: All I can tell you is, I look back five years and there was a Minister that issued a direction saying: wireless is a substitute, it doesn't matter how much market share has been lost to local. As long as there is an alternate provider out there and a wireless provider out there as well the market is competitive, open it up.
2674 All I'm saying is that was then, this is now, times are different obviously as well, but the reality is if I look at what is in front of us right now, I can see a train coming down the track.
2675 All I'm saying is there is a need for you folks, regardless of what we do here, to understand that downstream there may be alternatives, not just a duopoly, to protect the interest of users and then we are all going to be back facing each other again. So it is something you have to look forward to and figure out where your business is going to grow.
2676 MR. TACIT: We understand that and, you know, these smart business people here, if they adapt to the markets they will survive and if they are not as smart as I think they are then they won't, but that's a separate issue.
2677 MR. GAUDRAULT: This just highlights the need to move on with this. We have been stuck here for how long? Let's get on with it and let's focus on that next thing.
2678 COMMISSIONER KATZ: Okay.
2679 You mentioned in your remarks this afternoon that the Commission has deemed essential services to be near essential at 15 percent. Correct me, are there words in our decision saying there is an essential service that is deemed to be 15 percent or near essential services --
2680 MR. TACIT: No, sorry. I don't know exactly what we said, but that near essential services are priced at 15 percent.
2681 COMMISSIONER KATZ: That's what you said here.
2682 MR. TACIT: There are essential and near essential services that are priced at 15 percent.
2683 COMMISSIONER KATZ: And the services that you are looking to have at 15 percent we have deemed to be near essential?
2684 MR. TACIT: No, no. We are saying that when you split out the access cleanly from the -- for example, unbundled local loops are near essential pricing, right, so we are saying that the access portion of a service, if you actually split it out cleanly and you don't start introducing usage and other transport into it, becomes a pure access component and it should be treated as if it's near essential. That's what we are saying.
2685 COMMISSIONER KATZ: But currently ADSL was deemed to be mandatory, conditional, non-essential.
2686 MR. SANDIFORD: Aggregated ADSL was deemed to be that fact.
2687 MR. TACIT: Aggregated.
2688 MR. SANDIFORD: There was a finding for a non-aggregated ADSL as well that was deemed to be essential.
2689 MR. TACIT: So what we are saying here is that this is now going to look like the disaggregated service.
2690 If we haven't used precise enough wording to convey that I apologize, but that's the intent.
2691 COMMISSIONER KATZ: Okay.
2692 We talked about business risk and you spoke to Commissioner Molnar about business risk and the need to keep costs down, but on the reverse side of it where is the incentive for you folks to invest if in fact we agree with you that it should be cost that should be at peak, it should be 15 percent. Why would you ever invest a dime?
2693 MR. TACIT: Because we will have actual money to innovate. And for all the reasons that you said, to the extent for example that wireless substitutability may become an issue or whatever, none of these people sits still in the marketplace, they want to invest, they want to move ahead, but they need the revenues to do that.
2694 Again I repeat, we are not looking for a handout, we want to pay the fair costs that it costs to get the services and an appropriate return that compensates them for the risk as the Commission has measured using the cost of capital that is implied in the Phase II costing model, plus the additional 10 percent on mark-up that has been provided.
2695 We have no problem with that, we just want to make sure that when that is done that we don't end up cross-subsidizing the incumbents in addition to that and then we will invest.
2696 The other thing is, we need to have disaggregated services. We had an -- I know this isn't the place, but since you asked the question I'm going to take the liberty to answer it.
2697 COMMISSIONER KATZ: I'm sure you will.
2698 MR. TACIT: We had put forward an application for an ADSL-CO service that would have created opportunities for investment. Maybe something like that will come up again.
2699 COMMISSIONER KATZ: Okay. I think the Chairman wants to call a break.
2700 THE CHAIRPERSON: I will take the liberty and call a break. Some of my panellists need a call of nature so let's take a 10-minute break and we will get back.
2701 MR. TACIT: Thank you.
--- Recessed at 1438
--- Resumed at 1452
2702 THE CHAIRPERSON: Okay. I think, Len, you were in the middle of questioning.
2703 COMMISSIONER KATZ: Thank you.
2704 I want to move on to the different models that are out there and I want to understand from your perspective whether these different models actually yield a different result, because there is a unit cost and there is a volume and depending on how you measure it and where you measure and what you measure, at the end of the day if we are all based on the same principle, and that is recovery of costs -- as defined however we end up defining the costs -- does it really matter whether it is measured at peak?
2705 If you have a quantum you have to recover whether you divide it at peak or you divide it at various points along the way, the per unit may be different, but when you multiply it by the volume, whatever that volume is, either at peak or total, you are going to get the same number.
2706 Am I missing something?
2707 MR. TACIT: Well, I don't want to accuse you of missing anything, but we do think there is an important difference and I would like to explain why.
2708 As we said in our opening statement, what matters in terms of network usage is the costs that relate to expansion of capacity. When capacity is built moving data across it -- and I think Bell actually admitted it yesterday in response to Commissioner Molnar's question about: Wouldn't you agree with me that when you are not capacity constrained there is virtually no cost to moving those bits? The answer was yes and that's the correct answer.
2709 So there is no cost to recover in that scenario, so yes it matters.
2710 So even if there is some degree of correlation between usage and peak, why would you use the indirect thing that you have to estimate when you have the actual thing itself that you can measure and then you won't be into any economic misallocations and potential cross-subsidies.
2711 So yes, we think it matters a lot or we wouldn't have gone to this extent to advance the 95th percentile approach if we didn't think it mattered a lot.
2712 The fact of the matter is that under any kind of usage-base system where you are measuring bytes there will be an over recovery of costs over time.
2713 You can't get away from that. Even if it was an under recovery, that's no better, even if you could say: Well, maybe it will be under recovery. We are not trying to avoid paying for what is consumed, but what's consumed is peak capacity, not bytes travelling across a pipe.
2714 COMMISSIONER KATZ: And this will be done on a retroactive basis annually?
2715 MR. TACIT: No, it would be done on a monthly basis.
2716 Just like, you know, you get your phone bill today and you get billed for minutes, the peak measurements can happen over a one-month period and then you figure out where your 95th percentile measurements are and those are the measurements based upon which you apply the rate, whatever it is. It's a very simple mathematical calculation.
2717 COMMISSIONER KATZ: But the per-byte rate would change over time.
2718 MR. TACIT: It would be a megabyte per second rate. It's a speed rate.
2719 COMMISSIONER KATZ: Yes.
2720 MR. TACIT: It's not -- yes.
2721 COMMISSIONER KATZ: And it would never change?
2722 MR. TACIT: No. Well, yes, hopefully it would decline over time to reflect the efficiencies that will come up.
2723 We addressed that point in our opening statement. That rate over time should fall as efficiencies of networks increase due to not just economies of scale and scope but also due to technological advances, which are happening very quickly. You know, we are going from 1 gig to 10 and 100 in pretty short order and as that happens the cost is going to be driven down pretty quickly.
2724 So yes, you would want that rate to change over time.
2725 COMMISSIONER KATZ: And we are measuring speed, not volume?
2726 MR. TACIT: That's correct.
2727 COMMISSIONER KATZ: Okay. Okay, I have to work on this one.
2728 MR. TACIT: Do you require any kind of additional explanation on that?
2729 COMMISSIONER KATZ: If you can expand upon the fact that volume never comes into this even at peak and it's only speed at peak, I would certainly appreciate it.
2730 MR. TACIT: So if you go back to the comments that we made in our opening statement, if you think of bursty traffic which -- let's say you have a specific amount of volume of traffic that you want to have to go through your facility, you can have it going over a longer period of time or you can have it go as a burst. If it goes through as a burst there is going to be less room for other traffic at that time. If it goes through at a smaller level, other traffic will be able to go at the same time.
2731 So it turns out that the way you do that, then, is to measure the volume over 5-minute increments and to figure out what the speed of that link was during that 5-minute period and to add all those up over the month and then you discard the 5 percent, because those are the 5 percent that are spurious results, and you apply the 95th percentile measurements, whatever that group of measurements is, you basically average those and apply the rate to that.
2732 COMMISSIONER KATZ: Okay. Let me move on.
2733 How many of those of you that are on the panel are selling value-added DSL versus TPIA or how many are actually buying and selling both?
2734 MR. SANDIFORD: On the panel the only member that is selling both would be TekSavvy Solutions. The rest of us are all doing DSL.
2735 MR. GAUDRAULT: Again, we are actively trying to sell everything we can basically across the country.
2736 COMMISSIONER KATZ: So you were initially selling just DSL.
2737 MR. GAUDRAULT: That's right.
2738 COMMISSIONER KATZ: When you made the decision to upgrade to sell both, what was the -- I don't want to ask you what the cost was, but what was the magnitude of the decision that you had to take and what were the thought process and how big a decision was it?
2739 MR. GAUDRAULT: So I guess you guys had a big part to play in it, the uncertainty of that. The fact that we had been selling 5 megabyte for years without any upgrade, the fact that looking -- after doing the 261 proceeding it became pretty obvious that at some point people are just going to want faster speeds period. So the sellability of the product became, "Geez, you know, people are going to start wondering."
2740 So sure enough a 5-meg product is less sellable today than when this proceeding first started a couple of years ago.
2741 So all these things, looking into the future saying, "Wow, if we don't do this we won't have anything viable." They are not giving us the higher speeds. So all this put together.
2742 And then initially we tried looking at TPIA sooner, but it is really expensive. Like on any one connection the ROI is like two or three months. It's tough. So it was a big pill to swallow just to start TPIA the way we did and it's either kind of you have to go big or not go at all with TPIA the way it was set up.
2743 Even now there are a lot of problems with it, just it's tough.
2744 COMMISSIONER KATZ: Did you have to build an overlay network or could you use the infrastructure you had and simply adapt it?
2745 MR. GAUDRAULT: So we leveraged whatever we could out of whatever we were doing with DSL, but the input rates just from TPIA themselves were the real barrier. So the gear we had, for sure that helped, there is no doubt you know. So somebody coming in just to do TPIA, it's no mystery to me why there's not a lot of people selling TPIA already, it just doesn't make sense. You are shooting yourself in the foot.
2746 COMMISSIONER KATZ: And the costs are competitive for you to buy TPIA relative to DSL?
2747 MR. GAUDRAULT: Well, let's just say I'm banking that this is all going to work out pretty good.
2748 COMMISSIONER KATZ: But today are you paying a premium?
2749 MR. GAUDRAULT: For...?
2750 COMMISSIONER KATZ: For the same throughput on TPIA as for GAS I guess?
2751 MR. GAUDRAULT: So in the access part?
2752 COMMISSIONER KATZ: Total. Per customer. When you sell to a customer --
2753 MR. GAUDRAULT: Well, TPIA costs more, but it's hard to compare apples to apples because they are different speeds. That's the reason we went there.
2754 COMMISSIONER KATZ: So you don't sell the same speed TPIA and GAS?
2755 MR. GAUDRAULT: Well no, I got 5 megabyte. With GAS I'm stuck, like I have nothing left to do. How long can you sell the same thing? I have been selling it for five years, I have nothing on DSL.
2756 COMMISSIONER KATZ: Okay
2757 MR. GAUDRAULT: It's done. I'm done. I got nothing. If you don't give me nothing, I'm done.
2758 COMMISSIONER KATZ: Okay. But you will, with the matching speed decision?
2759 MR. GAUDRAULT: So I got it today, literally.
2760 COMMISSIONER KATZ: Yes, okay.
2761 MR. GAUDRAULT: Thank you.
2762 MR. SANDIFORD: Matching speeds launched in Bell territory this morning.
2763 MR. GAUDRAULT: Once I'm done here I'm going to post the thing to say to my customers --
2764 COMMISSIONER KATZ: It launched this morning so you know what your costs are going to be presumably. So your wholesale cost of a like, I don't know, 25-meg service, higher or lower when you are buying it wholesale from Vidéotron than buying it from Bell?
2765 MR. GAUDRAULT: It's all interim and it's fairly new so I don't know.
2766 COMMISSIONER KATZ: You don't know.
2767 MR. GAUDRAULT: But this is where I'm -- like really we need to get to cost-based and once we have cost-based, then I'm going to have the financial resources to go past that.
2768 I need to get past this. I don't want to be purely GAS or purely TPIA, I want to go ADSL-CO, I want to -- like we are building fibre in Perth, like we are doing everything we can.
2769 COMMISSIONER KATZ: Yes.
2770 MR. GAUDRAULT: I mean I live and breath this stuff, that's all I do.
2771 Am I going to invest? Yes, just give me cash and I'm going to go spend it.
2772 COMMISSIONER KATZ: The rest of you are sticking to DSL?
2773 MR. STEVENS: No. If I could add to that, Mr. Katz?
2774 As you may know, we actually are a cable TV provider as well so we are offering high-speed over cable as well and so we know the product very well.
2775 We looked at offering in the Rogers areas several years ago whether we should offer TPIA, we could never get beyond some of the obstacles that were there. This is five or six years ago.
2776 You had your proceeding last year which gave us great encouragement because a number of changes were made and we reassessed, but then decided to wait until these rates become final.
2777 Because right now these rates are interim and, being quite honest, as has been proposed by the cable companies, if you accept what they proposed we probably won't be doing it. If you actually look carefully at the costs and approve rates that we believe is where they should be, we will be launching TPIA.
2778 MR. ROCCA: We also have been working on this since February 2010 trying to establish TPIA with Rogers.
2779 COMMISSIONER KATZ: Those are my questions, Mr. Chairman.
2780 THE CHAIRPERSON: On this last point of Mr. Katz, I mean we have been assiduous is trying to make sure that you are there and that you provide competition to the existing duopoly. We did the high-speed access -- notwithstanding appeal to Cabinet and everything, we changed our rules for TPIA and made them -- et cetera. It was out view that really the last thing that is needed is to deal with this hearing and make sure to reestablish --
2781 MR. SANDIFORD: You are absolutely correct.
2782 THE CHAIRPERSON: Or is there anything else that is missing from making TPIA a viable alternative to ADSL?
2783 MR. TACIT: For TPIA, no, as long as you don't take away the local POI option as well.
2784 MR. GAUDRAULT: And of course the rates are --
2785 THE CHAIRPERSON: Unless -- I didn't catch the tail end.
2786 MR. TACIT: Unless you take away the local POI option, then that will be a setback in TPIA.
2787 MR. GAUDRAULT: For us, we are completely set up now, we have 19 POIs lit --
2788 THE CHAIRPERSON: Yes.
2789 MR. GAUDRAULT: -- and so now this is going to happen and if you take that away, basically I just --
2790 THE CHAIRPERSON: Who is talking about taking that away?
2791 MR. TACIT: Well, it's not clear. There is opposition to the notion of continuing the existing POI arrangements now that aggregated POIs exist.
2792 The cable carriers -- and I don't want to speak for them -- but they seem to be taking the view that (a) they will not offer local POI arrangements any longer under the existing tariffs, and (2) that they will -- I think they would like to move us off the existing POIs.
2793 But they can speak for themselves on that point.
2794 THE CHAIRPERSON: They were here this morning and I didn't hear anyone raise that point.
2795 MR. TACIT: Well, you weren't asking those questions in your letter.
2796 THE CHAIRPERSON: Oh, I see. So this is something outside this proceeding.
2797 MR. TACIT: It's part of this proceeding, it's just not the questions you asked.
2798 MR. GAUDRAULT: The details. The devil is in the details.
2799 THE CHAIRPERSON: Okay.
2800 MR. TACIT: That's why we appended Appendices A and B.
2801 THE CHAIRPERSON: Thank you.
2802 Commission Morin...?
2803 COMMISSIONER MORIN: Thanks, Mr. Chair.
2804 One of your members, Primus, has already indicated that it would be interested as an alternative to the CNOC model to a model that would allow an ISP to pre-purchase network capacity bandwidth on a monthly basis.
2805 Were you aware that the MTS Allstream model provides for the ISP to pre-purchase a bandwidth pipe and what are your views on such a model and are any other members of your group interested in this model?
2806 MR. TACIT: So I won't speak for Primus because they are going to be here tomorrow morning and they can speak to that point directly, but in terms of -- again, we are trying to follow sound economic principles.
2807 One of the things we really wanted to do in this proceeding is have this established because we see this as a rare opportunity to restructure all of these wholesale rates. We wanted to do it in a principled economic manner. We don't think that trying to pre-purchase capacity,l guessing how much we need is the correct way to do it.
2808 As I said earlier, we want a straightforward, simple model where flat rates follow non-traffic-sensitive costs, usage-based rates follow usage-sensitive costs, we get billed monthly, we pay, we get the service, we use the Lego blocks to create our wonderful services in multicolours and shapes and that is really what we want to see done.
2809 COMMISSIONER MORIN: But just on the principle, isn't it about time that small players in this industry have the ability to choose the size of pipe that works best for them?
2810 MR. TACIT: Yes. That we agree with. We don't have a problem with -- in fact, that is one of the things we would like, is to have multiple options available and not have any artificial constraints.
2811 One of the things we have been advocating is instead of the concept of matching speed, wire speed, which means if there is no underlying cost difference when we get a port from an incumbent, we should be able to determine what the speed on that port is that we offer to the customer as long as we are not imposing a new cost on the incumbent.
2812 So I fully agree with you, to the extent that we can have more building blocks, more of those Lego blocks, then we are happy with that.
2813 COMMISSIONER MORIN: Because what I'm interested in with the MTS Allstream model is that it is splits control evenly between the incumbent providers and the wholesale customers. The little guy knows exactly the price he will pay for a month and the big guy will have a predictable income stream, too.
2814 That's why I raised this issue since the beginning with all the numbers I provided for Bell, Cogeco, Rogers and Vidéotron in comparison with the MTS Allstream model.
2815 MR. TACIT: Yes. and we understood that we wold be in a position to address that more in reply, which is why we were kind of waiting to do it then rather than get into too many details of that right now.
2816 COMMISSIONER MORIN: Okay. I appreciate that. Thanks very much.
2817 THE CHAIRPERSON: Thank you very much.
2819 COMMISSIONER DENTON: Thank you, Mr. Chairman.
2820 So I'm in a situation where my reading so far is that the phone companies and the cable companies have serious and quite rational sounding objections to your pricing, peak period pricing model, and I hate it when I don't have a view on a technical matter that I should have a view on. So I'm going to go through their objections and I would like you to address them.
2821 The first objection from Bell was that the 95th percentile method does not provide proper incentives to reduce network congestion over the shared network. It provides an incentive for wholesale ISPs to maximize their usage to their 95th peak percentile.
2823 MR. TACIT: Yes, Commissioner Denton.
2824 It relates to a comment I made earlier in response to a previous question and that is if the rate is set properly we are actually paying for that entire capacity. This was the question that related to multiple users.
2825 So if we are paying for that whole pipe based on a peak, it doesn't matter -- it shouldn't matter to the incumbent if we actually use less than that peak. In fact, that is an opportunity for them to bill somebody else at the same time as they are billing us.
2826 They actually benefit from that under this approach so we are not causing them harm by doing that. Again, it comes down to how you set the rate but this is common practice in the industry.
2827 COMMISSIONER DENTON: Anyone else?
2828 MR. ROCCA: So for example, you know, if the peak rate Saturday night, 1 o'clock in the morning for example, happened to be 600 megabytes on the network, even when the ISP was using less than that, perhaps only 300 or 200 megabytes at some other time, they would still be paying for 600 megabytes the remainder of the time.
2829 MR. TACIT: If I could add one other thing in that regard, it is Bell is starting with the wrong premise. They are trying to use that charge to constrain our use to the network.
2830 What the real challenge is, we have to find a price for them to charge us that is fair for our use of the network. This is not a constraint. This is not a straitjacket that they want to put us in. This should be a charge that we pay for our fair use of it.
2831 Also, for Canadian consumers, we need to be able to offer the services that are there, not constrain what people use. We heard that before. This shouldn't be a constraint, so that people use less, this should be an ability, so that people pay fairly for what they use.
2832 I think that Bell started from the wrong premise to try to put that as a constraint on us.
2833 COMMISSIONER DENTON: Okay. Anyone else?
2834 Okay. Objection No. 2: With the 95th percentile, an ISP may have the incentive to reduce its peak usage, but once it reduces that to a level it finds acceptable, it has no incentive to reduce its overall usage below that level at any other time of the day.
2835 MR. TACIT: That is the same point we just discussed.
2836 COMMISSIONER DENTON: Got it.
2837 This one you should be able to bat out of the park.
2838 COMMISSIONER DENTON: No, I mean seriously.
2839 MR. TACIT: No pressure.
2840 COMMISSIONER DENTON: No pressure here.
2841 The second objection seemed to be that you would have to replace your legacy ATM technology.
2842 MR. TACIT: Bell has been actively trying to get ISPs to move off ATM anyway, and I understand there is some tariff, either just filed or about to be filed -- Paul has more details on that -- where they are going to be de-standardizing ATM.
2843 COMMISSIONER DENTON: Isn't it about time?
2844 MR. TACIT: Yes.
2845 Small ISPs and big ISPs are capable of going through that transition in order to incent -- in fact, Bell, I think, for its own internal efficiency and network reasons, is trying to do this. They have actually made ISPs various transitional offers available to reduce their costs for these kinds of transitions.
2846 So this is not an insurmountable issue at all.
2847 MR. ANDERSEN: My understanding is that Bell has conveyed to us that they would like to see all ISPs, anyways, that are on the ATM network migrate, and they are looking for ways to incent that in the near future.
2848 MR. SANDIFORD: We have very few members that are still left on ATM. Most of CNOC's members have already made the transition.
2849 COMMISSIONER DENTON: Yes.
2850 Objection No. 3: We expect that smaller ISPs will not be able to control unexpected temporary spikes in traffic demands, which may cause their 95th percentile fees to increase greatly.
2851 MR. ANDERSEN: As one of the small ISPs, as they described yesterday, at least on the residential side, we have been managing our network with 95th percentile, from the standpoint of our transit, for over 15 years. It has never presented a problem from that standpoint.
2852 There is a variety of technologies available. Either we can use technology to manage the flow down to an acceptable level, or we can pay for the traffic that we use, if we use more.
2853 COMMISSIONER DENTON: Fourth: There is no mechanism for us to differentiate between business and other non-usage paying traffic and GAS residential traffic.
2854 It was the separation of charges for each ISP.
2855 MR. TACIT: Yes, and I addressed that earlier, saying that if the method is applied properly, we would be happy to see it applied to all traffic nationally.
2856 COMMISSIONER DENTON: You would. Okay.
2857 A fifth flaw with the 95th percentile is that it would be more difficult to reconcile amounts to be billed, given that measures are unlikely to be made at the exact same moment in time by the wholesale ISP and network provider.
2858 MR. SANDIFORD: We have a couple of different 95th percentile billing systems in place in our company. In every single one of them, we run it at five-minute increments throughout the day.
2859 Every single one of our servers, despite being different systems, has their time synchronized within a matter of milliseconds. There are internet time servers that are employed by every operator on their network that ensure critical pieces of their network infrastructure are timed together and synced together.
2860 As long as everybody knows that we are going to measure at 8 o'clock, there is no reason that somebody else should be saying: Oh, I accidentally measured at 8:01 and I got something different.
2861 It's a very simple problem to overcome.
2862 MR. ROCCA: Additionally, it's not like that usage disappears or is significantly different. You are talking about fractions of a megabyte per second that may be off from month to month. It's a very insignificant number statistically that there would be any difference.
2863 COMMISSIONER DENTON: The thing that I am getting from the carriers is that the 95th percentile method measures at the wrong place, or that there need to be several points of measurement, and that usage is basically an effective, practical proxy for peak usage.
2864 In other words, what they are really saying is that, practically, despite all the theoretical difference, the method that they are proposing is better than rough justice, it is actually quite accurate justice, and it does a better job than what you are proposing in practice.
2865 MR. ROCCA: As Commissioner Molnar mentioned, there is no perfect model. We think that this is the best proxy for what actually happens, which is based on capacity.
2866 I would be happy to address the Bell figure diagram, if you are interested, and go through their three points of congestion. I have a few comments about that.
2867 COMMISSIONER DENTON: Sure.
2868 MR. ROCCA: Bell indicated, first of all, at Point 1 --
2869 I am not sure if everyone has the diagram still.
2871 At Point 1, first, Bell has indicated that there is no congestion at the FTTN node. The reason for this is that, at the FTTN node, they are fed by newer Ethernet technology, rather than ATM, and they are capable of driving line speeds.
2872 In the case of the legacy or ATM --
2873 COMMISSIONER DENTON: Just to be clear, we are talking about Figure 1, the company's internet access network from the Bell presentation?
2874 MR. ROCCA: That is correct.
2875 COMMISSIONER DENTON: Thank you.
2876 MR. ROCCA: Red Circle 1 there --
2877 COMMISSIONER DENTON: Yes.
2878 MR. ROCCA: -- was the point I was making.
2879 Do you want me to repeat that?
2880 COMMISSIONER DENTON: Yes, just very briefly.
2881 MR. ROCCA: Sure.
2882 Basically, Bell has stated that, on FTTN nodes, there is no congestion, and this is because they are fed by gigabyte, or 10 gigabyte Ethernet and they are capable of driving line speeds.
2883 In the case of the legacy or ATM nodes that are fed by slower speeds -- and that's where the claim of congestion is -- much of this would be taken care of through natural migration. The heaviest users that are using those nodes would, undoubtedly, be moving to higher speed services fed by FTTN, and as ATM is migrated and upgraded out of the network, the congestion on these nodes would rapidly disappear.
2884 In the second spot, the Metro network, they have a magnitude less links to manage. These are the connections from the Central Office back to the Metro core.
2885 Our method doesn't just pay for a link that's congested, it pays for hundreds of these links that are not congested, as well. We are not asking to only pay when a link is congested, we are saying that we are going to pay the rate that it costs all the time, and you manage the congestion and upgrades as needed, as growth on the network is placed there.
2886 This is really the incentive for carriers to avoid congestion, because it really maximizes what they can charge us by ensuring that there isn't congestion in that Metro core.
2887 The third point is really the point where there is the maximum capacity, and the interface that is selected from the ISP can really control the speeds that we purchase from them, and manage policy at that point.
2888 We see the congestion as -- not only is it something that is manageable, but something that we are paying for, to avoid, and as the capacity is upgraded, hopefully they will want to avoid congestion, so they can charge us lots and lots of money.
2889 COMMISSIONER DENTON: Okay. That answers my questions in relation to that. Thank you.
2890 THE CHAIRPERSON: Thank you.
2892 COMMISSIONER PATRONE: Thank you, Mr. Chairman.
2893 Good afternoon. Just as a note of clarification, did I hear you say to Vice-Chairman Katz that your incentive to build out your own networks and your own facilities would grow if your market share also grows, and that the model is designed to help that take place?
2894 MR. TACIT: What we are saying are a couple of interrelated things.
2895 In order to be able to invest, you have to have revenues, and even if you don't have all the money, you have to have enough to be able to leverage, so that you can do whatever you need to do, get debt equity, whatever. But you do need revenues.
2896 If you are paying too much for wholesale services, you are not going to have those revenues to get the money to be able to invest.
2897 Now, when competition is fair, you would expect that, over time, we would get a reasonable share of the market, more than what exists today.
2898 As that happens, the absolute amount of investment from our side will grow, as well, because there is more money floating around on the competitive side.
2899 COMMISSIONER PATRONE: And that investment would be facilities.
2900 MR. TACIT: Yes, exactly.
2901 I should also add that we are investing today, but we are investing -- in addition to what ISPs invest directly, they invest a lot indirectly in the incumbent networks, because those incumbents have to serve us. They have to plan that capacity, and presumably they use some of the revenues that they get from us to do that.
2902 COMMISSIONER PATRONE: I just wanted to understand that, in fact, you do draw a correlation between market share and incentive to roll out networks, which you appear to do.
2903 Is the flip side not also true, that the incumbents would be disincented to build as they lose market share?
2904 When I look at what you have written in paragraph 45, you say:
"...our model creates proper economic incentives. It leads to more efficient use of incumbent network capacity, and...incentives for investment in new capacity by incumbents."
2905 I am trying to reconcile those two facts, because if your model is designed to increase your market share, essentially at their expense, then does it not also stand to reason that they would be disincented to build out network as a result of that loss in market share?
2906 MR. TACIT: I have a response and an example.
2907 The response is, if you look at paragraphs 37 and 38 of Regulatory Policy 632, the Commission repeated what MTS Allstream and the cable carriers both said, that it is competition that is actually driving them to invest.
2908 So I think it's the reverse. I think if there was more and better competition in the marketplace, the incumbents would generally be driven to invest more.
2909 COMMISSIONER PATRONE: Sir, I don't see how this model can both incent and disincent --
2910 MR. TACIT: It's incenting people --
2911 COMMISSIONER PATRONE: The ILECs as I -- if I follow the logic that you are telling me, because you are saying your model is designed to increase your market share in order for you to --
2912 MR. TACIT: Market share and network investment --
2913 COMMISSIONER PATRONE: -- rollout your own networks.
2914 MR. TACIT: Market share and network investment are not -- they are not identical.
2915 COMMISSIONER PATRONE: But you told me that there is a correlation between market share and incentive to rollout.
2916 MR. TACIT: Well, there is for us because we are starting from a very small percentage. But I think at the economies of scale and scope that the incumbents have it's not going to be a material impact on them.
2917 COMMISSIONER PATRONE: So it applies to you but not to them.
2918 MR. TACIT: We are just saying that the economies of scale and scope are very different. We are going to be rapidly wanting to invest to reach greater economies of scale and scope. So the rate of growth of our investment will probably be faster for some period of time to come.
2919 There will come a point where it balances out. I don't know when and how that will be, at some equilibrium point of some sort. But at this point we are the ones that have the incentive to want to gain that scale and scope that we don't have.
2920 MR. GAUDRAULT: Plus, they are getting back cost-wise plus whatever we are doing anyways. So it's not --
2921 COMMISSIONER PATRONE: Well, we have heard consistently that they would do better if they were -- if you -- you know, with retail customers.
2922 MR. TACIT: But would Canadian consumers do better? That's the question. Is it about keeping the incumbents whole or doing what's good for consumers?
2923 MR. GAUDRAULT: It's about competition, though. We want competition. That's why we are here.
2924 COMMISSIONER PATRONE: Yeah, I was getting at this from the vantage point of incentives to build out.
2925 MR. TACIT: What I think we want is total incentive to invest, to grow, not to preserve Bell's ability to invest relative to us, if the point is to incent everybody to invest. And the way that happens is through competition, as I said, as the Cable Carriers and MTS Allstream said in the previous proceeding.
2926 COMMISSIONER PATRONE: Well, I simply raise it because of what you have written in paragraph 45 which is that you see your model as providing an incentive for investment and new capacity by the incumbents. And I would suggest that they would take exception to that.
2927 MR. TACIT: No, because I think they will be managing their congestion better. For example, Bell would be incented. If you look at the economic underpinning of peak capacity pricing, it incents people to have designed their networks in an efficient way that avoids congestion.
2928 So to me that tells me that a company like Bell will be more inclined to augment capacity and avoid the congestion problems that they seem to be having right now.
2929 COMMISSIONER PATRONE: Well, to me logic dictates that a greater market share incents your group to build out a network.
2930 Then, losing market share would disincent the ILECs from rolling out from rolling out a further network. That's just the flip side of the same logic to me.
2931 But those are my questions, Mr. Chairman.
2932 THE CHAIRPERSON: Thank you.
2933 Tom, do you have some questions?
2934 COMMISSIONER PENTEFOUNTAS: Yes, very briefly, just to follow up on what my learned colleague brought forward.
2935 You are currently as a group profitable, are you not?
2936 MR. SANDIFORD: Yes, we are.
2937 COMMISSIONER PENTEFOUNTAS: Then if I understand your logic, you are not profitable enough to invest in infrastructure?
2938 MR. SANDIFORD: That is not correct. We are investing in infrastructure.
2939 COMMISSIONER PENTEFOUNTAS: Okay. And what would it take to get -- you know, we talk about duopolies and the evils of duopolies. What would it take to get a third player in the game?
2940 My understanding and listening I think, to Mr. Rocca and Mr. Gaudrault, was that you need greater profits and to be able to put more capital aside, to be able to leverage that to invest in infrastructure in a more powerful sort of way. Is that correct?
2941 MR. TACIT: Yes, but what we are saying is it's not about a handout. It's just about ensuring that we are not paying excessively for what we use.
2942 If we are not overpaying then that money will be there and it will allow us the flexibility to differentiate ourselves in the market which will attract revenues because we will have things that are different to give to consumers that they don't have today.
2943 COMMISSIONER PENTEFOUNTAS: So you would like to become more profitable.
2944 MR. TACIT: Yes, of course.
2945 COMMISSIONER PENTEFOUNTAS: Okay. And in so doing you will be able to invest in your own network --
2946 MR. TACIT: Investing --
2947 COMMISSIONER PENTEFOUNTAS: -- the best of all worlds.
2948 MR. TACIT: Invest more.
2949 COMMISSIONER PENTEFOUNTAS: Invest more.
2950 MR. TACIT: There is significant investment already happening.
2951 MR. SANDIFORD: I can give you some examples if you would like.
2952 COMMISSIONER PENTEFOUNTAS: I understand that.
2953 So you are going to become profitable. Basically, your model is we become profitable on the investment of the incumbents: is that correct?
2954 The network -- the investment in network amongst the incumbents -- in other words, you will go on their network and become profitable in a sense -- there is nothing wrong with it -- on their backs and that will allow you to invest in your own network.
2955 MR. TACIT: Well, when you say "on their backs" you make it sound like it's a bad thing. It's public policy.
2956 COMMISSIONER PENTEFOUNTAS: No, I'm not saying it's a bad thing. That's what I'm saying.
2957 MR. TACIT: But it's public policy that we need regulation or this Commission wouldn't be here because they have market power. And all we are saying is since the market cannot work on its own in this situation because they can exercise market power, the regulator has to substitute itself for the market and make sure they don't gouge us. It's a simple thing.
2958 We are willing to pay what we need to pay to help them recover costs and earn a fair return on it and then how they use that money is up to them, whether they dividend it out, build their networks, whatever. It's up to them.
2959 But you know, it sounds to me like you're almost questioning the underpinnings of regulation to begin with.
2960 COMMISSIONER PENTEFOUNTAS: Not at all. I'm just asking a simple question.
2961 THE CHAIRPERSON: Okay.
2962 Tom, do you -- no.
2963 Okay, those are our questions. I hope you are prepared for next Monday to have a vigorous debate with Bell and the cable companies because their views on the proper model is diametrically opposite to yours.
2964 MR. SANDIFORD: I look forward to it.
2965 THE CHAIRPERSON: I am looking forward to hopefully as a result of that debate to discern who is exaggerating and who is telling the truth and what should be the right model that we adopt.
2966 Thanks very much.
We will take a five minute break before we deal with the last one.
2967 MR. SANDIFORD: Thank you.
--- Recessed at 1530
--- Resumed at 1542
2968 THE SECRETARY: A l'ordre, s'il vous plaît. Order, please. We will be ready to start again.
2969 So, Mr. Chairman, we will now proceed with the panel of consumers from western Canada who are appearing as a panel by videoconference, one from Vancouver, one from Edmonton. It is Mr. Daniel Finnis and Jesse Whitnack.
2970 We will hear each presentation which will then be followed by questions from the Commissioners to the panel.
2971 Just for the record, I would like to indicate that Mr. Christopher Bruin has indicated to the Commission he will not be appearing at the hearing today.
2972 So we will start with the presentation of Mr. Daniel Finnis from Vancouver.
2973 Mr. Finnis, can you hear me well?
2974 MR. FINNIS: I can.
2975 THE SECRETARY: All right. Welcome to the hearing.
2976 The panel is now ready to hear your presentation. You may now proceed. You have five minutes.
2977 MR. FINNIS: Okay. I believe most of my points were actually addressed yesterday. However, I will just briefly go over a few of my key issues.
2978 I don't believe that usage-based billing being imposed on independent ISPs by the incumbents is entirely fair. I think it's applying -- it's forcing a business model on the small ISPs by the large ISPs and it disallows small ISPs to differentiate their services as they may desire.
2979 Additionally, usage-based billing has been designed, or so Bell claims, to address congestion issues and I don't believe that it does address these network congestion issues because it doesn't allow people to use extreme amounts of data in off hours when no one else will be using it, whereas -- sorry, most data will be used or most congestion occurs during peak hours.
2980 However, most -- some people may download extreme amounts of data in off hours but they will be charged far more than those downloading small amounts during peak hours who will actually cause the majority of the congestion.
2981 My last point actually will be the access to the network.
2982 Bell's claim that the majority of their congestion issues occur at the differentiation of their own customers and independent customers -- and I believe that allowing independent ISPs access to the lines to these independent ISP customers before the central offices would be a good way to reduce this possible congestion.
2983 I believe that's actually all of my points because everything else was addressed yesterday.
2984 THE SECRETARY: Thank you for your presentation. Hold on. We will have some questions for you further down.
2985 We will now hear the presentation from Edmonton of Jesse Whitnack. Can you hear me well?
2986 MR. WHITNACK: Yes, I can.
2987 THE SECRETARY: All right. So the panel is now ready to hear your presentation. You have five minutes.
2988 MR. WHITNACK: Thank you.
2989 Hello, CRTC. My name is Jesse Whitnack.
2990 The other week I called up Bell asking questions about unlimited data packages for an internet account for a residential customer. On their website even in the business section unlimited does not exist.
2991 So I tried another method. I called them up and I told them I need unlimited data usage. After three reps demanded that I pay for data consumption, even one who hung up on me, I was transferred to a business account rep who then offered me a dry loop DSL 1 megabit up, 12 megabits down with unlimited data transfers.
2992 I then reiterated I must have unlimited as I send on average 6 to 10 terabytes a month of data to various other countries around the world. He assured me I would never incur a charge for my data usage.
2993 I then asked -- I asked him why he charges UBB to residential customers and he stated that because of OTT services that are causing people to disconnect their home TV packages and replace them with the likes of Netflix, Hulu and YouTube, they cause a lot of congestion.
2994 I then asked him, "What business is it of yours that your customer has chosen other service scenarios?" He laughed and said, "How would you like it if someone were to use one of your services and replaced all your other services with someone else's free or cheaper online services?"
2995 I replied, "That's what a free market is. The consumer chooses who to invest his monthly earnings in".
2996 He laughed again and said, "Whatever. Could you provide me your company's address so I can check and see if we can get you faster speeds?"
2997 I replied, "I don't have a company. I'm a residential customer. I'm outraged that you would speak with the majority of your customer base this way and Bell's stance on usage-based billing to wholesale customers, it now seems they are just trying to get them out of the business section of their market". He hung up on me before I was able to finish that statement.
2998 Cable and telecom companies argue that bandwidth is a scarce resource and imposing caps and overage fees will relieve pressure on high speed networks. Families pay more when they use more electricity, these companies point out, so why shouldn't families pay more if they use more bandwidth?
2999 The simple answer: Because a certain fee do not get electricity for free because they are a company or a privileged friend.
3000 Another answer: ISPs somehow got around any type of regulation on the equipment monitoring the data transfers. They are very flawed and very easy to manipulate.
3001 Another simple answer: Electricity, gas and other products we pay usage fees on are traded on an open market. Data is not traded on any market. It is not a physical product.
3002 Their pricing per gigabit is anywhere from 50,000 percent mark-up to a 200,000 percent mark-up. If anyone from any other country around the world aside from the likes of Australia decided one day to DDoS, direct denial a service of a Canadian IP, it would send that customer's data usage through the roof within hours.
3003 I'm not talking about gigabytes. DDoSes can send terabytes, even pedabytes worth of redundant packets to a victim's IP address. I personally asked IPs about issues such as this and the only response I have ever gotten was, "If you got hacked you more than likely provoked it".
3004 I even tried to tell them I got DDoS, not hacked. But it makes no difference as the reps have no idea what a DDoS is. They then link me to their little antivirus and see if that will help my problems, which it won't.
3005 I have been arguing against usage-based billing for years now. Usage-based billing in itself proclaims that only the rich people are allowed to use the internet as they wish.
3006 It is not a good turn to divert congestion but again to stomp on services like Netflix and third party ISPs who are gaining popularity among a mass population who wouldn't be able to afford these new charges on top of their regular monthly bill which is anywhere from 100 to 200 a month.
3007 I would also like to take this opportunity to correct Konrad von Finckenstein's statement, quote, "Over the top, OTT, refers to internet use over and above surfing and email like streaming television and movies through online services." End quote.
3008 Seeing as how the internet itself was created on the idea of transferring data back and forth and not emailing and viewing web pages, I can just as easily refer to emailing and surfing the web as over the top services as compared to data transfers.
3009 The CRTC has stated that Canada is one of the first network-neutral countries on this planet. Your OTT statement alone completely contradicts network neutrality. To be network-neutral you must adhere to a few simple cheaper rules.
3010 Network neutrality is a principle which advocates restrictions by internet service providers or governments on consumers -- consumers' access to networks that participate in the internet. Specifically, network neutrality would prevent restrictions on content, sites, platforms; the kinds of equipment that may be attached or the modes of communication.
3011 In Canadian eyes, these are broken, all the above. Shaw has recently even threatened to disconnect me because I'm not allowed to chat in more than two simultaneous chat rooms at one time.
3014 ISPs enforce on customers or the ISP would have simply charged him for the data usage and Google would have closed that garage door for good within a month and likely got a job at Future Shop.
3015 That is the innovation that you are taking from us with usage-based billing and lack of regulation.
3016 And that's all I have.
3017 THE CHAIRPERSON: Okay. Thank you for your presentations.
3018 Let's start with you first, Mr. Finnis. You are talking about off hours and pushing things. You have heard the -- I don't know whether you heard earlier the testimony from TELUS, for instance, which suggested essentially that peak hours is a concept of the past, that nowadays people are using the internet at various times and for various purposes.
3019 So there is really no off-peak to which you can push things, et cetera, that the network is loaded at most times, et cetera and the idea that you should charge less for off peak hours and on peak hours is something that makes sense in electricity, et cetera where people use the air conditioner during the day but not at night or whatever, but it doesn't really apply to telecom.
3020 Do you agree with that or do you disagree with that?
3021 MR. FINNIS: I disagree with that.
3022 I use my internet connection and at one point I'm on cable and the internet node was very congested from approximately four p.m. till probably midnight. At midnight my speed went from 3 megabit to 12 or 15 megabit which was what I was paying for. So clearly there is a difference between peak and non-peak hours in terms of --
3023 THE CHAIRPERSON: In terms of billing, undoubtedly but that's for -- he was talking about in terms of, say, through traffic on the net.
3024 MR. FINNIS: I believe that is what I just stated, is that the traffic on the network during those peak hours from four to midnight must have been much greater because my speeds were much slower than after midnight.
3025 THE CHAIRPERSON: You measure your speeds, do you?
3026 MR. FINNIS: I do.
3027 THE CHAIRPERSON: I see, okay.
3028 And then you mentioned at the beginning of your presentation that UBB limits the choice of innovation of the ISPs and it forces a middle business model on them that they may not want to adopt.
3029 Now, before us is all of the proposals. None of them, in effect, has a UBB function in the way the original UBB expression was in effect saying that you impose caps on the end-users the same way as the large company, be they a cable or telco, imposed on its own end-users.
3030 Basically, they are all now working on aggregated models, that here is the amount of money -- of benefits. You need more, you pay extra. But none of them insist on caps.
3031 So to that extent your issue of imposing business models on the user or the buyer of wholesale bandwidth is gone, is it not?
3032 MR. FINNIS: Okay.
3033 THE CHAIRPERSON: I'm just asking for your confirmation. That was really what you were --
3034 MR. FINNIS: Yeah, I will agree to that.
3035 THE CHAIRPERSON: Yeah.
3036 Now, Mr. Whitnack, what business are you in? I'm really fascinated by somebody who uses terabytes of data each day.
3037 MR. WHITNACK: There are many businesses around the world that use terabytes of data every day, many. There is no reason for even a child to go over to his grandparents' house at night and download freely without his grandparents having to learn the ins and outs of data --
3038 THE CHAIRPERSON: Yeah, but you are -- are you -- I am trying to figure out. Are you in the business or are you a residential user, because you have one --
3039 MR. WHITNACK: I am a network administration student as a residential customer right now.
3040 THE CHAIRPERSON: Yes, as a residential customer. So all your objections that you had in terms of caps and now your experience with Bell really relate to the retail market? You, as a retail customer, are not what you want from Bell?
3041 MR. WHITNACK: No. I am not even with Bell. That Bell statement was just an investigation of mine to see where they are heading with this.
3042 THE CHAIRPERSON: Yeah, okay, but my whole point is this -- sorry, go ahead. Finish.
3043 MR. WHITNACK: They are demanding usage-based billing on -- sorry.
3044 They are demanding usage-based billing on a basis that there was never congestion, whereas they offer their businesses absolutely unlimited no matter what. Even five, 10 years down the road they are promising these businesses unlimited data.
3045 Businesses use a lot more data than consumers ever will. When it comes to pirates and copyright infringers and people like that, they don't run their stuff from their basement connection. They get a job as a network administrator or a corporate IT professional and use 150 terabyte servers in the basement of a corporate IT or a corporation and that's how they run their pirating network.
3046 And right now at the moment they are the number one data consumption on this planet. They send on average 15 to 150 terabytes of data per day around the planet.
3047 THE CHAIRPERSON: So their contention that they need usage-based billing in order to get a fair return for the expense of running a network and to avoid contention, in your view, is not warranted. If there is a problem in contention it's caused by business, not by retail users, if I understand you correctly.
3048 MR. WHITNACK: Usage-based billing, I don't believe it belongs anywhere in Canada as many other countries have shied completely away from usage-based billing on the defence that it stops innovation like people building a website in their garage like Google or Facebook at college. None of that stuff will happen on a usage-based billing system.
3050 THE CHAIRPERSON: But surely in any business at a certain point in time if you have more users than you have capacity you have to expand that cost. That's a cost and that cost you impose on your customers. That's really what usage billing is -- in the way it is being used right now is really trying to make sure that people pay in accordance with the amount that they use.
3051 Now, we can argue about what a proper model is and whether the costing is properly -- but the basic concept is the more you use the more you should pay.
3052 You object to that?
3053 MR. WHITNACK: I do, 100 percent.
3054 THE CHAIRPERSON: What is the rationale for it?
3055 MR. WHITNACK: Right at the moment the cable companies and the telecom companies, their CEOs are the highest paid CEOs in North America, including in the States. They are one of the highest paid.
3056 They are in a very profitable market right now. They have no reason to whine about money or investing in their own infrastructure which every other country, every other ISP on this planet is doing with flat rates, not with usage-based.
3057 THE CHAIRPERSON: Well, I don't know on what basis you make that statement. I certainly have no evidence that we are the only country in the world that employs usage-based billing. It is a form --
3058 MR. WHITNACK: Iceland and Australia are the only other two countries that employ usage-based billing throughout the country.
3059 THE CHAIRPERSON: Well, even if that is true that's already two countries. You just said before no other country.
3060 But I gather basically your point is one of equity rather than of business principles. You feel that these companies are doing well enough they don't need to resort to this method of billing?
3061 MR. WHITNACK: Yes.
3062 THE CHAIRPERSON: Okay. Thank you.
3063 Okay, those are my questions. Do my colleagues have any questions?
3064 Okay, thank you very much for participating in this process. I think that's it for today then.
3065 Madam Secretary, I turn it over to you.
3066 THE SECRETARY: It is. We will resume tomorrow at 9:00 a.m., Mr. Chairman.
--- Whereupon the hearing adjourned at 1600, to resume on Wednesday, July 13, 2011 at 0900