I feel like Sherman & Peabody in the "Wayback Machine" when I read shit like this.....
Burstable Internet gives you the flexibility to choose a minimum commitment level, while allowing for burstability beyond that level when it is needed; its like having bandwidth on demand. Pay for your minimum commitment and only pay for overage if and when you exceed that commitment level. An industry-standard 95th percentile calculation is used to determine the overage. With near real-time utilization reports available via our secure web portal, Burstable Internet gives you the value and control you need with the built-in flexibility to support extra bandwidth when you need it.
Back when UBB/CBB fights were taking place I wrote
»2011-77 proposed rebuttal statement
When you read the following, think of the current CBB regime we have whenever you see "AVP", and keep in mind this was written 2-1/2 years ago, so don't get too hung up on the dollar figures I was throwing around back then.
Thomas Little, president of Bell Canada's wholesale unit said Bell spent more on its internet revenue infrastructure last year than it brought in from sales of internet services. "The model for recovering those costs is broken," he said. "The solution is to ensure that those who use the most pay the most."
Ok...let's deconstruct this a bit....
Bell spent more than it took in on internet.... well, how many businesses have 1st-year cost recovery on capital investments without some form of government subsidy? Very few.
But more critically, is Mr. Little's broad statement disingenuous? Is he wanting the Commission to believe that the totality of this investment is solely due to residential retail internet service - the very subject of these hearings? Or is he saying something else? Are Bell's expenditures for IPTV included in the statement he made? Is he including backbone infrastructure expenses designed to also support all his Canadian business customers internet access but only including his independent ISP customer sales? Just what is included?
Would the Commission be so bold as to issue a request such as, "Mr. Little, could you please provide a detailed record of exactly how much Bell earned by providing data services over its 'internet' infrastructure last year to all companies such as all the major banks, insurance companies, the government of Canada, provincial governments, and all other commercial, not-for-profit companies, internal charge-backs for your own retail internet services, and charges attributed to independent ISP's? Since Bell is making a claim in a public forum about the extent of its spending and revenues, please provide these detailed numbers for the public record. If Bell chooses not to provide detailed numbers for the public record, then your statement of expenses and revenues shall be ignored in any deliberations as hearsay."
Furthermore, if those that use the most pay the most, can we see what rates Bell charges large customers who push vast amounts of data through Bell's network - Air Canada, the chartered banks, television networks, provincial and federal governments, oil companies, etc.... on a per gigabyte
basis where charges are billed that way?
And while we're at it, why does business pay more for the same services as residential customers for the same speeds/usage when the traffic travels on the same backbone at the same speeds? Are retail business internet bytes somehow longer or fatter than residential ones? Do they follow different laws of physics? Is it that business bytes get free drinks and ride in a comfy chair as they shuttle through Bell's network, and as such need to be charged more?
An 'internet' byte transferred by a small business is no different than a byte transferred by a residential user. As far as the network is concerned, the 'internet' bytes of business and residential customers are fungible and are counted and transmitted the same way. The CRTC has erred in not considering business and residential internet issues at the same time - we now all have to make submissions to the Commission yet again over the same issues in the weeks ahead.
According to Bell's Mirko Bibic, AVP "is designed to ensure an adequate return on our network investments (and) to create incentives to limit congestion," He also said that wholesale customers now stand at 17 per cent of all users on the network but routinely account for a third of traffic volume.
Limiting core network congestion is the responsibility of network engineers, not billing departments. The Commission itself has said this on many occasions. Bell's and cable's profits over the past number of years have been more than adequate to support network expansion and upgrading without impairing their survival. Their returns on invested capital is far in excess of inflation, and in Bell's case also far in excess of what was granted by the CRTC when Bell was treated as a regulated utility in years gone by. Bell nor the cable companies are lacking in financial wherewithal to fund network capacity expansion.Notably, some incumbent presentations stated that they had little/no congestion in their core networks.
Should all Canadians be punished for either the deliberate lack of investment in core capacity by some operators, or by the shortcomings of their network planners? The CRTC should NOT be rewarding malfeasance or incompetence.
Returning to Mr. Bibic's statement, Bell has said that it has more load than it feels is fair. Maybe that's an accident of history, or maybe it's a result of having a non-viable TPIA infrastructure until now.
The CRTC has asked several times about evening of the 'burden' on telco vs. cable. But why should the Commission even bother worrying about this? It's similar to the VHS vs. Betamax issue of the late 1970's/early 1980's; maybe one technology, xDSL or cable will prevail, maybe it will be 50/50 or 70/30. Who cares? The market will sort it out, customers will vote with their wallets. For most retail users the technologies are, for the most part, fungible.
Incumbent management will have to reassess whether they are spending money on the correct infrastructure - it happens all the time. Perhaps FTTN is enough, maybe FTTP is required... maybe cable will decide that FTTP is the way to go tomorrow and replace coax with fiber, but that will be a decision for each of the telco/cableco's - not me, and not the Commission. Companies will sink or swim based on their decisions - as it should be. It is NOT
the Commission's job to save Bell or Rogers from the worst instincts of their management. However it IS the Commission's job to promote and ensure vibrant competition in the telecommunications market even if it means the incumbents lose market share, or even if it means that there are asymmetrical loses of market share between telco and cable.
If Bell feels it is losing money on retail internet services then perhaps they should consider exiting that market and concentrate on being the best, most cost effective dumb-pipe they can be - in partnership with indie ISP's -and together steal business away from Rogers, Cogeco, and Videotron,.Alleged Network 'Abuse'
Nobody is abusing anything. Each personal or business user is doing what is best for them and their needs. This is NOT about going out to harm by 'abusing' any other user.
My DSL or cable link is bought and paid for - I'll use it according to way I want. Bell and Rogers wants everything every which way. Buy the link, pay for usage, pay more for usage if you are an indie customer than if you are an incumbent customer.
If I own a house in Toronto and am away in Florida over the winter, does the city charge me less for garbage collection, property tax, not needing snowplowing, or a lessened likelihood that I'll cost them money by calling city hall and having a clerk spend time on the phone with me, etc.. while I'm gone? No, they don't.
Am I subsidizing people who live in the city over the winter? Maybe yes, maybe no. Are the other people in the city over the winter 'abusing' me by using services that I'm not? Yes ....but only if you view the world through Bell or Rogers coloured glasses.
If I'm away over the winter and have Bell internet service which goes unused, does Bell refund money to me? No. Do they let me 'bank' my unused bytes? No. Am I subsidizing others as a result? According to Bell's perverse logic of light vs. heavy users I am, but they won't acknowledge it - they will say that I am using the service as I intended. Well Mr. Chairman, that cuts both ways - if it's fair when I consume zero bytes then it's fair when I consume several terabytes.AVP Analysis
Albert Einstein was famous for his 'Gedanken' (thought) experiments. Let's use one of these to examine the impact of AVP on IISP's.
Assume for a moment that Bell's AVP is approved. As a result of 'symmetry' the same will be approved for TPIA.
Because of the inefficiencies in the way AVP works at a practical level vs. the simplicity of 95th percentile with or without 'commits', or even the MTS model, there is what is called an 'impedance mismatch' in engineering terms. The mismatch is between what is actually 'consumed' and paid for at the wholesale level vs. what can be charged/recovered at the retail level by the independent ISP's.
In an engineering sense an impedance mismatch typically results in the non-functioning or inefficient operation of the equipment. In the internet access millieu, the practical effect of the crude nature of the AVP cost vs. retail billable mismatch will result in the slow-to-rapid destruction of the independent ISP's finances - and the lessening of competition
.AVP is also crude, unnecessarily complicated, and non-reflective of improvements in price/performance of networks.
If adopted, AVP will never reflect the relentless drop in cost per byte transmitted which is the hallmark of the telecommunications industry since the earliest days of telegraphy - back in the 1830's.AVP is the cash cow that keeps on giving at unjustified rates.
The ONLY way AVP could be countenanced is if there is a mandated adjustment reflecting Moore's Law (or a close approximation to it) as a divisor applied to the AVP rate each year. But then you'd need an economics department, input from industry about chip fabrication yield rates, throughput rates, monitoring of selling prices of equipment, and so on to fix the divisor annually at a fair value. So why make things complicated?
Occam's Razor postulates the principle of 'lex parsimoniae', or roughly put for our needs here - the simplest solution that does the job is usually the best.
That is what 95th percentile does - it's widely used in the industry, it is an accepted standard everyone is familiar with, it also reflects the changing and ever lowering costs of data transmission. Best of all it is simple - take the maximum usage rate for the month in Gb/second x 0.95 x the price/Gb/second, add the tax, and send out the bill.
95th percentile works and has the added benefit of preventing double or even triple-dipping by the incumbents. 95th percentile captures usage in a fair manner for both the incumbent and its direct customer, and hence provide a fair return to the incumbent.
Even incumbents themselves are users/suppliers of either 95th percentile or MTS-type billing when they deal with other network providers outside their own territory or networks. If they are happy doing that for their own needs - though I'm sure that paying a competitor for anything never makes them happy - or for customers they have contracted to provide national services to - then why should they be unhappy providing 95th percentile billing to independent ISP's? The answer is simple - market power.
The closer the incumbents can make the independent ISP's resemble the incumbent, the less incentive retail customers of all types have to consider purchasing services from those pesky indie ISP's. AVP is as close to that as they can get in the absence of UBB as first proposed by Bell.
As in manufacturing - where the economics of the cost of producing the next incremental widget is next to nothing due to economies of scale, so too is the cost of transmitting the next incremental byte over a network next to nothing. In fact, in modern networks, the cost of transmitting an incremental byte approaches zero perhaps faster than in any other field of human endeavour. However, AVP prices the incremental bytes at the same cost as the first - or even more than that - ad infinitum.
Let's repeat that to make certain you understand ....... In fact, in modern networks, the cost of transmitting an incremental byte approaches zero perhaps faster than in any other field of human endeavour. However, AVP prices the incremental bytes at the same cost as the first - or even more than that - ad infinitum.
AVP will surely create the necessary conditions for a lessening and weakening of competition in the Canadian internet access market. If AVP is adopted, at some point in the not too far distant future, the Commission will be forced to return to something it has declared multiple times that it does not wish to do - declare internet access as 'essential' and regulate *retail* internet access services prices for both business and residential customers.Last-Mile Analysis
A simplified pricing model based solely on speed
ought to be employed for the last mile.**The Regional Duopolies vs. All Canadians
The uncomfortable truth which the Commission has to face is that the handful of large incumbent telco and cable company incumbents, by their very nature, will never fully embrace competition, and will at every turn attempt to subvert CRTC decisions by simply failing to comply; to co-opt the Commission to their will through veiled threats of denying London Ontario or another municipality some particular service; of making an example of their unbridled power to your political masters.
The incumbents take great glee in repeatedly pointing out to the Commission that the independent ISP's are 'eating their lunch' ..... as if the incumbents have a god-given right to it in the first place. They bemoan the 5-6% market share that the independents have.Well it may be that the independent ISP's don't have a right to a certain market share either, but what they do have a right to is an unrigged game, and so do Canadian consumers.
If the Commission and the government want a competitive landscape in Canada, then the market share of participants needs to be rearranged - by fiat, legislation, or other means, such that a given market has at least 4-5 competitors at the retail level, each with at least 15% market share
- this is the minimum in each geographic area which can be considered effective competition.
The incumbents will cry out that they are 'owed' better - which is hard to fathom given the vast infrastructure they have in place and competitive advantages they have accrued through years of monopoly or quasi-monopoly status, with their attendant guaranteed, or nearly so, rates-of-return on capital.
Then they will cry, "think of the shareholders". Well of course incumbent management must think of the shareholders - and the logical follow though of that will be to split their companies into those divisions they feel they can earn the best rate-of-return on, and spin the rest off to shareholders to let them decide whether to hold on or dispose of the shares of those new companies.If the policy direction is looking out for what's best for consumers then the CRTC and the Competition Bureau will move to functional separation requirements if shareholders wont do it themselves.
It's easy enough for the incumbents to spin-off their 'arms' into shareholders hot little hands and then let the market pick the winners and losers. The suggestion that the ATT breakup of the 1980's was the model to look at was wrong - for that decision only created a series of smaller clones - each with dominance within their own territories - the situation with frightening similarity to the Canadian telco/cableco market.
Instead, I would suggest looking at the 1948 United States v. Paramount case and the 1956 IBM Consent Decree as the best solutions to our situation in Canada. Considering precedents such as these would have led the Commission and the Competition Bureau to have asked the right questions in the first place. Had they done so, there would be no need for the hearings a few weeks ago, no need for endless internet hearings, no need for the business internet submissions and/or hearings to come - because the framework for a competitive landscape benefiting consumers would have been laid through functional separation.Commission Actions
If for no other reasons, the Commission must throw caution to the winds and rule more favourably on the side of more effective competition - widespread, and deeper at the wholesale level, resulting in more choice of cost effective solutions for individuals and businesses across Canada.
As long as true competitive conditions exist in a marketplace, capacity will be added and suppliers (incumbents, new entrants, indies) will compete for customer dollars, in their roles as network providers, service providers, network integrators, and possibly just as resellers in some cases. They will compete with types of services, reliability, location of service, price, and customer service. Supplier margins will necessarily decrease - that's how it should be. If the shareholders of a particular supplier don't like that, then they ought to hold a vote to decide whether to exit that business and focus on improving their rate of return elsewhere
The Banking Act reforms of the 1980's opened up what was considered to be a 'closed' banking market - then populated primarily by the "Big 5" banks - to further competition from Schedule 2 banks, and mandated their access to the Canadian Payments Association, Interac, and other infrastructure of the Canadian banking system. Similar bold rulings from the CRTC are necessary.
The Commission must further and quickly educate itself on the technical details of networks so it can't be 'snowed' by vague statements from fast-talking lawyers and lobbyists in shiny suits. Doing so just might result in asking the right questions and choosing the correct decision the first time, and obviate the need for endless hearings into the future.Put this issue to bed once and for all - require 95th percentile wholesale billing by all incumbents, and zero usage billing for the last mile connection.
Just some thoughts on how the last mile might be charged.Again, take the following numbers with a grain of salt - they're from July 2011.**
Something along the lines of the following should suffice:Rates Set Today
6Mbps or under profiles - $5/month
7-10Mbps profiles - $10/month
11-15Mbps profiles - $15/month
16-20Mbps profiles - $20/month
21-25Mbps profiles - $25/month
26-35Mbps profiles - $35/month
36-50Mbps profiles - $45/month
>50Mbps profiles - $60/month
Each time an incumbent introduces a new higher speed profile for its own customers it MUST offer the same speed profile to all other ISPs. What happens to the tiered profile pricing upon the new speed introduction is as follows:
Say a 65Mbps profile is introduced. Its price is, by CRTC dictat, set identical to the previous highest speed profile price, ie. the new higher speeds price cannot exceed the tier it supercedes [note this does not explicitly set the actual dollar amount of the price it merely caps it relative to what the incumbent priced the previous tier by tariff]. Speed tiers lower than the new higher speed then are automatically collapsed into the tier belows price pointRates Set Tomorrow Based on a New Speed Profile Introduction
10Mbps or under profiles - $5/month
11-15Mbps profiles - $10/month
16-20Mbps profiles - $15/month
21-25Mbps profiles - $20/month
26-35Mbps profiles - $25/month
36-50Mbps profiles - $35/month
51-64Mbps profiles - $45/month
>=65Mbps profiles - $60/month
Each existing user, whether incumbent or IISP is automatically lifted to the maximum speed at the new price tier applicable to them. This can be automated within the provisioning system. No service charge to the IISP or end user is applicable to the automatic speed uplift.
Thus a user previously at 6Mbps paying $5 for the link will automatically get bumped to 10Mbps for the same $5/month. The only exception to this would be if the line is not capable of handling 10Mbps. If the end user needs to be moved to a remote or different type of DLSAM or CMTS in order to achieve the new higher speed, then a one-time service fee of $100 may be charged. Once a users speed tier reaches the 7-10Mps speed profile it is deemed that the users connection is henceforth exclusively on FTTN and no further service charge may be charged as automated provisioning scripts can take care of successive profile changes.
However should a user request a change of their profile outside the automatic profile upgrade, then a service charge for the change can be charged. Example: a users speed profile is currently 10Mbps and the user requests a speed profile change to 25Mbps. The user incurs a service charge for this change.This last-mile model treats all Canadian users equally whether incumbent or IISP, and automatically gives them the benefit of better price performance as it becomes available.
However, incumbents, who may prefer to lock customers to a fixed speed for the duration of a contract period, should be permitted to NOT upgrade their retail customers until the end of their contract period but MUST automatically upgrade the customer to the highest speed tier at the end of each contract period should the customer wish to remain a customer of the incumbent.
IISPs may also elect to follow this contract/upgrade model for some or all of its customers too.
The incumbent is NOT free to substitute, in any way, shape, or form, its own retail provisions on any ISP or their customers. This includes caps, penalty clauses, service charges of any sort other than those approved via tariff application, contract T&Cs, or any other conditions or fees.