Professional Documents
Culture Documents
COMMENTS OF
Introduction
2. As confirmed by the Commission in its March 11th letter regarding the scope of
issues, this proceeding deals with the issue of billing practices for mandated
wholesale residential high-speed access services. The current review, which
was commenced by the Commission on its own initiative, follows earlier
proceedings in which the Commission approved the introduction of usage-based
billing (UBB) rates for incumbent telephone carriers’ wholesale residential
Gateway Access Services or equivalent services, and for incumbent cable
carriers’ third-party Internet access services, at a 15% discount from the carriers’
comparable usage-based billing rates for their retail Internet services. 1
3. In its Notice of Consultation the Commission stated that the objective of its
current review is that “Small ISPs continue to be afforded the flexibility to bring
pricing discipline, innovation, and consumer choice to the residential retail
Internet service market.” To this end, the Commission has posed the following
questions to interested parties:
1
See Telecom Decisions CRTC 2010-255, 2010-802 and 2011-44.
-3-
4. In addition the Commission has asked that parties proposing changes to the
regulatory approach demonstrate, as applicable, how such changes would:
5. In this submission, Rogers presents its views on these issues, and proposes a
change to the current regulatory approach that it believes will better satisfy the
Commission’s stated objectives. This proposal will enable Small ISPs to
structure their own retail offerings as they choose, with or without a usage-based
component, thereby enhancing their ability to offer consumers a broad range of
Internet service plans. In support of its position, Rogers has filed as Appendix A
of this submission the evidence of Dr. Jeffrey Church, Professor of Economics at
the University of Calgary, who has addressed the importance of rational
wholesale pricing in a market characterized by increasing demand and a capital
intensive distribution network that must continue to expand to handle this
2
TNC 2011-77, para. 12.
3
Ibid, para. 13.
-4-
6. The recent debate about wholesale pricing of residential Internet services has
sparked considerable public interest. This is apparent at the regulatory, political
and public level. It is also apparent from this debate that there is considerable
confusion in the public about the distinction between wholesale and retail pricing
of Internet services. This confusion is evident in the calls for flat rate pricing for
unlimited use of retail Internet services, as well as in claims that metered billing
of wholesale Internet services results in the wholesalers dictating the retail
pricing structure of Small ISPs. It is important for all parties to this debate to
realize that this proceeding is addressing wholesale prices for Internet services –
not retail pricing which is highly competitive and unregulated.
7. In Rogers’ view, it is also important to stand back from the recent rhetoric and to
consider the public policy objectives that relate to Internet access, and the
economics underlying efficient wholesale pricing.
8. It is relatively clear what the public policy objectives are. They have been
discussed extensively over the past few years in the context of the Government
of Canada’s digital economy initiative and in the Commission’s regulatory
proceedings. Key among these objectives are:
10. Both wholesale and retail pricing of Internet service have a direct impact on the
ability of carriers to build out their networks and increase bandwidth. Unless
prices are set at levels that support the maintenance and operation of existing
facilities, as well as investment in expanded and upgraded distribution facilities,
the Government of Canada’s and the Commission’s objective of expanding
access to high-speed Internet services and increasing the capacity and speed of
Internet access services in Canada will not be attained.
11. As discussed in the joint cable company submission, the retail high-speed
Internet market in Canada is characterized by steadily increasing demand as
consumers increasingly use the Internet to download high bandwidth applications
such as video and new services such as Netflix which offer attractive video
offerings. This demand cannot be satisfied without increasing the capacity of
distribution facilities. As demand is expected to continue to increase, there is an
on-going requirement to continue to make significant investments in facilities.
Increased capacity is critical so that Canadians can participate fully in the digital
economy – we recognize Canadians need access to the amount of bandwidth
and the speeds they expect. This network expansion cannot take place without
appropriate pricing. This applies equally to the wholesale and retail markets.
12. Pricing also affects consumer choice. If pricing is too low, it will hinder network
expansion and limit both capacity and access speeds. This will result in
increased network congestion thereby limiting consumers’ ability to utilize the
types of high bandwidth services and applications they want access to. If prices
-6-
are too high, they will inhibit demand – not only for Internet access – but also for
third party applications and services.
13. As stressed by the Commission in its Notice of Consultation, pricing should also
be reasonable and fair to consumers and not require low use retail subscribers to
pay for high use by other subscribers.
14. It is against this background of public policy objectives and economic principles
that Rogers believes UBB and other wholesale pricing models should be
assessed.
15. There are some clear advantages of usage-based billing for wholesale services
in the current environment:
• the user pay principle provides the correct supply and demand signals for
efficient consumption and investment;
• low use consumers are not required to subsidize high use users
16. In contrast, flat rate pricing for unlimited use of wholesale Internet access
services would require the Internet carriers to bear the financial risk associated
with the retail pricing of Small ISPs. Such a rate structure would send the wrong
economic signals to Small ISPs – encouraging them to increase market share by
lowering prices for larger or perhaps unlimited usage without regard to the
resulting impact on network congestion. This would shift the risk associated with
Small ISPs’ retail pricing policies to the carriers, who would have no control over
it. It would also push the carriers to implement more invasive technical ITMPs to
control network congestion – rather than the less invasive ITMPs favoured by the
Commission.
-7-
17. While a flat-rated system is not consistent with the policy objectives or the
economic principles discussed above, the current UBB pricing could be improved
upon to ensure that Small ISPs have flexibility.
18. One criticism of the current UBB model at the wholesale level is that it imposes a
pricing model on Small ISPs, thereby limiting consumer choice at the retail level
and limiting the ability of Small ISPs to offer different service choices to
consumers.
19. While Rogers does not believe that the current model dictates a UBB approach
by Small ISPs, as they are still allowed to price retail services on an unregulated
basis in any manner they wish, it acknowledges that the existing UBB model
makes it more difficult for Small ISPS to offer other pricing models to their
customers without increasing their own economic risk.
20. However, the answer to this drawback of the current UBB model does not dictate
the elimination of UBB as an economically efficient allocator of resources at the
wholesale level. Rather, it points to adaptation of the UBB model to provide
Small ISPs with more scope for innovative pricing and service offerings at the
retail level.
21. In this regard, Rogers proposes that Small ISPs be permitted to aggregate the
average monthly usage assumed in Rogers' TPIA tariffs for all of their customers
under a given UBB tier acquired at the wholesale level. This would increase the
flexibility of Small ISPs to offer their customers a variety of pricing packages. As
some customers use less than the average, and some use more, this would
enable Small ISPs to maximize the value of the tiers and offer distinctive pricing
options to their customers.
22. With this amendment, Rogers believes that usage-based pricing at the wholesale
level will fully satisfy the Commission’s objectives of: allowing consumers
fulsome access to Internet; respecting the principle that ordinary consumers
served by Small ISPs should not fund the bandwidth used by the heaviest retail
Internet service consumers; and ensuring that Small ISPs retain flexibility and
continue to be a source of innovation in the industry. Consumers will be well-
-8-
served by this solution since it will empower Small ISPs to offer innovative pricing
without reference to the cable carriers’ own retail pricing.
23. It will also satisfy the principles enshrined in the telecommunications policy
objectives in section 7 of the Telecommunications Act and in the Direction to the
CRTC on implementing these objectives. This approach will ensure that market
forces are used to the maximum extent feasible as the means of achieving the
telecommunications policy objectives and that the Commission uses regulatory
measures that are efficient and proportionate to this purpose and interfere with
the operation of competitive market forces to the minimum extent necessary to
meet the policy objectives.
Demand Characteristics
26. This growth shows no signs of letting up in the near term. According to Chart B1
in the joint cable company submission, CISCO VNI forecasts a compound annual
growth rate of 33% over the period 2009 to 2014. To put this in perspective,
Canadian Internet traffic more than doubled between 2009 and 2011 and is
forecast to almost double again between 2011 and 2014. It is clear from
CISCO’s forecast that it is primarily video that is driving this growth – growing
more than 550% between 2009 and 2014.
-9-
27. Internet usage varies significantly among subscribers with 10% of subscribers
generating 60% of traffic in 2009.
28. The distribution of Internet by time of day indicates that there is a relatively
lengthy and flat peak period. As Chart B3 in the joint cable company submission
indicates, traffic volumes reach 70% of maximum by mid-day and exceed 90% of
maximum by early evening. There is only a brief period during the day when
volumes are relatively low. At least 75% of subscribers remain connected even
during off-peak periods.
Supply Characteristics
29. As discussed in section 2.3 of the joint cable company submission, capacity
available for Internet services is a finite resource. This means that it is possible
for all capacity to be consumed at any given point when demand exceeds supply.
30. Rogers expands network capacity primarily by splitting nodes. This reduces the
number of subscribers that share available capacity on the node in question.
When a node is split, incremental fibre optic facilities are used to provide
backhaul to the new node as well as additional Cable Modem Terminating
Systems in the associated head-end.
31. While the number of subscribers affects the number of nodes required to serve
them, the amount of usage generated by those subscribers (or by a sub-set of
them) also drives the need for network expansion. Therefore, while the cable
4
Sandvine, Fall 2010 Global Broadband Phenomena Report, page 18; available at
http://www.sandvine.com/news/global_broadband_trends.asp.
5
Joint Cable Submission, March 28, 2011 para. 31.
- 10 -
drop used to reach a customer’s premise has a fixed cost, the network behind
that drop does not. Its cost increases whenever demand is reaching capacity
limitations. This means that Rogers must continue to invest in its network in the
face of rapidly increasing demand.
Network Congestion
32. Network congestion is a reality in the Canadian Internet market. Faced with the
exponential growth in demand that characterizes this market, the wide range of
usage patterns exhibited by its customers, and the need to make economically
rational investment decisions, it is not possible to simply build an enormous
network that will outstrip demand for the next few years. Because the increase in
demand is more related to the use of the Internet for video and other high
bandwidth applications than it is to an increase in the number of customers
subscribing to the service, it is not currently possible to finance network
expansion to a level where all users can make unlimited use of it, without raising
prices significantly.
33. It is for these reasons that Rogers uses a usage-based billing approach to make
its service attractive to both low usage and high usage customers and to act as a
non-technical ITMP to inhibit excess use of the service.
35. As several leading economists have explained, requiring users to pay based on
use is not only fair and pervasive throughout our economy, it is also essential to
the achievement of what is referred to in economic theory as allocative
efficiency. 6 Allocative efficiency is achieved when resources are supplied and
6
Make the heaviest online users pay their fair share, Leonard Waverman, Dean of Haskayne School of
Business, University of Calgary, Globe and Mail, February 7, 2011; CRTC’s UBB decision: pretty good
- 11 -
consumed at the most efficient level. This will only be the case if prices are set
so that users are required to pay their fair share.
economics, but a very bad explanation in a muddled regulatory environment, W.T. Stanbury, Professor
Emeritus, University of British Columbia, Hill Times, 21 March 2011
- 12 -
are not forced to subsidize heavy use that is in fact degrading their own service
and access to the Internet.
39. The Commission’s principle that ordinary consumers served by small ISPs
should not have to fund the bandwidth used by the heaviest retail ISPs is an
extension of the underlying economic principle of user pay and that no
consumers – whether served by a carrier or a reseller ISP – should be required
to cross-subsidize the consumption of other users.
40. As explained in the evidence of Dr. Church, a tiered pricing structure – involving
both a fixed access fee and additional usage-based charges for use above a cap
– is important both in terms of recovery of the large capital costs incurred by
carriers to provide broadband service, and also to ensure efficient management
of the peak-load problem and congestion on the network:
14. The second is the role of pricing in efficiently managing the peak-
load problem and congestion on the network. The peak-load
problem arises when service is provided across a number of time
periods and demand over those periods is cyclical but predictable;
output is not storable; and capacity is the same across time periods.
In these circumstances increasing capacity will reduce congestion
costs during peak periods, but at the cost of having excess capacity
in off peak periods. Installing less capacity lowers capacity costs,
but increases the costs of congestion in peak periods. Managing
the trade-off efficiently between the costs of congestion and the
costs of additional capacity requires appropriate pricing. The
magnitude and sunk nature of capital investment ensures that such
management is especially important with broadband networks.
41. Under a uniform pricing structure, all users pay the same amount, regardless of
how much they value or use the network. In this context, although the costs of
network expansion are driven by a small group of high value customers, these
costs must be recovered through all users, with the result some lower usage
customers are driven away from the service.
- 13 -
42. In contrast, a tiered pricing structure allows a service provider to recover more of
its costs from those consumers that value the service most highly. Specifically, a
tiered pricing structure, involving low access fees and high usage charges
23. … reduces the extent to which consumers with low valuations for
the service opt not to purchase access, while at the same time reducing
the incentives for high valuation consumers to engage in personal
arbitrage by masquerading as low value consumers. The high value
consumers are high value typically because they have a high willingness
to pay for usage, i.e. get lots of benefit from using the broadband
network. They will be heavy users. Charging a high usage fee with low
access fees discourages them from avoiding a higher access fee since it
reduces their usage. Hence the firm can charge higher access fees to
them and lower usage fees. The effect of the first assists in recovery of
fixed and sunk costs, the second contributes to expanding usage.
26. One result of using such a price menu is that more of the burden of
cost recovery is likely to fall on high value, heavy users. Besides
typically being more efficient than a uniform price, a price menu is also
likely to be viewed by many as more equitable. Those who benefit from
the network pay a greater share of its costs.
43. The Commission does not regulate retail rates of either carrier or reseller ISPs.
In both cases, it relies on the marketplace to provide the correct signals. In the
case of carriers, demand and costs incurred to accommodate high usage
generate signals for efficient retail pricing and investment decisions. In the case
of resellers appropriate cost signals must be flowed through wholesale rates,
since resellers do not incur the underlying network and capital costs directly.
This means that the costs of extremely high users must be flowed through to
resellers. If they are not, then resellers do not bear the true costs of the services
they are providing and will not factor these costs into their own retail rates. They
will set rates that attract a disproportionate number of high-usage customers and
other customers will have to cross subsidize the costs of this use. As
summarized by Dr. Church:
Rogers Proposal
45. One criticism of the current UBB model at the wholesale level has been that it
imposes a pricing model on Small ISPs, thereby limiting consumer choice at the
7
Appendix A, Economic Principles and Usage Based Billing, Dr. Jeffrey Church, para. 3.
- 15 -
retail level and limiting the ability of Small ISPs to offer different service choices
to consumers.
46. While Rogers does not believe that the current model dictates a UBB approach
by Small ISPs, as Small ISPs are still allowed to price retail services on an
unregulated basis in any manner they wish, it acknowledges that the existing
UBB model makes it more difficult for Small ISPS to offer other pricing models to
their customers without increasing their own economic risk.
47. However, the answer to this drawback of the current UBB model does not dictate
the elimination of UBB as an economically efficient allocator of resources at the
wholesale level. Rather, it suggests that the UBB model should be adapted to
provide Small ISPs with more scope for innovative pricing and service offerings
at the retail level.
48. In this regard, Rogers proposes an adjustment to the TPIA tariff that would permit
Small ISPs to aggregate the average monthly usage assumed in Rogers' TPIA
tariffs for all of their customers under a given UBB tier. Small ISPs would only
pay UBB for any excess usage at the wholesale level when their own customers
collectively exceed the combined monthly usage allowance. Under this
approach, Rogers would establish a new TPIA usage rate based on the average
volume of usage aggregated over the end-users of each pricing tier, thereby
passing on to Small ISPs a wholesale UBB threshold that is based on a much
broader base of customer usage. This would give Small ISPs greater flexibility to
offer their customers a variety of pricing packages. As some customers use less
than the average, and some use more, this would enable Small ISPs to maximize
the value of the wholesale usage purchased from Rogers and to offer distinctive
pricing options to their customers including their own minimum and maximum
usage levels.
50. In paragraph 12(b)(ii) of its Notice of Consultation the Commission has asked
interested parties to comment on whether it should set a minimum threshold level
for the sale of bandwidth by large incumbent carriers to the Small ISPs and, if so,
what it should be.
51. Under Rogers' proposal discussed above, and in section 4 of the joint cable
company submission, there would be no need to establish an explicit minimum
use threshold. Rogers would be offering Small ISPs a volume threshold for a
given pricing tier based on an average volume of usage aggregated over the
Small ISP end-users of that tier. Small ISPs would be able to balance off their
own high and low customers' usage and only pay for additional GBs if their
customers in aggregate exceeded the specified total volume. This would make a
clear distinction between wholesale thresholds and whatever retail thresholds
ISPs might want to establish. This would give Small ISPs flexibility at the retail
level to establish their own minimums and maximums for their own customers, or
offer flat-rated services.
52. In its Notice of Consultation the Commission stated that the objective of its
current review is that “Small ISPs continue to be afforded the flexibility to bring
pricing discipline, innovation, and consumer choice to the residential retail
Internet service market.” To this end, the Commission has posed the following
questions to interested parties:
• the user pay principle provides the correct supply and demand signals for
efficient consumption and investment;
• low use consumers are not required to subsidize high use users
55. In contrast, flat rate pricing for unlimited use of wholesale Internet access
services would require the Internet carriers to bear the financial risk associated
with the retail pricing of Small ISPs. Such a rate structure would send the wrong
economic signals to Small ISPs – encouraging them to increase market share by
lowering prices for larger or perhaps unlimited usage without regard to the
resulting impact on network congestion. This would shift the risk associated with
- 18 -
Small ISPs’ retail pricing policies to the carriers, who would have no control over
it. It would also push the carriers to implement more invasive technical ITMPs to
control network congestions – rather than the less invasive ITMPs favoured by
the Commission.
56. While a flat-rated system is not consistent with the policy objectives or the
economic principles discussed above, the current UBB pricing could be improved
upon. It is for this reason that Rogers has presented its proposal to permit Small
ISPs to aggregate the average monthly usage for all of their customers under a
given UBB tier acquired at the wholesale level.
57. With this amendment, Rogers believes that usage-based pricing at the wholesale
level will fully satisfy the Commission’s objectives of: allowing consumers
fulsome access to Internet; respecting the principle that ordinary consumers
served by Small ISPs should not fund the bandwidth used by the heaviest retail
Internet service consumers; and ensuring that Small ISPs retain flexibility and
continue to be a source of innovation in the industry. 8
58. Rogers’ proposal will also satisfy the Canadian telecommunications policy
objectives enshrined in section 7 of the Telecommunications Act and in the
Direction to the CRTC on implementing these objectives. It will also ensure
wholesale rates that support the on-going investment in Internet access facilities
that is required to implement Canada’s digital economy.
59. In particular, Rogers' proposal for usage-based wholesale rates will help to
sustain and develop a technologically advanced Internet access service that
responds to the economic and social requirements of users and renders reliable
and affordable service of high quality. By providing positive signals for
sustainable network development, Rogers' model will help to increase network
capacity in the face of rapidly increasing Internet usage. This will in turn increase
the ability of users to make use of new high bandwidth services and applications
8
Ibid, para. 13.
- 19 -
while offering them a variety of usage and pricing packages that suit their
individual usage profile and financial resources. With the added flexibility offered
to Small ISPs in the form of aggregated usage at each pricing tier, Rogers'
proposal will further empower Small ISPs to offer their own retail customers
distinctive pricing and service offerings, thereby expanding consumer choice.
This will be accomplished without distorting the economic principles underlying
usage-based wholesale pricing.
60. This approach will also ensure that market forces are used to the maximum
extent feasible as the means of achieving the telecommunications policy
objectives and that the Commission uses regulatory measures that are efficient
and proportionate to this purpose and interfere with the operation of competitive
market forces to the minimum extent necessary to meet the policy objectives. It
will therefore comply with the Direction to the CRTC on Implementing the
Canadian Telecommunications Policy Objectives.
***END OF DOCUMENT***