Last week, I had the pleasure of participating in an hour-long fireside chat with Scotia Capital’s Jeff Fan to discuss Canada’s current regulatory environment and, of course, we talked about the implications for the Rogers-Shaw transaction.
Jeff created a summary of our conversation and issued it as an Equity Research Alert. I have embedded the note below.
The session was recorded. You can watch it here (registration required).
“Delivering on the Copper Decommissioning Promise” is the title of a recent Scotiabank report, issued as part of its Converging Networks 2.0 series. Frequent readers know that I have often cited Scotiabank research. I thought the bank’s September 13 report merits highlighting.
North American incumbent local exchange carriers (ILECs) have been aggressively deploying fiber within their territories over the last decade. These upgrades have delivered sizable improvements in market share, churn, average revenue per user (ARPU), and cost to serve. Canadian ILECs have been more aggressive than their US peers in rolling out fiber; hence, they have been able to deliver stronger wireline metrics. But what about the āholy grailā of copper decommissioning? TELUS Corporation is the most advanced on the copper decommissioning path within our coverage. Why is this relevant to investors, and when should we begin to bake value upside into these names? In this report, we explore some of the regulatory differences between the United States and Canada related to copper decommissioning and provide an update on fiber rollout and decommissioning plans for companies under our coverage. Bottom line, we believe regulators should encourage ILECs to decommission copper while also making sure to protect vulnerable customers. Fulfilling the copper decommissioning promise will provide additional incentives for ILECs to invest in network upgrades down the road.
As this paragraph notes, Canadian phone companies have been more aggressive than their US counterparts in deploying fibre. Scotiabank estimates that fibre represents about 60-65% of the Bell and TELUS total footprint, while Verizon is about 60%, Frontier is approximately 47% and AT&T has the lowest percentage, despite covering close to 28 million of its premises.
The copper migration by TELUS is seen as enabling monetization of the scrapped copper cables, as well as permitting redevelopment of real estate as central offices are converted. “The saved space inside COs is being rented to cloud companies to install servers.”
Of course, this raises the question of how regulators view the network evolution to fibre. In the US, the FCC has had rules in place for nearly a decade. The US regulator has a web-page describing how technology transitions could affect consumer services. In Canada, the CRTC has indicated “it will shortly address issues related to decommissioning practices through further process.” In its wholesale broadband decision last month, the CRTC added “In the interim, to ensure that consumers are not negatively affected, parties are expected to avoid instances where competitors could lose access to higher-speed aggregated HSA. Should such situations arise, the Commission is prepared to address them expeditiously on a case-by-case basis.”
Scotiabank said “We believe it will be important for the CRTC to not impede ILECsā copper decommissioning initiatives, especially now that fibre to the home (FTTH) wholesaling will be regulated, while at the same time enforcing measures to safeguard users who need access to 911 services in case of power outages.”
Scotiabank noted two primary concerns with copper decommissioning: reduced competitive choice; and, emergency phone access during a power loss. Solutions exist to mitigate against each of these. The CRTC’s Telecom Regulatory Policy CRTC 2024-180: Competition in Canadaās Internet service markets, addresses the risk of reduced competitive choice by mandating fibre resale. Battery backup provides an option for emergency access, where customers do not have alternate means to call during a power outage.
The CRTC has a very full calendar of activities, so it is difficult to forecast the timing of a regulatory review of copper decommissioning policies. I’ll leave the topic with this caution from the Scotiabank report. “We understand why putting some guardrails in place for copper decommissioning is important; however, we hope that this review does not lead to a heavy-handed regulatory decision that would curtail ILECsā drive to decommission their copper.”
There are signs of accelerating progress in cellular and satellite convergence. A recent report from Scotiabank lead with an attention grabbing headline, “Mobile Carriers Poised to Create $1.0T in Equity in the Direct-to-Cell Revolution”.
Last year, I wrote about the FCC taking steps to explore “innovative collaborations between satellite operators and wireless companies”. In February, the FCC issued a 160-page “Report and Order and Further Notice of Proposed Rulemaking” to establish what the agency called the world’s first regulatory framework for terrestrial to space inter-connectivity. The intent is “to enable collaborations between satellite operators and terrestrial service providers to offer ubiquitous connectivity, directly to consumer handsets using spectrum previously allocated only to terrestrial service.”
A March 11 Scotiabank report discussed work underway between AST Spacemobile and Latin American operator AMX as well as TIM Brasil. The report provided capital cost metrics that demonstrate why there is such a high level of interest in interoperability. Scotiabank is calculating the cost of connecting a subscriber in a remote area with a traditional terrestrial tower at a minimum of US$111/subscriber. This estimate includes “civil infrastructure, radio access network [RAN], backhaul, core, software, and permitting”. In lower population density areas, Scotiabank says the cost could be as high as US$500/subscriber.
By way of comparison, the bank says “SpaceMobileās BlueBird satellites could do the job for US$10/sub”, based on 750 MB of data consumption per month, “an ideal solution for distant areas out of coverage”.
Recently, RCR Wireless News reported on developments from the Satellite 2024 conference. According to the RCR report, many of the players are talking about the need to get “orders of magnitude improvement in airtime pricing”. During one of the panel discussions, Mark Dankberg, CEO of ViaSat said that “while satellite thinks of pricing in terms of dollars-per-megabyte, terrestrial network operators are thinking dollars-per-gigabyte.”
Nonetheless, work on cellular and satellite convergence is progressing. A year ago, Rogers and SpaceX announced an agreement enabling Canadians to stay connected beyond the limits of terrestrial wireless networks. Last week, the Canadian government agreed to lend more than $2 billion to Telesat to help fund development of its broadband satellite constellation.
How will the economics work? According to Scotiabank, cellular and satellite convergence could be a “structural solution to a cash-strapped telecom industry.” The bank estimates that the global telecom industry invests more than US$310B in capital each year, in addition to US$68B in tower leases. Scotiabank estimates mobile carriers spend more than 15% of consolidated capital each year in RAN investments in lower density or non-profitable areas, “frequently to comply with license requirements.” As a result, Scotiabank believes satellite-enabled connectivity could represent US$46.5B in annual capital savings for mobile operators.
With a trillion dollars in equity potential, cellular and satellite convergence is truly the next space race.
Is it possible that ISED got it right? As Canada’s 3800 MHz auction closed yesterday, it appears that ISED’s use of “caps” (as contrasted with spectrum set-asides) may have contributed to an auction that kept prices internationally competitive.
The total money raised in the auction was $2.16B, far below the Bay Street financial analysts expected range of $4B – $10B. The average cost per Mhz-pop was $0.29, more than 60% lower than Scotiabank’s pre-auction estimate of $0.70. As I wrote earlier this week, the $0.29 is in line with Australia’s recent auction which worked out to C$0.26 per MHz-pop.
Recall that only 2 years ago, the 3500 MHz auction raised $8.9B, with an average cost of $3.28 per MHz-pop.
With significantly higher costs of capital, financial analysts were expecting bidding to be lower, but the results came in significantly lower than expected. BMO Capital Markets called it “A much more disciplined auction.” Scotiabank said “Finally a spectrum auction that does not break the bank”. TD Securities said, “In short, we are delighted with the outcome of the auction. Each of Rogers, Bell, TELUS, and Quebecor spent materially less than what we and the Street had expected in this mid-band auction.” A note from National Bank credits the cross-band spectrum cap and more available spectrum.
The lower spectrum cost means carriers will be in a better position to invest in physical infrastructure.
A number of carriers released statements last night:
Reading these, we might conclude that ISED got it right.
In any case, remember, there is a free webinar from the International Telecommunications Society next Thursday (December 7): “Optimizing spectrum auctions”. See you there!
Could the CRTC’s latest wholesale regulations stifle competition, precisely the opposite of what was intended?
The introductory section of the decision is filled with puzzling statements, like this:
In recent years, the Commission has noted declining competitive intensity in this industry. The number of Canadians who buy Internet services from independent wholesale-based competitors has fallen by 40%, even as the overall number of Internet subscribers in Canada has increased. In addition, a significant number of wholesale-based competitors have been bought by incumbent companies. When competitors exit the market, Canadian consumers are left with fewer options. It is therefore important that the Commission revise its approach to promote competition and protect the interests of Canadians.
The CRTC seems to believe that the most relevant measure of competitive intensity is by counting the number of smaller wholesale service resellers. The CRTC didn’t look at pricing as a measure of competitive intensity; despite rampant inflation, prices for internet services have declined nearly 8% in the past year according to Statistics Canada’s Consumer Price Index. The CRTC didn’t look at levels of investment; a recent PwC report found “the Canadian telecom sector has invested an annual average of $12.1 billion in capital on network infrastructure. This represents approximately 18.6% of average revenues, which is higher than the 14.2% average across the peer telecoms in the U.S.A., Japan, Australia, and Europe.”
Falling prices and high levels of capital investment are inconsistent with declining competitive intensity.
The CRTC claims there has been a 40% drop in the number of Canadians who buy internet services from “independent wholesale-based competitors”. (I am unable to find data that supports the CRTC’s 40% drop in wholesale subscriptions. The CRTC’s Open Data shows that wholesale subscriptions fell 30% from a peak of 1.3 M in 2019 to 0.9 M in 2022.) In any case, this is a small subset (roughly 6%) of the competitive marketplace for broadband services. Let’s look at the CRTC’s own open data (Tab N-I5 of the Data – Retail Fixed Internet table). This chart examines the share of the residential internet market by type of service provider.
What we see is that phone companies (termed “Incumbent TSPs”) and cable companies jointly held 87.9% of the market in 2015. That share declined each year through 2020, before moving up slightly in 2021. Newly released 2022 data shows a 2.5% jump in phone company share, largely at the expense of wholesale-based service providers and likely due to acquisitions. The independent wholesale-based segment dropped in 2022 by 2.5% of the market to end the year with 6.2% share. I’ll note that in most cases, the acquired companies continue to operate as flanker brands, so consumer choice has not been reduced.
What is interesting to see is the relatively steady growth in the segment called “Other facilities-based carriers”, rising from 4.5% to 6.0%. Other facilities-based carriers have grown their market share, going from 700,000 subscribers in 2018 to 900,000 in 2022. In other words, we are seeing evidence of the success of facilities-based competition.
The CRTC’s remedy for the declining presence of one type of market participant (wholesale-based) is to require Bell and TELUS to enable resale of their fibre to the home facilities in Ontario and Quebec only. Why Ontario and Quebec only? The CRTC Decision says “The record of this proceeding shows that the competitive presence of wholesale-based competitors has declined most significantly in Ontario and Quebec. These provinces are where competitors have historically attracted the largest number of subscribers, and where they are currently losing subscribers the fastest.”
Duhh. Of course these two provinces “have historically attracted the largest number of subscribers”; more than 60% of Canada’s population live in Ontario and Quebec. Doesn’t it follow that these provinces are where competitors would attract the largest number of subscribers? And it logically follows that where you have the most subscribers would correspond to where service providers would be losing subscribers the fastest. If the CRTC wants to change that state, why is the decision geographically limited? As the CRTC explains in the body of the decision, the Commission still isn’t sure if mandated wholesale access is going to be a permanent or temporary arrangement. “[T]he Commission considers that a narrower scope would reduce the potential impact on both competitors and the incumbent carriers if the Commission later determines that aggregated FTTP access is not required on a longer-term basis or is to be provided under different service configurations.”
In any case, the decision applies only in Canada’s two largest provinces and only to the two largest phone companies, not the cable companies and not the independent phone companies, such as TBayTel in Thunder Bay. In reality, this means the decision has a disproportionate impact on Bell because TELUS is the incumbent local phone company for a much smaller share of the total population, and only in Quebec. But it is worth noting that the cable companies, not the phone companies, are the biggest segment in the broadband market. Resale of gigabit services from the cable companies have been mandated by previous CRTC determinations.
In last week’s decision, the CRTC claims it recognizes and is supportive of investment in facilities.
At the same time, the Commission recognizes that continued investment by incumbent companies is crucial to ensuring that Canadians continue to benefit from robust and reliable Internet services. To achieve this, the Commission has established just and reasonable interim rates that wholesale-based competitors will pay those incumbent companies for access. These rates will ensure that large incumbent companies across Canada continue to have incentive to invest in their networks.
Unfortunately, just and reasonable rates for existing fibre facilities don’t necessarily support the business case for building fibre in new areas.
Even though the wholesale rate may be set to cover the “cost” of the facility, Bell won’t be financially indifferent when a customer chooses a resale-based service provider. Scotiabank estimates that FTTH households have an average $140 per month bundle, substantially more than the revenue associated with a wholesale customer. A major inhibitor of new fibre construction is that the CRTC rejected Bell’s proposal to base the rates on a combination of 5 years of historical and 5 years of forecasted capital. The CRTC set rates based on the historical capital costs.
As Scotiabank recently noted, the average cost per home for fibre has historically been in the order of $1500. This is expected to increase to $2000 per location as fibre plans move to less urban locations. These areas may already have a cable company providing 50/10 broadband and therefore are ineligible for government broadband subsidies. So phone companies look at whether the expected revenues from expansion justify the cost of upgrading to fibre in order to compete against the cable company. With higher capital costs to roll out fibre to less-urban households, the revenue requirements are logically higher. In any business, if the expected revenues from a proposed fibre area don’t provide a reasonable return on the investment, the capital project won’t get approved.
As a result, Bell responded by announcing a cut in its capital program. The capital cuts won’t impact the urban centres; those areas already have fibre. The budget cuts likely will not impact fibre rollout to areas that do not have 50/10 broadband, although there may now be a larger subsidy requirement. As I have written before, the biggest areas of concern will likely be in suburban areas, where the business cases for fibre is most fragile.
I can’t help but consider the irony of the CRTC’s wholesale fibre decision resulting in a reduction in facilities-based competition, the competitive choice of more than 90% of all Canadians.
Is the CRTC choosing to protect competitors when it should focus on protecting competition? Does the most recent decision micromanage some industry participants, attempting to support one group of competitors at the expense of others? To what extent should the CRTC be examining factors to incent intermodal competition for consumer broadband, leveraging 5G for fixed wireless?
The Commission needs to carefully consider whether its regulations stifle competition, and inhibit investment.