Carrier-Neutral, Open-Access Model enables fair telecom competition
1 The Carrier-Neutral, Open-Access Model
In this model, wholesale access to facilities-based TSP networks, most importantly ILEC networks who dominate Canadian telecom, must be neutral to all third-party providers who want access, meaning there is no favourtism and open-access meaning there is transparency and accountability in wholesale policies, practices and agreements to ensure fair-dealing.
1.1 The Carrier-Neutral Principle
The principle of carrier-neutrality is required to prevent TSPs wholesale division from favouring their own retail division over third-party providers’ retail divisions. Also, carrier-neutrality will encourage competition from non-facilities-based existing and emerging providers because they will have the confidence that they are competing on a level playing field.
1.2 The Open-Access Principle
These networks must also comply with the principle of open-access; meaning that the prices, specifications and agreements for wholesale access are open and transparent and there are significant consequences and penalties to the wholesaler for anti-competitive price and non-price behaviours. This information is published and is publicly available to achieve transparency.
There are two methods of achieving the carrier-neutral and open-access principles.
1. Operational separation – is the idea that vertically integrated facilities-based companies who are the dominant providers in their markets (ILECs, SILECs and cable companies) separate their wholesale and retail businesses by establishing a “Chinese Wall”[i] between them which is a barrier that separates wholesale and retail divisions as a means of restricting the flow of information. Regulations are established by the regulator to ensure that the Chinese Wall is maintained so that non-dominant third-party providers may enjoy open access on a carrier-neutral basis to the incumbents’ networks. In particular, the regulations are intended to prevent the incumbent from sharing information between their wholesale and retail divisions that would provide them an unfair competitive advantage over the third-party retail provider. This model is difficult to administer because it is extremely difficult to know and often impossible to gather evidence to prove that the information is or is not being shared between the wholesale and retail division of the incumbent provider.
2. Structural separation – is the concept that the wholesale and retail divisions of the dominant providers are separated into two or more companies. By physically and legally separating wholesale and retail services the inherent conflict of interest of the vertically-integrated company in the operationally separated model are eliminated. This is the case because the wholesaler can only earn wholesale revenue. Consequently, the wholesaler is motivated to attract retailers onto its network as this is their only source of revenue. Moreover, the regulatory burden to know or prove that the wholesale and retail divisions of vertically integrated telecom companies have shared information is reduced or eliminated, as the self-interest of the wholesaler and the retailer will see to it that they don’t.
1.2.1 Conclusion on the principles of open-access and carrier-neutrality
Therefore, a carrier-neutral, open-access wholesale regime is the most effective platform for ensuring more consumer choice, leading to better Internet services and lower rates. Subsequently, the most efficient and effective way to achieve the principles of carrier-neutrality and open-access is to structurally separate the wholesale and retail divisions of vertically integrated dominant telecom providers.
1.3 Operational separation of wholesale and retail services
The following is a brief examination of issues of the operationally separated model in Canada.
1.3.1 Canada
In Canada, we practice “operational separation” only. Operational separation is the idea that vertically integrated facilities-based who are the dominant providers in their markets (ILECs, SILECs and cable companies) separate their wholesale and retail businesses. The CRTC then regulates and enforces regulations to ensure that information is not shared between the two sides of the business that would disadvantage third-party non-dominant TSPs and virtual resellers who require access to the incumbents’ networks in order to provide services. But in practice, the regime to ensure this so-called Chinese wall between wholesale and retail is maintained, has repeatedly failed. As reported in the Financial Post on January 31, 2020, “The BTLR report also called calls for legislative changes to prevent lengthy appeals processes to CRTC. Distributel’s chief executive officer, one of the small telecom companies that relies on wholesale rate access to the big telecoms’ networks said the recommendations were welcome news, “the resellers won a favourable decision from the CRTC in 2019 on wholesale rates, but Rogers and BCE Inc. are now appealing that decision to the courts. “I especially like that they’ve called out the need to change and curtail the appeals process,” Stein said. “Parties have used the appeal process just to delay inevitable decisions.”[ii]
Since deregulation in 1993, TSPs, principally ILECs and SILECs and cable company incumbents have fought to maintain their market share dominance by thwarting retail competitors through price and non-price behaviours. The CRTC, through the administration and adjudication of the Telecommunications Act is supposed to ensure that third-party TSPs such as non-dominant carriers and MVNOs have equal and fair access to facilities-based providers’ networks and established wholesale and retail “operational separation” through regulations to accomplish that.
However, operational separation has largely failed as incumbents wholesale and retail divisions are in a permanent state of conflict of interest. This is because at the board level of each incumbent there is no operational separation between wholesale and retail. A telecom incumbent’s board has a fiduciary responsibility to shareholders to maximize value and shareholder returns. Consequently, telecom incumbents employ their vast resources to influence regulatory outcomes and employ anti-competitive price and non-price behaviours to roadblock retail competitors to maintain market share dominance. When they are infrequently caught, incumbent telecoms bring the full weight of their legal and political resources to litigate against competitors, petition the Commission to changes or stall decisions and lobby the political masters of the CRTC to reverse unfavourable decisions. These strategies, as documented above, have been very effective at stifling competition for over 100 years, protecting market share of incumbents and keeping retail prices for subscribers artificially high, such that Canada has some of the highest telecom prices in the world.
These price and non-price anti-competitive behaviors include misleading the CRTC, ISED and consumers about performance, coverage and prices and playing games with forborne services. For example, ILECs heavily influenced ISED and CRTC in developing the guidelines for the CTI and Broadband Fund programs including the criteria for determining if a hex is served. In addition, the incumbent telecoms provided virtually all of the date regarding claimed coverage of hexes in the CTI and CRTC Broadband Fund national map where there is clear evidence that many of those hexes do not meet the universal service objective. This is done in an effort to limit third-party competition and freeze the competitive market. Plus, the ISED and CTI have no process to validate the coverage claims of incumbents such as surveying residents and businesses to audit coverage claims.
Incumbent use forborne status to thwart competition. For example, a common practice of ILECs is to share information between their wholesale and retail divisions when a request for proposals or tender is issued by public sector organizations for multi-site contracts. These WAN/SD-WAN services are forborne from regulation. Because non-dominant providers do not have ubiquitous coverage, they must come to the ILEC wholesale division to provide transport, access and FTTP connectivity to some or many of the sites in the tender. Two things generally occur, 1) the wholesale price quoted a retail third-party retail provider by the ILEC is higher than the retail price the ILEC’s retail division is including in their own retail bid, and 2) the ILEC’s retail division is informed by its wholesale division of the third-party’s requirements giving the ILEC retail division a competitive advantage.
Another common practice is for ILECs to deny access and/or impose high make-ready fees and/or delay issuing permits to third-party providers for attachment to ILEC-owned pole and duct infrastructure and access to central office facilities and manholes and colocation facilities. This is a particularly effective technique for stymieing third-party retail providers when time sensitive projects such as fulfilling RFPs and tenders let by public sector organizations. The BTLR Panel stated, “Our new model is platform agnostic and technology neutral. It focuses on the activities being carried out and establishes consistent obligations to support Canadian cultural policy for all media content undertakings involved in similar activities.”[iii] Further, the Panel’s recommendations include:
· “establishing an obligation, as a condition of forbearance from retail rate regulation, either to mandate supply of related wholesale inputs or to explain why it is not necessary or appropriate to do so;
· replacing the tariff regime with a modernized system for reference offers that would improve wholesale outcomes by elevating the importance of terms and conditions of service, including service supply and quality conditions; and
· broadening the power of the CRTC to mandate interconnection.”[iv]
Therefore, to eliminate the inherent conflict of interest of vertically integrated ILECs, there needs to be structural separation between the wholesale and retail divisions of the companies.
1.4 Structural separation of wholesale and retail services
Structural separation of wholesale and retail would solve the problem we have today in Canada where ILECs exhibit anti-competitive price and non-price behaviours. For example, wholesale rates are often higher than retail rates for forborne services such as multisite bidding opportunities for public sector business of ILECs as described above. Third-party providers are often denied pole attachments and external network-to-network interfaces of ILECs making it uneconomical or technically unfeasible to access and traverse the ILEC network. Finally, structural separation of wholesale and retail telecom would reduce the regulatory burden on the CRTC and smaller TSPs, by eliminating the conflict of interest inherent between retail and wholesale service provision of vertically integrated ILECs that drives their anti-competitive price and non-price behaviours in the first place.
Structural separation would eliminate the inherent conflict of interest that leads to the ILECs misleading the CRTC, ISED and consumers about performance, coverage and prices and playing games with forborne services. A ubiquitous, unified and standards-based FTTP and LTE national wholesale network, based on the principles of carrier-neutrality and open-access, will lead to lower capital and operating costs and higher uptime availability to the benefit of wholesale providers and to all retail TSPs, ISPs and ASPs, which is critical to support reliability and usability of end-users’ differentiated applications. While the CRTC would need to be vigilant in regulating and enforcement as they should be today, the primary obstacle to fair and increased competition, which is the ILECs’ conflict of interest, would be removed. Indeed, based on the results of other countries who have implemented structural separation of wholesale and retail services, the wholesale provider becomes highly motivated to attract retail providers as wholesale customers, since wholesale services are their only source of revenue. This would lead to more retail providers emerging to compete for business and residential customers as the price and non-price barriers to entry into the market would be removed, lowering the cost of entry and time to market.
By structurally separating wholesale and retail service, the wholesale infrastructure owners could provide wholesale access to retail provider that exist in the marketplace in Canada today. For example, third-parties, such as resellers on wireline and MVNOs on wireless, may compete to provide services and apps; and because these “virtual network” providers do not own the telecom infrastructure they aren’t in competition with the wholesale providers.
A structurally separated, carrier-neutral, open-access wholesale and retail regime will lead to more competition for subscribers. More consumer choice will lead to better services and yield lower rates for everyone. In turn, this would give all Canadians world-leading access to public safety, healthcare, education, government and marketplaces; giving every Canadian business and resident a competitive advantage versus their OECD peers whom they presently lag behind. Importantly, this new regime will better enable federal Ministers to fulfil their mandates universally and effectively as they are statutorily obligated to do, as every business and resident would enjoy equitable access to the Internet.
The following are two example of countries who have successfully implemented forms of the structural separation models.
1.4.1 Singapore
Singapore[v] had much the same situation where two telecom companies owned and operated vertically integrated facilities-based telecom companies and dominated the market as represented by the pizza slice a shown in Figure 1. The model in Figure 1 is Singapore’s approach for breaking up the vertical control the telecom oligopoly had in order to enable third-party retail providers to compete for services more effectively by creating structural separation between the infrastructure providers and service operators within the pizza slice.[vi]
The government of Singapore mandated change setting up a “Passive Infrastructure Co. (AssetCo) which owned the conduit and pole line attachment infrastructure whose sole source of income depends on the Fiber Network Co (NetCo) to install/attach fibre optic cables on Asset Co’s assets. So, Asset Co is not motivated to thwart said attachments by NetCo. In turn, the NetCo is structurally separated from the Active Infrastructure Co. (OpCo), so the NetCo is highly motivated to attract OpCos to provide lit Wavelength, Ethernet, Internet, and Transport services (layers 2 – 4 in the Open Systems Interconnection or ‘OSI’ stack[vii]) onto the NetCo fibre optic network. Therefore, NetCo is not motivated to prevent said services. Finally, Singapore created operational separation between the OpCo and providers delivering Retail and Consumer services to end users. This was because there were not enough competitors in this part of the market to enable physical separation, so they needed the OpCo to be able to compete to deliver these services as well (this is the main weakness of the model as this compromise maintains the inherent conflict of interest between OpCo wholesale and retail with their Retail and Consumer competitors). However, this structural and operational separation AssetCo and NetCo overcomes the main inherent conflict of interest present in the vertical integrated telecom company, by physically separating the AssetCo from the NetCo and physically separating the NetCo from the OpCo.[viii]
This approach has yielded tremendous results, such that Singaporeans enjoy one of the highest FTTP penetration rates in the world. Singapore also claims the fastest speeds, highest availability, lowest latency, greatest scalability and reliability in their optical connections for subscribers and one of the lowest cost per Mbps for Internet in the world.[ix]
Figure 1 Singapore’s separation of vertically integrated telecom networks and services model
1.4.2 New Zealand
Through the New Zealand Government’s Ultra-Fast Broadband (UFB) Initiative, [x] the existing copper network is progressively being replaced by a fibre network. The network is financed on a concessional basis. Upon the launch of the UFB Initiative in 2009, the Government established Crown Infrastructure Partners (CIP) as a Crown-owned investment company to manage the Government’s investment in the new fibre network. Among other responsibilities, CIP’s role has included the negotiation of commercial agreements with the private-sector partners who are constructing the new network, namely Chorus and the Local Fibre Companies - Northpower Limited (Northpower), Ultrafast Fibre Limited and Enable Services Limited (Enable).[xi]
Wholesale and retail providers of fibre services are structurally separated as shown in Figure 2. The wholesale provider, Chorus, offers carrier-neutral, open-access to its network. “The Telecommunications Act sets out ‘line of business restrictions’ for Chorus, including a prohibition of providing retail services and restrictions on providing wholesale services beyond layer 2.”[xii]
Figure 2 New Zealand's Wholesale Structural Separation Model[xiii]
Government Communications Minister Amy Adams stated, “The UFB build is going from strength-to-strength, with fibre being rolled out to communities up and down the country (including many rural communities). The project continues to be on budget and well ahead of schedule with over 618,000 homes, workplaces and schools now able to connect to the UFB network. It’s held in good stead internationally, with [Nokia] highlighting the programme’s design and execution as well-managed and cost-effective.”[xiv]
According to Vox News, “The initial UFB project consists of 33 areas covering 75% of the population to be complete by end of 2019. UFB2 (January 2017) would provide fibre to more than 151 new towns bringing coverage up to 85% of the population to be complete by end of 2024. UFB2+ (August 2017) would provide fibre to more than 190 new towns bringing coverage up to 87% of the population with the UFB2/2+ project now to be fully rolled out by end of 2022. As of November 2019, the original UFB project is 100% complete, with an uptake of 55%.”[xv]
Some key fibre statistics reported are:
· Total length of fibre cable laid: 28,000kms
· Length of individual strands of fibre laid: 2.2 million kms
· Target for the uptake of fibre was 20 percent by the end of 2019, uptake today is 55 percent
· When Chorus started the fibre build in 2011:
· the average connection speed was 10Mbps - it’s now over 125Mbps
· the average internet data usage was 12GBs, it’s now 297GBs
· Chorus residential and small business fibre users are now averaging 360GBs of internet data monthly
· Peak time network throughput now regularly above 2Tbps
· New peak time record of 2.60Tbps coincided with the quarter finals of Rugby World Cup
· Gigabit connections growing rapidly, now nearly 70,000
· Recently launched 2Gbps and 4Gbps wholesale plans (Hyperfibre) available from February next year, an 8Gbps service is planned to follow
Similar best practice models have been adopted in Japan,[xvi] South Korea,[xvii] and Sweden[xviii] and many more examples of structural separation are in operation today.[xix] “The new European Electronic Communications Code (EECC) expressly encourages wholesale only operators by providing them with a specific, light regulatory regime as they naturally provide access to multiple telecom service providers without discrimination or abuses. For this reason, the wholesale-only model is the ideal vehicle for the roll-out of very high capacity networks as envisaged by the Code, as well as a facilitator of innovative services such as Internet of Things (IoT) and 5G.”[xx] The New York City municipal government has recently released an “Internet Master Plan” based on universal broadband throughout the five boroughs, relying on open-access fibre networks that can be used by multiple ISPs because they are carrier-neutral.[xxi]
1.4.3 Conclusion on structural separation versus operational separation
Therefore, the optimal mechanism to ensure robust retail competition based on carrier-neutrality and open-accessibility is to structurally separate the wholesale and retail businesses of vertically integrated telecom providers as it eliminates the inherent conflict of interest that drives the anti-competitive behaviours of these companies.
[i] https://whatis.techtarget.com/definition/Chinese-wall
[ii] https://business.financialpost.com/telecom/time-to-reinvent-and-rename-crtc-legislative-review-panel-recommends
[iii] https://www.ic.gc.ca/eic/site/110.nsf/vwapj/BTLR_Eng-V3.pdf/$file/BTLR_Eng-V3.pdf, page 12.
[iv] https://www.ic.gc.ca/eic/site/110.nsf/vwapj/BTLR_Eng-V3.pdf/$file/BTLR_Eng-V3.pdf, page 14.
[v] https://www.oecd-ilibrary.org/docserver/5k49qgz7crmr-en.pdf?expires=1552849345&id=id&accname=guest&checksum=BAC5B863E2EA4675B373B0C6165540D7
[vi] http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.903.8395&rep=rep1&type=pdf
[vii] https://www.iso.org/ics/35.100/x/
[viii] https://www.mci.gov.sg/portfolios/infocomm-media/initiatives/factsheets/pricing-of-telco-services
[ix] https://www.speedtest.net/reports/singapore/. Speed test data from Q4 2016-Q1 2017 is in, showing that Singapore has some of the fastest internet speeds on the planet. With an average fixed broadband download speed of 180.61 Mbps and an average mobile download speed of 44.37 Mbps, the country ranks second in the world for both.
[x] http://legislation.govt.nz/regulation/public/2019/0275/latest/LMS185107.html
[xi] https://comcom.govt.nz/__data/assets/pdf_file/0025/147841/Cambridge-Economic-Policy-Associates-CEPA-Cost-of-capital-for-regulated-fibre-telecommunication-services-in-New-Zealand.-Asset-beta,-leverage-and-credit-rating-21-May-2019.pdf, page 9.
[xii] https://comcom.govt.nz/__data/assets/pdf_file/0025/147841/Cambridge-Economic-Policy-Associates-CEPA-Cost-of-capital-for-regulated-fibre-telecommunication-services-in-New-Zealand.-Asset-beta,-leverage-and-credit-rating-21-May-2019.pdf, page 9.
[xiii] https://www.itu.int/net/itunews/issues/2011/07/43.aspx
[xiv] https://ufb.org.nz/new-zealand-ultra-fast-broadband-rollout-50-complete/
[xv] http://www.voxy.co.nz/business/5/352994
[xvi] http://japan.kantei.go.jp/it/network/0122full_e.html
[xvii] http://www.ftthcouncilap.org/wp-content/uploads/2019/04/FTTH-APAC-Panorama-Report-2019_Low.pdf
[xviii] https://www.government.se/496173/contentassets/afe9f1cfeaac4e39abcdd3b82d9bee5d/sweden-completely-connected-by-2025-eng.pdf
[xix] https://www.oecd.org/daf/competition/Structural-separation-in-regulated-industries-2016report-en.pdf
[xx] http://www.reykjavikfibrenetwork.is/news/europes-wholesale-only-and-open-access-operators-form-new-alliance-accelerate-rollout-fiber
[xxi] https://tech.cityofnewyork.us/wp-content/uploads/2020/01/NYC_IMP_1.7.20_FINAL-2.pdf