ARCHIVED -  Letter

Ottawa, 25 May 2012

File No.:  8665-C12-200807943

BY EMAIL

Mr. Bill Abbott
Senior Counsel of Regulatory Law
Bell Canada
Floor 19
160 Elgin Street
Ottawa, ON  K2P 2C4

Bill.abbott@bell.ca
Bell.regulatory@bell.ca

Dear Mr. Abbott:

Re:  Regulatory Policy CRTC 2009-430 – Accessibility of Telecommunications and Broadcasting Services

In Broadcasting and Telecom Regulatory Policy CRTC 2009-430, the Commission reviewed its relay services policy framework as part of its proceeding to address Unresolved issues related to the accessibility of telecommunications and broadcasting services to persons with disabilities. The Commission extended the existing message relay services requirement to include a requirement to provide IP Relay service. The Commission also determined that further investigation into the projected use, size of individual ASL or LSQ Video Relay user markets, and cost of providing Video Relay Service (VRS) in Canada is required. In this regard, the Commission considered that information resulting from the Final Report of the Feasibility of Video Relay Service (VRS) for Canada (Bell Feasibility Study) and the Final Report on TELUS’ Video Relay Services Trial (TELUS Trial), as approved in Telecom Decision 2008-1, would be vital to determining whether a follow-up proceeding on VRS is warranted. Please refer to the news release dated May 25, 2012 <http://www.crtc.gc.ca/eng/com100/2012/r120525.htm>.

The Commission has received and reviewed Bell Canada’s commissioned report, completed by Mission Consulting, Final Report of the Feasibility of Video Relay Service (VRS) for Canada originally filed 24 February 2012 (amended 4 April 2012) as part of the Decision 2008-1 process. The Commission has also received and reviewed the Final Report on TELUS’ Video Relay Services Trial filed on 14 March 2012. After reviewing the two reports, the Commission has identified a number of discrepancies between the two reports which require clarification. The Commission therefore requests that Bell provide responses to the questions in the attachment to this letter no later than 26 June 2012.

Please contact Kay Saicheua (kay.saicheua@crtc.gc.ca) at (819) 934-1358 should you have any questions in regard to this letter.

Yours sincerely,

ORIGINAL SIGNED BY /

Nanao Kachi
Acting Director
Social & Consumer Policy

Att.

Interrogatories regarding the Bell (Mission Consulting) Final Report on the Feasibility of VRS for Canada

1) Projected number of ASL and LSQ VRS users

In Phase 9 of the Final Report of the Feasibility of Video Relay Service (VRS) for Canada (Feasibility Study), it was projected that there are approximately 34,000 Canadian Sign language users¹. The Feasibility Study then projected that of those 34,000 Sign language users, 15,345 would become consumers of VRS. The projected number of VRS users was calculated using methodology that was based on estimates “without statistical basis” (phase 9, pages 16-17).

Statistics Canada published the 2006 Participation and Activity Limitation Survey which gathers information about adults and children whose daily activities are limited by a health-related condition or problem². The survey reveals that in 2006, 35,470 Canadian adults self-identified as both having a hearing or communication limitation and using Sign language as a means of communication. An additional 2,620 children with hearing limitations reported using Sign language to communicate. This totals 38,090 Canadian users of Sign language in 2006.

i) Footnote 14 (phase 9, page 13) states: “Using the formula that 1 in 1,000 of those identified as profoundly deaf use Sign as their primary language”. However, the footnote references an estimate found in Table 2: Canadian demographic estimates for the total population in Canada with some form of hearing loss. In addition, the ratio of total Sign language users to the total population identified as profoundly deaf does not match the above-mentioned ratio. Please clarify the assumptions used to derive the estimates in Table 2, phase 9, page 13 and their sources.

ii) Given the availability of Statistics Canada data, explain why the methodology used in the Feasibility Study to forecast Canadian Sign language users was selected as the preferred methodology.

iii) To derive the “Canadian Forecast” of ASL and LSQ VRS users (15,345 users), the Feasibility Study forecasted the number of VRS users in the United States which made use of:

• estimates of the Deaf ASL user population in the U.S which “is without statistical basis” (phase 9, page 17) and;
• the total number of VRS users in the U.S., which was calculated using a distribution method which had “no available statistical data to support the distribution” (phase 9, page 16).

The resulting forecasted number of VRS users in the U.S (155,748) represented 0.05% of the total U.S population. In forecasting the number of potential VRS users in Canada, the U.S ratio of VRS users to total population (0.05%) was then adjusted downwards by 10% to compensate for the effect of aggressive marketing by American VRS providers. The resulting Canadian Forecast of ASL and LSQ VRS users was 0.045% of the total population, or a projected 15,345 potential VRS users in Canada. (phase 9, pages 19-22)

(a) Given the lack of statistical data, explain why the methodology used in the Feasibility Study was chosen and justify how this projection is an accurate representation of the potential number of Canadian VRS users.

(b) Provide statistical data to support the 10% reduction applied, and explain how this percentage accurately quantifies the impact of aggressive marketing by VRS providers on the user base. 

iv) Should the Feasibility Study’s projections for the number of VRS users require revision, submit the revised projections highlighting the assumptions and methodology used.

2) Projected Canadian VRS usage (minutes)

The TELUS Final Report summarizes the results of their 18-month trial³. During the course of the trial, participants used the video relay service to make an average of 21 calls per month per user. The average call lasted 5.16 minutes, revealing that monthly demand was roughly 108 minutes per user.

The Feasibility Study’s forecast of Canadian VRS annual minutes is 6,820,853 for a fully deployed service. This represents an annual demand of 444.5 minutes per user under the Canadian Forecast, which equates to a monthly demand of 37 minutes per user. This forecast, in the Feasibility Study, is referred to as the “Canadian Forecast” and is used to represent “the most probable forecast for Canada for a fully deployed 24 x 7 VRS at an eventual maximum usage rate” (Phase 9, page 22).

i) Reconcile the difference between the Study’s projections and the results of the TELUS trial. Where the Feasibility Study’s projections require modification, submit any assumptions and explain your methodology and rationale.

ii) The forecast of Canadian VRS annual minutes was derived using a multi-step approach. First, an initial forecast of 10,826,750 Canadian VRS annual minutes was calculated using the estimated average annual VRS minutes per user in the United States (635 minutes). To derive this estimate of U.S. average annual VRS minutes per user, a number of calculations were performed, including calculations which made use of:

• estimates of the Deaf ASL user population in the U.S which “is without statistical basis” (phase 9, page 17) and;
• the total number of VRS users in the U.S., which was calculated using a distribution method which had “no available statistical data to support the distribution” (phase 9, page 16).

Then, using the initial forecast of 10,826,750 Canadian VRS annual minutes, the following downward adjustment was applied to account for various differences between the Canadian and U.S. markets (phase 9, pages 20-22) to derive the “Canadian Forecast” of annual VRS minutes:

• 10% to account for fraud, and;
• 30% to account for the timing of VRS availability in Canada versus the United States

(a)Given the lack of statistical data, explain why the methodology used in the Feasibility Study was chosen and justify how this projection is an accurate representation of the potential Canadian VRS usage (annual minutes).

(b)Provide statistical data to support the percentage reductions used, and explain how these percentage reductions accurately quantify the impact of both fraud and the timing of VRS availability.

VRS operators and effect on community interpreting

3)The Feasibility Study states that U.S. VRS policies will affect Canadian VRS as U.S. regulations “presently allow multiple VRS vendors to be reimbursed at relatively high per minute rates for allowable VRS traffic. Because the rates are currently high, it is cost effective for the dominant U.S. VRS vendor, Sorenson Communication, to establish multiple VRS call centers in Canada to serve U.S. VRS Consumers” (phase 11, page 10).  “There are presently 599 ASL Canadian Interpreters (members of AVLIC⁴); however an estimated one-third of these are already employed by Sorenson providing VRS for U.S. consumers. The result is a present significant shortage of ASL interpreters. Requests for community interpreting are not being met” (phase 9, page 25). The Feasibility Study also mentions that the U.S demand for Canadian ASL interpreters has contributed to “significantly reducing the availability of ASL interpreters for work within a VRS that will service Canadian consumers” (phase 11, page 10).

i) Provide your understanding of why U.S. VRS providers are hiring Canadian interpreters. Indicate whether the primary trigger is limited to cost-savings, or whether there exists an interpreter shortage in the United States.

ii) Quantify the wage differential between Canadian and American ASL interpreters, if one exists. Include the average wage rate for both, as well as the loaded labour cost.

4) The Feasibility Study states that “an outsourced ASL VRS would be expected to have a higher adoption rate by Canadian ASL users since the availability of the service would not be constrained by a lack of interpreters” (Phase 10, page 4), and later presents an option whereby “ASL is outsourced to the United States where there are plenty of ASL interpreters” (Phase 11, page 21). One of the advantages to this option was: “there may be minimal VRS competition for Canadian ASL interpreters, and therefore less impact to ASL community interpreting within Canada” (phase 11, page 21).

i) Given that U.S. VRS policies are expected to affect Canadian VRS (phase 11, page 10) and VRS providers are currently hiring Canadian interpreters, and thus already exerting pressure on Canadian community interpreting, explain why there may be “minimal VRS competition for Canadian ASL interpreters” in the proposed outsourced option.

ii) Phase 9, page 17 of the Feasibility Study highlights that there are 15,500 total identified Sign language interpreters in the U.S. Provide and justify your understanding of whether a Canadian VRS provider could set up a call center in the United States to serve Canadian consumers, and whether this would be a feasible option for a Canadian VRS, having regard to, among other things, privacy and cost considerations.

5) Estimated Cost of a Canadian VRS

The cost analysis contained in the Feasibility Study uses the current U.S. Federal Communications Commission (FCC) VRS reimbursement rates, which are then adjusted to represent applicable Canadian per-minute reimbursement rates. These Canadian per-minute rates are then applied to the “Canadian Forecast” of usage to generate a “Canadian Forecast” of VRS provider costs.

We note that there are a number of existing VRS models internationally, revealing numerous service provision models and service provider-compensation models.  We also note that providers may be reimbursed for service provision (e.g. an unlimited 24 /7 service-provision model similar to that in the United States) in different ways, such as  compensation per-minute of relayed called, compensation per hour of operator time, or funding a set number of operator positions to name a few.

The Feasibility Study identifies one disadvantage of a per conversation-minute reimbursement option to be:

“without carefully defined requirements, accountability and monitoring, payment based on minutes can lead to provider abuse and fraud, whereby the VRS provider makes and bills for disallowed calls, pays consumers or others to make VRS calls, or pays consumers to make calls to toll-free numbers known to have unusually long hold times” (phase 11, page 28).

These aforementioned sources of fraud have proven to be a significant problem in the United States, as recognized in the Feasibility Study, where VRS providers are reimbursed on a per minute basis. We also note that the FCC is currently undergoing a VRS Reform where the way VRS providers are compensated could be changed from per-minute reimbursement to per-user reimbursement.

i) Provide the status of the FCC VRS Reform.

ii) Indicate whether a per conversation-minute compensation model is a prerequisite to implementing a 24/7 service provision model.

iii) Indicate why, given the aforementioned disadvantages of the per conversation-minute reimbursement option, the Feasibility Study has chosen to forecast the cost of a Canadian VRS using this reimbursement option over a different model such as reimbursement per video interpreter hour which has the advantage of “Eliminat[ing] almost all potential vendor fraud by removing incentives for inappropriately increasing the number of minutes, number of calls, or number of users” (phase 11, page 31).

iv) The Canadian Forecast of VRS provider cost was based on the following elements:

• The “Canadian Forecast” of usage, previously discussed in question 2.
• The Canadian per-minute reimbursement rate, which was based on the current FCC VRS per-minute reimbursement rate.
• The assumption of a 25% operator efficiency rate, which implies that for each hour worked the operator would relay 15 minutes of conversation.

(a) Justify why a 25% operator efficiency rate was chosen. Indicate the advantages and disadvantages of implementing this operator efficiency rate for a Canadian VRS.

(b)We note that an increase in the operator efficiency rate would significantly decrease costs, decrease the number of operators required for VRS, hence minimizing the impact on community interpreting. Indicate what measures, if any, aimed to increase operator efficiency rates should be considered. Justify your response.

[1] Footnote number 14 states the assumption that 1 in 1000 of those identified as profoundly deaf use Sign as their primary language (phase 9, page 13).

[2] Refer to Statistics Canada, PALS 2006: A Profile of Assistive Technology for People with Disabilities.(Catalogue No. 89-628-x, no. 005, ISBN 978-0-662-48688-6)  pages 18 and 26.

[3] TELUS Communication Company. Final Report.  <http://www.crtc.gc.ca/public/partvii/2009/8678/c12_200905557/1686878.zip>

[4] Association of Visual Language Interpreters of Canada (AVLIC)

Date modified:

2012-05-25