«Delivered to | Northern Communications Information Systems Working Group c/o Government of Yukon Delivered by | Nordicity Date | Foreword The Project ...»
5-year trafc prediction module: This module is based on an incremental trafc pattern, year over year, of 30% for the 5 next years. The percentages can be changed on a per year basis. Based on the backbone network proposed in this study, the trafc generated in 5 years can be transported without any problem. The challenges will be on the satellite platform since it is based on a $/Mbps cost.
For a more detailed look at the optimization, including screen captures from the modeling spreadsheets, please refer to Appendix 2.
The following diagram provides the interactions between the diferent modules within the optimization model. Any changes that occur in any of the modules will be refected automatically on the fnal cost estimation.
5 Industry sources, including Telesat based on 15 Mbps trafc demand.
Figure : Dynamic Optimization Model
The total capital costs of achieving this report’s recommended service standards through upgrades to the current network are provided in the table below. For the NWT, these costs assume that the Mackenzie Valley Fibre project will be completed independently of this study and is therefore not costed in the model.
HOW MUCH WILL IT COST?
ACHIEVING THIS REPORT’S RECOMMENDED SERVICE STANDARDS COULD COST IN THE RANGE
OF $300 MILLION TO $1 BILLION (EXCLUDING LAST MILE COSTS), DEPENDING ON WHICH
PROJECTS PROCEED. THE ECONOMIC AND SOCIAL BENEFITS OF ACHIEVING THOSE STANDARDS
IS PROVIDED IN CHAPTER 4.Table 1 : Costs to Upgrade Current Network Infrastructure by Technology by Territory using baseline assumptions (Mackenzie Valley Fibre exists, and carrying only high priority trafc on redundant links.)
Table 1 shows the costs of achieving this report’s recommend service standards, based on the Dynamic Optimization Model. This is based on the most cost-efective ‘upgrade’ solution to meet the recommended connectivity standards – considering each of the three technology platforms and each of the 75 communities. The cost to connect each community to the transport network using microwave or fbre or satellite is aggregated, by Territory, and the total upgrade costs show the amount that would be required to upgrade the current infrastructure using the same technology in order to achieve the 9 Mbps service standard. The upper estimate and lower estimate are based on a +/- 50% adjustment factor used to account for possible variances between the costs determined for the Alaska project.
These costs will for the purposes of costing and for subsequent chapters. This reference scenario considered in this study will be used to compare against the possible proposed other solutions/scenarios that could involve other future project in the North.
7.1 What do the Communities Get?
A key output of the optimization model is the overall capacity increases experienced by individual communities in the Territories. The numbers shown in Table 1 - below refect upgrading the microwave capacity, fbre-optic capacity or satellite-carrying capacity for each community based on the technology that they currently use for backbone connectivity OR the assumed increase they will have received by virtue of completion of the Mackenzie Valley Fibre project.
When comparing the improvements in capacity per Territory, the reader will note that Nunavut, although displaying improvements, displays lower overall capacity upgrades in many communities. This is a result of the specifc technology platform being considered for upgrading.
When upgrading a microwave link, capacity increases by 1,000 Mbps, when upgrading a fbre link, capacity increases by 10,000 Mbps. However, in the case of satellite (the main technology in Nunavut), capacity increase is tied directly to capital spend – e.g., each Mbps added incurs signifcant CAPEX. This illustrates clearly the enhanced future capabilities presented by technologies other than Satellite.
Due to the overwhelming cost impacts of providing 100% trafc redundancy using mixed technologies - but heavily weighted on satellite, it is worth exploring the possibility of building new fbre-optic links in the three Territories. This is examined in the following section, and presented as a rough order of magnitude examination.
7.3 Possibility of New Fibre Builds?
Another consideration in the Territories is whether or not new fbre build initiatives could shape the future of communications. As previously noted, the reference model has incorporated the assumption that the Mackenzie Valley Fibre build will go forward independently of this study on the basis of the declared intent of the Government of the Northwest Territories. In addition to this project, there are other initiatives and/or plans underway which could see new fbre installations over time. In the course of modeling, these were not factored into the model, as there is no certainty that any project will occur. However, in order to provide additional information and insight into the overall scope of the project, Tables -2 serve to illustrate, at a high level, some of these possible fbre builds. The impacts of the fbre builds are also illustrated by showing the per-Territory resulting costs for capacity upgrades. For many remote sites, the impact would be minimal, but for sites on the main fbre trunk lines, these additional fbre builds could increase the presence of redundancy, and therefore increase the overall reliability and robustness of the network.
The costs in the tables are derived by surveying a number of recent fbre building projects and estimated costs, including the Alaska benchmarking values, the Mackenzie Valley Fibre project, and the Arctic Fibre undersea cable projects. These values are presented merely as guideline pricing, and not meant to represent detailed engineering costs, which are beyond the scope of this report.
Figure : Possible fbre builds in Yukon
In the case of Nunavut, the assumption used to price out the impact of new fbre builds was that the Arctic Fibre primary network project (shown in purple in the fgure below) goes forward and provides service to the communities of Iqaluit, Cape Dorset, Hall Beach, Igloolik, Taloyoak, Gjoa Haven and Cambridge Bay (e.g., assumes the existence of a subsea network from Cambridge Bay to Iqaluit). Using this as a backbone network, additional subsea fbre builds are added in as illustrated on the following page.
Figure : Possible fbre builds in Nunavut (Refecting the Existence of the Arctic Fibre Backbone)
In order to summarize the impacts of building new fbre to the three Territories, the following table illustrates the total CAPEX expenditures for network upgrades as well as new fbre builds to all Territories, and compares the base scenario to the scenario which assumes the existence of the Arctic Fibre build, as well as additional fbre deployments. Although the total costs for the fbre build scenario appear lower than the base scenario, it should be noted that there is no capital associated with Phase I of the Arctic Fibre build, as the corporation announced that no funding would be required to connect the frst group of seven communities in Nunavut (Iqaluit, Cape Dorset, Hall Beach, Igloolik, Cambridge Bay, Gjoa Haven and Taloyoak).5 Table 2 : Comparing to – total costs
The next chapter explores the economic and social impacts that can be expected to fow from achieving the service standards recommended in this chapter.
5 Confrmed in speaking directly to Arctic Fibre. Related news article: http://www.nunatsiaqonline.ca/stories/article/65674arctic_ fbre_ramps_up_cable_project_with_seven-community_nunavut_tour/ Grocery Store in Stewart Crossing, Yukon Chapter 3: Financial Model
1. INTRODUCTIONThis chapter estimates how much the infrastructure proposed in Chapter 2 would cost, and what amount of subsidy may be required to provide an operator with a reasonable business case for deploying and providing afordable service.
Specifcally, this chapter:
Explains the fnancial model developed to cost the infrastructure;
Identifes key assumptions used in the model; and, Explains the results of the model on a pan-territorial and Territory-specifc level.
The costs are modelled on the basis of how much it would cost a third-party incumbent service provider, and how much of a subsidy would be required to induce deployment in terms of covering the costs of that deployment and ensuring a reasonable fnancial return for the incumbent service provider.
The costs are modelled in reference to two scenarios. There are two roll-out scenarios to consider for the fnancial modeling for broadband implementation – the ‘slow’ roll-out scenario consists of implementation of the minimum recommended starting point service standard (9 Mbps download, 1.5 Mbps upload) over fve years whereas the ‘fast’ roll-out scenario consists of implementation of that service standard over three years.
A fow chart of the fnancial model is provided below to provide a brief overview of the most important elements of the fnancial model.
Figure 1 : Flow chart of fnancial model The costing is based on several assumptions that may change over time.
1.1 Methodology The fnancial model shows a cash fow fnancial statement. The fnancing model is thus based on recovery of cost and cash fow and quantifes only the incremental revenues and costs – due to the new investment and build-out of the network.
The fnancial year for network upgrade is postulated to commence September 1st, 2014 (pending approval before March 31st 2014, design work would then be April 1st - August 31st, 2014)5. The actual build would be September 1st – August 31st, 2015.
The capital costs are the same as they appear in Chapter 2 and are diferentiated between fbre, microwave and satellite (and redundant satellite, where the latter has been included in the model scenario).
The scenarios for the Territories are thus based on:
Service ofering: up to 9 Mbps. This corresponds to the recommended minimum starting point service standard.
Comparable type of frm: integrated (landline, cellular, ISP).
Subscriber totals and forecasting totals are the same as they appear in Chapter 2.
In addition, two fnancial modeling scenarios are used for each Territory according to refect diferences in roll-out timing. (Note: Costs and corresponding revenues were adjusted according to the roll-out schedule).
a. Slow roll-out of infrastructure: Based on a 5 year roll-out.
b. Fast roll-out of infrastructure: Based on a 3 year roll-out.
Additional scenarios are employed to refect the inclusion (or exclusion) of the redundant satellite costs and new fbre builds.
Output of Model
The key indicators for the fnancial model include:
Free Cash Flow (‘FCF’) and Earnings Before Interest, Taxes, Depreciation and Amortization (‘EBITDA’).
Net Present Value (‘NPV’).
Estimates of fnancial incentives required to induce an operator or other third-party 5 The September – August fnancial year corresponds with the CRTC fnancial reporting for the telecoms industry and might correspond with the funding cycle of the project in the hypothesis that funding would be sought from upper levels of government in 1st quarter 2014.
commercial entity to fnance network upgrades.
Estimates of the annual level of subsidy for households – general and low-income – in order to bring the average consumer cost of service in line with the national average.
2. FINANCIAL MODEL In this section, the various lines of evidence, relevant terms, and assumptions used in the fnancial model are explained. Assumptions that are common for all 3 Territories are presented frst, followed by additional assumptions - specifc to individual Territory’s particular situation.
2.1 Operating Revenue 2.1.1 Subscriber Segmentation
Residential fxed broadband Existing broadband subscribers The existing broadband subscriber number is based on a calculated average of the number of subscribers currently receiving standard broadband connectivity (1.5 Mbps+). This is calculated using the territorial broadband penetration rate, population and household totals. Once calculated, the model also predicts the trajectory of these subscribers to 2023.
The number of broadband households is then divided into households that can aford the higher connectivity and those that cannot. The calculation of broadband households is based on a household income threshold being somewhat higher than the Canadian household income average - taking into account isolation pay.
The demand for broadband is calculated using CRTC data and is then used to calculate the existing universe of broadband subscribers that are expected to migrate to the ultra-broadband service (e.g., 9 Mbps+) available on account of the upgrades to the transport network. A forecast of ultra-broadband (‘UBB’) subscribers is provided to 2023.
The incremental revenue for these migrating subscribers is based on the UBB premium: that is, the expected average increase in consumer payments for a UBB service in comparison to a standard broadband service. For example, the current average revenue per unit (‘ARPU’) is $60.00 per month. Analysis of CRTC data indicates that Canadian consumers would be willing to pay a 38% premium or approximately $23.00 per month for broadband services.
Thus the UBB ARPU would the $83.00 per month and the incremental revenue for each residential UBB subscriber would be $23.00.
New subscribers The time series for the annual number of new subscribers represents the number of residential users that will become subscribers for the frst time given this new and enhanced connectivity.
This number is contingent on the number of non-subscriber households and the ability to pay.
That is, the estimate of new subscribers is based on a subset of the total number of households that are above the average household income and are willing to spend a portion of their disposable income on new connectivity.