«Delivered to | Northern Communications Information Systems Working Group c/o Government of Yukon Delivered by | Nordicity Date | Foreword The Project ...»
Low-income households The number of low-income household subscribers is calculated using the total households that fall below the average income. The low-income households will only become a subscriber with an adequate subsidy to subsidize their connectivity. In the fnancial model, the subsidy for the low-income households is $30.00 in 2013 dollars (e.g., $31.84 in 2016, after accounting for infation at an average rate of 2% per annum).
With implementation of the subsidy, the consumption behaviour of low income households is assumed to mirror that of general households: that is, the low-income segment follows the same trajectory in terms of the penetration of broadband and UBB services as general households.
ARPU Calculations Where the current ARPU is calculated based on current existing subscribers with current
connectivity, the calculation includes:
Incremental beneft: The ARPU premium that can be levied on existing subscribers after they migrate to ultra-broadband.
The ARPU premium is calculated based on new customers obtaining their service from the higher speed broadband.
The ARPU subsidy is calculated based on the customers that are not able to pay for the higher broadband service but would like to have this service – thus there will be a subsidy ofered to subsidize the price and let those below the afordability threshold buy the higher speed broadband.
The subsidy does not rise with the CPI, post 2016. In other words the subsidy is gradually “phased-out” by not preserving its real-dollar value.
Rollout of ultra-broadband in the smaller communities (e.g., Beaver Creek pop. 104) will not happen without monthly subsidies).
9 Mbps (average) is defned as an essential need for users in the Territories.
Normally, this would lead to a subsidy from the territorial government.
2.1.2 Government The government purchases connectivity from the incumbent and thus is an institutional subscriber. However, the total budget for government connectivity does not increase over time.
The fnancial model assumes that the government will not be a source of incremental revenue.
2.1.3 Business The fnancial model also includes estimates of the incremental revenue associated with UBB services acquired by small and medium-sized enterprises (‘SMEs’). The fnancial model assumes no incremental revenue from enterprise business customers ; the assumption is that this customer segment is already connected to a high-capacity transport network, due to the concentration of these types of customers in urban centres (where fbre backbones are already installed ) or in areas that are likely connected in the proposed Mackenzie Valley Fibre Link project.
Incremental revenue from SMEs is assumed to be equal to 20% of incremental revenue in the fxed residential market. This ratio of 20%is similar to the ratio observed across the whole Canadian Internet access market. Data published by the CRTC indicate that broadband access revenue in the business customer segment was equal to 25% of total revenue earned in the fxed residential market in 2011.
2.2 Operating Expenses
The fnancial model includes estimates of three types of incremental operating expenses:
network operating expenses (commonly referred to as ‘OPEX’), subscriber cost of acquisition (‘COA’), and sales, general and administrative (‘SGA’) expenses. Each of the elements of operating expenses is calculated on the basis of diferent base number.
2.2.1 OPEX – Network operating expenses OPEX includes the costs of power supply, connectivity and maintenance of the network. For microwave and fbre infrastructure, OPEX for networking operating expenses was calculated in
the following manner:
Microwave: 10% of CAPEX;
Fibre (upgrade): 10% of CAPEX;
Fibre (build): 1% of CAPEX; and, Satellite: $50,000 per annum for ground stations in communities linked by satellite A lower rate for fbre (build) was used in comparison to fbre (upgrade), since the former involves signifcant construction costs to bury fbre along long distances, whereas an upgrade would involve running fbre through existing conduits or adding electronics to existing fbre cables.
2.2.2 Susbcriber cost of acquisition (‘COA’) COA includes marketing and connection costs for each new subscriber. It also includes the cost of subsidizing modems of new customers. The model assumes marketing and connection costs of $500 per new subscriber, and a modem subsidy of $100 per subscriber. Therefore, total COA is $600 (in 2013 dollars) for each new subscriber.
The total COA in the SME segment is assumed to equal to 20% of the total annual COA in the fxed residential broadband segment. This model could be further developed by taking into account such factors as churn, although the relative importance for that factor in particular would likely be less, given the current market structure in the three Territories.
2.2.3 Sales, general and administrative The sales, general and administrative cost category (‘SGA’) includes the business overhead and customer care costs associated with an ISP business. To estimate SGA, data published by CRTC was used to calculate the average monthly level of SGA per subscriber in the cable broadband market in Canada. The resulting fgure, $13.99, for 2011 was infated to 2013, by applying a 2.0% infation rate. The resulting rate, $14.55, was then grossed up by 75% to account for the diferent cost structures exhibited by large national communications companies (which comprise the majority of the data published by CRTC) and small communications companies operating in the North.
The resulting rate, $25.47, was then allocated across the standard broadband and UBB services.
The ARPU ratio was used to conduct this cost allocation. That is, the UBB ARPU premium, $23, accounted for 28% of the total UBB ARPU of $83. This ratio of 28% was applied to the gross-up SGA of $25.47 per subscriber per month to arrive at an incremental SGA of $7.06 per subscriber per month (in 2013 dollars).
2.3 EBITDA Earnings before interest, taxes, depreciation and amortization (‘EBITDA’) is a common metric for A comparison of the fnancial data published TELUS and GCI (Alaska) indicated that the latter’s SGA costs were 75% higher than the former’s.
proftability. EBITDA was calculated by subtracting the total operating expenses from the total revenue. The subsequent EBITDA margin is the ratio of total revenue and the EBITDA number.
2.4 Capital Expenditures The fnancial model includes two types of capital expenditures (‘CAPEX’): (i) network upgrade and (ii) access equipment costs.
2.4.1 Network upgrade The network upgrade CAPEX were taken from the optimization model and is the total upgrade dollar amount for all technologies – microwave, fbre costs and satellite costs (including the costs of implementing a satellite-based redundancy network, where included in the scenario). In the slow roll-out scenario, these costs are spread out over a fve year period. In the fast roll-out scenario, these costs are spread out over a three year period.
2.4.2 Access equipment costs The fnancial model also includes estimates of the costs of installing capital equipment to connect the upgraded transport network to the frst mile network. This connection is typically facilitated using digital subscriber line access multiplexer (‘DSLAM’) equipment. The model assumes DSLAM costs of $500 per new subscriber. New subscribers include existing subscribers who migrate to UBB service. The underlying assumption is that this migration will require new DSLAM equipment or other investments to provision the UBB service to the frst mile.
2.5 Free Cash Flow (FCF) FCF represents the cash or fnancial surplus (or defcit) available to suppliers of long-term investment capital, after deducting cash operating expenses (e.g., excluding depreciation and amortization) and capital expenditures from operating revenue. Any surplus in FCF would be available to distribute to suppliers of debt or equity capital; conversely any defcit in FCF must be covered by drawing down or issuing new debt or equity capital.
Figure 1 : Free Cash Flow
Note on the Assumption Regarding Sources of Financing for the Network Upgrades Average incomes of northern residents - with the exception of the Yukon are lower - and cost of living higher in comparison with the rest of Canada making it unlikely that household subscribers would be able to adsorb the full cost of the proposed connectivity upgrades.
In the focus group and survey research, front line managers indicated that territorial governments have limited fnancial resources even for connectivity improvements for delivery of government services. In the run up to the June 2013 CRTC Hearing, evidence fled indicated that NorthwestTel’s modernization plan would provide incremental improvement on current connectivity levels – nowhere near the minimum proposed connectivity standards. This and other research indicated that the fnancial resources of stakeholders in the Territories were likely to be inadequate for the recommended network upgrades and that some outside funding of an initial capital investment would be necessary. This was expressed as the fnancial incentive required to attract the incumbent to invest and operate the network. The implantation plan in Chapter 5 illustrates the action items for the Task Force to undertake for broadband in the Territories. The fnancial requirements outlined in this report will be a topic for consideration by the Task Force to fund broadband implementation.
All of the modelling results in this section are based on the fast rollout. The results for the slow rollout option can be found in Appendix 3. A rolled-up analysis of the modelling results on a pan-Territory basis is provided on Page below.
3.1 Yukon Figure presents the annual FCF that the incumbent operator would experience under each of the four applicable scenarios. The signifcant negative FCF in the early years, 2014 to 2016, are on account of the network upgrade CAPEX spread out over three years. While the magnitude of the negative FCF in the early years depends on the particular upgrade scenario, in later years, all scenarios yield slightly positive FCF.
Figure : Annual FCF, Yukon* (20) (40) $ Millions (60) (80) (100) (120) (140) Source: Nordicity estimates * Fast roll-out scenario with satellite redundancy costs included.
6 Assuming Arctic Fibre backbone is built.
Over the entire forecast period, 2014 to 2023, the total CAPEX for the primary network upgrade in is $21.82 million. An additional CAPEX of $6.13 million is also required to expand customer access to the upgraded network in the base-case and all scenarios.
In this particular base-case scenario, the NPV is positive; therefore, no fnancial incentive is required, because the present value of incremental revenue is expected to completely ofset the upfront CAPEX, and ongoing OPEX and administrative costs of the upgraded network.
Although no fnancial incentive would necessarily be required to attract the incumbent operator to invest and operate the network, household subsidies of $13.16 million, or $1.65 million annually would be required to maintain afordability for Yukon households – both below and above the low-income threshold. These household subsidies include payments to ofset the higher cost of ultra-broadband service as well as additional subsidies for low-income households.
In, the 2023 broadband penetration target is 96.1% for 1.5 Mbps+ service; for UBB service (e.g., 9 Mbps+), the penetration rate target is 76.9%. Also, the 2013 broadband ARPU is estimated to be $60, while the 2023 ARPU is $97.25 (before household subsidies).6 Adding satellite redundancy to the upgraded network raises the primary network CAPEX cost to $61.64 million. The access network cost remains at $6.13 million. In fact, it is unchanged across all the scenarios.
, the higher CAPEX costs result in a negative NPV, which entails that the required fnancial incentive would be $41.88 million. Note, however, that because the incremental revenue from upgraded network in later years yields a positive FCF, the required fnancial incentive is less than the primary network upgrade CAPEX.
the primary network upgrade costs to $390.14 million, and the required fnancial incentive to $340.08 million. The scenario with 100% satellite redundancy is, by far, the costliest to implement in terms of both CAPEX and required fnancial incentive.
, the primary network upgrade CAPEX is $104.44 million and the required fnancial incentive is $67.73 million.
6 The forecast ARPU includes the impact of infation; so it is not reported in terms of real 2013 dollars. The fnancial model assumes an average annual infation rate of 2%. So, the average ARPU for 2023 can be discounted by 22% in order to convert the amount to 2023 dollars. For example, a 2023 ARPU of $97.25 would be equivalent to an ARPU of $79.78 in 2013 dollars.
Table 2 : Yukon Financial Model Summary*
(100) (150) (200) (250) (300) Source: Nordicity estimates Over the entire forecast period, 2014 to 2023, the total CAPEX for the primary network upgrade in is $67.09 million. An additional CAPEX of $6.65 million is also required to expand customer access to the upgraded network in the base-case scenario and all scenarios.
In all scenarios, the NPV is negative; and therefore, a fnancial incentive is required, because the present value of incremental revenue is not expected to completely ofset the upfront CAPEX, and ongoing OPEX and administrative costs of the upgraded network. The required fnancial incentive for the base-case scenario is $73.62 million.
In addition to the fnancial incentive required to attract the incumbent operator to invest and operate the network, household subsidies of $14.93 million, or $1.87 million annually, would also be required to maintain afordability for NWT households. These household subsidies include payments to ofset the higher cost of UBB service as well as additional subsidies for low-income households.