Traverse County, Minnesota

Transcription

1 Traverse County, Minnesota Broadband Feasibility Study July 19, 2016 Finley Engineering CCG Consulting

2 Table of Contents Page Executive Summary...3 Findings...4 Recommended Next Steps...8 Section I Background Research...9 A. Incumbent Providers...9 B. Current Broadband and Other Prices...13 C. The Connect America Fund...15 D. The Consequences of Poor Broadband...17 Section II Preliminary Engineering...21 A. Network Design...21 B. Network Cost Estimates...33 C. Competing Technology...35 Section III Financial Business Plan Analysis...43 A. Business Plan Key Assumptions...44 B. Business Plan Results...54 C. Sensitivity Analysis...60 D. Financing Options...63 Page 2

3 EXECUTIVE SUMMARY Finley Engineering and CCG Consulting submit this report of our findings and recommendations for the feasibility of creating a broadband network in Traverse County. The county has a unique situation today where telephone cooperatives in the area have built rural fiber, but the county seat of Wheaton and about 600 other homes do not have fiber. The county is a textbook example of broadband haves and have-nots those living in the cooperative service areas have fiber and everybody else does not. We looked at three different business plan options for getting broadband to everyone: building fiber everywhere, building fixed-wireless broadband in the rural areas, or doing a mix of the two technologies. Finley Engineering developed estimates of the cost of deploying each of these options and CCG used these costs in financial business plans to see if there is an economically viable model for providing broadband in the rural areas. It seems to us that the only likely long-term broadband solution for the county is if one or more of the existing cooperatives brings broadband to the areas that don t have fiber today. This means that we then concentrated on looking at financial solutions that assume that the coops bring broadband. It does not look to be economically feasible to immediately build fiber to the areas that don t have it. There are not enough potential customer revenues to cover the cost of constructing and financing a network. The wireless-everywhere option looks easier to justify. This scenario requires a smaller investment and can be financially self-sustaining. However, the wireless option does not generate enough excess cash to fund future fiber projects. The hybrid model looks to be the most promising. This scenario brings fiber to Wheaton and to homes that live close to a proposed backbone fiber, plus it builds wireless elsewhere. This scenario has a low breakeven penetration rate of 44% and it generates significant cash at expected penetration rates that could eventually fund fiber everywhere. The long-term goal should be to build fiber everywhere and the hybrid scenario looks to be the only solution that can get funded today and that can meet this long-term goal. The first priority for the county upon receipt of this report is to get the process started to find a partner to file for a DEED grant. That grant process is already underway and the clock is ticking on submitting a grant request this year. Only the entity that will ultimately own the network and serve customers can make the grant request. This means that one or more of the cooperatives should be the ones to seek grant funding. Page 3

4 FINDINGS Following are the key findings of our investigation: THE PROBLEM Lack of Rural Broadband: As the county already knows, there is a glaring lack of broadband in the rural parts of the county today. The county has a unique broadband situation that we have never seen. Today you have fiber broadband provided in parts of the county by four telephone cooperatives. This leaves the county seat Wheaton and less than 600 other rural homes without fiber. We generally find a county seat with good broadband and the rural areas without. However, there are a significant number of rural customers in the county that have broadband making for a stark contrast to those that don t have fiber. The Small Study Area: The county has the smallest population and is one of the least densely populated in the state. One of the challenges in finding a broadband solution is the relatively small number of homes and businesses without fiber. The fiber business is an economy of scale business, meaning that the greater the number of customers the easier it is to be profitable. That makes for a challenge in Traverse County. Connect America Fund: The rural broadband picture is muddied because both Frontier Communications and CenturyLink accepted funding from the FCC to improve rural DSL in the county. This funding comes from the Connect America Fund (CAF II), which is part of the FCC s Universal Service Fund. The telcos have six years to improve rural broadband and there is no way to know during the six years when they will upgrade the rural broadband in the county. The telcos are only required to meet a minimum DSL speed standard of 10 Mbps download and 1 Mbps upload. This is much faster than most rural DSL today, which can be as slow as dial-up. However, this is not broadband as defined by the FCC or the State of Minnesota and will be obsolete within a very short time. Having these upgrades on the horizon complicates finding a solution since at least some rural households in the county might find this DSL to be an adequate broadband solution and elect not to take the faster fiber option if offered. BASIC FACTS ABOUT THE COUNTY Potential Customers: We looked at several different sources of data for counting homes and businesses in the county. We determined that the maps from the Minnesota Department of Transportation were probably the most accurate. The passings (potential homes or businesses) used for each of the scenarios are as follows: Fiber Everywhere (Wheaton and rural) 1,408 Wireless for rural areas 593 Road Miles: To serve fiber to the parts of the county that don t have it today would require building fiber along about 390 miles of streets and roads. It is probable that a final fiber design might find ways around building all of the miles, so we are probably being conservatively high in our estimated costs. Page 4

5 THE POSSIBLE SOLUTIONS The study looked at three possible engineering solutions to bring broadband to the parts of the county that don t have fiber today. Build Fiber Everywhere: We first looked at building fiber everywhere within the study area. That means building fiber in Wheaton and in about 600 rural homes and businesses in the rest of the county. Build Wireless Everywhere: Another option explored is to build fixed wireless for just the rural customers. This solution does not bring broadband to Wheaton. The key issue for providing good wireless broadband is to feed the wireless towers with fiber, so the solution includes fiber backbone built to feed wireless towers. This fiber could also be the basis in the future for extending fiber to residents. Hybrid - Build Some Fiber with Wireless for the Rest: The final option studied was a hybrid of the first two scenarios. This brings fiber to Wheaton as well as to customers that live close to the backbone fibers. It supplies wireless to everybody else in the rural study area. ENGINEERING FINDINGS Backbone Fiber Network: Finley Engineering recommends that all network solutions include the building of a fiber backbone. These would be primary fiber routes used to either feed fiber huts or wireless towers in the various solutions. The recommended backbone fiber in the analysis would be 36.2 miles long and built with 96 fibers to accommodate future growth. Aerial vs Buried Fiber: We designed the whole network using buried fiber. The soil in the county allows for relatively easy burying of fiber and so the cost to bury fiber in the rural parts of the county would not be any higher than to place the fiber onto existing poles. A buried network will last longer and have fewer maintenance issues. While buried fiber is generally more expensive than aerial fiber, make-ready work on poles can be very costly. Total Asset Costs: Following are the assets required to launch each of the three different business plan options. These assets assume the business having a 70% penetration rate (described more below). Fewer assets are required if there are fewer customers. Fiber Rural Everywhere Wireless Hybrid Fiber & Drops $11,289,493 $1,203,135 $3,602,942 Network Electronics $ 570,939 $ 376,450 $ 612,718 Huts/Towers $ 197,490 $ 903,079 $ 912,208 Customer Electronics $ 282,580 $ 248,000 $ 353,553 Operational Assets $ 333,711 $ 146,574 $ 330,663 Total $12,674,213 $2,877,238 $5,811,663 Page 5

6 BUSINESS PLAN RESULTS The goal of the financial business plan analysis was find solutions that were financially sound (would generate enough cash to be solvent) and that might eventually bring fiber to everybody. Penetration Rate: The penetration rate is the percentage of potential customers in the study area that buy service. We began the study process looking at an arbitrary penetration rate of 50% for Wheaton and 60% for the rural areas in the county. Based upon our work in other rural markets, these estimates are conservatively low. We then determined the needed penetration rate to make each option financially sustainable, and called this the breakeven penetration rate. Product/Rate Assumptions Telephone Basic Local Line $23.50 Line with Unlimited LD $33.50 Fiber Residential Broadband - Base product $ 55 - Faster Product $105 - Gigabit $150 Wireless Residential Broadband - 20 Mbps $ Mbps $ 99 Cable TV At Market Rates Business Plan Results There are detailed financial summaries of the various business plans starting on page 54 of the report. Following is a high-level summary of the three scenarios studied: Fiber Everywhere: We could not find a financial scenario where building fiber to the study area was financially self-sustaining with a reasonable customer penetration rate. The scenario required total financing (debt and equity between $13.8 M and $14.7 M) and the revenues generated by the potential customers were not great enough to cover the debt payments. The breakeven penetration rate (where a business plan would barely succeed) requires 67% customer penetration rate and 40% equity. That could be possible, but that is both a high needed take rate in Wheaton and also a larger amount of equity than most investors are willing to put into a project for this small number of passings. Rural Wireless: The rural wireless scenario can break even with a 71% customer penetration rate and 20% equity infusion. While that many customers is probably achievable, this scenario does not generate any excess cash for eventual investment in fiber. The business plan does better with a 40% equity infusion and then has a breakeven penetration rate of only 58%. The scenario only generates $3.5 M in excess cash over 25 years at a 70% market penetration. This scenario is feasible, but it would be a standalone business and would not be able to fund future fiber projects. Page 6

7 Hybrid Fiber and Wireless: This scenario builds fiber to Wheaton and to customers living along the backbone fibers and serves everyone else with wireless. This scenario does the best financially. It has a very modest breakeven penetration rate of only 44%. With a 50% penetration in Wheaton and 60% in the rural areas, the business would generate $6.5 M in cash over 25 years that could fund rural fiber. With higher penetration rates it could earn more cash, up to perhaps $10 M. This scenario has the added advantage of being one that can probably attract a DEED grant. Conclusions Drawn: It does not look like there is a sustainable business plan for fiber everywhere today. There are not enough customers in the study area to justify the cost of the network. We would contrast this business plan to one we did recently for another county that had similar network costs but with three times more customers. The low number of customers in the potential service area is the main issue to overcome in any business plan. The rural wireless scenario looks to be quickly profitable and can cover its costs, but because it is only covering around 600 homes, it does not generate enough cash for reinvestment in fiber. This would bring a rural broadband solution, but within a decade the wireless speeds will be inadequate and obsolete. The hybrid fiber/wireless business plan looks to be the best option, by far. It has a relatively low breakeven penetration rate of 44% and it generates sufficient cash that can eventually be used to bring fiber to the rural areas of the county. This scenario has the added benefit of being a plan that could attract a DEED grant (which might not be possible for the wireless-only option). Breakeven Analysis: Again, the breakeven broadband penetration rate for each scenario is as follows, meaning the percentage of customers needed for the plan to be cash self-sufficient: Fiber Everywhere, 20% Equity None Fiber Everywhere, 40% Equity 67% Rural Wireless, 20% Equity 71% Rural Wireless, 40% Equity 58% Hybrid, 20% Equity 44% OTHER ISSUES STUDIED Sources of Financing: We looked at the many different financing tools that are available to fund and construct one of the business plans. Of most immediate importance is scrambling to see if there is the possibility for winning a DEED grant this year. Comparing Technologies: The report looks at the pros and cons of fiber, wireless, and the other technologies in use in the county today. The Consequences of Poor Broadband: This answers the question why build broadband? Page 7

8 RECOMMENDED NEXT STEPS If you decide to move forward, the following tasks are the most likely next tasks to tackle. In your case, since there is the opportunity for some substantial state grants later this year, you need to make the next steps happen rather quickly if the county is to benefit from those grants. 1. Find a partner. The very first step is to look for an operating partner. In order to file for one of the DEED grants this year it s likely that one or more of the four telephone cooperatives would have to quickly decide on asking for a grant. That means that you must get this report to them quickly and take whatever steps needed to help them file for a grant. We know that you did not expend all of the funds you had allocated for this study process and perhaps the county could help to pay for a grant filing with one of the coops. 2. Consider a survey. Our experience today is that there will be sufficient demand in the rural areas for broadband and that you almost surely would get at least as many rural broadband customers as predicted by our financial analysis. That is not such a sure thing in Wheaton; you might want to do a survey to find out the level of demand. If you do a survey, it s essential to do it in such a way as to get statistically valid responses that you or one of the cooperatives can believe. CCG can help you make sure that a survey is done the right way. 3. What if Nobody Applies for a Grant this Year? It is already late into 2016 and it s possible that none of the coops around you would be ready this year to file for a DEED grant to serve the rural parts of the county. If that doesn t happen, you should have a plan in place to be ready for any grant opportunities for next year. That would include getting a firm commitment from one of more of the coops to consider the area, doing a survey to make sure the demand exists and to undertake public education so that the public is behind the efforts to get broadband. 4. Consider the possibility of some county funding. If a grant is not filed this year, the county should consider the possibility of pledging some financial support towards a new broadband venture. As an example, the new cooperative in Sibley County got a loan of 25% of the cost of the project from the cities and townships that are getting broadband. It was that pledge of financial support that made it possible for the cooperative to raise the rest of the money. We re not suggesting that the county should be prepared to put in 25% of the cost of the project, or that you should make this a 20-year loan like was done in Sibley County. However, some smaller amount of contribution might make the difference in finding a broadband solution. Page 8

9 I. BACKGROUND RESEARCH In this section of the report, CCG will look at the incumbent providers in the county, at the products and prices of existing service providers in the market, and at the impact of the Connect America Fund. A. INCUMBENT PROVIDERS The county has a mixed telecom and broadband situation. There are a number of different incumbent cable, telephone, and data providers today in parts of the county. First, there are four independent telephone companies that serve some portion of the county, and which have already built fiber to their customers or will soon finish fiber construction. The customers of these companies can buy the triple play over fiber. Park Region Mutual Telephone Company, associated with Valley Telephone, operates in the Browns Valley area. They currently operate a copper network and provide fiber to anchor institutions with speeds up to 60 Mbps. The company has built fiber along the lake and around Highway 27 and plans to complete the construction of fiber to every rural customer during this summer. But some existing customers will remain on fiber-to-thecurb technology for the foreseeable future. Federated Telephone Cooperative, part of Acira, operates in the Chokio area. They have built fiber in their service area and serve 17 customers using wireless technology with speeds up to 50 Mbps in the Graceville area. Runestone Telecom Association operates in the Norcross and Tintah areas. The company has built fiber within its telephone service area. They have also built fiber to several anchor institutions in Wheaton. The company owns a cellular tower in Dumont. Red River Communications operates around the Fairmount area and has built fiber to all of its customers. Next, both Mediacom and Frontier Communications serve the city of Wheaton. The broadband speeds of the two incumbent providers are lower than the FCC definition of broadband at 25 Mbps. Finally, the remaining rural parts of the county are without real broadband. Frontier Communications mostly serve these areas for telephone service. There is a small number of households served by CenturyLink. Customers in these areas have very limited and inadequate broadband options. Frontier Communications Frontier Communications is a large telephone company with corporate headquarters in Stamford, Connecticut. The company serves Wheaton and most of the rural parts of the county not covered by the independent telephone companies. Frontier is now the fifth largest telephone company in the US after the completion of a recent purchase of Verizon customers. The company changed Page 9

10 their name from Citizens Communications Company in Frontier Communications has grown through acquisitions and continues to buy customers. For instance, in 2015 they agreed to buy 2.2 million customers from Verizon in Florida, Texas, and California. The company spent $8.5 billion to buy a huge pile of customers from Verizon in 2009 and in 2013 bought the Connecticut operations of Verizon. Frontier is an incumbent telephone provider and is considered a provider of last resort, meaning they must try to reasonably provide telephone service to somebody within their defined service area. At the end of the first quarter of 2016, Frontier had 3.4 million total customers that included 2.5 million broadband customers. For the first quarter of 2016, the company had revenues of $1.36 billion. Much of Frontier s footprint nationwide is rural like in Traverse County. Frontier is working to maintain and offer services over aging copper cables in the county and elsewhere. About a third of their customers, in California and Texas, are on fiber obtained from the Verizon acquisition. Frontier provides service in rural areas using DSL, and like in Traverse County, much of the DSL is of older types that can offer maximum speeds of between 6 Mbps and 12 Mbps download for customers close to a Frontier central office, but much slower in rural areas. Frontier accepted money from the FCC s Connect America Fund to enhance DSL speeds in the county as well as in other parts of Minnesota. This will be discussed in detail elsewhere in this report. Mediacom Mediacom is large cable company with corporate headquarters in New Your City. Medicom is the incumbent cable provider in Wheaton and also serves a number of other small towns throughout Minnesota. They are an interesting company that serves some large markets like parts of the New York City metropolitan area also serves numerous small rural markets. The company reported earnings for the first quarter of 2016 of $445 million. The company has 1.33 million total customers and 1.1 million broadband customers. The company operates older cable systems in Wheaton and other rural markets and it has been reported to us that the download speeds in Wheaton are in the range of 12 Mbps and slower. CenturyLink CenturyLink is the third largest telephone company with headquarters in Monroe, Louisiana. Several years ago the company purchased Qwest, which was formerly Mountain Bell and US West, and was part of the Bell Telephone system. The company only serves a few customers in the rural parts of Traverse County, so we are not providing much analysis of them. Dish Network Page 10

11 Dish Network is a large satellite provider and has customers in Traverse County. The company has around 14 million customers nationwide and annual revenues of over $14 billion. The company has average customer revenues of over $80 per month. Dish Network can be bought as a standalone service and is also available as a bundle for Frontier customers. DirecTV DirecTV is one of the largest cable providers in the US with more than 20 million customers. DirecTV merged with AT&T in In Traverse County, DirecTV is available as a standalone service and is also available as part of a service bundle with CenturyLink. Satellite Data There are a number of satellite providers available in the county. In each case, the availability depends upon the ability to have a clear line of sight from a satellite dish to the satellites. The top four providers in the country are Exede (which also markets under the name of Wildblue), HughesNet, DishNet, and StarBand. In general, there are several issues with using satellite broadband. First is latency, which means delay in the signal. When an Internet connection must travel to and from a satellite, there is a noticeable delay; that delay makes it hard or impossible to do real-time transactions on the web. Any website or service that requires you to maintain a constant connection will perform poorly, if at all, with a satellite connection. The second biggest issue is the small data caps. These limit the amount of data a customer can download in a given month. All of the services require contracts of up to 2 years. Finally, the service can be expensive. Here is a short summary of the four providers: Exede (Wildblue): Exede uses the newest satellite and uses technology that has meant a significant increase in download speeds. Exede touts speeds up to 17 Mbps download although customer reviews say the average speed is more like 12 Mbps. Still, that makes it the fastest satellite service. They also tout an upload speed of almost 5 Mbps. Monthly plans range from $49.99 to $ per month and vary by the size of the monthly data cap. There is also a $9.99 monthly fee for the modem as well as a $ installation fee. The basic package comes with a monthly allowance of 10 gigabits of total download (same as the largest cellular plans). The premium service has a cap of 25 gigabits. This puts the price per gigabit at $5.50, about half the price of cellular data. Exede does allow unlimited download at night. HughesNet: HughesNet is the oldest satellite provider. They have recently upgraded their satellites and now offer speeds advertised as 8 Mbps download and 0.4 Mbps upload. Their prices range from $49.99 to $ The smallest package has a 10 gigabit download limit per month and the largest one is 20 gigabits. When including the $9.99 cost for the modem, the premium package equates to $7 per downloaded gigabit. Page 11

12 DishNet: DishNet is associated with Dish networks and can be bundled with their cable product. DishNet prices range from $49.99 to $ They also charge $10 monthly for the modem. They have download speeds of 7 Mbps and upload at 0.8 Mbps. The monthly caps range from 10 gigabits per month on the smallest plan to 50 gigabits on the larger plan. For the largest plan, this works out to $1.80 per downloaded gigabit, making them the most affordable satellite provider. StarBand: StarBand is a legacy satellite provider that works on older satellites. Their prices range from $59.99 to $ with a $14.99 monthly charge for the modem. Their data caps range from 1 gigabit for the smallest plan up to 5 gigabits on the largest plan. That works out to a cost of $27 per downloaded gigabit for the largest plan, making them probably the most expensive broadband per gigabit in the country. Cellular Data There are four primary cellular companies in the country AT&T, Verizon, T-Mobile, and Sprint. Only Verizon and AT&T have wide coverage in rural counties like Traverse, although there are exceptions. We expect that some households in the county use their cellphone data plans for household broadband. There are several problems with this. First, the speeds you can get vary by distance from the cellphone tower. Speeds for cellular data generally are not fast. There are two different cellular data standards in use: 3G and 4G. 3G data speeds are capped by the technology at 3.1 Mbps download and 0.5 Mbps upload. 4G can operate at about 12 Mbps download and the upload varies by service provider. There are slightly faster 4G networks which have speeds up to about 25 Mbps download, which you might think of as 4.5G, but those are available today only in urban areas. For both of these standards, actual speeds in the field will vary by distance from the tower as well as by how busy a tower is, meaning actual speeds in rural areas tend to be fairly slow for most customers. The biggest problem in using cellular data for home broadband use is that most of the plans have very small data caps. For example, with AT&T you can buy 10 gigabits in monthly download for $79.99 and then pay $10 per additional gigabit downloaded. While cellular data avoids the latency issue of satellite data, it is more expensive per downloaded gigabit than satellite data and for most customers will be slower. What Kind of Competition Should You Expect? Anybody implementing a broadband business plan should consider the kind of competition they might see in the marketplace. In Traverse County there are two distinct parts of the market. It has been our experience that there would be very little reaction from the incumbent provider, Frontier, in the rural areas. With that said, Frontier is in the process of using federal funding to improve the rural DSL speeds. Even if the county were to get a better technological solution than Frontier s, the faster DSL would probably attract some customers if it is priced affordably. Page 12

13 Frontier and CenturyLink have up to six years to upgrade the rural DSL and it s possible that if you find a solution quickly enough they wouldn t do their upgrades. They have nationwide goals to upgrade 40% of their rural customers by the end of 2017 and 60% by the end of There is no way to predict when Traverse County might see the DSL upgrades. You would get more pushback in Wheaton. Yet it s unlikely that Frontier or Mediacom will commit the capital dollars needed to provide significantly faster service in Wheaton. Somebody building fiber in Wheaton would be expected to get a considerable amount of customers. However, price is always an issue and if the incumbents maintain some low-priced products, it s expected they would keep some share of the market. B. CURRENT BROADBAND AND OTHER PRICES This section of the report examines the broadband prices available to customers today in the county. It used to be easy to analyze the prices of services. Just a few years ago you could go to the web and find the prices charged by any telco or cable provider, and except for the rare special, most customers in a given town paid about the same thing for service. This is no longer true. Most telco providers have removed their standard prices from the web and so there is no baseline cost you can compare. Further, companies have developed strategies to charge different rates to different customers. We know from experience that prices will vary widely by customer. Over the years, customers have purchased various specials or other promotional pricing and might be charged differently than their neighbors. It seems almost counterintuitive, but the customers paying the most from most incumbents are those that have been with them the longest. This means that there is no longer anything that can be considered as a standard price in the market. Nevertheless, if you want to compete against these companies, you must understand that there will be a range of prices. Frontier Communications Prices Frontier Communications is the incumbent telephone provider that serves Wheaton and rural parts of the county. Frontier s rates are no longer tariffed, meaning that they can offer special prices or put products into bundles. Frontier offers cable TV through bundles with Dish Network. Telephone Rates Frontier offers a base price of a basic residential telephone line for $ They also have a line with features and unlimited long distance for $ Both of those lines also have an extra charge of $6.50 for a Subscriber Line Charge and up to $1 for an Access Recovery Charge (ARC). Frontier DSL Page 13

14 Frontier only advertises residential DSL nationwide at speeds of 6 Mbps download and 768 kbps upload. Residential DSL pricing is listed on the Frontier website as follows: Add DSL to an existing phone line for $19.99 per month. Standalone DSL with no phone line is $34.99 per month Bundled DSL with a phone line with voice mail, caller ID, and call waiting is $ Mediacom DSL customers can also add a 100-channel line-up including local channels from Dish Network for $ Again, there are many customers paying prices that are different from these. There are customers who might be paying lower rates due to past specials and customers paying more than the current specials. The bad news for rural customers is that the DSL costs the same everywhere, but in some places near the edges of the DSL coverage area customers might be getting speeds that are not much faster than dial-up. Mediacom is the incumbent cable TV provider in Wheaton. They offer the triple play products either standalone or in bundles. Following are their most recent prices for Telephone Rates Mediacom offers a phone line with unlimited long distance calling and 17 features. Standalone Phone $49.95 Bundled with one other product $39.95 Bundled with TV and broadband $29.95 Voic $ 4.95 Residential Broadband Launch Up to 3/512k Mbps $29.95 Monthly 150 GB data cap Prime Up to 15/1 Mbps $49.95 Monthly 250 GB data cap Prime Plus Up to 50/5 Mbps $59.95 Monthly 350 GB data cap Ultra Internet Up to 100/10 Mbps $79.95 Monthly 999 GB data cap Ultra Plus Up to 150/20 Mbps $99.99 Monthly 2 TB data cap Ultra 400 Up to 400/20 Mbps $ Monthly 4 TB data cap All broadband products also require the lease of a cable modem for $7.50/month. We don t believe there is any service in Wheaton faster than the Prime product. Page 14

15 Cable TV Basic $29.95 Family TV $72.95 Prime TV $88.95 Local Surcharge Varies by market, up to $8. Regional Sports Surcharge Up to $3. Bundles Cellular Data The company has very large bundles such as the following: Xtream Silver Prime Plus Internet, DVR, Family TV, Phone, Voic $ Xtream Gold Prime Plus Internet, DVR, Movies, Family TV, Phone, Voic $ Cellular data is some of the most expensive broadband in the US. Most data plans charge about $10 per downloaded gigabit of data. To put that into perspective, about 25% of US households now routinely download over 100 gigabits of data per month, mostly video. Comcast says that 10% of their customers download over 300 gigabits per month. If billed at cellular data prices, 100 gigabits would cost over $1,000 per month (and there have been reports from people living in rural areas who have gotten gigantic bills from the cellular providers). Satellite Data We have also talked about this elsewhere in the report. Satellite data is very expensive, but not quite as costly as cellular data. The best broadband prices for downloading 1 gigabit of data from the four major satellite providers are: Exede at $5.50 per gigabit, HughesNet at $7.00, DishNet at $1.80, and StarBand at an incredible $27. All of them have tiny monthly caps on how much a household can download. C. THE CONNECT AMERICA FUND In the fall of 2015, both Frontier Communications and CenturyLink accepted funding from the FCC to improve rural DSL in the county. This funding comes from the Connect America Fund, which is part of the FCC s Universal Service Fund. The Universal Service Fund today is funded primarily from surcharges on telephony revenues. Originally, the USF was funded by surcharges on landline telephones and special access circuits only, but eventually a surcharge was also placed on cellphones. Following are the uses of the Universal Service Fund for the last two years (2015 not yet released): Page 15

16 High Cost Fund $4.17 B $3.75 B Lifeline $1.80 B $1.60 B Rural Healthcare $0.16 B $0.19 B Schools and Libraries $2.20 B $2.27 B Total $8.33 B $7.82 B The Connect America Fund is a component of the High Cost Fund. The FCC has set aside $1.7 billion per year for the six years starting with 2016 to build or upgrade rural broadband. These funds were made mostly available to Census blocks that have little or no broadband today. The funding was available to the largest telcos automatically. Both CenturyLink and Frontier elected to take this funding for Traverse County. The FCC identified 27 customers in the CenturyLink footprint that don t have broadband and awarded them $33,807 per year for six years. That s a total of $200,000 which equals over $7,500 per customer. The FCC identified 325 customers in the Frontier footprint and awarded them $259,921 per year. That s a total of $1.5 million and works out to $4,800 per customer. The amount of the awards are based upon nationwide cost models that have been developed to estimate the cost of upgrading rural areas to broadband. Now that these two companies have accepted the funds, they must use the money to increase rural data speeds. All of the customers in those rural areas must be upgraded to data speeds of at least 10 Mbps download and 1 Mbps upload. The companies have six years to make the needed upgrades. Note that those speeds are far slower than the FCC s own definition of broadband 25 Mbps download and 3 Mbps upload. The companies can use the money to implement any broadband technology that will achieve the desired speeds. Both companies have said that they will use the money to upgrade or add DSL in the rural parts of the county and the rest of Minnesota. In order to implement the DSL, the telcos are going to have to extend fiber deeper into the rural areas to support the DSLAMs (DSL transmitters). These rural areas have no broadband today; the customers there will be glad to have something faster than dial-up. However, the DSL speeds that are required by the program are already inadequate for many homes. When considering that household demand for broadband has been growing at a rate that doubles every three years, by the end of six years these areas will have four times the demand for broadband than they have today. Altogether, the FCC is spending $8.6 million in the county to end up with relatively slow DSL. It s a shame that this money couldn t have instead been made available to projects like the ones contemplated in this report. Page 16

17 Following is a comparison of what the upgraded rural DSL will be able to do compared to other parts of the county and state: Upgraded DSL Cable modems in the county today Cable modems in the Twin Cities Cable Modems with DOCSIS 3.1 CenturyLink fiber in Twin Cities US Internet in the Twin Cities 10 Mbps Up to 100 Mbps (10 times faster) Up to 250 Mbps (25 times faster) 1 Gbps (100 times faster) 1 Gbps (100 times faster) 10 Gbps (1,000 times faster) In conclusion, the Connect America Fund is both a boon and a curse. It will bring faster DSL to the rural parts of the county and to customers who have no broadband today. But the broadband that will be delivered will be slow even by today s standards and will feel really slow within a decade. We think band-aid is the best way to describe what the FCC is doing. It has industry experts all shaking their heads, because the FCC is throwing a lot of money at a solution that doesn t solve the problem of having inferior broadband in rural America. D. The Consequences of Poor Broadband As the county is already aware, there is a glaring lack of broadband in the rural parts of the county. The homes without access to adequate broadband represent almost half of the households in the county, so this is a significant issue. We found a unique situation in the county there is a significant amount of rural fiber already built in the county today (or being finished this year). This means that there are quite a few fiber haves in the rural area while the county seat of Wheaton and about 600 other rural homes are fiber have nots. That is the reverse of what we find in most places, as county seats tend to have better broadband than everywhere else. Wheaton does have some broadband today. There are two providers there Frontier Communications with DSL and Mediacom with cable internet. Additionally, Runestone Telecom Association, a cooperative, has built fiber into Wheaton and serves a few customers. However, the broadband in Wheaton is not nearly as robust as the fiber provided to the customers of the four cooperatives. Not only that, but the rural customers without fiber have no real broadband alternatives. Customers within a short distance of Wheaton can probably get very slow DSL, the speed of which drops quickly with distance. The other alternatives for connectivity in the rural areas of the county are dial-up, satellite data, or cellular data, all of which are inadequate for a variety of reasons. We talk about those technologies and the providers in more detail elsewhere in the report. There are economic implications to having parts of the county without fiber broadband. Lack of broadband causes all kinds of problems for rural homeowners including: Lower Property Values: There are now numerous studies showing that homes without broadband are worth less than similarly placed homes with broadband. Realtors have been reporting across Page 17

18 the country that broadband is at or near the top of the wish list for most homebuyers today. This means it is going to become hard to attract people to live in the rural parts of the county, and more significant is that homes without broadband are going to become harder to sell. Without a broadband solution, the rural parts of the county are going to become undesirable places to live, and this is only going to get worse over time as broadband speeds keep increasing in the places that have broadband. In Traverse County, this might mean that the rural areas without broadband are going to fare poorly against those that have fiber. It is likely to become easier to sell a home or to build a new home where there is fiber. And it is likely that this will lower the property values in the areas without broadband. You also have an interesting situation today where citizens might elect to live in the rural areas with fiber rather than in Wheaton. We ve never really seen this situation before. If Wheaton doesn t get better broadband, then over time you might see the reverse of what most places see, with more financial activity in the rural areas than in the county seat. That would have all sorts of implications over time for economic development, planning, and the tax base. Education: It s incredibly hard to raise kids today in a home without adequate broadband. The issue is not just data speeds, but also the total amount of data that even elementary students have to download. For example, some satellite broadband has speeds up to 15 Mbps, but the plans have tiny data caps that make it impractical for a home with children. The same is true with cellular data; we have heard horror stories of people with kids ending up with astronomical broadband bills for using broadband from cellphones for home use. Schools want to be able to use broadband for education outside the school. An increasingly common practice in places with adequate broadband is to have students watch video content at home as homework and then discuss it in the class. That frees valuable classroom time from watching video in class. The whole education process is increasingly moving to the web and kids without access to the web are lacking the tools that their peers take for granted. Working at Home: More and more jobs today can be done at home, even if only part time, but people living without adequate broadband can t participate in this newly expected part of the economy. Increasingly, companies are willing to hire people who work out of their homes. The beauty of such jobs is that they can be done from anywhere. Many of your residents commute to jobs in other counties and many of those employers would allow commuters to work a few days a week from home if they had a good enough broadband connection. Telecommuting is good for everybody. Avoiding a commute to a distance office saves a lot of money for employees. Companies have seen after years of trying this that employees are often more productive from home due to missing the various distractions that are in the work environment. Commuting is also a greener alternative, saving a lot of gasoline and cutting down on carbon dioxide emissions. Taking Part in the Modern World: People with good broadband have access to features of the web that require bandwidth. Households with good bandwidth routinely use broadband for things like Page 18

19 watching videos on services like Netflix, talking to friends and family on services like Skype, shopping at sites that have videos, playing video games (many of which have largely moved online), taking online courses from numerous colleges, or even just browsing today s video-rich Internet. Many people s social lives, for better or worse, have moved to the web; it is not uncommon to now have friends all over the country based upon some shared interest instead of based upon geographic proximity. Medical: There has been talk for well over a decade of the Internet improving medical care in rural areas and for the elderly. We are finally starting to see some of this come to pass. There are now the beginnings of telemedicine in rural Minnesota and other rural areas where patients are able to connect to specialists in the urban areas without having to make the long drive in for an appointment. Over the next decade, this is expected to become routine. For residents without good broadband in their homes, telemedicine is being done from doctor s offices in county seats and other towns with broadband. In the last few years there have been over 100 start-up companies exploring technologies that will allow people to stay in their home longer as they age. Most of the new technologies being explored involve the need for real broadband. There are dozens of different approaches being investigated and it s certain that some of these technologies will be in play within the coming decade. This is one use of broadband that looks to be sufficiently funded because these new technologies are competing with the extraordinarily high cost of moving elderly people to institutional care. Future Services: The broadband world isn t standing still and the need for broadband has been growing at an exponential rate since the mid-90s. Residences in the US have doubled the amount of bandwidth they use about every three years since 1995, and that trend is expected to continue at this pace for the foreseeable future. To put this into perspective, if a home needs 6 Mbps download today to be happy (the typical DSL speed), then nine years from now they are going to want 48 Mbps download speeds (doubles three times). It s hard for people to visualize the impact of anything that grows at an exponential rate. Look around at other things that increase at exponential rates. Computer processing speed has been doubling about every 18 months since the early 60s. This phenomenon is known as Moore s law, named for the engineer who noticed it. It is this exponential growth that means that the computing power in your smart phone is far faster than the best PC you could buy for your home a decade ago. Because of exponential growth, your smartphone is now far more powerful than the best supercomputer of twenty years ago. Every industry expert expects the need for broadband to keep growing. Every year the size and the amount of things we do online gets larger and faster. It s been that way since we ve gotten online and there is no end in sight for new uses for broadband. Programmers are already starting to invest in making 4K video, which is 8 times larger than HD video. It s not just video that s causing the bandwidth growth. Everything we do online takes more bandwidth over time. Files get larger, web pages get busier, and we use more and more applications at the same time. As an example, when social media sites began embedding video in the last few years the bandwidth needed to be on a service like Facebook grew much larger. Page 19

20 Another growing area of broadband usage is surveillance cameras. These are particularly useful on farms to allow for monitoring of animals and remote parts of the property. We are just now starting to see the use of smart home devices things in the homes that connect to the internet. Just this Christmas saw the release of virtual reality headsets for the first time the precursors for home devices that will allow for immersive entertainment similar to the Start Trek holodecks and big users of bandwidth. These things are just the beginning. There are future things still on the drawing boards. Just as we couldn t have known 20 years ago how the Internet would change our lives and our homes, we can t imagine today all that s coming in the next 20 years. Page 20

21 II. ENGINEERING DESIGN AND COSTS Finley Engineering did the engineering analysis. There were several engineering solutions considered including fiber everywhere, wireless everywhere, and a hybrid version that served Wheaton and some other areas with fiber but most of the rural area with wireless. This section also provides a general primer on the technologies considered for the rural areas. A. NETWORK DESIGN A study was done only for the parts of the county that don t have fiber broadband today. That means that the following areas were not included in the study: Park Region Mutual Telephone Company, part of Valley Telephone, operates the Browns Valley exchange. Currently, they are operating a copper network and have fiber to anchor institutions with speeds up to 60 Mbps. The company has already built fiber along the lake around Highway 27 and uses the fiber with copper drops to reach customers. The company plans to complete the construction of fiber to every rural customer during this summer. But some existing customers will remain on fiber-to-the-curb for the foreseeable future. Federated Telephone Cooperative, which is part of Acira, operates the Chokio exchange. They have built fiber in their service area and also serve 17 wireless customers with speeds up to 50 Mbps into the Graceville Exchange. The Runestone Telecom Association operates the Norcross and Tintah exchanges. The company has built fiber within its telephone service area. They also have built fiber to Wheaton and serve several anchor institutions there on fiber. In addition, the company owns a cellular tower in Dumont. The Fairmount exchange is operated by Red River Communications and is currently served by full fiber-to-the-home. After excluding the above areas, the study consists of Wheaton and the surrounding rural areas. This includes Arthur, Collis, Monson, and Dumont. We considered several different network designs: We looked at providing Fiber-to-the-Home (FTTH) for the whole service area. We looked at providing FTTH to Wheaton and to customers that live along a proposed backbone fiber. The remaining customers in this scenario are served with wireless. This is referred to throughout the report as the Hybrid scenario. Finally, we looked at a wireless-only scenario for the rural parts of the county (not including Wheaton) using point-to-multipoint wireless technology. Before looking at the specific network designs, we gathered information about the county for use in all of the scenarios. Following is a description of the most important of the generic data we used. Page 21

22 Passings: The telecom industry uses the term passing to mean any home or business that is near enough to a network to be a potential customer. We verified passings in several ways. First, we talked to Traverse Electric Cooperative. They serve only rural customers that include the areas covered by this study. The coop has 718 rural customers in the county. About 85% of the electric customers are served using an aerial electric network constructed on poles, with the rest buried. About 75% of the customers have an aerial electric drop with the remaining customers with buried electric drops. We also acquired a detailed map from the Minnesota Department of Transportation. The map is a current map and shows the location of what they call active building sites meaning occupied. Their map shows a total of 529 active buildings in the rural study area. We also were able to get counts of homes in Dumont (64 passings) and Wheaton (815 passings). This means that the whole study area has 1,408 passings, or potential customers. The MDOT map has no details on the nature of each building. We have to assume that there are some rural businesses for example, we know there are some rural farms that employ a significant number of employees. The number of estimated businesses are guesses made based upon conversation with the county. Unless there is a big difference between our estimate and the actual number of total businesses, the study results will not be materially affected. Wheaton Town Residences 692 Wheaton Town Businesses 123 Wheaton Rural Residences 123 Wheaton Rural Businesses 3 Arthur Rural Residences 109 Arthur Rural Businesses 2 Collis Rural Residences 177 Collis Rural Businesses 3 Dumont Town Residences 62 Dumont Town Businesses 2 Monson Rural Residences 110 Monson Rural Residences 2 Total 1,408 The Wireless-only scenario excludes Wheaton and has 593 passings. The Runestone Telecom Association told us that they currently own a fiber that goes through Wheaton and terminates at a cellular tower in Dumont. The company serves the schools and the library using fiber. They say that Mediacom provides services on fiber to the county Social Services Building and to the Courthouse. There doesn t appear to be any large multi-dwelling units in the study area, or at least not any that would be large enough to require the study to consider serving them in a different manner than serving single family homes. Page 22

23 Road Miles: Analysis of the MDOT maps show that there are about 390 miles of streets and roads in the study area. These are roads that are maintained all year, meaning they are plowed when it snows. Our study is conservative in that it assumes that fiber would be built along all of these roads. It s likely in a detailed design that some efficiencies could be found that would result in small reductions in the road miles that need fiber. Basic Network Design Fiber Backbone All three network designs utilize the construction of a backbone fiber. As shown on the map on the following page, we ve designed the backbone fiber in a star configuration with Wheaton as the hub. On the map, the backbone fiber is marked by blue lines that go from Wheaton to three different parts of the rural portion of the county. This configuration was chosen for several reasons. First, all of the network designs require the placement of electronic equipment in the rural parts of the county and this design fulfills that function. Ideally, fiber supplying electronic huts or wireless towers would be designed using a ring configuration. In a ring the fiber is self-healing, meaning that the network doesn t go dead in the event of a fiber cut. However, in this case the service area is not large enough to justify a ring. A ring would also add to the cost of the project. It is possible that if one or more of the existing telcos decide to build the suggested solution, they could eventually create a ring that would include fiber from their existing service territories. The total length of the backbone fibers shown on the map is 36.2 miles. The fiber routes were designed using 96-strand fiber, which should provide adequate pairs for current and future needs. The three routes on the fiber backbone stretch between Wheaton and the new electronics sites: North of the community of Collis, approximately 4.5 miles south of Dumont At the intersection of County Highway 4 and County Road 3 At the intersection of US Highway 75 and County Highway 16 In all scenarios, we based pricing upon recent quotes we got from vendors like Calix, AdTran, Clearfield, Cyan, and others. Neither Finley nor CCG are proposing any specific vendors and we are both vendor neutral. The costs chosen are representative of current electronic costs. In pricing the fiber construction, Finley used pricing from recent construction of fiber in similar conditions (soil type). The labor in the forecasts was done at current market and did not include the prevailing wage rate. In recent years, the DEED grants required prevailing wage, which in Minnesota means labor rates the same as those in the twin cities. But since that is no longer a requirement for the grants, we did not include it in the cost estimates for fiber construction. Page 23

24 Figure 1. The Proposed Backbone Fiber Network. Proposed fiber backbone shown in blue. The All-Fiber Network Scenario The first option studied was an all-fiber design. It is universally accepted that fiber is the ultimate network to provide big bandwidth going into the future for rural areas. All four of the telcos in or around the edge of the county have either already built fiber to their customers or will soon do so. The all-fiber design considered building fiber to Wheaton as well as to the rural areas that are not served by anybody else. We decided to include Wheaton for several reasons: Page 24