Article Info

Fairpoint & Occam Networks

Fairpoint Communications (FRP) plans to spend $781M over the next 5 years maintaining and upgrading the network they purchased from Verizon. Drilling into the details of the network spending suggests Occam Networks (OCNW) is well positioned to materially benefit from this proposed deployment. Details provided by Fairpoint and fundamental metrics of the Broadband Loop Carrier business lead us to believe Occam will recognize up to $125M in revenue, of which $80M will come in the next 18-24 months, provided Fairpoint executes it’s current capex plan.

Fairpoint provided an in-depth picture of how they plan to manage ‘SpinCo’ during a conference call April 17th. SpinCo represents the newly acquired residential and business subscriber assets & infrastructure from Verizon (VZ) in Vermont, New Hampshire, and Maine. During this call Fairpoint provided detail on their plans and expectations for growing broadband penetration within the subscriber base including capex forecasts for network upgrades. A full copy of the presentation is available here: (Link: Fairpoint IR)

Occam Networks was the first announced access equipment supplier (Jan 08 release here), and Cisco was chosen to provide the back-end IP routing infrastructure. Pannaway was selected as the second vendor of access equipment following the close of the Fairpoint/Verizon transaction.

Fairpoint ‘SpinCo’ Demography and Finance

Fairpoint’s newly acquired New England assets (SpinCo) have the following subscriber profile:

  • Residential Access Lines – 929k
  • Business Access Lines – 450k
  • High-Speed Data (mostly DSL) Lines – 223k

Other key information about SpinCo:

  • Broadband penetration in the SpinCo network is 16%, peer average is 25%. Fairpoint’s non-SpinCo network is 28%. Fairpoint believes 25% is an achievable target over 5 years.
  • Broadband is available to approximately 68% of the existing SpinCo customer base. Fairpoint has committed to increase this to 90% within 5 years and has contractual capex obligations with the states to achieve this.
  • These regulatory obligations (made in order to secure merger approval) require them to spend $633M over the next 5 years in the NH/ME/VT area. Their current spending plan increases this to $781M. This works out to about $566/line.
  • Fairpoint plans to spend $205M in 2008 (excludes back-office integration capex of $128M) specifically to deploy hardware assets in the field. The balance of the capex is allocated over the following 4 years.
  2008 2009 2010 2011 2012
Capex $205M $161M $155M $137M $123M
% total 26% 21% 29% 17% 16%

Top Down Analysis

In Q108 VZ spent $12M to broadband enable 30k lines in Maine – $400/line (link). AT&T is spending $361/line on Project Lightspeed (see AT&T Lightspeed Gets More Expensive), which is similar to what Fairpoint is planing. But Fairpoint is dealing with a much more rural and less dense subscriber base, and therefore a larger capex/line figure is reasonable. Fairpoint also includes the costs of a new Ethernet backhaul network and Cisco routing infrastructure – something not included in the Verizon and AT&T numbers.

Using the Verizon Maine deployment as a template would put Fairpoint’s access related capex at around $400/line.The remainder is backhaul at $166/line ($566-$400), or roughly 30% (approx $235M)of the total capex, allocated to Cisco (CSCO). The other 70% (approx $550M) is access spending, allocated to Occam and Pannaway.

Fairpoint’s capex numbers account for all the capitalized costs involved in installing a network: labor, fiber, copper conditioning, and the hardware itself. A general rule of thumb in the carrier industry is equipment/hardware capex is about 20%-30% of total capex. This would result in hardware revenues to Cisco of $50-$75M and $110M-$165M for Occam and Pannaway.

Bottoms Up Analysis

Fairpoint’s demography data can be used to make estimates about the quantity and type of equipment they will purchase. They will buy access equipment for three purposes.

  1. Increase broadband penetration in existing coverage areas
  2. Increase the broadband coverage area
  3. Upgrade equipment in existing coverage area

#1. Increased Broadband Penetration:

SpinCo, at 16%, is under-penetrated relative to the rest of Fairpoints holdings. It is reasonable to see this increase to 25% in 5 years. A 9% increase in subscribers among the 68% of the served area is 84k lines (68% * [928k Res + 449k Business] * 9%).

Assuming a per-port cost of $60, anticipated subscriber gains in areas already covered by broadband will require approximately $5M in hardware capex (84k * $60) over the next 5 years. This capex will flow to the vendors of the existing broadband equipment- either legacy vendors, or as we propose in #3 – Occam/Pannaway.

#2. Increased Broadband Coverage:

Fairpoint will expand broadband availability from 68% to 90% of the served customer base, increasing the potential broadband market by 302k lines (22%*[928k Res + 449k Business]). Using the broadband penetration target of 25%, we can expect to see 75k additional broadband subscribers over the next 5 years.

It is virtually certain that this deployment will scrap existing equipment in the field and migrate all lines, including existing POTS subscribers, to new converged Ethernet/IP access hardware. This will require the deployment of 300k POTS lines with 75k broadband lines added incrementally over the next five years. This is consistent with the Occam announcement that highlighted 50k in broadband lines expected to be deployed in the near term.

Therefore, we can expect Fairpoint to purchase equipment for a greenfield installation of 300k POTS lines and eventually 75k broadband customers. This capex will be front loaded, as all POTS customers will be migrated early on, and the new access chassis must be purchased and installed. We expect it to be complete by EOY 2009.

We assign a per port value to such an installation of $150, for total capex of $45M (300k * $150).

#3. Existing Network Upgrades

The other 68% of Fairpoint’s network is already broadband enabled. There are a total of 936k lines (68%*[928k+ 449k]) that currently support broadband. We believe that the majority of this equipment will be replaced with next generation Ethernet access equipment for the following reasons:

VT Cental Office Locations

VT Central Office Map

  1. Fairpoint has committed to spend nearly $800M in capex. It simply isn’t possible to rationally spend this much money on upgrading 300k POTS customers and building a new Cisco Ethernet/IP backhaul network. The money must be flowing to the existing broadband base for the Fairpoint capex numbers to make sense. 
  2. Occam announced that the first stage of the Fairpoint deployment entails the constr
    uction of 10Gb Ethernet rings interconnecting 200 central offices (CO’s) throughout the three states. We know that these rings are terminated by Occam equipment in each office, which requires the installation of a new Occam chassis. A cursory review of NH/MA/ME central office locations (source NH / VT / ME) clearly shows that 200 accounts for significant proportion of existing central offices. Fairpoint wouldn’t be placing Occam equipment in the existing broadband footprint if they didn’t plan to use it.
  3. Fairpoint needs to upgrade it’s DSL offerings to ADSL2+ in order to remain competitive with cable broadband. In some cases FTTH will be desired. It is difficult to see Fairpoint sinking more capex into legacy Verizon DSL infrastructure.
  4. Fairpoint has made it clear that high-margin Enterprise customers are a priority. The Ethernet services enabled by the Occam and Pannaway BLCs would provide the means for Fairpoint to pursue these customers with compelling services.
  5. Fairpoint doesn’t think like a traditional telco. Telcos typically shy away from capex unless it drives revenue growth. They do not spend money to save money. A tour of an average central office surprises many tech tourists, as much of the equipment is 20 years old. It is depreciated, the technicians know how it works, and if it isn’t broke, you don’t fix it. Fairpoint must build a simpler network capable of being managed by it’s relatively diminutive size. While normally we would be skeptical of carrier plans to spend money replacing perfectly good equipment, the nature of Fairpoint & SpinCo allows us to cautiously claim “This Time is Different” (also known as the four most expensive words one can say).

Unlike the broadband coverage increases, upgrading the existing network will take a full 5 years. And risk exists that such an effort would be halted if Fairpoint encountered financial difficulties.

A complete upgrade of the installed broadband capable base will require 940k lines (68%[928k+ 449k]), of which 250k must support broadband in addition to POTS. At $100/port, a number applicable to installation in an existing CO, this is $94M. This includes the 10GbE ring deployment that appears to be underway. The timing of this revenue should mirror the capex percentages shared by Fairpoint, with 26% being spent in 2008 and 21% in 2009.

However, it is clear that some of this network was recently upgraded by Adtran, and is unlikely to see new hardware immediately. Verizon spent the bare minimum given their long term plans to dispose of the asset. We assume 80% of the existing network will be upgraded in the next 5 years – or $75M

Summary

  • Increased Broadband Penetration – $5M over 5 years
  • Increased Broadband Coverage – $45M over 18 months
  • Existing Network Upgrades – $75M over 5 years, ($35M within 2 years)
  • Total: $125M in equipment, $80M within next 18-24 months.

Note: The top down figure based on industry ratios yielded a range of $110-$165M.

Market Share – Occam vs. Pannaway

It is difficult to provide a quantitative estimate of how this business will be shared by Occam and Pannaway. However, the following facts are known:

  1. Occam is an existing supplier to Fairpoint
  2. Occam was selected first and has provided quantitative details of it’s participation in a press release
  3. Pannaway offers no details of the Fairpoint announcement on it’s web site. The only disclosure made regarding Pannaway was made during the Fairpoint call, where they mentioned a dinner with the supplier the night before.
  4. The Maine business that generated our $400/line capex estimate was awarded to Adtran (ADTN). Their Q408 conference call transcript (see here, search for “Fairpoint”) indicates acknowledgment that future Fairpoint business was won by Occam. No mention of Pannaway is made.

Conclusion

The demography here is clear, and demography is destiny. Fairpoint has regulatory obligations and the business motivation to spend nearly $800M over the next five years upgrading their network. They have announced partner vendors and indications are that work has already begun.

Our biggest concern is why Fairpoint would bet their entire company on Occam and Pannaway. This is a significant piece of business that major equipment vendors would want. Perhaps tier 2/3 carriers simply “Think Different” and see equal risk in being a small customer fish in the big pond of Alcatel as they do working with critical vendors that are small companies.

Still, Fairpoint must have a risk plan in place, perhaps similar to the deal brokered by AT&T that put World Wide Packets into the hands of Ciena, or the deal brokered by British Telecom that partnered Covaro networks and ADVA AG. The logical acquirers of Occam/Pannaway would be Cisco or Ciena, as both lack Telco DLC solutions, and would be complemented by the addition of Occam’s products.

Author holds positions in OCNW and ADTN.

Discussion

Comments are disallowed for this post.

  1. I am aware of why you believe that Pannaway is a vendor, which is certain side comments made by FRP management on their 4/17 confernece call. To me, it isn’t entirely obvious as there has been no press release by either Pannaway or FRP as there was in the case of Occam. Certainly, I think Pannaway would want to announce such an important contract so the fact that there has been no announcement leads me to believe either that the contract is not a done deal, or that Pannaway’s role will be sufficiently marginalized so that a conscious decision was made not to press release it.

    Posted by osxman | May 6, 2008, 9:40 AM
  2. In their most recent conference call, Occam made it clear they were quite comfortable they would get the considerable majority of the contract.

    Posted by osxman | May 6, 2008, 9:40 AM
  3. Andrew,

    Do you believe that your 20-30% equipment estimate for capex is low? Did you vet this with either Occam or Fairpoint? The number seems low to me as I am presuming there will be significant reuse of property and plant.

    I don’t think there will be that much fiber and I don’t know much about copper conditioning but how much can that be?

    Posted by osxman | May 6, 2008, 10:40 AM
  4. On Pannaway:

    Certainly all qualitative data points to them not participating. If I had to guess I would agree that Occam will take most of the share. Occam praised another NH company, Wilcom, in their Fairpoint release, which leads me to believe Pannaway was tossed in as another token NH company.

    On Capex:

    The 20-30% was not vetted with Fairpoint or OCcam. The $781 figure includes ‘operating’ capex. Fairpoint would incur significant capex even if they bought no new hardware simply by maintaining the existing equipment in the field. While the number seems low for a ‘new build’ you have to take into account the top line also includes ‘routine’ capex not related to new builds.

    20%-30% is a rough guideline and is why a bottoms up approach is really more accurate. Taken together, one gets the best range of revenue one can expect when working with limited information. If the range was 30-40% then the bottoms up analysis would diverge, which means per-port cost assumptions are too low.

    Posted by Andrew Schmitt | May 6, 2008, 11:20 AM