This article continues a series of features profiling the thought-leaders at Digital Colony, business model innovators and pioneering investors in the convergence of towers, data centres, fibre and small cells. The series was initiated by Marc Ganzi’s “(Re)defining towercos” vision, followed by Raul Martynek’s guide to the history and future of data centres, and Steven Sonnenstein’s view of digital infrastructure through an investment lens. This latest article looks at Digital Colony’s opportunities in fibre and small cells, guided by subject matter expert Warren Roll.
TowerXchange: What is your role at Digital Colony?
Warren Roll, Principal, Digital Colony:
I am a partner in the fibre team and based in Boca Raton. I joined Digital Bridge in 2014 and have both led and participated in a number of key communications infrastructure deals including ExteNet Systems, Vantage Data Centers and Digita Oy. I have developed a specific sector expertise in small cells and fibre services. I am very focused on fibre, small cells, and wireless towers globally. I currently sit on the board of directors for ExteNet Systems and Digita Oy, our newest tower investment in Finland.
TowerXchange: Please summarise the opportunities you see in fibre and small cells?
Warren Roll, Principal, Digital Colony:
Globally, we see a growing demand for metro fibre networks, as well as a consolidation of national fibre, which will provide the future backbone for applications such as the Internet of Things (IoT) and smart cities. We believe that in the U.S. today, there are around 100,000 deployed small cells nodes compared to less than 10,000 in Europe today. We expect small cells in Europe to proliferate over the next 5-10 years.
As for fibre, major investment in the space is needed. The FFTH Council for Europe highlights as part of its cost model study that €156 billion is needed to deploy FTTH across 28 European countries.
Analysis like Deloitte’s “Communications Infrastructure Upgrade – The Need for Fibre,” suggests that the United States requires an investment of US$130–$150 billion in fibre infrastructure over the next five to seven years to adequately support broadband competition, rural coverage, and wireless densification. This massive investment will likely need to come from a variety of sources, including communications service providers, financial investors, and public/private partnerships. Unfortunately, the current wireline industry structure does not provide sufficient market incentives for fibre and broadband deployment.
Private equity is also moving into this space: Antin Infrastructure Partners who bought FirstLight Fibre in the U.S. and CityFibre in the UK has also added a Spanish wholesale fibre company Ufinet to its portfolio. Arcus Infrastructure Partners has reached financial close on the acquisitions of a 93.1% stake in E-Fibre, and 100% of Swiss4Net.
With small cells, rollout has started globally. This rollout enables operators to achieve advances in capacity and network speed. Small cells help patch coverage holes, optimise signal strength, and to increase capacity to meet consumer demand.
The U.S. is one of the most advanced markets. For example, around 62% of Verizon’s wireless deployments in 2017 were small cells. On the other hand you have numbers from the CTIA estimating that the number of small cells deployed is predicted to rapidly increase over the next few years from about 13,000 small cells in 2017 to over 800,000 total deployments by 2026. We expect this trend to continue globally.
TowerXchange: What specific fibre services, what kinds of fibre, and in what locations do Digital Colony have appetite to acquire / deploy? And what balance do you anticipate between acquiring existing fibre and laying new fibre to achieve your growth objectives?
Warren Roll, Principal, Digital Colony:
In the long term, we continue to like certain sector themes. Given we are focused on communications infrastructure, we are concentrated on certain guiding principles such as long-term contracts, investment grade credit, low churn, owned (not leased) and underutilised fibre assets with co-location opportunity.
Small cells and dark fibre would typically be the most interesting. These contracts tend to range from 7-10 years with multiple five year renewals. We like certain enterprise business as well, it just depends on the mix and the geography. If we see enterprise businesses in great locations with low churn and a good customer mix (larger enterprise tends to have less churn), then there are opportunities to grow into other services like small cells and sell dark fibre.
Overall, fibre connectivity is becoming critically important in today’s increasingly digital world, with next generation applications such as 5G, IoT, smart vehicles and ultra HD video driving the need for lightspeed data transmission. Therefore, a good enterprise business can make a great platform to provide these services to future customers.
We typically don’t focus much of our time on long-haul fibre and spend most of our efforts looking at metro plant in tier 1 and tier 2 cities.
Over the last three years we’ve seen M&A valuation multiples skyrocket. It’s very hard to make an economic argument to buy businesses that cost double, if not more what it costs to build. At ExteNet we have continued to stay disciplined and continue to build our own networks
Sometimes when it just isn’t economical to buy, we build. A great example of that is in the U.S. Over the last three years we’ve seen mergers & acquisition valuation multiples skyrocket. It’s very hard to make an economic argument to buy businesses that cost double, if not more what it costs to build. At ExteNet we have continued to stay disciplined and continue to build our own networks.
We are currently looking at smaller platform opportunities in the European region where we can grow through smaller acquisitions and organic construction.
TowerXchange: What kind of fibre is Digital Colony interested in; long distance, metro fibre rings, or last mile FTTT / FTTH / FTTX?
Warren Roll, Principal, Digital Colony:
We are most interested in Metro Fibre. We think everyone has heard of the word “densification” at this point but that is the major theme that continues to play out in the U.S. in the early innings and we see the same theme playing out worldwide in some of the most populated cities. As we continue to consume wireless data, our networks are getting clogged. The only way to effectively increase capacity is through more spectrum, and optimal use of spectrum. More antennas at a closer proximity to the end user allows for the most optimal use of spectrum. We will continue to see millions of antennas or “nodes” being constructed as we continue to consume data.
As we start thinking about 5G, small cells will provide denser networks and coverage and enable the use of new, higher frequency spectrum bands. They will also enable the delivery of new services that rely on close proximity to the user. Metro networks will take advantage of this “densification”.
FTTT, is very interesting too given the length of the contracts with wireless providers. Geography is everything so it’s a hard question to answer on its own, but you can usually layer on numerous other services like small cells and enterprise on FTTT footprints. The lowest priority of interest is usually FTTH, given the lack of services you can co-locate.
TowerXchange: Can we unpick what specific ‘fibre solutions’ you’re selling – how much service, maintenance etc is included in what you’re offering to carriers and other clients?
Warren Roll, Principal, Digital Colony:
Generally, the carriers want a full solution as it relates to monitoring and maintenance, regardless of what they are buying (turnkey small cell solution, dark fibre, or node-only). If the network shuts down, the customer wants it back up yesterday.
We take great pride in the services we provide to our customers. A cornerstone of our business is providing five nines standards; we make sure our networks are up and running 99.999% of the time. When networks fail, we are able to get them back up and running within hours. Maintenance services are typically outsourced.
TowerXchange: Roughly what proportion of small cell sites you’re leasing to customers are already along your existing fibre routes? And what proportion of the time are you laying new fibre?
Warren Roll, Principal, Digital Colony:
We target to own almost 100% of our fibre. We look to have very little IRU fibre which is our preference. Given the early innings of densification, we continue to build and lay new fibre every year. We see small cell growth growing at over 30% every year for the last three years and we don’t expect this to slow for the foreseeable future, if anything it could continue to increase as we enter into the initial stages of 5G rollouts in 2020.
TowerXchange: Can you break down the ‘bill of material’ to lay fibre – what equipment and services are required, what do you do in-house versus outsource?
Warren Roll, Principal, Digital Colony:
The primary costs / materials involved can vary based on the construction methodology (aerial versus underground versus third party conduit). Generally, there will be costs related to the cabling, splice boxes, new conduit (in the case of underground builds) and fibre attachment gear (in the case of aerial builds). Services related to the construction and attachment of the fibre and splicing and are largely outsourced to third party contractors. In many cases we will complete the fibre design work ourselves and the construction and laying of fibre is outsourced completely.
There are four major areas to consider for fibre builds:
1. Design/engineering
Usually done in house via fibre engineers
2. Permitting/licensing
Usually outsourced to an engineering firm or construction company
3. Construction
Usually outsourced to a fibre construction company
4. Splicing
Either in-house or outsourced to a splicing company
Bill of materials:
1. Aerial
a. Fibre
b. Strand (1/4” or 5/16”)
c. Splice case/trays
d. Anchors/guides
2. Underground
a. Fibre
b. Conduit
c. Innerduct
d. Manhole/Handholes
e. Splice cases/trays
TowerXchange: Crown Castle has described how they have found macro towers lease up at around 0.1 tenants per year, whereas small cells lease up approximately twice the speed - how has Digital Colony found co-location growth of small cells?
Warren Roll, Principal, Digital Colony:
This is true, for all the reasons discussed earlier. Carriers’ capex budgets are skewing more and more toward the construction of small cells to increase the capacity of their networks. We are seeing more carrier antennas added onto networks which is pure co-location. We expect to see more and more co-location as networks get denser and we begin 5G rollouts. Co-location should see lease up rates well in excess of 0.1 per year.
TowerXchange: How do you co-locate a small cell, for example do you place multiple antenna at the same site and is there a limit as a function of physical space or interference? Or do you simply add a “co-locating” small cell to an adjacent street lamp or other structure? Or are the antennas themselves shareable?
Warren Roll, Principal, Digital Colony:
There are some use cases where actual pole locations can be shared but the fibre infrastructure is the most expensive component of the build and therefore it’s quite valuable to utilise existing infrastructure even if due to technological or municipal limitations you can’t co-locate on the same pole.
We will now build backbones with 864 strand-count allowing for significant capex leverage for subsequent builds (generally a minimum of two strands per node per carrier are required for small cells).
Most wireless carriers only want one antenna (theirs) located on a street light/traffic signal/monopole. Along with the antenna, you place radios in a shroud underneath the antenna.
Technology is advancing on antenna use, but today most designs call for an antenna and radio(s) at each individual site. But, usually only one antenna is used at any given site.
TowerXchange: How do the costs and cash flows from fibre and small cells compare to macro towers?
Warren Roll, Principal, Digital Colony:
Typically compared to a single macro tower, the carriers are paying anywhere from 5-6x less for a single small cell which will generally cover a much smaller distance compared to a macro tower as they are primarily deployed to solve capacity issues in urban locations.
However, despite the delta in revenue per unit due to the competitive environment for new BTS towers the cash yields for a single tenant tower and single tenant small cell network are comparable in the mid to high single digits. For tower lease up there are negligible incremental costs and capex involved which substantially improves the cash yields with incremental tenants. Small cell yields will improve as well with incremental tenancy but not at the same rate as there may be incremental opex and capex involved with lease up (potentially new fibre and node deployments on new poles).
TowerXchange: Finally, please sum up your vision for the future of Digital Colony’s fibre and small cells business.
Warren Roll, Principal, Digital Colony:
We believe fibre is the essential part of all networks. It is the connective tissue that links internet infrastructure and data centres, wireless infrastructure, macros and small cells together. The next generation of wireless networks will need many types of communications assets - from macro cells, small cells and DAS to centralised radio access networks (C-RAN) and data centres and more and more fibre optic cable.
Digital Colony is determined to build a variety of communications infrastructure assets to serve our wireless, cloud and content and large enterprise customers’ needs. We believe that data consumption will continue to increase globally at an exponential pace and the necessary infrastructure required to meet this ongoing demand will need to be expanded and upgraded quite dramatically all across the world.
Introducing Warren Roll, Principal, Digital Colony
Warren Roll is a senior financial professional with 20 years of experience building businesses together with management teams as a private equity investor and investment banker. Roll has extensive sector experience in technology, media, telecommunications, services, and healthcare.
Roll served as a Senior Director within Private Equity at PSP Investments, one of the largest Canadian Pension Investment managers, with US$153bn of assets under management, which he joined in February 2011. He led PSP’s TMT and Healthcare initiatives and was responsible for investing over US$1bn in numerous successful investments.
From 2007 to 2011, Roll served as a Director at Aver Finance (BMO Financial Group), an international media fund specialized in financing media and new media companies globally. Prior to Aver, he was Vice President, Investment Banking, at Blackmont Capital and an investment banker at Desjardins Securities, where he helped raise in excess of $800 million for Canadian public and private companies.
Roll received his MBA degree with distinction and his BA with honors from the Richard Ivey School of Business and he currently sits on the Board of Directors and the Executive Committees of ExteNet Systems and Digita Oy.
