Month: October 2021

Trends in Home Broadband Usage

Before the pandemic, iGR had forecast that wired home broadband usage would reach 365 gigabytes (GBs) per household (HH) per month on average in 2021. Now, we expect average home wired broadband usage in 2021 to reach 403 GBs per HH per month. Moreover, we expect it to remain 10-15 percent higher than we had originally forecast. By 2025, we forecast the average U.S. household will use approximately 558 GB per month of wired broadband.

Average home wired broadband usage jumped because everyone was forced to be home and since everything was closed, the Internet became the main source of continued schooling, work, seeing doctors and finding entertainment. So, data usage increased: video streaming (Netflix, Hulu, etc.), game downloads, audio streaming, video conferencing (Zoom, Teams, etc.) and chatting (FaceTime, etc.). With people having found more convenient ways to get everyday “stuff” done, we do not think those changes will be reversed.

Another trend is the probably permanent change to how corporate America functions. Consider the following data from this paper (Has the COVID-19 Pandemic Accelerated the Future of Work or Changed Its Course? Implications for Research and Practice ( which I paraphrase and summarize here:

  • In 2019, the U.S. Bureau of Labor Statistics (BLS) reported that 29 percent of wage and salary workers could work at home and 25 percent had opted to do so at some point, but only 15 percent of employees reported working full days exclusively from home.
  • At the pandemic’s outset, many employees were forced to transition from primarily working onsite to working from home. The BLS estimated 35 percent of the U.S. workforce worked from home due to the pandemic at least once in May 2020. By year end 2020, those forced to work from home because of the pandemic had fallen to 24 percent. By July 2021, 13 percent reported that they had been forced to work from home at some point in the past four weeks due to the pandemic.
  • Between October 2020 and April 2021, a Gallup study found that of those white-collar workers working from home, 71 percent wanted to continue working from home. Similarly, a Harvard Business school study found that 81 percent of professionals who had been working remotely from March 2020 to March 2021 either did not wish to return to the office or wanted to telework via a hybrid schedule (e.g., working 2 to 3 days at home per week).

A permanent increase in the number of corporate workers who work from home one or more days per week will also decrease the total data usage on the wired (and cellular) networks in cities where they work. That change will ripple into other areas, as well: the cellular networks along the highways, congestion on those roads, ridership of public transportation, businesses (e.g., restaurants) in those cities.

More people staying at home for work means that data usage is more distributed – i.e., down to the subdivision level. This obviously changes the timelines for when wired broadband operators add throughput to their networks while increasing the breadth and depth of their coverage. And, of course this also means continued expansion of data centers, introduction of more edge compute nodes, pushing fiber networks (in the HFC networks) closer to end users.


Source: Novacap

October 21st, 2021, Novacap, a leading North American private equity firm, announced today that it has entered into a definitive agreement to acquire Stratus Networks (“Stratus”).  The transaction is expected to close in the last quarter of 2021, following the satisfaction of customary closing conditions, including required regulatory approvals. Headquartered in Peoria, Illinois, Stratus is a leading provider of fiber-optic bandwidth infrastructure services. Founded by Co-CEOs John Petrakis and Kevin Morgan in 1998, Stratus has grown to become a leading data and voice networking services provider for enterprises with a full suite of products including data, cloud and voice solutions. John Petrakis and Kevin Morgan will continue in their roles as Co-CEOs leading Stratus and remain significant shareholders in the business.

“The investment in Stratus continues Novacap’s deep experience of successfully partnering with strong founders and entrepreneurs to support long-term growth in the fiber industry» said Pascal Tremblay, President and CEO, Managing Partner at Novacap.  

“The accelerated digitization of our economy is driving the demand for robust digital infrastructure services to feed the digital value chain. The transaction will enable Stratus to expand its presence in growing market segments and larger geographic areas to meet the increased demand for connectivity,” commented Ted Mocarski, Senior Partner at Novacap.  “We look forward to applying Novacap’s strategic, operational and financial expertise to propel Stratus’ growth.”

“We are excited to partner with Novacap and take advantage of their substantial experience as a successful investor in regional fiber-optic providers such as Horizon Telecom, FirstLight Fiber and Fibrenoire,” said Kevin Morgan, Co-CEO and Co-founder of Stratus. “We are delighted to have a partner who will be able to provide the resources necessary, and share our vision, of being the premier provider of internet and voice services in our region. Novacap has already begun sharing ideas about ways to invest in our business to aggressively pursue the significant growth opportunities available to us ” added John Petrakis, Co-CEO and Co-founder of Stratus.


Houlihan Lokey served as the exclusive financial advisor to Stratus. Morgan Lewis & Bockius LLP served as legal counsel to Stratus.

Paul Hastings LLP served as legal counsel to Novacap.

About Novacap

Founded in 1981, Novacap is a leading North American private equity firm with over C$8B of AUM that has invested in more than 100 platform companies and completed more than 150 add-on acquisitions. Applying its sector-focused approach since 2007 in TMT, Industries, Digital Infrastructure and now Financial Services, Novacap’s deep domain expertise can accelerate company growth and create long-term value. With experienced, dedicated investment and operations teams as well as substantial capital, Novacap has the resources and knowledge that help build world-class businesses. Novacap has offices in Brossard, Québec and Toronto, Ontario. For more information, please visit

About Stratus Networks

Founded in 1998, Stratus has grown to become a leading data and voice network services provider for enterprises across Central Illinois. Since inception, Stratus has invested in and built superior network, product offering, ancillary services, partnerships and proprietary operational systems in order to deliver the best available service to its customers. The Company owns, leases and operates a large fiber network to deliver a full suite of solutions including data, cloud and voice solutions to reach significant scale in order to serve nearly 1,000 highly diversified enterprise customers primarily across the financial services, healthcare, telecom industries. For more information about Stratus, visit

Are the Data Centers Growing Increasingly Edgy?

The edge continues to gain momentum and the industry giants are paying attention. Just last week (on 10/14) we saw Digital Realty (DLR) – the staid “REIT-y” data center company – continue to move on this effort with a strategic investment in AtlasEdge. AtlasEdge came together through a partnership with Liberty Global and DigitalBridge. This is not DLR’s first move into the Edge space. In 2020, DLR teamed up with Vapor IO’s software-defined Kinetic Edge in some of its US facilities (including Atlanta, Chicago and Dallas). Equinix, DigitalBridge, and Switch all have their own version of edge offerings.

This should not come as a surprise as the data center moves are driven by the needs of their largest customers. And for the largest customers – namely the cloud players – the Edge is much more than just a white board concept.

Last week, I had the opportunity to be a guest speaker for a class at NYU to go through the general concept of what a data center is. During my talk, a student asked me what I thought of Microsoft’s underwater data center. I was quite confused, but sure enough I Googled it and it does indeed exist. (It is true – “teachers” often learn from their students.) If you look at the picture of this online, you will see that this ‘underwater data center’ is really more like a small tubular type structure than a true data center, and in reality, is yet another edge point of consumption – albeit two miles deep!

The “smart money” behind some of these edge concepts don’t question the need for the edge; what I have found is they question how to make money from the concept. Perhaps they need to look a bit further down the lines of the income statement to find the real value of the edge – costs savings vs. revenue enhancements.

A 2015 McKinsey & Company study found that an offshore oil rig generated data from 30,000 sensors — but less than one percent of that data was used to make decisions. The question becomes now – many years later – has that changed? It seems to me companies have a tremendous opportunity to use such data to make much smarter decisions from a cost side.

There have been several case studies that show edge deployment can result in tangible cost savings for enterprises deploying such initiatives. One recent case study put out by IBM showed its edge solutions for a manufacturing client resulted in a 5x increase in inspection efficiency, a 20 percent reduction in false-positive defect detections, and a 20 percent savings on software maintenance.

For the cloud players who support these enterprises, such statistics will make any CFO sit up and take notice. Given the fact that cloud companies continue to be the critical and growing customer for the data center companies (note: cloud accounts for 28 percent of DLR’s revenue and seven of EQIX’s top 10 customers), the edge cannot be ignored. But we all know that a heavy lift is needed on the infrastructure side to extract all the benefits of the edge. It is a similar ‘chicken and egg’ scenario as a 5G discussion.

However, although an edge opportunity in isolation is more of a cost savings driver for the end customer, how data centers position this edge to best support their cloud customers could be an important needle mover for their own top-line growth. If cloud companies are able to pitch the ‘realness’ of these cost savings to customers, they (cloud players) will need help from their 3rd party data center partners.

I remember the professor of my Entrepreneurship 101 class in business school started one class with this saying on the chalkboard (yes, it was still really a chalk board!): “Always remember – you can mine for gold or you can sell pickaxes.”

I thought of that quote when thinking about the edge value chain. It seems to me the edge play may be the new pickaxes for the data centers to pick up and run with in a much bigger way right now. Let their customers go hunting for the gold at the end of the rainbow. It is those customers’ customers that seem to finally be waking up to the benefits that the edge can bring.

Building the Hyper-Communicating Future

An adaption of Asimov’s Foundation series is currently airing on Apple TV. It was a good excuse for me to re-read the first book, the premise of which is basically this: by building a foundation that contains all human knowledge, a predicted thirty-thousand-year-long “dark age” can be shortened – potentially – to a mere thousand. Asimov then relates the implementation of this plan (the Seldon plan) across multiple chapters. It’s an intriguing concept that goes well beyond the merely literal spin I take here – that today’s effort and activity around building wired (and wireless/cellular) networks is foundational to our hyper-connected, hyper-communicating future.

The fiber builds happening across the U.S. are due to many factors – bridging the “digital divide,” increasing the overall availability of broadband and enabling 5G NR services, to name three. Fiber is key because it provides high throughput and scales well (particularly via C/D-WDM). Moreover, the high strand counts being deployed suggests that there will likely be a high ceiling of capacity for (hopefully) many years to come.

Some of the past week’s news include:

• Frontier entering a fiber partnership with AT&T
• In Indiana, AT&T partnering with Vanderburgh county to expand fiber broadband access
• Comcast Business expanding its network in several mid-Atlantic states
• TDS Telecom expanding its fiber network in a 7,300-strong town in Maine
• Consolidated launching gigabit broadband service to more than 6,500 homes and small businesses in Pennsylvania
• Also in Pennsylvania, Shentel began expanding the reach of its FTTH Glo Fiber network to another 6,000 homes and businesses
• Ziply Fiber securing $350 million in debt funding to continue its fiber expansion in the U.S. northwest
• Zayo entering the final phase of constructing its all-underground dark fiber routes.

Fiber is foundational to all networks, whatever the “last mile” connection is – FWA, cellular, Wi-Fi, satellite, cable, xDSL, mmWave or, ideally, fiber itself. Layered on top of these networks we’ll see all the apps we’re accustomed to (social, mapping, gaming, video calling and conferencing, messaging, etc.) along with those that are yet to come (connected vehicles, autonomous vehicles, AR/VR, smart city, etc.) and still more that no one has yet dreamed up.

Today, these are discrete networks operated by different companies and/or different units within the same companies. Over time, we expect that we (the industry and the people) will stop focusing on the individual networks and more on interconnecting the various foundations into a hyper-connected network across which everyone and everything communicates. It’s not Asimov’s Foundation. Not yet, anyway.

American Tower logo

American Tower expands data center footprint with DataSite acquisition

Source: RCR Wireless News

American Tower’s portfolio now consists of nine data centers

American Tower has closed the acquisition of data center owner/operator DataSite. The tower company will now assume responsibility for all management and operations of DataSite’s two data center locations in Atlanta, Georgia and Orlando, Florida. This deal expends the company’s current Atlanta presence and marks its entrance into Orlando.

“The acquisition of DataSite augments our Data Center portfolio. Since entering the data center market in 2019, we’ve been committed to meeting the growing demands of our customers by expanding critical connectivity services,” says Eric Watko, VP of product line management for American Tower. “These Metro Data Centers are critical to our edge strategy, as they contain the carrier hotel, providing interconnectivity access to internet exchanges and cloud services.” 

American Tower customers will have access to services across the DataSite network, including its dedicated internet access, carrier-neutral cross-connects, cloud ramps and point-to-point data center connectivity and cloud options, including WAN and SD-WAN options, according to the company.

American Tower’s portfolio now consists of nine data centers. Last year, as part of its Data Edge Center Initiative, the company launched six data edge centers in Denver and Boulder, Colorado; Pittsburgh; Atlanta; Jacksonville, Florida; and Austin, Texas.

The newly acquired data center in Orlando is strategically located to complement the Jacksonville location and will open “additional connectivity avenues in Florida” and offer “a vehicle for customers to conveniently connect to facilities in Miami that touch networks across Latin America,” said American Tower.

The Atlanta data center, located in Marietta, will allow for the extension of enterprise customer networks.

“DataSite is in good hands with American Tower,” says Jeff Burges, former president of DataSite. “They have shown a commitment to expanding in this industry and providing our customers the same level of service, with the same team they’ve come to expect. 

EQT logo

EQT Infrastructure and Stonepeak to acquire DELTA Fiber

Source: CISION

EQT and Stonepeak are pleased to announce that EQT Infrastructure V (“EQT Infrastructure”) and Stonepeak have agreed to acquire DELTA Fiber (the “Company”) from EQT Infrastructure III. Following the closing of the transaction, each party will hold a 50 percent stake in the Company and co-control DELTA Fiber through a strong industrial board.

Headquartered in Schiedam, the Netherlands, DELTA Fiber provides high-speed broadband, TV and fixed and mobile telephony to Dutch households and businesses connected to its superior fiber-to-the-home (“FTTH”) network. DELTA Fiber owns and operates approximately 50,000 km fiber-based network infrastructure that connects approximately 900,000 households and businesses across the Netherlands. The Company employs approximately 600 people and was established as DELTA Fiber in 2018, following a combination of DELTA and CAIW, which were acquired by EQT Infrastructure III in February 2017 and January 2018, respectively.

With 20,000 new connections per month, DELTA Fiber today is one of the largest and fastest growing fiber companies in the Netherlands and is on its way to reaching one million connections by the end of 2021. DELTA Fiber benefits from a rapid growth in data consumption and an increased demand for fast and stable internet. The Company’s new network rollout will be based on the latest fiber technology (XGS-PON) that enables speeds up to 10 Gbps. This is the prelude to its 25G-PON technology that enables speeds up to 25 Gbps.

DELTA Fiber will benefit from EQT’s and Stonepeak’s significant combined expertise in the digital infrastructure sector and vast track record in fiber rollout across the Netherlands and Europe. Both parties are committed to investing significantly in the continued digitalization of the Dutch society by accelerating nationwide B2C and B2B FTTH connectivity in suburban and rural areas. Moreover, the Company’s fiber broadband is more sustainable and energy efficient than the legacy networks, with approximately 40-60 percent lower energy consumption.

Together, EQT Infrastructure and Stonepeak will support DELTA Fiber and its management team in its ambition to reach a footprint of two million fiber connections by 2025, thereby covering a quarter of the country. The Company will also be supported by a strong advisory board with seasoned industry experts who possess broad expertise within digital infrastructure and FTTH rollout. 

Matthias Fackler, Partner within EQT Infrastructure’s Advisory Team, said, “We are deeply impressed by DELTA Fiber’s management and employees’ strong performance over the past few years. EQT Infrastructure is excited to support their continued journey of digitizing the Netherlands by providing high quality broadband infrastructure to Dutch households and businesses. EQT Infrastructure shares this vision with Stonepeak whose vast experience in the digital infrastructure space makes them an ideal partner to support DELTA Fiber in its next phase of evolution and growth.”

Brian McMullen, Senior Managing Director at Stonepeak, said, “Stonepeak has long recognized the mission critical nature of broadband in today’s society and we look forward to working with Marco and the team to accelerate the additional rollout of DELTA Fiber’s network across the Netherlands. We are delighted to partner with a like-minded peer in EQT Infrastructure on this transaction, which will accelerate DELTA Fiber’s ability to connect households throughout the country with reliable broadband.”

Cyrus Gentry, Managing Director at Stonepeak, added, “DELTA Fiber, with its unique asset base and industry-leading management team, represents a compelling investment opportunity that will complement Stonepeak’s existing global portfolio of residential broadband-focused platforms.”

Marco Visser, CEO of DELTA Fiber, said, “Two leading international investors joining DELTA Fiber confirms our success in recent years. That EQT is choosing to invest in our company again, together with Stonepeak, shows confidence in our ambitious plans for the future. Together they provide us with a solid foundation for further growth.”

The transaction is subject to customary conditions and approvals. It is expected to close in December 2021. With the acquisition of a stake in DELTA Fiber, EQT Infrastructure V is expected to be 60-65 percent invested (including closed and/or signed investments, announced public offers, if applicable, and less any expected syndication).

Analysis of DIGITAL – looking at the hard numbers

October 4, 2021

It’s official: investing in digital infra stocks really has beaten the big benchmark, the S&P 500, over the last five years or more! I, like many, had always made the assumption that investing in the right tower co, for instance, was a smart strategy, but would the ‘average’ investor have done so well if they’d invested in a diverse basket of digital infra stocks, i.e. would a basket of tech infrastructure stocks have outperformed, say, the S&P 500?

Until recently it was very difficult to answer this question because there was no real world basket to compare with, but a few weeks ago Digital Infrastructure Investor launched their Digital Infrastructure Investor Global Index, ticker DIGITAL. This index went live June 28th and contains 32 mostly well known constituents. I’m sure you’ll hear much more about this new equity index over the next few months, but for now I’m simply curious about what the numbers tell us. What do the hard numbers tell us about investing in tower cos, data centres and fibre operators?

The big, important number is that of returns over the medium term. Usefully, the index has been put through the trusty old time machine that is called a quantitative backtest and the numbers that have come back look impressive. Since 12th April 2015 $5000 invested in the S&P 500, for instance, would have returned $10,400 to date, whereas this index returned over $12,000.

Obviously back tests come with all sorts of expected caveats, so for me the more interesting story requires a bit more digging around in the numbers playground. In particular, what are the key technical and fundamental numbers supporting this curated assembly of digital infrastructure equity players?

To help answer this I’ve taken the top eleven constituents in the index which trade on either the US or European exchanges and then run that smaller sub set of stocks through the various sources of stock data including Bloomberg, Yahoo Finance and Sharepad in the UK.

Note that these eleven stocks do comprise well over 65 percent of the aggregate value of the index, so although they don’t give us the complete story about the index, they are nevertheless bound to dominate any narratives that do emerge. On the issue of geography, many of the longer tail of 21 stocks in the wider index are not listed in either the US or Europe, and thus underlying financial data is patchier.

So, digging around inside this shorter list of 11 (out of 32) names reveals some interesting insights, not least the fact that these stocks have collectively beaten the S&P since the middle of the last decade. In terms of longer time scales, more than a few of the businesses haven’t been around more than five to 10 years, but for the six that can boast a 15-year share price track record, the total shareholder return including dividends has been a stonking 783 percent versus 229 percent for the S&P 500.

That said, I think the most interesting insight is a technical one. The average beta across these large cap tech stocks over the last five years is a relatively lowly 0.4 (to the S&P 500), while the average correlation over the same time is also a relatively lowly 0.486. These two numbers would suggest that the big digital infrastructure names aren’t really your classic momentum players and thus they might offer some diversification benefits for an equity investor. Yet there is a price to pay – the average volatility of these top 11 stocks is double that of the benchmark S&P 500 index. So, these infra stocks potentially offer some diversification benefits but at the price of greater price volatility.

What about valuation metrics? The consensus view has long been that these stocks are collectively more than a tad overpriced. The numbers tend to back this up. The average price to earnings ratio – if there is one – is well over 100 times earnings and the average price to net asset value is a slightly eye watering 19.5 times. But before we mark this down as a win for the valuation critics, it’s also important to consider some other hugely important numbers.

The first is that the dividend yield, where dividends are paid, is a generous 2.57 percent versus 1.28 percent for the S&P 500 benchmark, and that doesn’t include any share buybacks. In addition, it is clear that these businesses are veritable cash machines, producing lots of free cashflow based off surging turnover growth. Using analysts’ estimates for the coming year, this cross section of 11 stocks is expected to grow turnover by around 20 percent in the coming year.

That in turn translates through into an average forecast earnings per share growth of 81 percent across the 11 stocks in the coming financial year. So, whilst it’s true that buying into these digital infrastructure stocks does mean paying a very substantial premium against the benchmark, you’re also on the receiving end of very strong both top and bottom-line growth which is feeding through into cash pay-outs to shareholders.

Now, many investors might still think that these valuations are aggressive (and they are in my view) and they may also point to high gearing across the sector – running in the hundreds of per cent for these 11 stocks on average. But I would suggest that for many investors the obvious growth characteristics of this space plus its collective diversification benefits might be exactly what they need to ignore these valuation concerns. And certainly, most analysts don’t seem to be too put off by the rich valuation metrics – the average broker rating across the sector is 2.2, with one representing very bullish and five very bearish. It’s also worth noting that the average short ratio is five, which doesn’t seem terrifically bearish.

Stepping back from this tsunami of numbers, a couple of points do stand out for me. The first is that these valuations do demand that the digital infrastructure giants deliver on their growth promise. A quarter or two of disappointing numbers or even a year or two of underwhelming growth could spell trouble because of those rich valuations. My sense is that given the secular demand profile of this sector, that shouldn’t be too great a risk, but given the obvious scramble to build new operating assets, there’s always an ever present risk of a supply glut which could hit margins.

The other obvious point is that high gearing is fine if growth in the bottom line – profits and free cashflow – is surging. But that situation could turn nasty if interest rates and bond yields start to rise and cashflow growth slows down. That could be a dangerous double whammy that could in turn smash those dividend payouts. On that point I’d note that the average dividend cover for these 11 stocks is a lowly 0.5, which suggests that dividends could be vulnerable if free cashflow starts to wilt.

Is Cable Ready to Spend?

The September conference cycle is never dull and always full of interesting tidbits for those who really listen to the undercurrents of what CEOs and CFOs are saying. I have found that history has shown this time and time again in 20+ years of going into “September listening mode.”

One of the key themes this past month came from the cable sector. In short – cable may be down…but they are sitting back up. Comcast stock has struggled to find its footing since September 14th when its CFO previewed that broadband adds would be less than Street estimates. Specifically he noted they have seen a slowdown since late August in Broadband adds and growth would be down from the (record) Q3’19 levels. ATUS also previewed weaker than expected Q3’21 adds – now guiding to 15K – 20K in broadband losses (with plan to finish flat / slightly up on net adds for FY’21). Its stock has also felt some pain – down approximately 25 percent since September 1st.

The question is what is driving this weakness for cable? The CMCSA CFO acknowledged the times in the competitive front are indeed a changin’. Specifically he noted: “we’ve been mindful that competition’s coming in one of our best and favorite businesses, frankly, for quite some time.”

One of the things we have often discussed in our write-ups is that telecom has very much “woken up” to the need for more fiber. When telecos were not as much focused on broadband and fiber-deep, life was very good for cable. When I was an analyst one of the things I used to say was ‘never underestimate the cable cos’. Two years later, we would continue to say this.

In the September webcasts from the cable cos there seemed to be very purposeful messaging being sent. Perhaps this was best seen by listening to the cable Godfather himself, Brian Roberts (Chairman and CEO of Comcast). When asked about the balance sheet at the Goldman Sachs conference he made a notable point. Specifically, he said: “You’re going to have a lot of free cash flow and your first goal would be to invest in the business.” This tails off some comments from the CFO two weeks prior who noted: “We’re going to step on the gas a little bit….and continue to invest in the network on the path to DOCSIS 4.0 which will give us multi-gig speeds up and down.” This was not surprising, but what was surprising was the follow up to this statement. He went on to say: “no, no, no trigger other than that’s the path we want to get our business on and with the success of trialing 4.0 technologies in the last bunch of months….its for real and we want to get ourselves on the path to doing that.”

The “no other trigger” seems hard to me. The trigger is kind of hiding in plain sight, isn’t it? In the last nine months, AT&T, Frontier, Windstream, Consolidated, Lumen, Apollo and many many other smaller players (too many to list) have been spreading the “fiberized” gospel. And this will only be entering hyper-drive mode as a massive amount of broadband infrastructure capital enters the states’ pockets.

My guess is the history books will show that this past month will indeed be a “September to Remember” because it will mark the point where cable showed they are not backing down. They may have been hit, but they are no shrinking violet and if telecom is moving, cable is saying “game on – we are ready!”