Miguel Dajer, Senior Vice President, Futurewei Technologies, USA
Published: 13 Apr 2021
CTN Issue: March 2021
A note from the editor:
This month we examine the recent US C band auction and look for some winners and losers. Certainly, it made a bunch of money for the US government and is seen as a major boost for 5G in the US. But how did the operators fare and was it worth all of that money? Our guide through the maze of spectrum allocation this month, Miguel Dajer, explains the game, identifies the players and awards grade points for effort and success.
We are excited to announce that as well as untangling spectrum policy for us, Miguel will also be bravely stepping into the position of Editor in Chief for CTN. Your poor current EIC, having spent too much time in the bright glare of online content editing, will now quietly retire to the IEEE luxury retirement home for old editors (we know you always suspected this is how we spent your membership dollars). As always, your feedback is most welcome.
Alan Gatherer, CTN EIC Emeritus
C-Band Spectrum Auction, Was It an A or an F?
Although 5G builds on existing 4G infrastructure, 5G networks deployed at scale will require a complete redesign of communications infrastructure. Industry experts generally agree it may take a decade to completely roll out 5G networks and to realize their full value through the Internet of Things, automated driving, telemedicine, artificial intelligence and virtual and augmented reality. So, we are in the very early innings of the 5G game. With this background the big issue that will make or break the future of many 5G services is spectrum. 5G Standardization defines two sets of spectrum “ranges”, in standards lingo we call them FR1 and FR2 or low spectrum generally agreed to be below 6 GHz and high spectrum generally to be in the 27+ GHz, what the industry is generally calling millimeter Wave (mmW) spectrum.
Before delving too deeply in to the spectrum auctions topic lets provide a general overview of the spectrum definitions of FR1 and FR2 and the bands being discussed. Figure 1 shows the general definition of FR1 and FR2 bands as per 3GPP R16. A more broad definition sets FR1 at “below 6 GHz” and FR2 at “above 6 GHz”. In reality the spectrum is well defined by standards organizations.
Within FR1 there are 45 defined bands, many of them used for legacy services like 2G, 3G and of course 4G/LTE but there’s also spectrum reserved for unlicensed operation for Wi-Fi as an example. Within the FR1 block there’s further division of spectrum, in general terms these are defined as low band and mid-band frequencies (figure 2) where mid-band is the focus of new auctions in the US and globally. Every 3-4 years the world comes together to try to harmonize spectrum, the last one called WRC-19 or World Radiocommunications Conference provided the latest agreement on the bands being defined for 5G, the next meeting is in 2023: WRC-23.
Source: With Thanks to Stephen Wilkus of Spectrum Financial Partners, LLC
Both CBRS and C-band have been auctioned recently, with CBRS receiving smaller total bids given the shared nature of the spectrum as can be seen in the figure. On the other hand, the recent auction of C-band included 280MHz of licensed dedicated “mid-band” spectrum, much more valuable to operators. For completeness, band 41 which sits in the 2.5 GHz portion of the spectrum (lower mid-band or higher low band?) covers 180 MHz and is mostly owned by T-mobile.
Globally, regulators have been rushing to find ways to clear spectrum. China and Korea have led but the US and other EU countries have quickly taken action on this front. Originally US operators focused on mmW only to realize that this spectrum is more suitable for some very specific use cases given the propagation and deployment challenges that it imposes. As such, lower band spectrum continues to be highly desirable. I remember attending a conference a few years back where there was a discussion on the issue of spectrum for the upcoming 5G systems. While everyone on the panel was gung-ho about the possibilities of huge bandwidths in the mmW bands, the CTO at that time for Tele2, Joachim Horn, took a contrarian position saying (paraphrasing): “I will be happy getting all of your sub 1Ghz spectrum to deploy my networks”. Of course, C-band is not sub-1Ghz but the point is that lower band spectrum is still the foundation for broadband wireless services because of its inherent propagation characteristics.
In the US the C-band auction drew over $80B in total bids, as expected Verizon and AT&T dominated with a combined total of approximately $68B, Verizon alone an estimated $53 B when factoring in all the spectrum clearing and ancillary costs. T-mobile only spent around $9B, a validation of its Sprint acquisition which included a large swath of nationwide spectrum in the 2.5GHz band (band 41). The prior higher auction bid was for the AWS-3 band which fetched $43B. In order to compare the apparent higher cost of C-band Vs AWS-3 we need to look more closely at the price paid per MHz/pop (price per one MHz of bandwidth passing one person in the coverage area of the license). This is the metric most used when assessing cost of spectrum. In the case of AWS-3, this cost was USD $2.20, whereas C-band sits at around USD $0.95. Figure 3 provides a complete view of the cost of the different spectrums auctioned over the years in the US.
So, is the C-band spectrum auction an A or an F for the operators and the US Government (USG)?
Let’s start with the USG. In most analysts’ perspectives the auction was a massive success from the USG. The auction had many roadblocks from the start, but the FCC took pro-active action and got the process off-the-ground. Now that it is completed the proceeds can be used to support other communications related initiatives. I give the USG an A for their efforts and positive results.
How about from the operator’s perspective? It depends on who you are as an operator and also depends on who is analyzing the results. There are many opinions as analysts debate how good (or not) the combined spectrum assets of different companies, post C-band auction, are. Below I post a couple of opinions from industry experts to make the point.
In her article in Forbes “C-Band Spectrum Is Not The Beachfront; It’s Affordable Housing” , Roslyn Layton makes the argument that C-band will make high quality service available to pretty much every one in the US and provide true 5G capabilities along with it. In her opinion, C-band is beachfront because it provides this broad access and can provide a windfall to operators. Although I somewhat agree with her perspective, I believe that from the operator’s point of view this spectrum is perhaps affordable housing. Beachfront spectrum should be the best possible spectrum available for the service, in the case of C-Band Vs Band 41 as an example, I would argue one if better than then other. So, who is sitting on Beachfront, T-mobile, AT&T or Verizon?
Another Forbes contributor, Bob O’Donell, on his article titled “C-Band Auction Points To Dramatic Shift In 5G” writes: “… you don’t spend this kind of money unless you intend it to be the absolute core part of your network. Based on the results of this auction, I’d argue that it is now abundantly clear that the future of 5G in the US (as it is in other countries) is in mid-band.
That’s great news for T-Mobile, who specifically purchased Sprint to get access to the 2.5 MHz mid-band frequencies that it had previously licensed and, as a result, is the only US carrier with currently deployed mid-band frequency holdings. The news is clearly less good for both Verizon and AT&T, but it also explains why those companies likely spent so much more on the C-Band auction to get access to those frequencies—they really need them. It also explains why Dish believes it can make a go as a mid-band 5G-only carrier.”
I agree with the assessment from Mr. O’Donnell: 1) The future of 5G is really in mid-band spectrum (given the need from large bandwidth and reasonable coverage), 2) Agree that T-Mobile sits on great spectrum Vs competitors and 3) 5G is real and operators will build networks that will yield terrific services over time, including mmW related applications. The other good news is that after so much money spent on spectrum, this spectrum needs to be deployed and monetized. This means lots of capex for the development of new services, which in turn means opportunities abound for equipment manufacturers. In the end consumers and enterprises alike will benefit from this bounty.
So, as we attempt to grade the result of the Auction, in my opinion everyone passed with at least a “C”, an operator got an A, another one a B and a couple got a C. All in all, everyone now has sufficient spectrum to deploy a very capable 5G network. Who actually got the grades? Well, I have my opinion and I’m sure the reader has its own too.
Statements and opinions given in a work published by the IEEE or the IEEE Communications Society are the expressions of the author(s). Responsibility for the content of published articles rests upon the authors(s), not IEEE nor the IEEE Communications Society.