The much anticipated CBRS Auction 105 is over, and it did not look much different from any other cellular auction. Mobile operators and broadband providers paid for over 90% of the proceeds, getting the most valuable licenses. Verizon alone paid for 41% of the proceeds, but only got 3% of the licenses paying $0.30 MHz/pop, compared to an average of $0.22 MHz/pop, as we discussed in a recent Sparring Partners and in this article.
Enterprises, venues and even city agencies largely stayed out of the auction. This was the right choice and a predictable one. No enterprise, venue or city fills a county, so why get a PAL, when you can use GAA?
Utilities: a notable exception
While most non-service-providers politely declined to enter the auction, utilities were the most aggressive bidders. San Diego’s Sempra (San Diego Gas and Electric) paid the highest price at $21 million for 3 licenses, and Southern California Edison paid $118 million for 20 licenses. On a per MHz/pop, they paid more than Verizon ($0.57 and $0.51, respectively, compared to Verizon’s $0.30). The use case for utilities has always been a popular one among the CBRS crowd, but are the PALs worth that much?
Not all utilities saw the same value in the PALs. According to some excellent analysis by Burns & McDonnell, only 11 utilities (5.5% of them) got PALs amounting to 1.64% of CBRS licenses – almost a rounding error, especially if we consider the fact that they do cover the country, unlike other enterprises. I will get back to this later.
A tale of two utilities: location, location, location
Utilities like Sempra and Southern California Edison had an extremely sharp focus (see map below), defined by their service area. For utilities, PALs are valuable only if they have them across the area they serve. If a utility operates in five counties, and gets a PAL only in two of them, their value is reduced. This can explain their willingness to pay a higher MHz/pop price: this was not the metric they were using to guide their bidding strategy.
At the other end, Alabama Power was able to get a bigger footprint at a much lower price (but it still made an $18.8 million investment), presumably due to less competition in these areas. But the location focus is the same: the PAL footprint covers its Alabama footprint, plus some counties in Georgia Power’s service area (both utilities are owned by Southern Company).
For utilities location is everything. If they want to build a CBRS wide-area network, they need to cover their service area. This is different from a mobile operator, for which CBRS is a welcome add-on to other bands and not a band essential to establish a network.
Why do utilities need PALs?
Among enterprises, utilities have a unique set of features and requirements that increases the value of PALs above and beyond that to mobile operators in some areas and makes them the ideal candidate among enterprises. Utilities have a capillary network of customers, facilities and staff across the entire PAL area. A CBRS network, anchored by one or more PALs, can provide the reliable coverage that licensed access guarantees and that utilities need to provide services to their subscribers (e.g., to monitor and optimize energy consumption), but also to ensure mobile connectivity to their staff (e.g., for remote assistance during service disruption or upgrades, to manage workflow), and to remotely control and manage their infrastructure and facilities.
Because reliability and real-time operations are crucial to the quality of the service and to avoid major service disruption, utilities have a strong preference to deploy and control their own private networks, and to minimize their dependency of mobile operators which cannot grant utilities the level of control they want.
PAL access gives utilities limited capacity (each PAL is 10 MHz and each licensee can have up to 4 licenses in the same area), but its main value is enabling consistent and reliable coverage. PAL access becomes the anchor on which GAA access can be added as needed, within the same network – same access technology, same devices, same applications. For instance, within a power plant or office building we could expect a utility to use GAA access more liberally than it would in the WAN and that it would need more capacity at those locations. So, the value of a PAL is increased by the ability to use PAL and GAA in conjunction. GAA alone would not be sufficient for utilities to provide reliable coverage across their footprint.
A new way to share infrastructure, learning, and procurement leverage
Even though only a few of the utilities got licenses, there is a tantalizing opportunity for utilities in different areas – which provide similar services but do not compete with each other – to collaborate as they roll out networks and services. They could work together on procurement for instance to ensure they have access to the equipment they need and to develop the applications they need, or work with vendors to do so.
Within their service area, they can share the infrastructure with other non-competing utilities (for instance water and electricity), which may have different requirements but a similar need for good coverage across the PAL area. This approach would help utilities spread the cost of network deployment across a larger number of use cases, services and, even more importantly, funding sources. Network slicing may provide utilities the comfort they need to enter in infrastructure-sharing agreements.
The challenges of funding, planning, deploying and operating a wide-area network should not be underestimated, however. This is the likely reason many utilities stayed out of the auction. And it was the right choice for utilities that are more cautious, more cash strapped, or lack the resources they need. However, if the 11 utilities with the new PALs can show a convincing business models, others may follow, and infrastructure sharing and a secondary market for PALs may be a way for the other 189 utilities in the U.S. to get into the game. And Dish may have enough licenses and motivation to lend its support to utilities’ plans to deploy CBRS with a PAL anchor.