New regulatory filings from Dish Network indicate that the company may divest its Boost Mobile assets. Boost is the largest of Dish’s four retail wireless MVNOs, which will be migrated to its own network as that is built out and as subscribers upgrade to 5G (they currently run on AT&T and T-Mobile networks). Dish chair Charlie Ergen, and treasurer Jason Kiser, have set up a special purpose acquisition company (SPAC), currently a popular M&A instrument in the US hi-tech business, called CONX, which would acquire Boost.
Some analysts speculated that this was a sign of what skeptics have believed since Dish first invested in 4G spectrum – that Ergen has no real interest in building a long-term mobile operator business, but just wants to wait for the best time to sell spectrum and retail wireless assets. However, the spin-off of Boost would also fit well into Dish’s stated strategy of building a wholesale-only network business, in which its own retail wireless brands, of which Boost is the largest, would be customers, along with enterprise partners. Each would be enabled, by Dish’s fully cloud-based network architecture and core, to have a specially configured slice.
In that context, it would make good sense to separate a low-margin and low-value retail business from an infrastructure business with higher margins and market value, and would also reduce potential conflicts of interest between Dish’s retail brands and future wholesale customers, if these were also on the consumer side (though its main stated objective has been to build a platform for enterprises and their service providers).
If Dish is seen primarily as a wholesale, infrastructure-centric, network-as-a-service operation, splitting off Boost to maximize the value of the core business is logical, and reflects a general move among telcos to unlock value in infrastructure – though this is more commonly done by carving out towers or fiber rather than the retail operations. It would also, potentially, make it easier to build up new customer bases and maximize the revenue and ROI on the 5G build-out.
Dish has other MVNOs including Gen Mobile, Republic Wireless and Ting Mobile, but these are not mentioned in the SEC filing document, which just says that CONX will announce additional details if, and when, a definitive agreement is executed.
Dish said in a statement: “We regularly evaluate ways to enhance our business. As part of this process, we are in regular dialog with interested parties who may assist us in accomplishing our goals, including recently preliminary conversations with CONX Corp. There can be no assurance that these preliminary discussions will lead to a transaction nor as to the structure or terms of any such transaction. We do not intend to provide further updates unless and until those discussions conclude in agreement as to a transaction.”
Of course, regulatory approval would be necessary. Dish acquired Boost via a complicated transaction largely designed to appease antitrust agencies in the USA, when T-Mobile and Sprint were seeking to merge. Dish acquired Boost from Sprint, along with its customers and various network and spectrum assets, as the basis of its own 5G launch, which ensured that the number of national MNOs in the USA would not be reduced by the Sprint/TMO marriage.
Jonathan Chaplin, a Wall Street analyst with New Street Research, commented in a client note: “If the transaction is completed, it would give Boost capital to accelerate growth, in a structure where the investment doesn’t deprive Dish of capital for funding the network deployment. Boost will become the anchor tenant on Dish’s network as it gets built out. With little more than growth in MVNO payments from Boost, Dish’s fully deployed network will nearly break even; any additional wholesale or enterprise revenues would be gravy.”
He added: “Giving Boost capital to grow through this structure increases the odds they will be successful, which in turn increases the prospects for the success of networks. Though splitting off Boost cuts Dish off from an easy source of cashflows, it should improve visibility into the margin profile of the underlying network infrastructure business; this could in turn improve Dish’s access to capital.”
A deal would also somewhat distance Dish from the travails of the MVNO business, especially in the low cost markets that Boost targets. Its business model for the networks business would, if successful, transform its potential margins, growth and position in the value chain, and there is likely to be significant demand for enterprise 5G capacity on an as-a-service basis, as some enterprise providers start to look for more flexible alternatives to the current widespread build-out of dedicated private networks. Those are an expensive option for all but the largest companies, and so a cloud-based, slicing-focused approach would be welcome.
However, it must be noted that there have been failed attempts at this wholesale-only approach in the USA before (albeit without the benefits of cloud-based RAN and core or 5G) – Clearwire and Ligado (formerly LightSquared) among them. And the new business model will take some years to gain scale, as Dish waits for market demand to emerge and mature, and for its network to be fully built out, and supported by developers and enterprise trials.
In the meantime, it is reliant on its main, but ailing, pay-TV business, and on its retail wireless units, for ongoing revenue and to reduce the amount of finance it needs to raise to continue its 5G roll-out (however capex-efficient it claims that will be because of its Open RAN, cloud-native architecture). Boost has been shedding users – in the most recent quarter, it lost another 210,000 net subscribers, bringing its total to 7.87m, and the MVNO has lost 1.1m wireless customers since the acquisition by Dish. Recently, Dish has introduced plans to drive Boost upmarket with its first postpaid option, Boost Infinite.
Ergen was asked during Dish’s Q2 earnings call in August why he didn’t divest Boost, and he acknowledged that the MVNO is not essential to the 5G business case, which derives its value from the network assets themselves. “I guess the answer is it’s not a necessity but today we prefer that it belongs with us,” he said.
Another potential new element in Dish’s future 5G business model might be a far more powerful anchor tenant than Boost. A wholesale-only wireless business is hard to build from scratch without a significant tenant to get predictable revenues flowing and instil confidence. Various possible partners have been mooted since Dish unveiled its plans, but the most likely is Amazon, which has conducted joint trials with Dish, as well as several other initiatives to build connectivity in its own spectrum, or in CBRS shared spectrum, to support a fully smart logistics and IoT operation across its vast US delivery network, and potentially to gain improved control of connectivity to the AWS cloud for enterprise customers.
Speculation of a strategic alliance between the two companies was only intensified in April 2021 when Dish said it would implement some of its 5G core and RAN functions in the AWS public cloud (though the more demanding Open RAN distributed unit functions are mainly deployed in Dish’s own Dell/VMware-based edge cloud). AWS claims that its platform delivers, on average, an 80% reduction in deployment and update times for cloud-native functions “when compared with traditional deployment models (days to minutes)”, and that would be essential to Dish’s aim to deploy a far more cost-efficient network than its most established 5G rivals.
Dish will make heavy use of AWS Breakout Edge Data Centers (BEDCs) for its highly distributed RAN. These BEDCs are deployed in AWS Local Zones and will host the network functions that need particularly low latency, and “also provide Internet peering for general 5G data service and enterprise customer-specific private network service”. The latter is important given the wholesale-only model, which aims to support enterprise network operators and enterprise slices.
The BEDCs will host the centralized units (CU) within the disaggregated Open RAN as well as the user plane function (UPF) that “anchors user data sessions and routes to the Internet. The BEDCs leverage local Internet access available in AWS Local Zones, which allows for a better user experience while optimizing network traffic utilization. This type of edge capability also enables Dish enterprise customers and end users (gamers, streaming media and other applications) to take full advantage of 5G speeds with minimal latency. Dish has access to 16 Local Zones across the US and is continuing to expand.”
Dish’s Local Data Centers (LDCs) and individual cell sites will send traffic into Passthrough Edge Data Centers (PEDC), which will then connect to Breakout Edge Data Centers (BEDCs) that sit in an AWS Local Zone.
At a more centralized level, Dish’s network will also use AWS Regional Data Centers (RDCs) and National Data Centers (NDCs). Three RDCs are deployed in an AWS compute ‘Region’ and each in a separate Availability Zone (AZ) for redundancy.
The RDCs will host Dish’s signaling processes, such as authentication and session management, and voice calling services. The NDCs will host the subscriber database, IMS, and OSS/BSS, while the AWS Direct Connect service will link the RAN’s distributed units (DUs) with AWS Local Zones, promising sub-10ms midhaul connectivity between the BEDC and the on-premise DU, which may be located at a cell site or central office, often using the Dell platform.