RESEARCH

Regulators and investors are making a mistake in their analysis of Dish

We published a report this week showing that if Dish deploys a network, they will have the capacity to take at least 15% of the market’s traffic, and an incentive to price aggressively to do so (LINK).  We argue that their unit cost would be so low that they could sell capacity in the wholesale market well below current wholesale rates (below even the existing carriers’ unit costs).  This could open the market to competition that would dwarf anything we have seen from Sprint in a decade.

A neutral-host wholesale capacity platform does more than just drive competition in the retail wireless market though; it paves the way for innovation in new business models that wouldn’t otherwise be possible.   Companies across industries will be able to purchase wireless network capacity, either as an input to, or bundled with, a product or service they sell to consumers or enterprises.

It would also drive investment: the existing carriers would have to accelerate investments in fiber, spectrum and 5G equipment, and they would have to innovate, in order to remain competitive with a new network with a big cost advantage.

Consumers would benefit directly from faster, better networks, more choices, and lower wireless prices.  Companies with products or services that are delivered over a network benefit for the same reasons; the value of their businesses should rise as connectivity costs fall.

So, what is the problem with the T-Mobile / Sprint deal?

If Dish can propel the kind of competition we describe here, the market doesn’t need Sprint to keep prices low.  Sprint doesn’t seem to be having much impact anyway; they offer service at half the market rate, and they can’t hang onto subs.  The big guys are taking price up.

In fact, if Sprint getting acquired facilitates Dish building a network, we would argue that Sprint brings much greater benefit to consumers by disappearing than by staying independent and impotent.

The market doesn’t seem to share this view; neither do regulators.

If everyone agreed with us, there would be 13 state AGs enjoying a more relaxed summer, Dish and T-Mobile’s stock would be much higher (see Slide 6 for why T-Mobile is still a winner in this scenario: LINK), and AT&T and Verizon’s stock would be much lower.

There are three reasons why investors and regulators might disagree; they might believe:

  1. Our analysis of network economics is wrong
  2. Dish may choose not to build a network (“look at their record!”)
  3. Dish may try to build a network, but it is hard, and they may just fail

Taking these objections one at a time:

Objection #1: If Dish builds a network, their unit cost will be low, and they will have a strong incentive to price disruptively

We tackled this comprehensively in our report (LINK).  The inputs are simple.  We know what it would cost Dish to build and run a network ($18BN to build; $6BN annually to run), and we know how much capacity they would have to sell, so we know what their unit cost would be once the network is filled.

We know the unit costs for the four national carriers too.  Dish’s unit cost is 50% to as much as 75% lower than the established carriers.  This piece shouldn’t be controversial.

Dish has a strong incentive to price well below the market too.  Once the network is built, they have $6BN in annual fixed cost to support.  And the longer they take to cover their fixed costs the more cash they will burn, and the more capital they will need.  The marginal cost of each GB is $0.  Every GB they sell is profitable. No matter what they charge.  They will want to create an avalanche of demand to fill the network quickly, and the way to do that is to price low.  Disruptively low.

Given that they can have a business worth $100BN[fnote]Our estimate is actually $93BN, but rounding to the nearest $100BN, it is $100BN.  And a few more pennies in price per GB would get you there anyway.[/fnote] selling capacity at prices that are close to 90% below current retail rates and 75% below current wholesale rates, we would posit that they have a very strong incentive to price at levels that would be wildly disruptive…if they build a network.

So, the real question is will they?

Objection #2: we know Dish intends to build a network and launch a business

We know this almost beyond reasonable doubt.

We know it for three reasons:

  1. The value of the business could be $100BN (it could even be much more)
  2. The consequences of not building could be ruinous
  3. They have given up their best alternative to building a network (which would be selling)

The carrot: $100BN in value (at least)

It is best to read the report to understand the value of the business (LINK).  The mechanics are simple; we touched on them above.  Once you know how much capacity they have an what their fixed cost is, you can make up prices per GB, and cash flow and valuation fall out of the model.  At $0.75 / GB, you have a business worth $100BN (Dish’s market cap is $18BN today).

This assumes they sell all their capacity at that $0.75, but with retail prices at $6 and wholesale prices at $3, they ought to be able to sell all their capacity at $0.75.

The stick: if they don’t build, they could lose everything

Dish agreed to a new set of buildout requirements with the FCC as part of the deal.  The buildout requirements give the company more time to build a network, but they are also stricter, they impose harsher consequences for failing to meet the new deadlines, and they take away flexibility to pursue options other than building a network.

The new deal stipulates exactly what kind of a network Dish needs to build, down to the technology they must use, the number of cell sites, and amount of spectrum they have to deploy (at a minimum).  These stipulations require a much larger investment than Dish had to make under the old deal - $10BN vs $1BN, according to Dish.

The new deal also comes with $2.2BN in financial penalties if Dish fails to comply.  If Dish does nothing and loses the spectrum in 2023 and they have to fork over $2.2BN, and it all has to be paid for by a declining DBS business with too much debt, it could push the company over the edge.  Ergen could lose everything.

He would never let this happen.

The even bigger stick: they have given up the option of selling

The best alternative to deploying a network is selling the spectrum (many would argue that selling is far preferable to building).  The most likely buyers have always been one of the three successful national carriers.  Dish has effectively relinquished their ability to sell to one of the national carriers by entering this deal.

Some of our clients argue that Dish could still sell to Verizon.  The DOJ may allow the T-Mobile deal on condition that Dish is established as a new fourth carrier, and still allow Verizon to acquire the “new fourth” carrier on the basis that there is no material increase in market concentration.  This strikes us as unlikely.  We are fairly certain that Dish isn’t counting on getting away with a sale to Verizon.

If Dish wanted to keep the option of a sale to a carrier open, they could have easily not done this deal.  It would have probably resulted in T-Mobile’s deal being blocked.  This would have created at least one bidder for Dish’s spectrum (T-Mobile), and possibly more.

There could be other buyers for the spectrum from outside the industry, but we doubt Dish is counting on this either.  And as far as investors and regulators are concerned, if Dish’s spectrum ended up in the hands of someone other than the national carriers, like Cable, or Amazon, the impact to competition and pricing would be the same as if Dish launched a network, and perhaps more so.

Dish doesn’t just have an incentive to build; they are compelled to build.  If this is true, all the risk lies in them trying to build a network and failing (not from not trying).

Objection #3: Dish could fail; wireless is hard; but there are things that they and the regulators can do that would lower the risk of failure tremendously

If we accept that a new network would be disruptive, and Dish intends to deploy a disruptive network, regulators and investors should really be focused on the prospects of Dish succeeding.  We think there are four categories of factors that will determine Dish’s success:

  1. Capital
  2. A strong team
  3. The right partners
  4. Execution

We think the first three are connected, and easily solved.  If we are right about $100BN in value, Dish should be able to raise the capital they need.  If they have the capital and they are willing to invest, they ought to be able to assemble a strong team and acquire the right partners.

The real risk is execution.  There are never any guarantees with execution.  The best occasionally stumble.  This is the single most important issue…perhaps the only issue that regulators and investors should be focused on.

For investors, if Dish can execute, they will create a good deal of value for themselves, and destroy a good deal of value for Verizon and AT&T.  For regulators, if Dish can execute, consumer’s prices will fall, and investment and innovation will rise.  And if Dish fails to execute, the opposite is true.

Which brings us to what Dish and regulators can do to mitigate the execution risk…

Network share!

T-Mobile will be deploying new spectrum on 85,000 sites as they integrate their two spectrum portfolios onto a single network (assuming the deal is approved).  It would be relatively easy for them to deploy equipment for Dish at the same time.  The companies could split the deployment costs.  Both companies save money.  And Dish would get a network up and running far faster than if they deployed one independently.

Dish mgmt. suggested that they didn’t want to network share because they are planning a different network architecture than T-Mobile.  Having T-Mobile deploy the network for them shouldn’t interfere with this though.  Dish can still deploy a virtualized network with T-Mobile’s help.

We estimate that Dish needs 29,000 towers to meet their 70% POP coverage obligation, and 56,000 towers for nationwide coverage (see Slide 22 LINK).  We would imagine Dish can find a tower configuration among the 85,000 towers that T-Mobile will retain that would suit their needs.  Radios and antennas still need to be deployed on towers for a virtualized network.  The crew deploying equipment for T-Mobile can deploy radios and antennae for Dish.

Dish may want to deploy base stations in data centers rather than at the base of towers (the virtualized bit).  This is where their network architecture will diverge from T-Mobile’s.  T-Mobile could still do this for them (at a price, of course).  Or Dish could do it themselves.  Or they could find another partner for this piece (Amazon?).  Either way, having T-Mobile deploy the equipment at the towers would massively de-risk the deployment, while accelerating it and cutting its cost.

Think about what this means for the regulators!  They don’t have to wait in hopeful anticipation until 2023 to see if Dish deploys the network, and then figure out how to extract cash or licenses (of both) from them if they fall short.  The deployment will be in T-Mobile’s hands.  It will be mechanical.  T-Mobile could probably deliver a site-by-site deployment schedule to them by the time the deal closes.

T-Mobile can take on the management of the network too (at a price, of course).  It is unclear whether this will be necessary.  Dish may be perfectly capable of doing it themselves.  But there may be an incentive for both companies to do this (lower costs), and if the regulators are worried about Dish’s ability to manage the network, they can eliminate this worry by having T-Mobile do it.

The benefit for competition is powerful

Two networks with 400MHz of spectrum sharing passive infrastructure, would drive the unit cost for those networks lower still.  This might be the most capacity-rich network set-up, with the lowest unit cost, on the planet.  There would be at least two retail companies, and possible many more, feeding off that low unit cost.  That ought to guarantee low prices for consumers.

It ought to guarantee accelerated investment from the incumbents too.   And innovation from different kinds of companies that will have cheap access to managed wireless capacity for the first time ever.  Low prices; investment; innovation – that is the trifecta!

This is a complex undertaking, to be sure.  There is a tremendous amount of detail than need to be worked out.  But there are at least a dozen templates for these deals across the planet (DT should be able to get their hands on a few), and the companies have four months to figure it out.

If T-Mobile and Dish figure it out, what does it mean for the rest of the industry?

Well, we think Dish can be successful without any help from T-Mobile, but the market clearly doesn’t.  If the prospect of Dish being successful is demonstrably higher, because much of the execution risk has been removed, then investors will start thinking about winners and losers differently.  The consequences ought to be obvious.

Dish is a winner.  We have penciled out a $100BN business for them.  Some have argued that they may be a loser because it will take time to build the business (financial asset; operating asset).  We think that is drivel.  But we will save that argument for next week.  Even if it were true, it would be a statement about timing, not outcome.

T-Mobile is a winner.  This will be more controversial given that the entire wireless industry will lose value if pricing falls.  With Sprint’s spectrum, T-Mobile can create far more value with share gains than they will lose to price pressure (see Slide 6 for more detail LINK).

Cable is a winner.  As are all other MVNOs, and companies purchasing wireless capacity today, and any that might want to in the future.  Their costs go down.  The value of their business goes up.

Spectrum and fiber are winners.  Verizon and AT&T will need to invest aggressively in both to remain competitive.  They will face a tough fight for spectrum.  Dish and T-Mobile will have businesses whose advantage is driven by having more spectrum riding on a fixed cost network.  They need to get their fair share of all new spectrum to maintain this advantage.  And cable wants spectrum.  Five real bidders.  This is good for C-Band (Intelsat & SES).  And good for all the other spectrum sources that the market has forgotten about, like CBRS, L-Band (Ligado), and Globalstar.

Verizon and AT&T are losers.  I think everyone has underestimated how disruptive T-Mobile will be if they get the deal done.  If Dish builds a low-cost network, as we think they will, it will feed a more aggressive retail wireless business at Boost.  It will also feed Cable and a host of other MVNOs (existing and new).  Verizon and AT&T will lose share and face price pressure, and they will have to double down on spectrum and fiber at the same time.  Rough times for FCF and EPS.

By the way, T-Mobile isn’t the only possible network partner for Dish.  Verizon and AT&T could do it just as capably, and they have an incentive to do it.  They might get preferential access to much needed capacity.  And cash payments from Dish would help soften the blow from all the disruption.

The remainder of the Global Weekly Review can be found HERE along with all past and future reports.  Our updated valuation comp sheets can be found HERE.


Full 12-month historical recommendation changes are available on request

Reports produced by New Street Research LLP, 18th Floor, 100 Bishopsgate, London, EC2N 4AG. Tel: +44 20 7375 9111.

New Street Research LLP is authorised and regulated in the UK by the Financial Conduct Authority and is registered in the United States with the Securities and Exchange Commission as a foreign investment adviser.

Regulatory Disclosures: This research is directed only at persons classified as Professional Clients under the rules of the Financial Conduct Authority (‘FCA’), and must not be re-distributed to Retail Clients as defined in the rules of the FCA.

This research is for our clients only. It is based on current public information which we consider reliable, but we do not represent that it is accurate or complete, and it should not be relied on as such. We seek to update our research as appropriate, but various regulations may prevent us from doing so. Most of our reports are published at irregular intervals as appropriate in the analyst's judgment. This research is not an offer to sell or the solicitation of an offer to buy any security in any jurisdiction where such an offer or solicitation would be illegal. It does not constitute a personal recommendation or take into account the particular investment objectives, financial situations, or needs of individual clients.

All our research reports are disseminated and available to all clients simultaneously through electronic publication to our website.

New Street Research LLC is neither a registered investment advisor nor a broker/dealer. Subscribers and/or readers are advised that the information contained in this report is not to be construed or relied upon as investment, tax planning, accounting and/or legal advice, nor is it to be construed in any way as a recommendation to buy or sell any security or any other form of investment. All opinions, analyses and information contained herein is based upon sources believed to be reliable and is written in good faith, but no representation or warranty of any kind, express or implied, is made herein concerning any investment, tax, accounting and/or legal matter or the accuracy, completeness, correctness, timeliness and/or appropriateness of any of the information contained herein. Subscribers and/or readers are further advised that the Company does not necessarily update the information and/or opinions set forth in this and/or any subsequent version of this report. Readers are urged to consult with their own independent professional advisors with respect to any matter herein. All information contained herein and/or this website should be independently verified.

All research is issued under the regulatory oversight of New Street Research LLP.

Copyright © New Street Research LLP

No part of this material may be copied, photocopied or duplicated in any form by any means or redistributed without the prior written consent of New Street Research LLP.