Will Wall St. Put The Kibosh On The AT&T/T-Mo Takeover Before the DoJ Does?

The more I see AT&T frantically spend money like water and call in every political chip it has to try to pressure the Department of Justice to settle its case, the more I become convinced that it will ultimately be the Wall St. financial community that will finally persuade Randal  Stephenson to give up before AT&T gets to trial. Oh, I expect to see more wild gyrations. There’s perpetual whispers that AT&T will find a dance partner in the form of MetroPCS (the current favorite of the rumor mongers) or Leap or U.S. Cellular (one even occasionally hears Sprint, but that doesn’t even pass the laugh test) and they will publicly announce some big proposed settlement so that AT&T’s political friends and its cadre of honest politicians can howl some more for DoJ to settle. Who knows? We have five months until trial, and AT&T seems infinitely capable of making all sorts of political noise.

But the more I look at it, the more convinced I become that the upper management at AT&T and that of T-Mobile’s parent, Deutsche Telekom (DT), have not really thought through just what kind of a settlement they would now have to offer and how radically different it is from what AT&T expected to offer before DoJ brought suit. A settlement now is far, far more expensive than anything AT&T envisioned and quietly vetted with Wall St. analysts back in March. Back then, AT&T expected to divest from 30-50 midsized markets via a divestiture trust (allowing them to sell licenses at profit-maximizing prices over time), some wussy roaming and deployment conditions that could be easily evaded or ignored. Now, AT&T will need to divest enough to create a “T-Mo Lite,” something that can at least pretend to replace the loss of a national carrier. As I explain below, that becomes so expensive and complicated that even if AT&T can find the financing to make it happen, its stock is likely to tank on the mere announcement of such a deal.

Mind you, I am not saying a settlement is desirable or good policy. I continue to believe that AT&T’s take over of T-Mobile is so thoroughly awful as a mater of both antitrust and telecom policy that no conditions or divestitures can save it. But even discounting my opinion on the matter, there are certain practical realities that make a settlement at this point not merely bad policy, but so expensive and complicated to manage that it is effectively impossible.

If I’m right, the only question is how much shareholder money and political capital AT&T spends lobbying for a settlement that can’t be done for financial reasons before enough officers on the AT&T and DT Boards sit down AT&T CEO Randal Stephenson and DT CEO Rene Obermann and explain to them that the time has come to face reality, renegotiate the break up fee to let DT out early, and cut their loses before AT&T stock starts to tank big time.

I demonstrate why below. Warning, as this is a “show your work” thing, it’s kinda long . . .

Let us conduct the following thought experiment to create AT&T’s “ideal case,” on the grounds that if even the ideal case for an AT&T settlement is too expensive for AT&T to finance or Wall St. to swallow, then it is certainly true for the real world. Assume the Mighty AT&T Lobbying Machine performs entirely as advertised. Bill Daley, Valerie Jarrett, David Plouffe, and whoever else the cynical keep saying sagely will weigh in as a favor to Communications Workers of America (CWA) or AT&T, make a highly illegal call to Eric Holder on their behalf. Holder promptly unrecuses himself and orders Sharis Pozen to reverse herself, which she totally does, and there is no backlash or political stink from meddling in an enforcement action (as I say, this is a fantasy thought experiment). We will also assume the FCC is utterly browbeaten into submission by AT&T/CWA’s allies in Congress and, just for good measure, two former AT&T employees are nominated and confirmed to replace Copps and Baker. Every single possible political wheel is so greased in favor of AT&T that there are skid marks all over Washington.

Even assuming all this, any settlement must meet certain parameters. First, no matter how politically greased things are, a settlement must at least appear to address the concerns raised in the complaint. More importantly, Judge Huevelle must review any settlement under the Tunney Act. A settlement so favorable to AT&T that it doesn’t even pretend to address the DoJ’s concerns as outlined in the complaint simply does not pass muster, despite the broad discretion parties generally enjoy. Heck, Congress passed the Tunney Act precisely to prevent the kind of politically-based resolution of enforcement lawsuits we’re talking about here.

What AT&T Had Expected Prior To DoJ Filing

The gentle reader may ask: “Surely AT&T knew going into this it would need to make some divestitures and accept some conditions. Why can’t it just force a settlement along the terms it previously expected?”

I must confess, I think one of AT&T’s biggest strategic blunders was the failure to put conditions on the table given that they fully expected they would need to make some concessions. Comcast used that tactic very effectively in its acquisition of NBC. By putting a comprehensive settlement package on the table, Comcast converted the resistance to the merger largely into a discussion of price. It also gave Comcast’s supporters something definite to point to as real conditions that ensured the merger would serve the public interest and address anticompetitive concerns. Verizon used a similar tactic in its acquisition of Alltel. From Day 1, Verizon controlled the discussion by putting a serious package of conditions addressing a range of concerns on the table.

But, as I noted in a post some months back, AT&T not only failed to put a serious package of conditions on the table, it actively opposed binding conditions in its FCC filings even while publicly giving the impression it would accept conditions. This not only deprived AT&T of a valuable rhetorical tool, it seriously hurt AT&T’s credibility with regulators and undermined the arguments of its allies. Given AT&T’s abysmal reputation for non-compliance (to take just one example, in 2009 AT&T settled contempt charges for its willful violation of conditions imposed when it acquired Dobson wireless), playing cute with ‘voluntary commitments’ and no offer of actually enforceable conditions simply confirmed for anyone who cared at DoJ that AT&T had no intention of keeping any behavioral conditions and had no interest in making the necessary divestitures.

Still, in its heart of hearts, AT&T clearly had an ultimate strategy for divestment and conditions. While I am not privy to AT&T’s inner workings, I would wager that the strategy followed the same model as Verizon/Altell. Agree to maintain T-Mo roaming agreements for some term of years, reduce the amount of Universal Service Fund subsidy, and divest from a set of mid-size markets.

Unfortunately, AT&T’s settlement strategy relied on DoJ (followed by the FCC) accepting three basic premises of AT&T’s argument. First, DoJ would need to maintain the relevant market definition as local and ignore all impact on the national market (or accept that any impact on the national market was cured with relatively modest roaming conditions). But even that would not be enough, given that traditional concentration metrics show dangerous levels of concentration in 97 of the top 100 markets. Accordingly, DoJ would also have had to accept that: (a) concentration in the largest markets did not matter because of the presence of potential competitors and because of the “efficiencies” of combining AT&T and T-Mo in areas of spectrum constraint; and, (b) concentration in rural markets did not matter because AT&T needed these “efficiencies” to deploy 4G in rural areas.

If DoJ had accepted these as the basis for negotiation, it would have left AT&T and DoJ haggling over divestitures in a bunch of mid-size markets like Salt Lake City and Austin while ignoring divestitures in places like New York or LA where it might really hurt AT&T. It also would have limited the number of markets for divestiture to something not too expensive for AT&T. Further, even after DoJ and AT&T agreed on the right list of markets, AT&T would have had plenty of time to sell them off gradually to the highest bidder. The usual way of handling license divestitures involves putting them in a “divestiture trust” over which AT&T in theory exercises no managerial control, but allows AT&T to sell the licenses over time in a way that maximizes profit to AT&T.

Unfortunately for AT&T, that is no longer even a vaguely possible outcome no matter how much political pressure it brings to bear.

How The DoJ Filing Changes Everything

DoJ’s antitrust complaint rejected all of these premises. DoJ argues, quite forcefully, that the national market matters and that concentration in local markets (including the top markets) also matters. Most devastating of all for a settlement that ignores the traditional concentration metrics, the complaint states that the parties “cannot prove their claims of efficiencies,” so there is no reason (according to the complaint) to ignore all the traditional danger signs in the top markets.

Any settlement must make a nod to addressing the concerns in the complaint. Even with AT&T supposedly writing the script for Bill Daley and CWA writing the script for Valerie Jarrett and David Plouffe and all of them putting the screws on Eric Holder to give AT&T everything its little corporate heart desires, the Tunney Act requires Judge Huevelle to review any proposed settlement to ensure it is in the public interest. No judge is going to sign off on a settlement that just abandons the entire basis of the complaint and says: “Sorry judge, we meant to say the market is ‘local’ not ‘national’ and please forget all that stuff we said about T-Mo being a maverick firm and how local concentration matters.” So any proposed settlement must at least pretend to address the concerns in the Complaint.

That means that even at the offer stage, AT&T needs to propose something far more expensive than what it promised Wall St. analysts in private last March. Anything less than a “serious” settlement offer that goes to the concerns in the complaint has no political value — especially if the rumors that AT&T plans to announce some proposed “serious” settlement publicly as a pressure tactic. Sure, AT&T’s cadre of honest politicians and chorus of supporters will applaud and bark at the Moon about how ‘fair’ and ‘reasonable’ it is whatever AT&T proposes. But that won’t create any real political pressure of the kind that could move DoJ, since DoJ and merger opponents will simply note that such a proposal could never pass muster under the Tunney Act review.

To address the Complaint, any “serious” settlement offer from AT&T must construct a new national competitor, a “T-Mobile Lite” if you will. It must also make divestitures in the top 20 markets, not merely the middle markets, and offer roaming conditions in the major markets so open they might as well be wholesale. As an added bonus, under the new data roaming rules, whatever roaming deal AT&T gives its “T-Mobile Lite” it needs to give everyone else as well. Finally, keep in mind AT&T can’t just divest licenses. They will actually have to give up market share by transferring customer accounts to the new T-Mobile Lite.

Assuming AT&T can even get to that point (and, as I explain below, there are serious hurdles before getting even that far), Wall St. will have conniptions.

I pause to stress again that I am not saying such a proposal would actually be a serious settlement offer worthy of consideration. As I have repeated on numerous occasions, there is no set of conditions that can fix this deal, given the current concentration in the marketplace and AT&T’s legendary history of ignoring behavioral conditions. That’s why I put “serious” in “quotes.” This is a thought experiment envisioning the least costly settlement AT&T could hope to get at this point, despite its lobbying power. To maintain even vague plausibility, it will have to fit the parameters described above. As a consequence, actually constructing a settlement creates some real world problems.

Problem 1: Find A Dance Partner

First, AT&T needs to find a partner to build into the new T-Mobile Lite. It can’t be Sprint, even if Sprint were interested, because: (a) Sprint is too close to being a real competitor; and, (b) that is still going from 4 to 3 national competitors. While a strong 3rd competitor is still better than a de facto duopoly and a bunch of also rans, it does not address the Complaint’s concern that the number of national competitors will go from 4 to 3.

That leaves MetroPCS, Leap, and U.S. Cellular as the possible dance partners. (Of these three, MetroPCS is the largest, and therefore the easiest, and therefore the one most often whispered to be in”deep discussions” with AT&T.)  The problem is that you would need to more than double any one of these in size (triple in the case of Leap) to create a competitor approximately half as large as T-Mobile. This would still leave AT&T with a ridiculously large marketshare. But again, as a thought experiment, let us assume that you would only need to double one of these in size to create a credible “T-Mobile Lite.”

The average person probably thinks “whoa, someone offering to double or triple the size of my company? Coolness! How hard can it be to find a dance partner for that?” People who run companies, however, understand that such a sudden expansion is more often a nightmare rather than a dream come true. Leaving aside financing (which I will turn to next), such an expansion would require massive restructuring of systems, integrating new technologies and new customers into the existing structure, and numerous unanticipated and unplanned for changes in business strategy, business practices, and staff. Sprint nearly choked on NexTel, losing billions of dollars and millions of customers while it struggled to figure out how to absorb a company with an entirely different network configuration, customer base, and business plan.

Yes, Metro, U.S. Cellular and Leap all signaled in their FCC comments that they would be open to the merger given an opportunity to snarf some divested licenses. But these companies were envisioning the same sort of pre-complaint divestiture scenario envisioned by AT&T – a bunch of licenses in a divestiture trust. In that scenario, the companies could have picked up licenses in markets where expansion made sense in light of their existing customer base and business plans. They did not envision playing the role of the goose in AT&T’s effort to produce T-Mobile fois gras by being force fed enough licenses and customers to more than double in size.

Yes, AT&T leaked that they were in discussion with MetroPCS and Leap about possible divestitures as part of a possible settlement. (As an added irony, the article claims that Bank of America is handling the negotiation. You all remember how well Bank of America’s hastily negotiated takeover of Merril Lynch worked out, right?) But no one has heard any of these companies signaling their unbridled delight with the prospect or seen any further leaks on how things are moving. That may just be the usual silence around negotiations – except that AT&T has been about as silent as a dancing elephant when it comes to anything that indicates it might have a settlement offer for DoJ to consider. More likely the silence flows from the fact that MetroPCS and Leap (or US Cellular) needs to think long and hard before signing on to an agreement that would  radically alter the nature of their company. Like the proverbial dog that caught the car, AT&T’s dance partner would find itself on a very uncomfortable and bumpy ride.

Mind you, one of the three in question may convince itself that it can swallow a huge chunk of T-Mobile and come out ahead of the game. So lets assume AT&T finds a dance partner. Now we come to the next significant hurdle – how does AT&T’s dance partner pay for this huge chunk of T-Mobile?

Problem 2: Financing and Price

AT&T will not give these systems away. It will need to sell these systems to its dance partner. Even leaving aside the FCC approval problems I have discussed previously (we are still in fantasy thought experiment where all this stuff is rigged to fly through), financing such a huge purchase by one of the possible partners raises some very practical problems. If Metro or one of the others could have afforded to buy T-Mobile in the first place, it would have done so.

In addition, AT&T is essentially conducting a “fire sale,”  because it needs to reach an agreement (at least an agreement in principle) before it can claim to have a “serious” settlement offer. While that helps T-Mo Lite to buy the chunk of T-Mo AT&T must sell, it hurts AT&T’s bottom line. Even with a fire sale price, however, it is almost certain that T-Mo Lite would need to take on a heck of a lot of debt to finance the deal.

This raises two practical problems. First, who will actually finance such a deal? In case you hadn’t noticed, despite the Federal Reserve lowering the price of borrowing money to practically zero for the foreseeable future, financial uncertainty has made it extremely difficult to raise money. To give an example of how this financial uncertainty impacts the price a business like AT&T can get for its assets, consider this recent story of Citi Group trying to sell the music company EMI. According to the article, the financial world has changed dramatically even since March. “Tightened credit markets have made it more difficult for potential buyers to finance their offers, and nervous banks have required bidders to put up more equity.” As one analyst observed: “[e]verything you read in the news indicates that it is difficult to raise financing for any business, let alone music. It’s not the greatest time to be looking for debt financing” (emphasis added).

Recall that even back in March, when financing was easier, AT&T’s $20 billion bridge loan to buy T-Mo set off alarms in the financing community. Now, in addition to that, AT&T is going to need to scare up several billions more. Even worse from the lender’s perspective, the new loan will not be to AT&T, reliable blue chip titan with numerous lines of business in case this deal doesn’t work out, but to the newly created and far less creditworthy T-Mobile Lite.

As if this were not enough of a problem securing a loan, most analysts agree that AT&T and Verizon already enjoy a very significant advantage in the mobile phone market. Most buy into Wall St. analyst Craig Moffet’s “barbell” analysis, which holds that the wireless market has diverged into a high end/smartphone market dominated by AT&T and Verizon and a low end cheap voice market dominated by firms such as Trac Phone. In addition, while AT&T must remake its dance partner T-Mo Lite into a vaguely credible national competitor, it does not want to remake T-Mo Lite into an actual competitor. But if T-Mo Lite does not look like it can effectively compete with AT&T and make back the money it must borrow any time soon, if at all, then who will lend it the billions of dollars needed to buy the systems from AT&T?

AT&T could try some form of seller financing, or otherwise guarantee the loan to T-Mobile Lite. But this raises a different set of problems. First, if AT&T holds T-Mo Lite’s debt, it rather obviously compromises the independence and ability to compete of the new T-Mobile Lite.  In addition, AT&T financing the deal or taking on additional debt exposure to get financing for T-Mo Lite raises problems for AT&T and its stock. AT&T is a publicly traded company, and its officers (and shareholders) actually care about how these things impact share price. Financing T-Mo Lite hugely drives up the cost of the merger and AT&T’s long term cost in the same way that taking on any additional debt would. Worse for AT&T’s stock, everyone would recognize that some form of “seller financing” is a sign that financial markets regard this debt as high risk. AT&T can help out some by selling the T-Mo assets at a below market value price. But that also effectively raises the cost of the merger, with the negative impact on AT&T’s stock.

Finally, one needs to consider the impact on the stock of AT&T’s dance partner. MetroPCS, Leap and US Cellular are all publicly traded companies and their boards and management also care deeply about the value of their stock. Analysts will have all the same questions that lenders have about how a company like MetroPCS can possibly hope to integrate the new systems and customers into its network, or how they can ever hope to make enough to pay back the debt from the purchase. Unless analysts believe the optimistic story (and analysts are pretty pessimistic these days), the stock of Metro (or whoever else is AT&T’s dance partner) will also tank – a factor likely to be very unattractive to the potential T-Mobile Lite.

Problem 3: Other Conditions – Roaming, Handsets, Interoperability

Just divesting spectrum and customers won’t be enough to create even a faux competitor. As anyone who has studied the wireless industry for the last few years knows, wireless providers cannot attract customers without cool smartphones. Also, since only AT&T and Verizon even come close to having sufficient spectrum to support their operations without data roaming agreements, any “T-Mobile Lite” will need affordable roaming with AT&T. They will also need interoperable handsets, so that T-Mo Lite customers can take advantage of roaming rights.

This makes things awkward for AT&T because AT&T either has to continue to pretend it has no ability whatsoever to influence the handset market or come clean and say ‘yeah, we can pretty much get everybody but Apple to drop through hoops, and even Apple has to tread a bit warily given our market share.’ If AT&T opts for the former, they cannot build T-Mo Lite into a potential competitor. If they opt for the later they pretty much prove the government’s case.

But it’s not just going to be DoJ demanding access to interoperable handsets and cooler smart phones. Remember, we’re pretending the wheels are uber-greased and DoJ staff so thoroughly crushed that they will sign anything AT&T puts in front of them. The party that will insist on these conditions will be AT&T’s dance partner, T-Mobile Lite. Why? Because T-Mo Lite – whether it’s MetroPCS or someone else – is not stupid. They know they will need roaming and access to cool interoperable handsets to survive post-transaction, and they know the only way to get them is to nail them down now while AT&T is desperate to make a deal.

So lets assume AT&T bites the bullet, uses its market power to give T-Mo Lite access to cool interoperable handsets, and also gives T-Mo Lite some sweet roaming agreements. Now AT&T faces a new hurdle. How does AT&T avoid giving the same sweet deal to everyone else? First, under the Data Roaming Order the FCC adopted last spring, AT&T must offer “commercially reasonable” roaming terms, and the benchmark for commercially reasonable is “what have you offered anyone else.” But more importantly, when this “settlement” runs through the FCC, everyone in the industry is going to demand equal treatment, and it is really hard to see how Genachowski gets three votes for an Order that does not extend similar terms to competitors for as long as AT&T maintains its deal with T-Mobile Lite no matter how greased the wheels will be.

At What Point Does Wall St. Pull The Plug?

Lets keep in mind that for all the blah blah about spectrum and synergy, a key reason AT&T will pay $11 billion more than what the market would value T-Mo in a stand alone IPO is the ability to extract monopoly rents and exercise market power. That’s what makes Wall St. like (or at least tolerate) the deal, despite the high price tag.

But every one of these things discussed above drives up the cost of the deal, and the cost of future compliance, much higher than what AT&T told everyone to expect back in March when they announced the deal. And while the conditions discussed above don’t mitigate the market power AT&T will acquire, it will make it more expensive for AT&T to extract the monopoly rents. Add to this concerns about the cost of the merger to AT&T and its dance partner the risk described by Yankee Group that permitting further consolidation inevitably means much more intrusive regulation of the industry as a whole, and it combines to make the analysts that track wireless generally and AT&T specifically very unhappy.

At some point, the analysts that track AT&T’s stock will decide AT&T is offering to pay way too much, incur way too much long term debt, and risk too much in terms of future regulatory costs to justify holding on to the deal. They will insist that AT&T drop the deal or get downgraded. At that point, an increasing number of officers and members of AT&T’s board, people whose personal wealth depends on AT&T stock having a high valuation, are going to start asking how much money they and other shareholders have to lose so that Randal Stephenson and Jim Cicconi can avoid the embarrassment of taking a loss.

Application In The Real World

Needless to say, the real world is even less friendly to any possible settlement than our imaginary world where DoJ is thoroughly brow-beaten into submission. If we assume AT&T gets past the initial hurdle of finding a dance partner and announcing a low-ball opening settlement offer, it will undoubtedly need to repeat the exercise several times to try to pressure DoJ– each time offering larger divestitures and more conditions. That won’t play well to investors on Wall St. already spooked about poor potential growth prospects for the sector in a continuing bad economy. Even if analysts remain supportive (which I doubt), regular investors are likely to start abandoning AT&T as the litigation/negotiation drags on until trial. Once AT&T stock starts doing worse than the market generally, AT&T officers and board members are going to start pushing very hard to get out of the deal.

Problems for DT As T-Mo Wastes Away

Deutsche Telekom is not entirely indifferent to the health of AT&T’s stock, or to its future performance. Some of the $39 billion value of the merger comes from AT&T transferring significant stock to DT and giving it a permanent seat on the Board of Directors. But more pressing for DT is the wasting away of T-Mo as an asset they can sell to another buyer if they still want to exit the U.S. market.

As discussed previously, DT is working quietly on a “Plan B” that would involve reviving negotiations with one of the 5 jilted suitors from last fall or spinning T-Mo off through an initial public offering (IPO). But as long as DT remains contractually obligated to support AT&T’s efforts to win approval of the merger, T-Mo remains in a bad place wrt to churn. Some folks estimate T-Mobile might lose three times the number of subscribers it has already lost since AT&T announced it would purchase T-Mo if DT waits until September to get its break up fee. For DT to realize maximum value from sale of T-Mo to someone else or from an IPO, it needs to get out of AT&T’s death hug as soon as possible.

European analysts who follow DT know this as well as anyone else. Right now, most of them accept AT&T/DT’s assurances that DoJ is just “playing hardball.” But as we get closer to trial and AT&T’s public settlement offers get increasingly desperate, European analysts are going to start questioning the assurance that DoJ is just “playing hardball” and begin seriously pressuring DT to maximize value for sale of T-Mo to someone else rather than drown in AT&T’s embrace.

Bottom Line: “Settlement” Far More Costly Than What AT&T Sold Wall St. Analysts Back In March.

Here is the final take way from this little thought experiment. AT&T has spent so much time and political capital fighting to force DoJ to even begin settlement discussions that neither AT&T nor any of the analysts that follow AT&T paused to consider how expensive such a settlement would need to be, and how very different it would need to be from what AT&T described to analysts back in March. At some point, however, that reality will start to set in and AT&T will need to find financing and get possible settlements vetted by analysts or risk seeing their stock tank. When that happens, pressure will ratchet up on both AT&T and DT to stop throwing good money after bad or risk seeing their stock seriously downgraded – a threat they take far more seriously than either the DoJ or the FCC.

Until that day, we will continue to see AT&T make a lot of noise – potentially even finding a dance partner to open up the T-Mobile Lite Polka to the applause of the assembled honest politicians and AT&T cheerleaders. The key decisionmakers at AT&T and DT are extremely personally invested in making this deal happen. Neither Randal Stephenson nor Rene Obermann wants to walk away after spending millions of dollars lobbying and after assuring everyone on Wall St. and in Europe that this was absolutely for sure going to happen. If they need to spend millions more in shareholder money to try to save face, why should they care? But in the end, it may be Wall St. rather than DoJ or the FCC that finally pulls the plug on what has become an expensive and destructive personal crusade by corporate executives unwilling to admit they bet wrong and with the freedom to spend shareholder money like water rather than own up to a mistake.

Stay tuned . . .


  1. Great analysis. Since I used to work in telecom, I had a vague sense about some of these scenarios, but this really fleshes it all out.

    I assume that given all ATT’s blunders it would be impossible for DT to sell TM to Verizon when ATT does finally back out. After all, even though VZN aren’t quite the sociopaths that ATT are (at least if one ignores LEC divestiture), nearly all the arguments against the current sale would also apply to one to VZN.

    My question is, would it be conceivable that DT could sell TM to Sprint, or does any buyer really have to come from another industry?

  2. Jess:

    Verizon is a non-starter under the same theory, too much market power. And meshing CDMA networks with GSM, even though both will eventually get to LTE, makes the efficiencies even more questionable.

    I don’t see Sprint as viable because of the money problems and because it would still be a 4-to-3 merger. OTOH, it would be 3 roughly equal sized rather than two dominants and an also ran, so that is somewhat less bad. But still not a likely combination.

    We know there were 4 other possible bidders from statements made by DT. Be good if we knew who those might be . . .

Comments are closed