The blockbuster T-Mobile / Sprint merger was approved the Department of Justice in July, along with a complicated set of conditions aimed at turning Dish Network into a national wireless carrier. Instead of blocking the merger and maintaining all four major wireless carriers, the government is essentially trying to manufacture a fourth carrier to replace Sprint, by requiring T-Mobile to run a new Dish Network-branded wireless service for seven years while Dish builds its own network that will eventually compete with T-Mobile. If Dish doesn’t build that network by 2023, it will have to pay a $2.2 billion fine.

If that sounds like a complicated scheme that doesn’t quite make sense, you’re not alone. In a new filing, a group of seven economists and antitrust experts say the court should reject the DOJ’s proposed solution, calling it “doom[ed] … to failure” and “a remedy that does not meet the standard of restoring the competition currently provided by Sprint.”

For at least the next seven years, anyone buying service from Dish will just be getting rebranded T-Mobile service, and that’s not actual competition. Beyond that, it’s not a sure bet that Dish will ever actually build its own network, and even if it does, it’s all but certain that network won’t reach as many people as Sprint.

The filing is intended to convince the court to block the Department of Justice merger approval before it takes effect — technically, the DOJ filed a complaint in court to block the merger, and simultaneously filed the settlement containing the deal with T-Mobile and Dish. That settlement still has to be approved in federal court as serving the public interest. If the court isn’t convinced that the deal restores competition, then it could be rejected, potentially leaving T-Mobile back at square one.

Hal Singer, an economist at Georgetown University and one of the authors of the paper, says he thinks there is a strong case for blocking the deal. “What is so remarkable about this case is that DOJ’s complaint doesn’t pull any punches on the competitive harms” of reducing the number of carriers from four to three, Singer says. “That puts a huge burden on the DOJ’s settlement” to prove the complicated requirements on T-Mobile and Dish will actually preserve competition.

The filing, which is remarkably readable for a legal argument from a bunch of academics, says that betting on Dish to compete is fundamentally wishful thinking. Dish “would be attempting what no company has ever done before — to build and operate a nationwide wireless network, at a cost of at least $10 billion, from scratch, and in a short number of years.”

And Singer just doesn’t think that’s going to happen, because the business logic simply isn’t there. “Any rational firm in Dish’s position would sooner milk the cushy access arrangement, pay the penalties for not building out, and then flip the spectrum back to an incumbent carrier before incurring the massive investment costs to build out,” he says.

Since Dish has a history of breaking promises to build a network with the spectrum it already has, the chances of the DOJ’s plan working are even lower, the authors say. “This significant undertaking exceeds what Dish has promised regulators before, but failed to deliver time and again,” they write. “The DOJ’s aspiration to create a new competitor in these circumstances is fraught with risk that will surely doom it to failure.”

“The DOJ has accepted a consent decree with Sprint / T-Mobile whereby Dish — a company with no history or presence in this industry — will for the foreseeable future try to compete as an MVNO reseller with no network, and in the less foreseeable future may acquire and develop assets sufficient to become a full-fledged wireless carrier,” the authors write. “For that to happen, however, Dish will have to rely on T-Mobile’s vague and non-credible promises to behave counter to its economic incentives.”

To solve the problem of T-Mobile having to support its own new competitor, the deal contains a long list of things T-Mobile must do, including specific traffic management requirements, operational support, and three years of handling billing and customer support, but the authors say that list “discloses by implication the enormous difficulties that arise in having one company assist its direct competitor.”

The government traditionally prefers “structural” solutions to merger problems, or simply selling off assets and businesses without further oversight, while the DOJ’s plan is a “behavioral” remedy that tries to tell Dish and T-Mobile how to do business under the eye of regulators. And that creates a lot of opportunity for funny business, say the authors. “The extreme dependency of Dish on the good graces of New T-Mobile creates abundant opportunities for … strategic pricing, slowdown of provision, alteration of terms or quality of the assets and services,” the authors write.

The authors also point out that the deal doesn’t actually require Dish to build a nationwide network — the coverage target in the deal is 70 percent of the population, not 70 percent of the country. “It is clear that certain parts of the country will lose out,” write the authors. “[Dish can] cover 50 percent of the population by just targeting 15 percent of the most urban areas in the U.S. Even if Dish hits that 70 percent goal, the resulting network likely will not fully replace Sprint’s ubiquitous nationwide network, leaving nearly 100 million Americans with one fewer facilities-based carrier.”

The filing, along with other comments on the proposed merger deal, will first be submitted to the DOJ, which will then post them publicly and submit them to the court. There’s no timeline on when the court will make a decision yet, but the government will have to respond — and figure out how to prove that making T-Mobile help Dish Network build a network is somehow less complicated than letting anyone else buy Sprint.