The breakup AT&T chose: how Ma Bell traded 22 local phone companies for the right to compete
How AT&T negotiated its own 1984 breakup: it traded the local telephone monopolies that confined it for the freedom to compete, and then a federal judge added the terms that decided which half kept the value.
On November 20, 1974, the Department of Justice filed an antitrust suit against the American Telephone and Telegraph Company, its manufacturing subsidiary Western Electric, and Bell Telephone Laboratories, alleging that the Bell System had used its monopoly position to shut competitors out of domestic telecommunications services and equipment. 1 AT&T was then the largest company in the world, with roughly one million employees and $149.5 billion in assets. 2 The case ran for seven years and produced no verdict. The breakup happened anyway, because the defendant offered to dismantle itself — and the judge rewrote the offer before he accepted it.
That is what makes this the most instructive antitrust negotiation on record for anyone who deals with a regulator. AT&T did not lose its case and then get carved up. It chose which parts of itself to surrender, traded them for the right to compete, and then watched a third party who was never at the bargaining table add conditions of its own. The question the case leaves open: was that a settlement bought at a fair price, or did AT&T pay for the same outcome twice?
Deal snapshot
| Parameter | Terms of the settlement |
|---|---|
| Parties | The United States (Antitrust Division) against AT&T, Western Electric and Bell Telephone Laboratories. The 1974 action was filed in the U.S. District Court for the District of Columbia; the 1956 consent decree it also settled had been entered in New Jersey. 13 |
| Negotiators | Assistant Attorney General for Antitrust William F. Baxter for the government; Chairman and Chief Executive Charles L. Brown for AT&T. Both signed the agreement of January 8, 1982. 14 |
| Assets in dispute | The pre-divestiture Bell System reported $149.5 billion in assets, 87 million access lines, and almost one million employees at the end of 1983. 2 |
| Remedy | Separation of local exchange service from long distance and equipment. AT&T divested the local exchange business of its 22 Bell operating companies, which were regrouped into seven independent regional holding companies. It kept long distance, Western Electric, and Bell Laboratories. 13 |
| Consideration | No cash changed hands. AT&T's shareholders received one share in each regional company for every ten AT&T shares they held, and retained their adjusted AT&T stock. 2 |
| What the government gave up | The 1956 consent decree was vacated in its entirety, freeing AT&T to compete in markets it had been barred from, including data processing and equipment sales outside the Bell System. 3 |
| Closing path | Settlement signed January 8, 1982; consent decree entered August 24, 1982; reorganization plan approved August 5, 1983; divestiture effective January 1, 1984. 2 |
The choke point: a decree that fixed the boundaries without breaking the company
The negotiation of 1981 and 1982 only makes sense against a 1949 complaint. The government had sued AT&T that year seeking the divestiture of Western Electric and the separation of manufacturing from telephone service. Seven years later the case ended in a consent decree that imposed no structural relief at all. AT&T accepted an injunction confining it to common carrier communications, confining Western Electric to manufacturing for the Bell System, and requiring it to license its patents on reasonable terms. 3
At the time, being barred from "any business other than the provision of common carrier communications services" looked like a small concession. It became the company's central strategic problem. The clause kept AT&T out of the computer market as computing and communications converged. 1 In his 1979 annual report, Brown wrote that AT&T needed "to free the Bell System from the constraints of the 1956 Consent Decree," adding that it no longer made sense to deny the Bell System the chance to compete in unregulated markets. 4
The perimeter was shrinking at the same time. The Federal Communications Commission's 1968 Carterfone decision let customers attach equipment AT&T had not made to its network. In 1969 the FCC let Microwave Communications, Inc. sell private-line service between two cities. A 1977 appeals court ruling in the Execunet case required AT&T to let MCI and others compete in ordinary long distance. 1 By 1981, sixty private antitrust suits against AT&T were pending. 4
AT&T's public case was that its monopoly funded universal service. Above-cost long-distance rates and business rates subsidised cheap local and rural service, and only control of the whole system made that transfer work. A competitor free to attack the profitable segments would strip the subsidy and push local rates up. 1 Strip out the rhetoric and the two sides wanted different things. The government wanted the local bottleneck separated from everything that could be competed for. AT&T wanted out of a decree that capped its future.
Decision point 1: the government chooses a remedy, not a case
Date: 1981 into early 1982.
Decision: Baxter kept the case alive and made structural separation the only currency he would accept. On taking office he told AT&T's chief counsel that past negotiations were over: AT&T should divest the operating companies and compete with the rest. 4 His reason was not the price of long distance. It was cross-subsidy: a regulated local monopoly earning steady profits could fund unregulated businesses at prices rivals could not match, and no regulator could police that closely enough. 1 On the same day the AT&T agreement was announced, the Justice Department dropped its thirteen-year case against IBM, calling it "without merit." 5
Viable alternative: Accept a conduct remedy. AT&T could have been left whole and bound to interconnection and pricing rules enforced by the FCC — the approach the FCC itself was developing in its Computer Inquiry II proceeding, which would have required separate subsidiaries rather than divestiture. 4 Baxter could also have accepted what several cabinet departments wanted and dropped the suit.
Constraint and power asymmetry: Baxter carried less institutional weight than he appeared to. A task force of cabinet departments, including Defense and Commerce but not Justice, formally recommended that the case be abandoned; the Defense Department's own position paper, entered into the trial record, argued that breaking up AT&T would threaten national defence communications. 1 The Attorney General insulated him partly to defend the department's independence. On the other side, the enforcement argument cut against a conduct remedy: two former chiefs of the FCC's Common Carrier Bureau testified that the Commission "is not and never has been capable of effective enforcement" of the rules governing AT&T. 3
Negotiated term: Baxter would trade the whole case for one thing: the local exchange function out of AT&T. He was willing to leave AT&T Western Electric, Bell Laboratories, and long distance, and to give up the 1949-vintage demand to break up the manufacturer.
Consequence: The government's opening position became the shape of the final remedy. By naming the asset it would accept instead of the verdict it wanted, the Antitrust Division converted an eight-year trial it might lose into a negotiated separation it could enforce.
Decision point 2: AT&T decides which part of itself to surrender
Date: late 1981 to January 8, 1982.
Decision: AT&T offered the local operating companies and asked for release from the 1956 decree in return. The company's own "planning model," circulated in February 1982, grouped the local companies into seven independent regional holding companies. 2
Viable alternative: Fight the case to judgment. Baxter had not proved liability when AT&T settled; the defence had put roughly 250 witnesses on over five months and expected to finish its case within weeks. 3 AT&T could also have waited for Congress; it had pursued legislation for years and a telecommunications deregulation bill had passed the Senate in October 1981. 1
Constraint and power asymmetry: AT&T's alternative was worse than it looked. The company had already absorbed a signal from the bench: on September 11, 1981, Judge Harold Greene denied its motion to dismiss, writing that the evidence showed "the Bell System has violated the antitrust laws in a number of ways over a lengthy period of time." 16 A verdict could have cost AT&T Western Electric and Bell Laboratories, the two assets management regarded as the company's future. The legislative route was closing as well: opposition from the House telecommunications subcommittee chairman promised to derail the Senate bill, and a stricter House alternative was taking shape. 1
Negotiated term: The local exchange business, and the twenty-two operating companies that ran it, in exchange for a vacated 1956 decree. AT&T kept the businesses the decree had made most valuable to it by exclusion: long distance, manufacturing, and the laboratories. It also kept the option to sell computing and equipment wherever it liked.
Consequence: AT&T surrendered the assets that were most exposed to rate regulation and least able to earn a competitive return, and bought the freedom to compete where the growth was. Contemporary press read the deal as a win for AT&T. 1
Decision point 3: the third party who was not at the table
Date: January to August 1982.
Decision: Judge Greene refused to enter the agreement as written and attached his own conditions. The parties tried to make the settlement a modification of the 1956 decree, filed in New Jersey, so that it would escape the Tunney Act's requirement that a court find a consent judgment in the public interest. Greene called that reasoning, "most charitably described as disingenuous," applied the public interest procedure anyway, and took comments from over six hundred interested parties. 3
Viable alternative: The parties could have refused the modifications and resumed the trial, which Greene made explicit: if they did not accept his changes, the 1956 decree would stand and the case would be relisted. 3
Constraint and power asymmetry: Neither signatory controlled the closing. Because the case had been settled rather than decided, the judge held a veto over the remedy, and he was accountable to a different set of interests: ratepayers, state regulators, competitors, and the courts of appeal. His stated concern was that divestiture not produce local rate increases. 2 AT&T's leverage was thin. Walking away meant going back to a trial whose judge had already found the government's basic contentions sustained.
Negotiated term: Ten modifications, accepted by both parties on August 19, 1982, and entered with the decree on August 24. 2 The divested companies were allowed to sell but not manufacture equipment, and to keep the profitable directories, with the assets needed to print them transferred to them. The decree required the operating companies to give every long-distance carrier access equal in type, quality and price to what AT&T received. And AT&T was barred from electronic publishing over its own transmission facilities for seven years. 3
Consequence: The directories went to the divested companies, and the electronic-publishing ban shut AT&T out of a market it had wanted for seven years; the judge added that provision over the Justice Department's opposition. 1 The case was dismissed on August 24, 1982. 2
Decision point 4: the closing path is a second negotiation
Date: December 1982 to January 1, 1984.
Decision: AT&T filed a reorganization plan in December 1982 setting out how the separation would actually work, including the boundaries of the local access and transport areas where the divested companies would operate. 2 Judge Greene approved the boundaries in April 1983, then ruled in July that he would accept the plan only if the parties agreed to six more modifications. 27
Viable alternative: AT&T could have litigated the plan, or tried to implement a version the court had not approved. It did neither; the decree as amended made implementation conditional on judicial approval, and AT&T had already conceded that principle in August 1982. 37
Constraint and power asymmetry: The economics of the separation were still being negotiated, and the equal-access obligation gave the operating companies an opening. AT&T had assured the court that the access charges long-distance carriers would pay would keep divestiture from raising local rates, and the operating companies would have to spend billions building the switching capacity to deliver equal access. The court held AT&T to its own assurance: it required AT&T to reimburse the operating companies for any equal-access and network reconfiguration costs they had not recovered through access charges by January 1, 1994. 27
Negotiated term: The six modifications, accepted on August 3, 1983, included that reimbursement guarantee and the transfer of the Bell name and trademark to the divested companies. 27
Consequence: AT&T took a one-time charge against 1983 earnings of $5.2 billion, later revised to $5.5 billion, to write down its post-divestiture assets. 2 The new AT&T and the seven regional holding companies began trading on a when-issued basis on November 21, 1983, and the divestiture took effect on January 1, 1984. 12
What was actually divided, and who ended up holding it
The scale of the transfer is the reason this case is still cited. The reorganization assigned about 77 percent of the Bell System's assets — $115.5 billion of the $149.5 billion total — and 60 to 70 percent of its nearly one million employees to the divested operating companies. The remaining AT&T started with roughly $34.3 billion in assets. 2
On the terms as signed, the trade looked favourable to AT&T. It shed rate-regulated local service, kept long distance, and, at the time of the settlement, still held roughly 77 percent of the interexchange market on its own concession. 3 What that reasoning missed was where the value would sit twenty years later. The computer business AT&T bought its way into failed, and the equipment and laboratories it had fought to keep were spun off in the mid-1990s. 4 The divested local companies, meanwhile, turned out to be durable businesses: they did well on their own, helped by higher local rates and the access charges they levied on long-distance carriers. 1
The industry then reconsolidated under the name of the company that had been broken. Bell Atlantic merged with NYNEX and then with GTE to become Verizon; Southwestern Bell, renamed SBC, bought two other regional companies and in 2005 acquired AT&T itself for $16 billion, took its name, and bought BellSouth the next year. The last regional company, US West, is now part of Lumen Technologies. 1
The measured effects are more mixed than either side's brief. Consumers paid more for local service after the cross-subsidies ended and reported lower satisfaction with service quality. 1 On innovation, Martin Watzinger and Monika Schnitzer found that patenting in the telecommunications technologies where Bell had been active rose about 19 percent a year relative to comparable fields after the breakup — an increase of roughly 1,065 patents a year — while Bell's own patenting fell by about 24 percent. The overall gain came from companies outside the Bell System. 8
Five reusable frameworks
1. Separate the remedy question from the liability question
In a case against a regulator or a court, ask early what change would end the matter, not who is right. Baxter did not need a verdict; he needed local exchange out of AT&T. Once he named that, the negotiation turned from a contest over evidence, which he could lose, into a design exercise he controlled. A remedy defined that narrowly is also what made the deal closeable after eight years of litigation.
Use it when: A regulator, court, or counterparty can block your deal, and you are assembling arguments rather than a remedy.
Diagnostic question: What single change would make the other side stop, and can you deliver it without losing the value you are actually protecting?
Operating boundary: A remedy narrowed to the blocker can leave the underlying problem in place; the operating companies were barred from long distance and manufacturing precisely because separation alone had not been enough in 1956.
2. Price the litigation alternative before you surrender anything
Both parties' willingness to settle tracked what each expected from a verdict, not what each wanted. AT&T moved after the court denied its motion to dismiss; the government moved when a settlement, rather than a trial, could deliver a structural remedy it might not win. A BATNA is only a number once you have attached a probability and a cost to it.
Use it when: You are deciding whether to settle a dispute that has a contested, delayed, and binary outcome.
Diagnostic question: If this goes to judgment, what is the range of outcomes, and which of them can you not survive?
Operating boundary: A settlement locks in a specific shape; a verdict may be better or worse, and the judge's findings are not a forecast. Greene's denial of the motion to dismiss was a signal, not a result.
3. Negotiate with the veto holder, not only with the signatory
The settlement of January 8, 1982, was not the deal. The decree that mattered was the one entered on August 24, after a judge who was not in the room extracted ten changes. A negotiation is over when every party who can reopen, condition, or refuse to implement it has said yes, and a settlement that bypasses such a party invites a larger revision later.
Use it when: Your agreement requires approval, ratification, or implementation by a party that is not at your table.
Diagnostic question: Who can add terms after we sign, and does the package still work if they do?
Operating boundary: Anticipating the veto holder can mean conceding value early to a party who may not demand it; the cost has to be weighed against the risk of a late rewrite.
4. Trade the regulated asset for the unregulated freedom
AT&T traded the part of itself that was most constrained — local service, priced by regulators — for the freedom to compete where the returns were. When one of your businesses is capped, taxed, or fenced by regulation, the question is not how to defend it but what you can get for letting it go.
Use it when: A regulated or low-return business is the obstacle to competing somewhere more valuable.
Diagnostic question: Which of my businesses is the one the regulator most wants to constrain, and what becomes possible for the rest once I let it go?
Operating boundary: The freedom may be worth less than the asset. AT&T's entry into computing failed, and the regulated local business it surrendered became the more durable one.
5. Design the separation so it holds after you leave
A divestiture is not a signature; it is a set of operating rules that decide whether the two halves can compete fairly. The equal-access obligation, the local access and transport area boundaries, the ban on the divested companies manufacturing equipment, and the reimbursement guarantee for network construction are the terms that made the separation real. Without them, the separated entity can be recaptured through pricing, standards, or procurement.
Use it when: You are separating a business and the separated entity must be genuinely independent of its former parent.
Diagnostic question: After separation, what stops the two sides from re-creating the relationship through pricing, technical standards, or purchasing?
Operating boundary: Every rule needs an enforcer. The restrictions on the divested companies stayed in place until each could show the court that it could not use its monopoly power to impede competition — a standard that took years and repeated petitions to satisfy.
Manager's takeaway
AT&T's divestiture was not a defeat imposed on a company that fought to the end. It was a settlement the company designed — and one that a judge then improved on terms the company had not offered. Three questions are worth carrying into your next encounter with a regulator, a court, or any counterparty who can veto your deal:
- What single change would make the other side stop, and is it worth less to me than the outcome I am protecting?
- If this goes to judgment, what is my range of outcomes, and what is the probability attached to each?
- Who can rewrite this after we sign, and does the deal still work if they do?
The uncomfortable lesson is the fourth. AT&T got exactly the freedom it negotiated for, and it did not keep the value. The company that emerged from the breakup with the strongest position was the regulated local business it had been most willing to give away.
References
- 1
- 2
- 3
- 4Antitrust, Computer Inquiry II and the Break-up of AT&T, 1973-1984
historyofcomputercommunications.info
- 5U.S. Ends Antitrust Suits Against AT&T, IBM
washingtonpost.com
- 6
- 7
- 8The Breakup of the Bell System and its Impact on US Innovation
martin-watzinger.com
This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.