The night before he left, my old firm’s biggest rainmaker took the associates in his group out for steak and told us nothing. Wagyu, a Napa cabernet with a three-figure markup, a lot of loud talk about the Knicks. Two mornings later his corner office was cleared out, his assistant had already been reassigned, and the eight of us who reported to him learned the news from a Bloomberg Law alert that landed at 6:11 a.m. He had moved his entire practice to a competitor across the street. Books of business, two senior associates he had quietly warned, and roughly forty million dollars in annual originations walked out the revolving door before most of the equity partners had finished their coffee.
Nobody sued him. Nobody could.
I thought about that morning a lot in 2024, when the Federal Trade Commission announced it was going to abolish noncompete agreements across the American economy and half of LinkedIn treated it like the Berlin Wall coming down. Associates in my old orbit were forwarding the news with fire emojis. In-house friends were asking me whether their sales teams would suddenly be free to bolt. And I kept thinking: none of this touches us. Lawyers have been living in a post-noncompete world for decades. We just built a more polite cage and called it professional responsibility.
what the FTC actually tried to do
Let me set the table, because a lot of the coverage was sloppy about the mechanics.
On April 23, 2024, the FTC voted 3 to 2 to issue the Non-Compete Clause Rule. The final rule was sweeping. It declared that entering into or enforcing a noncompete with a worker was an unfair method of competition under Section 5 of the FTC Act. Existing noncompetes for most workers would become unenforceable. New ones would be banned outright. Senior executives, defined as people earning more than $151,164 in a policy-making position, got a narrow carve-out that let their existing agreements survive, but even they could not be bound by new ones going forward.
The agency did not undersell it. The FTC estimated the rule would free roughly 30 million workers, about one in five Americans, from these clauses. It projected higher earnings on the order of hundreds of billions of dollars over a decade, along with new business formation and, according to the agency, thousands of new patents a year. Whether you buy those numbers or not, the ambition was enormous. This was the biggest single attempt to rewire American employment law by administrative fiat that I have seen in my working life.
The effective date was set for September 4, 2024. For about four months, the business bar lost its mind.
And here is the part that got lost in the excitement. The rule had a specific exclusion for nonprofits, which meant a lot of hospital systems structured as nonprofits arguably slipped the net. It had questions around banks and other entities outside the FTC’s jurisdiction. It said nothing that would have changed the rules of the road for lawyers, because lawyers were already governed by something the FTC has no say over at all: the state rules of professional conduct. More on that in a minute.
Ryan LLC walks into a Texas courtroom
The challenge everyone remembers came fast. Ryan LLC, a Dallas tax services firm, sued the FTC on April 23, 2024, the very day the rule dropped. The U.S. Chamber of Commerce and business groups intervened. The case landed in the Northern District of Texas in front of Judge Ada Brown, a Trump appointee, which is roughly the judicial equivalent of drawing pocket aces if you are the one trying to kill a federal regulation.
In Ryan LLC v. FTC (N.D. Tex. 2024), Judge Brown first granted a preliminary injunction on July 3, 2024, but limited it to the named plaintiffs and intervenors. That was the appetizer. On August 20, 2024, she set the rule aside entirely under the Administrative Procedure Act, with nationwide effect. Two holdings did the work.
First, she found the FTC lacked substantive rulemaking authority to issue this kind of rule. The agency had hung its hat on Section 6(g) of the FTC Act, a provision that lets the Commission make rules and regulations for carrying out the Act. Judge Brown read that as a housekeeping grant, a power to make procedural rules, not a mandate to ban an entire category of contracts used across the whole economy. Second, she found the rule arbitrary and capricious. Her reasoning there was the kind of thing that makes a litigator wince on behalf of the government: the ban was too broad, imposed a one-size-fits-all prohibition without evidence that a narrower rule would not work, and relied on studies of state-level policy changes that did not support a categorical national ban.
The APA gives courts the power to vacate unlawful agency action, and vacatur runs against the rule itself, not just against the plaintiffs. So the rule was dead in the water everywhere the moment she signed the order.
It did not help the FTC that the courts were split. In ATS Tree Services, LLC v. FTC (E.D. Pa. 2024), Judge Kelley Hodge went the other way and declined to block the rule, essentially reading the FTC’s authority more generously. And in Properties of the Villages, Inc. v. FTC (M.D. Fla. 2024), Judge Timothy Corrigan granted a preliminary injunction on the major questions doctrine, the idea that an agency needs clear congressional authorization before it decides a question of vast economic and political significance. Three courts, three different flavors of skepticism and support. That kind of split usually means the appellate courts, and maybe the Supreme Court, get to sort it out.
Except they never did.
The FTC appealed Ryan to the Fifth Circuit and the Properties of the Villages matter to the Eleventh. Then the administration changed. Then the calculus changed. In early September 2025, the FTC, now chaired by Andrew Ferguson, voted to abandon its appeals and walk away from defending the rule. The agency signaled it would go back to case-by-case enforcement against noncompetes it considers abusive rather than trying to ban them wholesale. The grand rewrite of American employment law ended not with a Supreme Court showdown but with a voluntary dismissal.
So the headline for most workers in most states: nothing changed. Your noncompete is as good or as bad as your state’s law makes it. If you are in California, you were already protected. If you are in Florida, good luck.
why none of this ever governed lawyers
Now the part I actually care about, because I lived it.
Lawyers do not sign enforceable noncompetes. We are one of the few professions in America where a blanket ban on post-departure competition has been the settled rule for a very long time, and it has nothing to do with the FTC and everything to do with ABA Model Rule 5.6.
Rule 5.6(a) is short and blunt. A lawyer shall not participate in offering or making a partnership, shareholders, operating, employment, or other similar agreement that restricts the right of a lawyer to practice after the termination of the relationship, except an agreement concerning benefits upon retirement. Rule 5.6(b) does the same thing for settlement agreements, so you cannot buy off opposing counsel by making them promise never to sue your company again.
The stated reason is client autonomy. The client, not the firm, owns the relationship. A client gets to choose their lawyer, and a noncompete on the lawyer is really a restriction on the client’s ability to follow that lawyer out the door. The comment to the rule says the restriction limits the freedom of clients to choose a lawyer. That is the theory, and it is a good one. It is also, conveniently, a rule that lets partners take their books and run, which is why the profession has never been in a hurry to change it.
The case law here is older than the FTC rule by a generation. In Cohen v. Lord, Day & Lord (N.Y. 1989), New York’s highest court struck down a partnership provision that withheld a departing partner’s earned but uncollected earnings if he competed. The court held the financial disincentive was an impermissible restriction on the practice of law under the predecessor to Rule 5.6. Money penalties count. You cannot dress a noncompete up as a forfeiture clause and expect it to survive.
New Jersey went the same direction in Jacob v. Norris, McLaughlin & Marcus (N.J. 1992), voiding a provision that denied departure compensation to lawyers who competed. The through line across most states is that both direct bans and indirect financial punishments for competing are off the table.
California, being California, went its own way and made it more interesting. In Howard v. Babcock (Cal. 1993), the state Supreme Court upheld a partnership agreement that imposed a reasonable financial cost on partners who left and competed, treating it more like a toll than a prohibition. California is the state that generally voids noncompetes for everyone else under Business and Professions Code section 16600, and it recently tightened that further with SB 699 and AB 1076, both effective January 1, 2024, which made even out-of-state noncompetes unenforceable against California workers and required employers to notify workers that their clauses are void. So you get the strange picture where California protects the checkout clerk and the software engineer more aggressively than almost anyone, but lets law firms attach a reasonable price tag to a competing partner’s exit. The legal profession always seems to end up in its own bespoke category.
Put the map together and the point is simple. The FTC’s blocked rule would have been a big deal for a sales rep in Ohio, a nurse in Texas, a line cook whose franchise made him sign a noncompete for reasons that make no sense to anyone. For lawyers it would have been a rounding error, because Rule 5.6 already gives us more mobility than the FTC rule would have handed most workers. We could always leave. The question was never whether. It was how, and how expensively.
how talent actually moves
Here is where the PR and the reality diverge, and where my war stories live.
The absence of an enforceable noncompete does not mean lawyers float freely between firms like pollen. It means the friction moved somewhere else, into a set of soft controls that are, in their own way, just as sticky. A partner cannot be legally barred from competing, so firms manage the exit through economics, timing, information, and the raw social pressure of a small market where everyone knows everyone and remembers who behaved badly.
Start with capital and deferred compensation. Equity partners have money tied up in the firm. Capital accounts, unvested deferred comp, sometimes a chunk of the current year’s distributions held back. The partnership agreement dictates how and when you get it back when you leave, and while a firm cannot condition that money on your promise not to compete, it can absolutely make the mechanics slow and painful. You get your capital returned over eighteen months in installments. Your final distributions get trued up next spring after the audit. Nobody calls it a noncompete. Everybody understands the incentive.
Then notice provisions. Many partnership agreements require sixty, ninety, sometimes a hundred and twenty days written notice before withdrawal. Some jurisdictions and some agreements flirt with garden leave concepts borrowed from finance, where you are technically still employed, still paid, and expressly not allowed to work or solicit during the notice window. The ethics rules get uncomfortable when notice periods get so long that they effectively restrict practice, and bar opinions have pushed back on abusive versions, but a reasonable notice period is generally fine and it does real work. It gives the firm time to get in front of the clients before you do.
Which brings us to the actual battleground: the clients and the files.
Under the rules of professional conduct, the file belongs to the client. The client decides who represents it. When a partner leaves, the ethical script, spelled out in ABA Formal Opinion 99-414 and echoed in state guidance, is that the departing lawyer and the firm should ideally send a joint notice to affected clients letting them choose. In the real world, that joint letter is a hostage negotiation. Both sides want to be the first voice in the client’s ear. The partner who has spent ten years taking the general counsel to dinner has a head start. The firm has the institutional relationship, the other partners who touch the account, and the ability to slow-walk the file transfer while it makes its pitch.
So the exit gets choreographed. This is the word that matters. Choreographed.
The rainmaker who left my firm did not improvise that steak dinner. He had been in talks with the new firm for months. He had lined up his conflicts checks quietly, because the receiving firm has to run the incoming business against its existing client base and cannot take work that conflicts. He had sounded out the two associates he wanted to bring, without formally soliciting anyone in a way that would blow up in a lawsuit, because associates are employees and the poaching rules are murkier and the firm can get territorial. He had a sense of which clients would follow and had probably received soft assurances from the two or three that mattered most. When he pulled the trigger, the whole thing executed in about seventy-two hours, and the speed was the point. Speed denies the old firm the chance to counterprogram.
I have since watched this from the other side of the table, as the client. When I moved in-house and a partner who did our work announced he was jumping firms, I got the joint letter. I also got a very warm personal call from the partner within about an hour, and a slightly stiffer call from the firm’s relationship partner the next day. Both wanted to know if the company would move its files. And the honest answer, the one that Rule 5.6 is supposedly built to protect, was that I would decide based on who did good work at a fair rate, not based on any contract either of them had signed. The rule worked exactly as designed. I picked. In that case I split the work, kept some at the firm, sent some with the partner, and let both of them stew about it.
what the rule would and would not have changed for clients
People ask me whether the FTC rule, had it survived, would have shaken up legal services for clients. Short version: not really, and where it would have mattered, it would have hit around the edges of law firms rather than at their core.
Think about who inside a law firm signs noncompetes that Rule 5.6 does not reach. Not the lawyers. But firms are businesses, and businesses have marketing directors, chief operating officers, pricing analysts, business development professionals, IT staff, and knowledge management people. Some of those roles come with noncompetes, especially the senior business-side hires who see the firm’s rate structures, client profitability data, and lateral recruiting strategy. If you are the person who knows exactly how much margin the firm makes on its top twenty clients, the firm would very much like you not to walk that across the street. Those clauses would have been in the FTC rule’s crosshairs. The lawyers’ would not.
There is also the alternative legal services provider and legal tech world, which does not run on Rule 5.6 at all. The people building contract review platforms, the reviewers at managed services companies, the analysts at the ligation finance shops, the operations staff at the Big Four’s legal arms. That is a market full of ordinary employment contracts and ordinary noncompetes, and it is the market that is actually eating into traditional firm work. A national ban would have loosened talent there and probably sped up competition against law firms from the outside. Quiet, unglamorous, and more consequential than anything happening inside the partnership.
For the ordinary corporate client picking outside counsel, the practical effect of the blocked rule was close to zero. You could already follow your lawyer. You still can. The FTC’s death in Texas did not shrink your options by a single name.
Where I do think clients should pay attention is the confidentiality and non-solicitation side, which survives all of this intact. A lawyer’s duty to keep client confidences under Rule 1.6 does not end when the lawyer changes firms. When your partner defects and wants your business, the information he can use to court you is bounded by his duties to his old firm’s other clients and by the conflict rules. If he was staffing a matter adverse to a company that is now a target of his new firm, that conflict travels with him and can create imputation problems for the whole new firm under Rule 1.10. Sometimes a lateral hire the receiving firm wanted badly gets blocked, or ethical-walled into uselessness, because of a conflict nobody flagged until the offer was out. I have seen a deal partner’s move fall apart in the final week because his book conflicted with the receiving firm’s biggest private equity client and the client refused to waive. He did not have a noncompete. He had a conflict, which is worse, because you cannot negotiate your way out of it with money.
the rule everyone forgot about while cheering the one that died
I want to leave you with the thing that actually bugs me, now that I sit on the client side and watch the profession pretend it is a meritocracy of mobility.
Model Rule 5.6 is sold as a client-protection rule, and at the level of principle it is. But look at who it protects in practice. It protects rainmakers. It protects the partners with portable books who can walk into any firm in town and name a number, because the rule guarantees no contract can stop them from taking their clients along. The rule is a mobility subsidy for the people at the top of the pyramid who least need one.
It does very little for the associate. Associates do not have portable books. When a partner leaves, associates are chips, invited along or left behind based on a calculation they do not control, and if they are left behind they inherit a diminished practice group and a partner who is now openly job-hunting to backfill the loss. The service partners, the ones who do excellent work but never learned to sell, are similarly stuck. Their value is tied to the platform. They cannot walk because there is nothing to walk with. Rule 5.6 frees the free and pens the penned.
And the business-side employees, the marketing chief and the pricing analyst, get the ordinary noncompete that the FTC tried and failed to abolish. So inside a single law firm you can have three tiers of mobility. The rainmaker who is legally uncageable. The associate who is technically free and practically trapped. The COO who is contractually locked down. All in the same building. All under the same brand that sends around a press release about its collegial one-firm culture.
The FTC rule dying in Ryan did not change any of that, and if it had lived it would have changed only the bottom tier, the people the profession thinks about least. Which tells you where the friction really sits and who benefits from the current arrangement. It is not the client, most days. It is the person with the biggest book, who gets to eat the steak, say nothing, and be gone by Tuesday. I used to want to be that guy. Now I am the one who gets the 6 a.m. call, and I have learned to enjoy making both firms wait for my answer.