Orly Lobel – Enforceability TBD: From Status to Contract in IP and Human Capital Law
Noncompete agreements: used to blur lines of IP regimes, make it harder to tell what is owned/ownable. Trade secret lawsuits are used in anticompetitive ways, to put startups out of business.
Amazon’s standard contract: during employment, “employee will devote employee’s entire productive time, ability, attention, and effort to furthering Amazon’s best interests” … amazingly broad. Innovation assignment contracts don’t care about definition of IP; they list everything, whether patentable or not, copyrightable or not, reduced to practice or not, trade secrets, confidential business information, know-how, etc. NDAs are equally capacious.
New empirical study: 34% of employees asked to sign noncompete after they accepted the job; large majority have no negotiation/bargaining is rare; electronic contracts with pop-up notifications to “accept” all terms—“cubewrap”; handbooks, policies, manuals containing noncompetes.
80% didn’t know the state law; majority didn’t know if noncompetes were enforceable; actual signing wasn’t more likely to occur in enforcing states than nonenforcing; even among CEOs, 60% in California were required to sign noncompetes.
Notice is important: but we have lots of rules that have to be available to employees. (Cue discussion of disclosure literatures.)
Hot moment for reform: complete ban in California; Mass. bill (unsuccessful so far). Dozens of state statutes restrict assignment clauses. Some jurisdictions differentiate between high salary earners or tech v. other industries—Hawaii banned noncompetes in tech industry; MOVE Act would ban noncompetes for low-paid workers. Statutory notice requirement: 2007 Oregon law requires 2-week advance notice or an adequate salary raise when the employer introduces a noncompete; MOVE Act would also require employers to disclose noncompete restrictions.
Would notice help? Traditionally: Employment contracts are vague, open-ended. Sides prefer flexibility inherent in evolving relationship. Actual terms unfold over time; not presented at point of entry. Relational opportunism. Unknown facts: Q about what knowledge is secret, what’s been developed, what competition looks like, employee’s plans and options, training, skills. Law notice is interlocked w/fact notice. Courts considering reasonableness consider reasonableness at the time of drafting of contract plus everything that happened since.
James Bessen: Most of the discussion about notice/notice failure is about regulators/administrative agencies that fall down on the job somehow; here we see private actors having an incentive to obscure notice. Harm to wages is one negative aspect of noncompetes; drags on innovation is another aspect. Reduce’s employee’s incentive to invest effort in training.
Typewriter’s success was delayed for decades until keyboards were standardized, at which point it made sense to invest in learning how to type. Changed that industry and role of women more generally. This pattern repeats over and over again: those issues are related to portability of skills. Social value of large, trained workforce goes beyond particular employers/employees.
Lawsuits over noncompetes has tripled over past few years—litigation pace is increasing.
R. Anthony Reese – Reforming Termination Formalities
Termination of post-1978 transfers requires complicated formalities. How clearly does the law notify the would be terminator of how exactly to accomplish termination? In the law as usable, or as window dressing? If the former, we want the law to be clear. If the latter, the hoops should be very hard to jump through. Second, termination causes a change of ownership of some copyright rights—how well do formalities provide notice to others of the changes that result?
Fairly complicated: identify party/parties entitled to terminate and their shares; calculate time period in which termination can occur and choose a valid effective date; calculate when advance notice can be served; identify the party/parties who must be served; properly draft and serve a termination notice; timely record the termination with the Copyright Office.
Timing of notice is timed based on the execution of the grant, and when that was can be hard to find—undated grants; oral grants of nonexclusive licenses; implied in fact grants of nonexclusive licenses—the brawling, boisterous world of facts. 20-30 years later, can you figure out the date on which the oral license was granted/the conduct of the parties gave rise to the nonexclusive license/the check was endorsed? What about grants in yet-to-be-created works? Are they completed when signed or when the work was created, and when did it finish being created? CO takes the view that it’s not completed until the work and its © come into existence. How easy will determining that be?
Even if you have the simplest case: a signed, dated agreement for an existing work—there’s still a question about what the “end of 35 years from the date of execution of the grant.” The legislative history contradicts itself. Transfer of Sept. 2, 1987: does the window open Sept. 2, 2022? Sept. 3, 2022? Or, puzzlingly, Sept. 1, 2022? Legislative history mentions Sept. 1 and Sept. 2 in the same paragraph. Luckily, you can pick a date that’s not right at the beginning if you know that’s a problem.
We could (1) allow terms to run to the end of the year, as w/duration; (2) allow judicial reformation: judge could pick a valid date; (3) not close the termination window, letting them pick a new date.
Service of notice on the grantee or grantee’s successor in title. Which does that mean? Does the serving party get to choose? Unclear. Judicial interpretation limited. Even if we can identify who’s served, we have to find that person. The grantee doesn’t have to tell the terminating party about any subsequent transfer. Here, there is some guidance from CO: service is ok if you do a reasonable investigation, if there’s no reason to believe there’s been a transfer then you can serve a grantee by first class mail to the last known address. If you have reason to believe there’s been a transfer you serve the transferee.
What about notice to the person whose rights are being terminated and to the world at large? If a grantee is really interested in notice, they can record at the CO. If they don’t, we may feel less sympathetic to them. As for the rest of the world, notice must be recorded at the CO. May be the first time there’s any entry on the public registry about this work. (How often is that true for the works that are now terminated?) But there’s no requirement of any contact info for the terminating party. Also it need not identify w/specificity the rights that have been recaptured—must specify the grant, but can say “publishing agreement,” and we won’t know the content.
Could (1) require contact info to be included; (2) encourage identifying w/particularity rights claimed, though terminating party may not know what they are.
Alfred Chueh-Chin Yen: Extreme act of gov’t paternalism. There are any number of doctrines outside termination that exacerbate problems of notice failure Reese identified: work for hire doctrine for example; contract will say it’s a WFH but if not you assign anyway. Or you may be the joint author and get a contract that says you acknowledge you don’t have authorship rights. These are both ways of getting people not to terminate. Not only do the doctrines themselves create ambiguities, but the contracts make ambiguities worse. Ordinary author will not have the sophistication Reese has brought to bear. The termination provisions are thus backwards; termination should automatically happen, whenever. The acquiring party is better equipped to figure out what’s going on than an individual author.
Second solution: judicial. If we understand these measures as ways to get people to drop termination rights, we can have courts reject them.
Mark Lemley: Notice fatigue. (I feel it too.) Does notice solve anything? We’ve placed a lot of faith in notice, which can be ineffective, maybe merely because of cumulative effect of all these notices.
[RT: My comments on Lobel’s paper were the same as Lemley’s: Compare to wage theft: overt violation of the law, like inclusion of noncompete clause in California. Notice in this sense has the classic problems of disclosure. Lauren Willis: incentives to avoid providing effective notice of the rights of the consumer (or the employee). Instead of trying to educate employees further, especially given the fluidity that Lobel identifies, what about actual enforcement with penalties—deterrence is the real goal.]
Lobel: This is a big issue. On the particular issues I’ve been thinking about: notice about the terms of the contract—if you’re signing a significant agreement, there is value in inducing awareness of its provisions. (But how do we do that?) Employees in many circumstances can ask questions about the provisions. Also, if what you sign isn’t enforceable, that’s effective. Experimental study: people reported that they wouldn’t even consider a job offer when they sign restrictive covenants—needed a much higher raise to consider it than a control group. But once they learned the covenant wasn’t enforceable they behaved like the control group. But if a “reasonable” restriction was enforceable that wouldn’t help them much.
Kate Darling: what if termination is a terrible answer to the starving creator, not because of paternalism, but because 35 years later is a terrible time to renegotiate/create a holdup situation (unless they set their works free).
Reese: normatively I’d prefer a reversion system where renewal happened automatically—either the work goes into the public domain or it gets renewed and the rights return to the author. It’s not implausible to conclude that the continuing value of the work comes more from the author than the publisher; in many cases people make these contracts when they have a difficult time valuing their work. If the statute required ongoing royalty payments, we might feel comfortable saying that you could transfer rights away forever.
But if termination is window dressing, we should just remove it. Make a good deal or suck it up, rather than pretending that you’ll have another chance. (Termination as the disclosure of copyright transfers?)
Q: similarities b/t noncompete contracts and confidentiality agreements? Most trade secret law goes against former employees. But contractual provision just says “stuff that you learned is confidential if valuable” and there’s no notice about what’s protected—difficult problem to solve because how do you specify it in advance?
Lobel: problems in espionage prosecutions—courts and juries may just accept that what companies say is confidential is so: companies stamp “confidential” on everything.
Loren: recognized that valuation is difficult before exploitation; not paternalism to allow revaluation. Another notice problem: we get these assignment contracts that say they’re in perpetuity/not terminable. But they are. You need to know these things aren’t enforceable!
Reese: so we need to get info out to authors. Guilds can help. Of the 8000 terminations studying, a lot of them are music, and a lot use forms provided by the songwriters’ guild. Authors’ Alliance, etc.—sites to tell authors/successors that this is available, notwithstanding anything in the contract that says it’s not terminable.
Bone: Limit termination to higher value works. Is that wrong
Reese: some of this is a problem that it’s technical; another problem is that it is badly drafted and ought to be better; another problem is that the relevant facts are long-ago and far-away—look at the amount of digging required to answer (or not answer) questions about the Happy Birthday copyright—might want to tie rights to dates that are more obviously discoverable. Modern possibilities for termination might make us nervous about termination as a screen—if academics are terminating transfers to release them, and if we make termination difficult as a screen, they’ll be screened out so it only happens for blockbuster songs and we don’t get the CC releases we want.
Litman: Low-value works are exactly the ones that should be terminated b/c the current proprietor lacks incentive to make them available b/c they’re low value and might compete w/ high value newer works; author might have more incentive to make them available. Public may get better access.
Q: Notice window opening isn’t a huge issue b/c you can pick any time w/in a five year window. Shoot for day 5 to be sure.
Gordon: Speaking as someone who practiced copyright law—termination notices are incredibly ambiguous all the way through. Empirical implications: assume that incentivewise, there’s no effect on the amount of money they’re willing to pay, what then?
Reese: on average, value for women ought to be higher b/c women live longer and we’re now measuring by life plus. You could disprove the proposition that a lump sum price really represents value over the full length of the copyright term—lots of confounding factors. Not aware of anything but publishers’ anecdotal statements that they’re interested only in the short term.