Oral argument — Schuman v. Microchip Technology Incorporated

Good morning, and may it please the Court, Michael Rubin for the plaintiffs. I'd like to reserve three minutes for rebuttal, please. This is an appeal from a Rule 54b judgment entered after the District Court granted summary judgment to defendant ERISA fiduciaries on the grounds that the claims releases that they obtained from the two plaintiffs, ERISA beneficiaries, were valid and enforceable. Because it arose on summary judgment, all of the evidence that we submitted as the plaintiffs in opposition to summary judgment has to be accepted as true and inferences drawn in our favor. And what does that evidence show? That evidence shows that the fiduciaries knew when they obtained these releases from plaintiffs that they owed, under ERISA, the severance payments that had been made available to those individuals because they had worked for years for the company that Microchip, the defendant, acquired, Atmel. That Atmel had created an ERISA severance plan that entitled any employee who was terminated by an acquiring company within a certain period of time and was then terminated without cause to severance benefits. The reason Microchip and the other fiduciaries knew that they owed these benefits to plaintiffs and others was because the plan had expressed language, and that's in page 65 of the excerpts of record, and I'll quote the pertinent sentence. That sentence states that any decision made or other action taken by the company, that's Atmel, that's the predecessor company, prior to a change of control, before the acquisition becomes final, with respect to the program, the ERISA plan, and any interpretation by the company prior to that change of control of any term or condition of the program or related documents will be conclusive and binding on all persons and will be given the maximum possible deference allowed by law. The reason that provision was enacted or was adopted as part of the ERISA plan, as stated in the undisputed testimony provided to the district court, is because Atmel, when in July 2015 was looking for a merger partner, was looking to be acquired, wanted to ensure that its long-term loyal employees would stay with the company, would remain Atmel employees through the acquisition to enable the company to survive, and then, if they were fired without cause by the acquiring company, would have payments, severance payments protected by ERISA once they lost their jobs. Counsel, let me ask you this. I am largely sympathetic to your client's position, but I want to be sure I understand something. If I understand the record correctly, your clients were told that because of the way things had occurred, that they were not entitled to benefits under the original agreement. Your clients objected, and it seems like they entered into what I'm going to call a settlement agreement, wherein they were going to get a certain amount of money in exchange for releasing their claims. Is that correct or is that incorrect? That is largely correct, Your Honor. What distinguishes this case from the run-of-the-mill case involving the enforceability of a release, or even an ERISA case that involves the enforceability of a release that has to be scrutinized with extra care under Briscano is because to obtain that release in this case, the defendants independently violated their fiduciary duties, the highest fiduciaries known under law and equity and ERISA, against self-dealing. This was a self-funded plan. The money came out of Microchip's own treasury and the duty of undivided loyalty to the beneficiaries. What distinguishes this case is that the trustee, the fiduciary, knew that it owed the money, yet told these individuals that the plan had expired five months earlier. It hadn't. They were entitled to no benefits. They were entitled to benefits. And if they did not accept initially 25 cents on the dollar and then 50 cents on the dollar, they would have to sue under ERISA. Recall that under ERISA, the only remedy available is payment of the money that was owed in the first instance. So they forced them to the choice. Either you take what ended up being 50 cents on the dollar, or even though the fiduciary knew it owed this money, 100 cents on the dollar, take half of it or you're going to have to hire lawyers, undergo what has turned out to be nine years of litigation and uncertainty, where all you're going to get in the end is the money that was owed to you in the first instance. And the reason we cite provisions like the restatement second of trusts, paragraph section 217, and the restatement third of trusts, section 97, is because those provisions clearly state that what a fiduciary cannot do under the law of trusts and under ERISA is rest a settlement from a beneficiary by refusing to provide to that beneficiary what is clearly owed, and instead taking part of it, imposing the compromise. So it doesn't matter, Judge Smith, that whether or not these individuals knew that they had rights under the plan, because the fiduciary had no authority under the language of the plan to deny the payments to those individuals. I take your point, but where in the record does it say that the trustee in this case knew absolutely that the 100 percent that you're asking for was in fact owed? Well, that's why it matters that this is coming up on summary judgment, because, first of all, the judge recognized it. If you look at ER page 28 to 29, as well as page 10, the judge recognized the facts. We know that the fiduciary had a copy of the Atmel plan, had copies of this language that I quoted at the outset. We know that the letters were sent on January 14, 2016, to these two plaintiffs and others that expressly told these individuals and others that their rights under the severance plan continued and that those rights would continue whether the acquiring company turned out to be Dialog, which was the first company, or Microchip, which is the defendant. So that constitutes a conclusive and binding construction of the plan under the language. It was communicated. Those January 14 letters were given to Microchip's CEO, CFO, and others, so they were aware that Atmel had repeatedly made these promises. And because this is summary judgment, the trial judge needed to credit the testimony, the evidence that was provided by us, to show that Microchip did in fact know when it told these people, sue me, go to court if you need to, but we're only going to give you first 25 cents, then 50 cents on the dollar. We are not going to pay you those benefits we know that we owe. At trial, on remand, we will in fact show, as the evidence has turned out, we've taken depositions since much of this had happened, that Microchip did in fact know its obligations. And because the CEO, Mr. Sanghi, characterized the plan as overly generous and unnecessary, and because the money was coming out of Microchip's own pockets, this was a way for the fiduciary, in violation of its strict fiduciary obligations, to save itself millions of dollars by not paying the money that it knew it owed, but decided that it would not pay. Let me take you to a slightly different subject. As you know, of course, on 54B, we're here to try to decide what the test should be. As you know, the Eighth Circuit's test in Levitt goes up to nine factors. So far, we've been looking, I guess the district court looked at six. What do you think should be added, if anything, to the Eighth Circuit? Or what kind of a test should we have if you have your way? If we have our way, the only test is the test set forth in the restatement of trusts and the Bellows v. Bellows case from the California Court of Appeal. And that is a simple test. If a fiduciary under ERISA violates its fiduciary duty of undivided loyalty and against self-dealing in obtaining a release from beneficiaries and thereby avoids paying benefits that it knows it owes, the trial court must invalidate the releases. If, Your Honor, you were to revise the Eighth Circuit test under Levitt, you would add as an additional factor, and in fact it would have to be the controlling factor, whether there was an independent breach of fiduciary duty that resulted in obtaining these additional releases without paying any valid consideration. Because if they owed 100 cents on the dollar and all they paid was 50 cents on the dollar, they didn't even provide consideration to the beneficiaries. But rather than tinker with the common law test, our position is the common law test doesn't even apply in a case like this because there is a knowing breach because microchip knew or at a minimum should have known, by not sticking its head in the sand, but it knew because it had the documents, that it owed money, and a trustee simply under ERISA cannot tell a beneficiary, we know we owe you the money, but we're not going to pay you unless you give up all the rights, and then we're just going to give you a fraction. I gather that your comment about the under restatement of trust fiduciary role, that would perhaps be the first test, right, before you even get the rest of them. Correct. A violation of fiduciary duty under trust law. That's exactly right, and that goes along with what the court did in Washington, yes. You're down to about three and a half. Do you want to reserve? I would. Thank you so much. It pleases the court, Mark Kosicki, on behalf of Appley's Cross Appellants and Defendants below. As they have since their initial filing of the initial complaint, plaintiff's appellate arguments are based on a lot of vitriolic verbiage and disregard or distortions of precedential decisions that control this case. Do we have a precedential decision that controls it in the Ninth? Yes, Your Honor. Which one, in your view? I mean, we have the six-part test from the first and the second the district court applied. Other circuits have done other things. As Judge Smith mentioned, the Eighth Circuit has an eight-part test. Your Honor. Why do you say those are controlling? Because there has been no court that has adopted a standard other than knowing involuntary when assessing a release, and plaintiffs have not contested that their releases were knowing involuntary. They may argue on appeal, as they have. There should have been an additional factor that should have been considered by the district court, but they did not ask the district court to do that. The question is whether or not, when they sign these releases, they knowingly and voluntarily waive their claims under a plan that this court has held was ambiguous. So do you think that there are no additional duties owed in the fiduciary context of ERISA concerning releases? Is that your position? No, Your Honor. It's not at all. There are certainly remedies for breaches of fiduciary duty. They are provided to individual plaintiffs under Section 502A3. 502A3, as the Supreme Court has repeatedly stated, is not the grant of broad discretion for court to exercise its powers that it ordinarily has in equity to fashion remedies. That has never been the court's decision or statement. Following up on my colleague's comment, ERISA is a special law. It's shot through with trust concepts. I wrote a case a number of years ago called TIF. It's dealing with the fees that can be charged by fiduciaries under ERISA. And it was very clear that they're bound by fiduciary law. And in this particular case, it seems like you're just saying the fact that this is a specialized trust arrangement doesn't matter. You're just going to treat this like an ordinary knowing and voluntary kind of situation when you would have in a non-trust situation. As my colleague pointed out, I don't think there's any Ninth Circuit law that's directly on point. Would you agree with that? Yes, Your Honor. Okay, so we need to decide what the controlling law is in this case, right? In terms of adopting the knowing and voluntary waiver standard. Right, and what in your mind would dissuade us from seeing this through the prism of trust law principles? Your Honor, the Supreme Court's authority, the Supreme Court has made it clear that for a breach of fiduciary duty, individual claimants are not without remedies. They have the remedies available under Section 502A3, which is the section that plaintiffs suit on. I respect that, but what we're talking about here is you're saying that the two plaintiffs, I suppose there are more people involved here, but they knowingly and voluntarily entered into this settlement agreement. Your opponent says, well, wait a minute. These are fiduciaries. They knew that they were violating fiduciary law, and that would need to be disclosed to your client's owner for them to have knowingly and voluntarily waived their rights. What's wrong with that concept? Well, first of all, Your Honor, it's not one that they made below, and so it's been waived if they ever had it. Well, that's one point. Why then did the court send this up on a 54B? Well, Your Honor, that is a difficult question to analyze. I certainly don't understand why the court did that. We challenged it certainly by motion to dismiss, but the reality is there are remedies for breaches of fiduciary duty. That's what ERISA provides. The Supreme Court says under Section 502A3, plaintiffs have the remedies that were typically available in equity, and those remedies are equitable estoppel. Plaintiffs pled equitable estoppel. But they have to show that they detrimentally relied and that there are extreme circumstances to justify it. So they abandoned that claim. There is the potential for surcharge under Section 502A3. The Supreme Court has talked about it. That is a potential option. But plaintiffs abandoned their claim for surcharge. Plaintiffs had equitable remedies that the Supreme Court has made very clear are available, but they did not meet the standards that equity imposes on them. Plaintiffs are essentially seeking a claim of rescission. They want out of a contract that they claim was obtained fraud, duress, and breach of fiduciary duty. Equity provides a way to get out of a contract that was so induced. It's rescission. But the problem, I think, is the task the district court gave us was to formulate a rule. So what's your position on what the analytical rule ought to be? The district court applied the test from the Second and the First Circuit, and, of course, we've got the Eighth Circuit. What's your position? Your Honor, I think that the correct test is the test that the district court applied. Why? Because it addresses the factors that are reasonably considered in determining whether a waiver is knowing involuntary. Plaintiffs were free to ask the district court to add other factors. Oh, I understand. But I'm asking for your position on what the rule ought to be. And I guess you think the First and Second Circuit. Absolutely, Your Honor. It demonstrates that the person releasing an ERISA claim knew exactly what they were doing, knew there was a dispute, and decided to accept benefits that were different from what they thought they deserved. Plaintiffs claim the plain language of the plan entitled them to benefits. They had that plan, Your Honor. They had this letter that they incorrectly claim was an interpretation that was binding and conclusive on microchip. That's not true. But they had that letter. They had microchip's position that the definitive agreement that it entered into did not apply to the Atmel plan because it wasn't entered into by the deadline the Atmel plan required. They believe they still were entitled to benefits. Nobody forced them to accept the compromise, Your Honor. They were free to reject the compromise offer, not release their claims, and pursue them. That's exactly what the plaintiffs did in Berman. This Court is well aware of that. There were claimants who did not sign and pursued their ERISA claims. That's the right way to do it because if you're going to come to equity, you have to come with clean hands. You cannot accept benefits that aren't what you think they should be, keep them, and go ahead and turn around and then sue after having granted a release and pursue that for 10 years, a great burden to the Court and cost to defendants. There are ways to get out of a release. Plaintiffs alleged them, but they could not demonstrate detrimental reliance, Your Honor. They could not show they relied upon a false statement or an omission of fact to their detriment. And under ERISA, they also have to show that there are extreme circumstances that justify it. Your opponent says that the trustee here knew absolutely that what was being offered was a violation of the agreement and that the microchip was in violation. What's your response to that? It's absolutely false, Your Honor. They're making leaps of conclusory assumptions or simple assertions based on how they interpret certain facts. The district court correctly held that the plan is ambiguous. This court itself held the plan was ambiguous. The plan does not give the woman who wrote the letter that plaintiff's claim was a binding and clear construction or interpretation of the plan. It never gave her the authority to interpret the plan. It gave her the authority to adopt and formulate a plan and implement it. She never had authority to interpret the plan. Now, plaintiffs can claim otherwise. There are arguments that they've made certainly that she did, but that is not the law, and it's certainly not correct to say that there was not a good-faith dispute about that. There, as the court below recognized, was a good-faith dispute. The employees, despite plaintiffs claiming there were all these representations by both Atmel and microchip that the Atmel plan applied, over the 10 years of litigation, all of those have fallen by the wayside. What we have is a single letter that went to only a couple dozen of the members of the class that is not what plaintiffs claim it is. That's it, that one letter. And had microchip done anything else, there was nothing else to do. Its interpretation of that letter is different than plaintiffs'. So we can argue about that. Plaintiffs could have pursued their ERISA remedies, and they did, under Section 502.1b. But they can't. Let's say you're right. Our court said it was ambiguous. So why wouldn't this go back before a finder of fact to determine what the trustee understood, perhaps what the law was, and decide whether or not there was, in fact, a breach of a fiduciary obligation that would change how this case comes out? Why shouldn't that happen? Because, Your Honor, that's not what the plaintiffs asked the court below to do. They simply said we're entitled to an injunction precluding the enforcement of our releases. And there is simply no way, there is no case that plaintiffs can point to that an injunction against the enforcement of a contract, which is what a release is, is a remedy that is typically available in equity. The Supreme Court has made it very clear courts under Section 502.a.3 do not have this broad discretion that plaintiffs argue to fashion remedies. Plaintiffs have the remedies typically available in equity. If they relied on a misstatement, they had equitable estoppel to estop it. If they thought they were lulled into this by misrepresentations and, as they characterize it, lies, there's rescission. There are numerous equitable remedies that this group of plaintiffs could not meet. You cannot come to equity seeking the remedy that equity provides and say, well, I can't meet those requirements, but I still get out of my release by an injunction that no court has ever granted and the Supreme Court certainly has not suggested is available under Section 502.a.3. If I may, I'd like to reserve the remainder of my time to address the cross appeal. Thank you. Mr. Rubin. Thank you. Judge Smith is right. ERISA is a special statute, and this Court and the Supreme Court have stated many times that the duties not to engage in self-dealing and the duties of undivided loyalty to beneficiaries are the highest duties known to trust law. The record on summary judgment, including the letters that were sent to the plaintiffs in January 2016, which expressly stated the plan is still in effect and it doesn't matter whether the acquiring company is dialogue or microchipped, is sufficient, as is the SEC proxy statement, as is the testimony of Suzanne Zamoris, who said that Atmel repeatedly took the position and informed employees and took actions and interpretations as set forth in the plan prior to the merger, that if the merger went through and microchipped was the acquiring company, then the Atmel plan would continue in effect. They would be entitled to benefits. That is, quote, conclusive and binding on everyone, including the acquiring company. So the record on summary judgment establishes there was no good-faith dispute. When you talk about the First Circuit's test or the Second Circuit's test, that's similar to the Ninth Circuit's test in Viscano where special scrutiny has to be provided where there's an ERISA settlement to resolve a good-faith dispute. That is an entirely different line of case law than what we're asking the Court and what the restatement of trusts and what Bellows v. Bellows says is the law and is the law under a California law of trusts as well, which is that a trustee has no right, no power, no authority, no ability to tell a beneficiary, we are not going to pay what we know or should know we owe you unless you give up your right to most of that, release claims, and we'll give you pennies on the dollar. A trustee may not do that under any circumstances, certainly not under ERISA, and the summary judgment in this case shows that that is precisely what happened, which is why the district court erred in applying the six-part common law test. Now, let me just ask you this. As you know, the issue of whether the plan was ambiguous came before our court. We concluded that it was ambiguous. What role, if any, does that determination play on how we are to interpret what I guess the trustee sent to these folks? Right. If you look at the memorandum disposition in the Berman case, which is the case brought by nine individuals, the court focused on one clause and one clause only, which was the basis for summary judgment by the district court in favor of plaintiff beneficiaries. That was what is the phrase initial triggering event mean? In the memorandum disposition, the court said that Judge Gillian was wrong in ruling in our favor on summary judgment in that other case because there was some ambiguity and microchip had not been given sufficient opportunity to take discovery. In that memorandum disposition, the panel did not address, although we had presented it, did not address the conclusive and binding language, presumably because the panel recognized that since it was ordering further discovery, it was going to allow the parties to determine and eventually the trial court to determine whether in fact microchip knew or should have known that Atmel had conclusively entered into a conclusive and binding determination, interpretation, and action showing that the plan applied. So the ambiguity that was identified by the panel in the Berman case simply goes to what is initial triggering event mean? Our position is that now that the discovery has been taken, now that the Berman case was remanded and eventually settled before trial, there is evidence in the record, which was incorporated in the Schuman record, which shows that there was in fact repeated conclusive and binding determinations made by Atmel, so whether or not the clause initial triggering event was ambiguous no longer matters because under the plan, microchip had no power at all to reject Atmel's conclusive and binding determination. So it doesn't make any difference in the end, and the fact that a different clause had been determined earlier to be ambiguous doesn't affect the result here. We don't know a lot of the case issue here? No, none at all. Okay. Thank you, counsel. Thank you. I'll try not to take a breath in the minute and 26 seconds I have. The district court correctly entered judgment. It incorrectly refused to apply that judgment to members of a certified class who had joined the class by not opting out based on plaintiff's strategy throughout the case, which was we are not going to challenge these releases on the grounds that we're not knowing. I thought you were going to limit this to the cross-appeal. Yes, Your Honor. The fact is that was plaintiff's litigation strategy and it remains the strategy to this day. Plaintiffs claim the knowing involuntary standard did not apply and the class members opted into that. The court certified the class despite defendant's arguments that individualized showings would be necessary under the test to determine the enforceability of the releases. Plaintiffs disagreed. At that point, at least at that point, by the time they moved for class certification, plaintiffs had elected a litigation strategy that they would not address their releases under the knowing involuntary standard. They would seek their other potential equitable remedies under 502A3. Thank you, counsel. Thank you. Case discord will be submitted for decision.