Oral argument — Michael Buhl v. Abbott Laboratories

Good morning. My apologies there. Good morning. Therese Lawless on behalf of the appellate Michael Buhl, your honor. This is an appeal of a granting of summary judgment by the magistrate judge. So the rule that we're really following here is federal rule of civil procedure 56C, which holds that when a moving party establishes there are no genuine issues of material fact, that party is entitled to summary judgment. Abbott failed to meet that burden in this case. And I would like to reserve two minutes, if I may, for my rebuttal. I forgot to ask you. This is a case with numerous, numerous hotly disputed issues of material fact. I think it's important for the court to consider who Michael Buhl was before his career tumbled. He was a high-performing employee at Abbott from the time he commenced employment in 2009 until he complained about discrimination in 2016 and blew the whistle in June of 2016. He was a confident employee for this very long period of time, and his competency had never been questioned. He developed two patents for Abbott. He received a retention bonus in 2013. He always received excellent performance reviews. He was selected to manage the project, which he declined. He was given a promotion in 2014. He was given a rating of best of Abbott in December 2015. And he was awarded performance-based restricted stock units in March of 2016, only two months before he filed his first complaint. Management liked him so much that he was the one that they chose to make presentations to customers about the hematology analyzer that he was working on in May and June of 2016. If they really had issues with his performance, as a magistrate judge said, why would they have asked him to do that in May and June of 2016? These are questions of fact for a jury to decide whether or not the actions taken against him were discriminatory and retaliatory. His career came tumbling down after he brought these claims. And while it may be unpopular, Michael Buell, a Caucasian male, is entitled to the protection of the California Fair Employment and Housing Act based on national origin and race. And he is protected from retaliation if he complains about discrimination. Buell also has the protection of the California Labor Code, which prohibits employers from retaliating against employees or refusing to partake in activities that violate the law. Since both of the claims of retaliation involve the shifting McDonnell-Douglas test, I would like to discuss some of the critical material issues of disputed facts with respect to both the FEHA claim for retaliation and the whistleblower claim. Let's look at that. The statute is, the statutes are, the law is, that you can prove your prima facie case by showing that you made a complaint, you suffered an adverse employment action, and a causal link exists between the protected activities and the adverse action. What happened? He complains in May that Xi was discriminating and showing bias towards him and mistreating him. Later in May, Xi sends an email to Burgess, Buell's boss, saying that he is not planning on using Buell's algorithm for the whole blood in the HSQA analyzer, but will use Lin Hong's. I'm sorry, if I could just interrupt you. I'm going to focus kind of at a 30,000-foot level here, but by the time that Mr. Buell sends his first protected email in early May, the relationship between him and Mr. Xi had already deteriorated significantly, correct? Yes. I mean, I can't recall whether it was the May 6th email or the May 9th email, but Mr. Buell says the relationship is irreparably broken. Mr. Buell has no respect for Xi. Mr. Buell accuses Xi of creating a toxic work environment. So before the first protected emails ever sent, it's clear that Xi at least has serious, serious problems with Mr. Buell. Fair? I would disagree with the court's assessment that they were serious problems. It is our and Mr. Buell's belief that he was being treated differently because of his race and his national origin. So I would disagree, and I believe that's a factual issue for the jury. Well, I asked the question, did Xi have serious, maybe I should have made it clearer, but it's clear that Xi, his immediate supervisor, was having serious, I guess at least was expressing serious issues with his work before Mr. Buell engaged in protected conduct. Is that right? Again, I would say that Xi did that when they were in private, but whenever Mr. Buell presented his algorithm in the data review group or any place else where people were doing a critical analysis, Xi never expressed any concerns. And in fact, Elsa Virgis' testimony is, and she was Xi's boss, that as of May of 2016, Xi had never expressed to her any concerns about Michael Buell's performance. So Mr. Buell is complaining in early May that Xi has made his life hell, and he can't work with him anymore, it's irreparably broken. And I mean, we've all had bosses that made our life hell and we didn't like working for. That's not illegal. What's illegal is doing it for racial animus. What do you think is the one or two pieces of evidence that are most powerful in showing that when Xi was treating Buell badly, he wasn't just a boss who treated an employee badly like lots of bosses do, but that this was someone motivated by the fact that Mr. Buell was white? I think one of the most important things, and first for the retaliation claim that I was just discussing, we don't have to prove the underlying discrimination, you understand that, okay, is that he did not treat him in the same manner that he did the Chinese algorithm engineers. In other words, he secretly was creating another algorithm, never shared that with him, and then accuses Buell of not sharing his information. He excludes him, which is a common discriminatory tactic often found in gender discrimination cases, where women claim they're excluded from certain activities. I think another one that's very important, and although you could say that it's not itself racially biased, is constantly referring to these three gentlemen who were he allowed to work on the algorithm as my guys, and not making that same reference to Buell. I had a question about that, because Buell's declaration, I think it was paragraph 29 or 27 maybe, the my guys phrase referred to, in Buell's telling, the people Xi had known at Princeton, some of whom were Chinese American, but not all of whom were. So why isn't that evidence of discrimination in favor of Princetonians, but not racial animus? That's a good point. It is because the three that were actually being brought in were of Chinese descent. The other ones were software people. So when he was making that reference, he was talking about the algorithm engineers, and those were the three that we know that are Lin Hong, Mr. Yu, Mr. Wu, and Mr. Hong. So that is the difference right there. But it's still that they were people that he knew and had worked with before at Princeton. I guess what I'm struggling with, you make a lot of the my guys comment, but that's not an inherently racial term. I think you understand the issue. I totally agree, and I think in the court, the International Brotherhood has all of the case laws, so you have to look at the totality of the circumstances. In this case, he did express to Curcio, if you recall, about the mistreatment that he was receiving, that Xi wouldn't let him annotate the data. Xi was doing all of these things to him and treating him differently than the others, and Curcio actually, which I'll get to, never really investigated those claims of mistreatment, only investigated the issue of the people that were hired in Xi's group. So if I could just go on, I want to talk a little bit about what happened after he makes his complaint. So if you recall that in May, after Xi had sent the email saying that he's not going to use his algorithm, then in late June, Buell is asked to make presentations regarding the algorithm to customers who are thinking of leaving, and he's praised for these presentations. He then reads about the Theranos case and learns that their executives at Theranos have criminal charges against them for falsely presenting what their analyzers did, and so he expresses on June 28th, I think reasonably, that he doesn't think that it would be ethical for the company to be presenting to the customers that it was his algorithm being used in the analyzer, and he does that in a memo, and he says, I think that the other gentlemen should do it, and he sends it only to those people who were in this small group. The day after he sends that, the day after, Burgess and Buell tell Human Resources they support a review of his termination. Interestingly, Abbott no longer uses Buell's algorithm when they make presentations going forward because he had called them out. He continues to complain about retaliation, and he rebuts the review because the review has false information in it. On September, he informs the OEC investigators. Sorry, Ms. Lawless, something you said wasn't consistent with my memory, and I just want to make sure. You say that when Mr. Buell, in the end of June, says he doesn't feel comfortable doing this anymore, that he just contacted kind of the need-to-know people, and before this, of course, he'd been warned that he needs to stop going outside his chain of command, outside of his department, that that was unprofessional. Now, in my notes, and I took these notes myself, I could have gotten it wrong, I have him in his June 28th email copying 17 people on that, many of whom were outside his department. In other words, doing exactly what his superiors at Abbott had told him not to do, airing grievances widely and outside of his department. Is that wrong? Your Honor, it's my understanding that he was basically told that he should stop sending any inappropriate emails, and it was Mr. Bolton who agreed that this email was not inappropriate. What he was trying to do was to tell people, I don't think I should be doing this, I think that these other algorithm writers should be doing it, I don't think he used any inflammatory language, and they said the next day, well, we're going to write you up for termination, but then they didn't terminate him for those. They did not terminate him. I think it's very important to understand the cases that are cited by Abbott. You have a person who allegedly engaged in inappropriate behavior, that person's terminated right away. That did not occur with Mr. Buell. What they did is they said, we're going to, you know, later they terminated him for his performance. And that's one way to show that the defendant's reasons for terminating him are pretextual. And we showed that in many different ways. One, they didn't follow the PIP. I think it's a factual. Counsel, you're cutting into your rebuttal time. Do you want to save? Yes, I would just like to say the dishonest reasons that we set forth in great detail show that the performance review was not accurate at all. And I also think it's very important for the court to look at those investigations because they did not do adequate investigations pursuant to the law. So I will stop now and reserve my rebuttal time. Thank you. Good morning. Karen Selden on behalf of Abbott. Before I respond to counsel's specific points, I'd like to zoom out to the broader sequence of events that led to plaintiff's discharge. In late fall 2015 and early 2016, long before any HR or ethics complaints, and at a time when Mr. Shih was managing the Alinity project only on an interim basis, that's when plaintiffs became unhappy with the direction that the white blood cell algorithm was taking. Now, plaintiff's algorithm, the neural network-based approach, was conceptually innovative, but in practice it had difficulty detecting disease cells, and he doesn't dispute that. So Mr. Shih, who was brought in to bring this project back on track and had to meet an ambitious December 2016 launch date, made the decision to pursue a hybrid approach and to have his Princeton team develop what they called special case algorithms based on legacy technology. Now, plaintiff was clearly unhappy about this hybrid approach, as we can see in emails going back to late 2015 and early 2016. But the fact that Mr. Shih landed on this hybrid approach before he became the permanent manager, before plaintiff had made any HR or ethics complaints, completely undercuts plaintiff's central theory, which is that developing this alternate algorithm was part of some racist conspiracy to bring in Chinese developers from Princeton, or that it was done in retaliation for his later complaints. But the more plaintiff became convinced that Mr. Shih was moving away from his algorithm, the worse his performance got. As Judge Silch pointed out, by May 9th, he had told Ms. Burgess, Mr. Shih's supervisor, that my working relationship with men is irreparably broken. I can't work for someone for whom I have no respect. And his performance subsequent to that, and even before, was consistent with somebody who didn't respect his supervisor. He became increasingly disengaged and passive aggressive. He missed meetings. He resisted coaching and critique. And all of this we don't have to take just Mr. Shih's word for. We have contemporaneous emails. We have performance management documents that were done with the input of HR. We have an HR investigation as to whether those performance critiques were retaliatory, and HR concluded that they were not. And so all of this amply supported Mr. Shih's discharge in October of 2016. Now I'd like to just turn to some of the specific points that counsel raised. With respect to race discrimination and the My Guys comment that Judge Miller, you were pointing out, you're correct, it doesn't carry any racial connotations. The plaintiff hasn't shown specifically that it was intended as such. And his characterization of the term in his declaration, he says Mr. Shih said it constantly, but he doesn't provide any specific examples of it. And, in fact, if you look at Mr. Shih's deposition testimony, which preceded this declaration, and if you look at his response to Mr. Curcio during the HR investigation, he said in his depo he would constantly refer to his team around the organization as My Guys. That's at 3868 of the excerpts. And to Mr. Curcio he says he always refers to his team as My Guys. And as you pointed out, his team was not solely consisting of folks with Chinese-sounding names, and it was a diverse team. And with respect to picking the algorithm, what plaintiff is essentially asking this court to do is to sit as a panel of algorithm engineer experts and to decide whether Mr. Shih made the right decision and made the right call. But that's not the standard. What he has to show is that his algorithm was picked for reasons that were based on race. Plus he has to show that not having his algorithm picked was an adverse action. Just one second. My recollection is that you have conceded the prima facie case as to retaliation on part of the allegations, including the whistleblower allegation related to the algorithm that we've been talking about. Is that correct? Your Honor, yes. That is correct. Only with respect to the marketing of the algorithm, not with respect to the other two complaints regarding the reagent and the launch date. Right. And then you've also conceded the prima facie case on the retaliation claim based on discrimination, right? On race discrimination, Your Honor, we had not conceded that. And, in fact, the district court found that a plaintiff had not satisfied his prima facie case on race discrimination. Okay. And we believe that that decision was correct. But, of course, Your Honor can also go on and examine pretext on a de novo basis as well. And so with respect to the picking of the algorithm, as I said, it is not an adverse employment action to not have your algorithm picked for a project. That was not the reason Mr. Shee was, I'm sorry, plaintiff was discharged. It was because he could not get on board with the direction of the project once it was determined that his algorithm would not be used. And with respect to the timing of when that decision was made, plaintiff's counsel said that it was definitively made in May. And that's simply not correct. Whole blood means using the algorithm for testing human blood. There are other uses for algorithms. The record shows that the final decision about what to do with plaintiff's part of the algorithm was not made until August of 2016, and it was communicated to plaintiff in September. So the die was not cast that early, and it was not cast for reasons of the complaint. Isn't there a question of fact on when a decision was actually made? Because there are earlier communications that indicate that this shift is happening. There are early indications that they are moving away from plaintiff's algorithm. That is correct because of these concerns that Mr. Shee had noted going back to late 2015 and early 2016 about the inability of the algorithm to detect diseased blood cells. And again, this is while Mr. Shee was on the project on a temporary basis and would have no incentive to torpedo his career, Mr. Buell's career, in favor of his Princeton team. And, in fact, we have communications from late February 2016 where Mr. Shee tells Ms. Burgess, you know, I've developed some special case algorithms here, and I'm ready to hand them back off to the Santa Clara team when I go back next month in March. So these were dynamic decisions that were ongoing. There was not a definitive date on which it was determined before August 2016 that Mr. Buell's algorithm is out. And you can see in the emails back and forth continuously between Mr. Shee and Mr. Buell about tests that he's running and analysis that he's performing that are not up to par. And Mr. Shee is giving him feedback that Mr. Buell is continuing to resist. I would like to also address the point that counsel made with respect to the timing of the June 28 complaint followed by the notation about a review for termination. The fact remains that, yes, that notation is there, but Mr. Buell was not terminated in June of 2016 or in July or August or September. And instead what happened is Mr. Curcio reached out to Mr. Buell to get his side of the story and invited him to provide any further information he might have about any concerns he has with respect to Mr. Shee. And they referred his complaint to the ethics office, which is what they were supposed to do. The ethics office investigated that complaint. And the fact of the matter is the marketing presentations were not a big deal. Once Mr. Buell complained about them, Mr. Shee, who incidentally had no idea that Mr. Buell was making these presentations, said, our guys shouldn't be doing this. It should be the medical affairs people. And that's what happened. The following month, Mr. Hoffman out of medical affairs started making those presentations. So the idea that Mr. Burgess and Ms. Bolton would be so exercised because of this marketing complaint is just not a rational inference. What is a reasonable inference is what Judge Shilts pointed out, which is this was now the second time that Mr. Buell had made a mass e-mail complaint, putting his supervisor on blast for his supervisory decisions, calling him cunning and operating for political reasons. That was the May 6th e-mail. And in this June 28th e-mail, telling a group of 17 people, some of whom are in Europe, many outside of his department and his reporting structure, telling them that for some unknown reason, his supervisors reinvented the wheel. And he had specifically been counseled on those points. And that was the reason why they were upset. And that is clear and fairly uncontroverted on this record. I'll just make sure I don't have other points on the... Oh, and with respect to the investigations, I know counsel's critiquing the scope of the investigations, but that is not the standard again. Investigations don't have to be perfect under this Courts of Alamont case. They merely have to have been conducted in good faith and not be so inadequate as to raise an issue of pretext. And the court can see there were three separate investigations here, done by three separate investigators, multiple witness interviews, detailed reports. And so just nitpicking and cherry-picking the conclusions doesn't raise a tribal issue. With respect to the PIP issue, the district court was correct to point out that this idea of a performance improvement plan, there was no competent evidence in the record to show that that, in fact, was the process. We have hearsay evidence from one manager. We have another manager who stopped managing people in 2015. The only competent evidence here is from the HR person, which is Mr. Percio, who said, we changed our process and we applied the current process, which is a performance memo, to Mr. Buell, and it's really an issue of semantics because whether you call it a PIP or a performance memo, the performance was managed for months at a time, and it did not lead to improvement, and that was the reason for the discharge. I was going to ask you about that. What is the difference between a PIP and a performance memo? The performance memo looked like every PIP I've ever seen. It looked like a PIP on all things but names. What did Abbott consider to be the difference between the old PIPs and the new performance memos? I think it's exactly what Your Honor said. It was a semantics issue because the concern that Abbott had, and this is in Mr. Percio's declaration, is by calling it a PIP, there was concern that they were setting up an expectation of some sort of progressive discipline, and that was never Abbott's policy, and so they moved away from that wording and moved to a more holistic, what they called a single performance memo, which laid out in more detail what the expectations were. So I think it was more a concern about not setting up an expectation of progressive discipline. Unless Your Honors have any other questions, I will concede the balance of my time. Thank you, Counsel. Thank you. I think it's very important to really look at the facts of this case because at the time that Mr. Buell was given the performance deficiency memo and told that he had made all of these errors, he contested that these were wrong. These are factual issues for a jury, and you may be able to sit up at the high level but here is an employee who has been a fantastic employee his whole entire career, and he has not done the things that the performance review says that he's done. We have submitted substantial, and I'm sorry, very technical evidence showing that the criticisms of him were inaccurate, and that once they gave him that performance memo, he did not do anything wrong that was insubordinate or unpopular. And I would really ask the Court to look at the reputation of all of the various tasks that they claim that he didn't do properly and he claims that he did do properly, whether it's the hazard score test or the GR 481, all of those, because they're claiming that they terminated him for poor performance, and we have rebutted that evidence. It is for a jury to determine whether or not they had legitimate reasons to terminate him, especially after he had complained. These are factual issues that have to be determined, and a jury should be allowed to do them. Finally— Thank you. You can finish your sentence. Okay. Finally, I do believe that the investigations—it's easy to come into court and say, we put him on a performance review, we did everything. We investigated, we did everything. But an investigation doesn't mean talking to someone for 20 or 30 minutes, going out and getting data and not coming back to see if the data that you have is correct, which is what Olson did, which is what—and Curcio didn't even investigate his claims of mistreatment. Santini had information that came to her, and she never went back and followed up with him, and they relied on false information. These are also evidence of a pretext. Thank you, counsel, and thank both counsel for their helpful arguments. Thank you, Your Honor. Thank you. The case is submitted. Thank you.