# Oral argument — Joseph Curry v. Yelp Inc. > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A31619 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** September 11, 2017 ## Text Okay, I believe you're ready. Good morning, Reverend Smith. This is the Court against the Law for the Appellants Joseph Curry, and City of Miami Firefighters and Police Officers, the Tireman Trust. The issue of this appeal is whether appellants officially pledge securities fraud based on Yelp's status about the credibility of their reviews, which is their only product, as they say, their entire value proposition, and their denial of any relationship between business advertisements and the quality of reviews. There's been an unusual feature of this case, and where I think that it's helpful in assessing the plausibility of the allegations is that we have erroneous reactions in real time to the developing story of the FTC disclosures of Yelp complaints, beginning on the morning of April 2, 2014, when thinking when a media began to leak the news that the Wall Street Journal was going to publish an article about the FTC disclosures. The next day, as his complaints were reviewed and digested, and if I could briefly outline this timeline, I think it would be helpful to show what the market understood with the FTC disclosures in real time. Okay, that's fine. But I should tell you from my own standpoint, before your time runs out, I'd really like to hear your response to the lost consumption argument. I know you get to that last year, but I want us to look at it first. I'm sort of inclined to think you're willing to try to close up stories at a certain time point. In fact, I believe that the timeline will show that the market's reaction and the analyst's reaction to the disclosures in real time shows that that's not a problem with your lost consumption argument, of course. I'm thinking of the loose case, if that's how you pronounce it. Yes. Okay. I know I didn't get you off track. If you want to go through the timeline, that's fine. I'm going to be waiting to hear what you have to say about lost consumption. I'll just tell you that. I'm going to talk about lost consumption. Okay. So, what this seems to be is an insurmountable barrier for your argument here because there we said that the disclosure of an investigational document is never going to be enough. You're going to have something more. And here it seems that you have something less. You don't even have the disclosure of an investigation by the FTC. At least that was the fact that was publicly revealed. All that was revealed is that the FTC kind of collated a bunch of complaints from folks. So, it's not even the FTC provides the level of any kind of formal government investigation. So, how could that possibly be the case? So, that's where I get confused a little bit. And, Lewis, what the court said was that the launch of an investigation wasn't enough because the market could not have any idea of what the investigation would disclose. And I would say, Your Honor, that here what we have is actually the fruits of the investigation that we have more, that we don't just have, excuse me, the disclosure of a number of complaints. But here we have sort of the fruits of that investigation we have, but there was no conclusion reached. All it was was a bunch of people that complained and lodged complaints. There had been no determination by the FTC, correct me if I'm wrong, that in fact the complaints had merit and that in fact there was some company-wide practice that it was inconsistent with the representations that it demanded, was there? Well, Your Honor, I believe that the conclusion by agency about establishing fraud is a sort of philosophization. It's just a coxie of cause between the statements, the alleged misstatements and a reaction by the market. And here we have, well, not just that, you've got to have a revelation of facts that show that the representations that the company was making are not true. And here, it's not like, if the company had said in its public statements, not only do we not engage in these despicable practices, but we've never received a single complaint from anybody that we've done that. We are just, we've got a pristine record. Nobody's even complained about that. If they had said that and then it was disclosed that actually 2,000 plus people have complained, that would be different, right? But in fact what they disclosed was, yeah, people have been complaining about this for a while. The company certainly did not engage. We don't think there's merit to these complaints. And so the mere disclosure that, well, actually, yeah, there are, we do have an exact number of the complaints that have been lodged with one particular government agency doesn't seem to me that shows that the statements that the company made are false. I think what's tricky here is the fact that the company did say that these complaints were not true, that there was absolutely no relationship. And at the beginning stage here, we have a plausible case given the number of complaints that are consistent across the country. 2,000, a majority of 2,046 complaints stating that the oil firms that have called up businesses, businesses denied, agreed to pay for advertising. Good reviews would disappear, bad reviews would show up. This is across the country. This is a story from my counsel. This is just a little if I can interject a question. There are a lot of people selling this advertising around the world, right? Right. A lot of offices. And those people doing the sales, they're presumably trained up in some way by the company. But were there any allegations in the complaint? Because I didn't notice this, that some executive had said and would testify that they were training people to extort funds from people for advertising revenue under the threat of bad reviews. No, there's nothing explicit about that. Although, the inference that was happening based on the number of complaints and as an appellee's brief, pages 15 and 16, they disclosed or could see that there were 30 representatives who claim that they told business owners that there was a connection between paying for advertising and the quality of reviews. So that's 30 there that corroborated the other 2,000 complaints, which is possible to suggest that this was going on nationwide and that it's 30. And then there's of the 2,046, there is too much where the Yale representatives are named, which shows that it is a problem. The inference is that this is a company-wide policy given that this is happening. These are parts of the country. So a related question on that point is, did the FTC ever send a subpoena to Yelp asking them for PowerPoint presentations or videos or evidence of how they trained their marketing people? There are not any actions by the FTC. But what we do have, and I think this case is, as we said, the Lloyd v. CBB case, where initially there is, in that case it was a subpoena, and analysts assumed that there was an issue in that case, which was subsequently confirmed. And I want to take a quick ride down here. What we have are these complaints, and then as analysts reviewed these complaints, it became clear that this was a major problem, as well as Fargo predicted two days before the revelations. So, yes, there are these complaints out there, Yelp denies the complaints, but if this growing number of businesses come to the conclusion that this is a problem, then Yelp's business is going to suffer. It's actually that a growing number of complaints can alter the total mix of information available, and that's exactly what happened. And so you have this timeline where beginning of the morning of April 2nd, as these FTC disclosures are being leaked, such as in the morning there's a leak by what Bloomberg discusses, the FTC disclosures that most journalists have read the article, this clock starts to come down. Well, how do we do it at that point? There are these two graphs, they're going to give you a disclosure of the contents of it. Correct. So how does that get you anywhere? That initial 6% drop, that just keeps increasing? Well, so then as that evening of the next day, as the complaints were scrutinized, an analyst was saying now that we have looked at these complaints, we know that a majority of them are about this manipulation or abuse. And so it confirms the initial understanding that the complaints, and in fact, one of the complaints, one of the media reports that came forth from the street says, right, we don't know what the complaints are yet. But given the past complaints about manipulation, there's this presumption that that's what it's about, and then in fact that's confirmed that evening of the next day, which is also part of the law specialization. So I'll just focus on April 2nd, but it's important that the law specialization claim also involves April 3rd. As the complaints were digested by the market, and the market understands that, in fact, this is a problem, and it doesn't require anything else that explains that result. Excuse me. You said anything else that explains the result, anything else in the record that would be an explanation for the problem. There's nothing else in the record. And the analysts looked at this both on April 2nd, April 3rd, April 4th. As the stock continued to go down, the market understood that the problem was that if there was a sufficient number of consistent corroborated complaints, that this was a problem, and it didn't require a government finding a fraud to do this. Let's say that instead of the disclosure of this database of complaints, let's say that on April 2nd, the FCC itself had come in and said, people, we've received a large volume of complaints about this particular type of misconduct, and, you know, we've opened a formal investigation. We'll get back to you in six months. Right? Stock prices have gone up. We've got the stock going up. That would be just like a lease, right? And you would concede that you would use it in that scenario. It's not a lease. The difference with this, with Messler, I'm not going to have to remind you of that at all. That case that I just described would be covered by the rule of lease, and if you were the plaintiff there, you would be doing this, right? Yes. Well, the difference, I think, I'm trying to understand your, I'm trying to get, I'm trying to understand how you think the hypothetical I proposed is any different. I think it's actually a stronger case for the plaintiff than the one that you have, and I'm trying to get, because that's why you, where am I getting, where am I going wrong? Okay. I think the difference is a launch of an investigation as it leans, it's sort of a preliminary indication as they say in this, the market has no idea what will be found. It's sort of before you have all these complaints. Okay. Let's say the FTC analysis, though, doesn't disclose that that has been used, but just says we've received a large volume of complaints about this specific type of misconduct that the resident has been thinking that the company might have been interested in, and therefore, on the basis of this large volume of complaints, we've launched a formal investigation. Right. It seems to me that is exactly what Goose is talking about. It says it's insufficient without something longer, right? And I think what we have more is what happens with CBD, where you have a launch of an investigation with a number of complaints, and then the market's speculation that this is actually a problem, which is confirmed once the complaint is found. Absolutely. You said a launch of an investigation with a bunch of complaints, but I thought the agency, the FTC, never launched an investigation that received complaints, but they didn't open a formal charge for CBD. It's not on the record whether they did or not. It's not in the records. That says that they did. I would point out the notes follow, I believe, a paraphernalia which says that a launch of an investigation by the SEC is not relevant. The issue is, you know, at the pleading stage, that this is company-wide policy, this is an isolated incident, and that the defendant's statements are plausibly false and material, and that once the revelation of their falsity was disclosed by these consistent corroborative complaints, then the stock drops. It's a substantial cost of the job at the pleading stage. Let me suggest this. I'm sorry if I got it. I'm sorry to talk about lost causation. You should take the rest of your time to address any other issues, and I'll give you additional time for me. So I think for the materiality we have, at MSA, the nature of this is it helps own your clients, and the credibility of their reviews, as I said, is entirely your proposition. They have on that lead, on their website, the common questions, the zero relationship, and so they warrant that this is a problem. This is a crazy issue for them. And as Will Svago pointed out two days before, if there's a growing number of complaints, it's going to alter the total piece of information. So I think materiality, especially by, I don't need to know what to talk about now, but it shows that the analysts in real time understood that this was a huge problem for them. In terms of standard, we have, again, this is the only product. They're on notice that there are these allocations about manipulation, talking about lawsuits, prior media reports, and they make statements that certainly indicate that there is a position to note that this is, what they say, is a true zero relationship, and this is one of those co-operations instances where how important this issue was because it's the only issue the company has, and it would be absurd to suggest that the company is not aware of the problem. So those are my questions. Okay. If you have any questions, let's hear some last words. All right. Good morning. My name is E.C. Cordes. I wrote up Arnold Porter K. Scholler for the Yeltsin Foundation. Before I come to the Las Clases, which I think is a showstopper here, let me put a little bit of this in perspective. Okay. In front of Judge Tiger, his lower court was undisputed evidence about the size and scope of Yelp. Yelp operated an online platform where any member of the public could post a review, a review of a restaurant, of your guitar teacher, of your plumber, whatever you wanted to comment, you had access. The scale was enormous. Okay. As of December 31, 2013, there were 50 million businesses who were subject to reviews. There were over 66 billion reviews. Okay. And the reason I say that is to put it in perspective with the release of 2,000 unverified accusations. The 2,000 unverified accusations, they were good enough to summarize in Exhibit 6 to the amended complaint. 500 of those had nothing to do with either of the two issues they raised. And very importantly, for materiality purposes, 700 of those had been released two years earlier from a different FOIA request. And we know that materiality turns on a significant alteration in the mix of information available to investors. I guess I don't see why you think this is terribly helpful for you. I mean, that audio note, April 2nd, whatever that day was, if that wasn't a good day for your client component, nobody wants to have a review to the public when there are thousands of complaints with the same request. No, what we're talking about is Yelp itself, as you noted, had anticipated this and had several times. Price is concerned. Everyone's worst fears about the shady practices your client was engaged in. So it's not a good day. April 2nd was not a great day. He threw a Yelp there and ran around saying, hey, this is wonderful news. Look what the FTC has disclosed. So I don't know why you're taking up time going through this, because I actually think this is not helpful for you. I think that you should shift to the actual arguments that are in play. That would be more helpful. Well, Judge Teicher dismissed non-materiality as well as lost causation, CN, and falsehood. I'll move on from materiality, although I do think it's quite important. The analyst reports that came out contemporaneously thought that it was something between the centuries. The analyst report said these are unverified accusations. So let me turn to lost causation. Do you flag that as an issue? The main service standard under Zeus is that you need a proper stock price, and it has to be caused by the revelation of fraud. Potential for fraud, possibility of fraud, accusations of fraud are not enough. That's what Zeus holds. And in this case, you've had nothing but accusations. There is no corrective disclosure, which is really the hallmark of most issue or D&B-5 cases, in which there is a corrective disclosure. There was never a corrective disclosure here. There was never a finding by a regulator. And, for example, one of the cases, Justice Gould just wrote an opinion in the Dose of Genetics. Dose of Genetics is a classic lost causation case, because the issuer's management had said, our products are cleared by the FDA. And the FDA then sends a letter disputing that, and the company has to recall a whole line of products. The recall of the whole line of products is the corrective disclosure that established lost causation in the Dose of Genetics. There's no such thing here. And as you pointed out, that Wall Street Journal article in the 2000 complaints, which, by the way, accumulated over six years, one-year, six-year range of complaints, had come out. Let's say Yeltsin said, we're going to conduct an internal investigation to determine whether or not these things have any substance or not, and the stock price goes down. The cases in the Ninth Circuit are crystal clear. That's not enough for lost causation. There wasn't a lot of inside selling, and about that we'll be discussing here. There was a lot of inside selling. However, the records show that the inside selling occurred in equal amounts before the class period and after the class period. And it's very interesting to think about the nonsense involved in inside selling. The classic inside selling case is the company knows it's going to release bad information, and the insiders dump their stock before they release the bad information. What here is, there's no tying in the allegations between our people knowing that there were going to be these complaints released through a FOIA request, so they couldn't have anticipated the bad information and sold the stock. And that's why the stock selling in this case is completely irrelevant, on top of the fact that the people were selling equal amounts before and after, and almost all of them pursuant to a 10b-5-1 plan, which allows insiders to sell fixed amounts over a period of time without regard to the news. In other words, your stock broker is told you're going to sell 10,000 shares a month, irregardless of the news. And that's a safe harbor under the federal securities laws, under 10b-5-1. All these people with significant sales in this case have 10b-5-1 plans. And as Justice Wofford pointed out, in the hypothetical, if the Wall Street Journal article and the FOIA information had been revealed, and the FTAs, I'm sorry, FTC, said we're going to launch an investigation, or even file a complaint, even if they had filed a complaint, it would not be enough for loss causation, because it simply is on the contingency for fraud, not the actual finding of fraud. And I wonder if I have to see it or not. Okay. Ethel, let me ask you a question, if that's okay. And by the way, we appreciate the raise, but we're judges, not justices. So here's my question. Does the record show an answer to my query to your friend on the other side of the case? As to whether the FTC ever sent a subpoena to Yelp about their marketing practices. During the pendency of this case, we put it into the record in the second motion to dismiss. The FTC announced and told Yelp, and Yelp announced, that the FTC was not going to pursue any investigation or charges against Yelp. And although they may very well have requested documents, they very well may have done some private, non-public investigation, it was undisputed in the record that during the pendency of this case, it was announced they were not going to do anything, and they haven't done anything since. There's never been a finding by anybody, not by a court, not by a regulator, not by anybody, even in private litigation, that the allegation that Yelp died advertising to refuse is true. Yeah. Let me pause right there, because that is the contrary guess that seems, I don't know, sympathetic about the plaintiff's argument, is that they're saying, well, you're basically requiring us to prove, right, before we can even get past that, as well, we see it's functional, prove the existence of these kinds of practices in the company, and how could we possibly be expected to do that without the benefit of any discovery? So how would a plaintiff ever prove, ever plead, if that's the world in which I cover? Let me answer that in two ways. First of all, the PSLRA was specifically designed to address your question, and it says you have to prove with specificity. You can't do it on information and belief, but you can't do it on the basis of third-party accusations. Remember, these are not plaintiff's allegations. Plaintiff wasn't a customer, plaintiff wasn't a business, plaintiff is relying, in this case, 100% on unverified, mostly anonymous, third-party submissions to the FTC. That's not enough. The second answer to your question is this. In a typical 10b-5 case, the knowledge of those practices comes in two or three ways. One, it comes from corrective disclosures, where a company says, we admit it's a problem. The second way it comes is from confidential witnesses with personal knowledge and information about the company's practices. Many of the cases that are cited by the plaintiffs in this case use inside confidential witnesses to establish the minimum seating requirements of the PSNRA. And that's the answer to your question about what they're expecting to do. If the Wall Street Journal article, instead of reporting on a collection of complaints that actually seem out of place, I don't think I've interviewed 10 or less former Yelp sales folks who said, yeah, this is exactly what we did and what we were told to do, right, with the same exact conduct. If that had been the case, the Wall Street Journal article, it's not going to talk to them. You can see that they had proof of that. It would have to be in their readings. The Wall Street Journal, if they said, we've interviewed Confidential Witness 1, who management instructed to do this, or Confidential Witness 2, who was trained to do this level of extortion, or Confidential Witness 3 that saw it happening, none of that is here. In fact, if you actually go to H6, which is the summary of all these claims, 90% of the ones that have anything to do with advertising are based nothing but inference. Our reviews changed. You know, there are, as Judge Tiger found in his opinion, maybe seven or eight that even cite to an unnamed Yelp salesperson, and not one of them ever connects back to management. Yeah, I know, but, I mean, God, if a thousand people have the same thing happen to them, or, oh, it's just a coincidence that when, you know, a week after we told the Yelp salesperson, no, we're not interested, all of a sudden all of my five-star reviews just disappeared, or some new, you know, negative one-star review appeared, that they're lying. See, that's probably not a coincidence that this kept happening over and over and over again. If Yelp did not have 50 million businesses on its platform, maybe a couple, 10, 20 people who said something like that might have credibility. Yelp has a filtering mechanism that constantly changes reviews, even without a call by advertising. If some business went to their reviews, they would change from time to time, even if they had never gotten a call asking them to advertise. And that was never tied, you know, the allegations here were never tied to anything. And, in fact, the only case that ever addressed this, the Levin case, which we cited, which was a state court case, found that there was nothing. And the last thing I would say on that point, which is really a very interesting point, at the end of the Wall Street Journal article, that is the foundation of this loss causation, they report on a Harvard study by a Harvard Business Administration professor who studied 3,000 businesses and summed hundreds of thousands of reviews and concluded, objectively, that there's no evidence of a tie between advertising. And I thought that was interesting because they completely ignored that part of the Wall Street Journal article. One last point on the Wall Street Journal article, if I may. In the amended complaint, they cited to an academic who had supposedly done an event study. And the event study said it was the Wall Street Journal article of April 2nd, which was the causation of the loss. And Mr. Leff has completely altered that position, but that puts him in the operative cleaning is that allegation. And Judge Tigers saw on the article itself that it didn't come out until 4.23 p.m. Eastern Time, which is after the market closed, and it had already reported on a 6% decline. So that was a completely bogus event study, but that was the one that was relied on in the amended complaint. Now you pointed out that the shortcomings of the complaint, in your view, and how some of those things may have been corrected. Isn't it logical then that the District Court was not exercising the appropriate discretion when it declined to allow an additional member of the RISD? Secondly, because the amended complaint was dismissed by the Census. Was that really appropriate? Well, they had had two chances, right? We litigated one complaint that was dismissed. They got a second chance to file the amended complaint. We litigated that again. What's critical about asking for Thief to file a third amendment is you're supposed to tell the court how you're gonna fix your plea. And they never did that. They never did that until they asked for reconsideration of Judge Tigers. So it's very important that when they asked for Thief to end their opposition to our second motion, they didn't say anything about how they were gonna fix their plea. And, Counsel, as I understand it, when they filed their motion seeking to amend, they didn't, at that time, attach a second amended complaint or say what they would amend. That's correct. That's what the motion to reconsider later. That's correct. And Judge Tiger denied the motion for reconsideration. And I would say that if you look at the pleadings here, they say that the second amended complaint merely contains all the operative facts in the operative complaint. They don't really point out how it would have changed the court's opinion. And I think you go down a very dangerous road if you start addressing a court's discretion to allow amendment when he's already allowed one. Counsel, I think that probably took you over your time a little bit with my question, but if Judge Watford and Judge Sands don't know exactly, I'd like to ask you to at least take 60 seconds to address the Sander issue, the big picture on that issue. Sure. So the Sander issue here really has two elements. The first thing that they pleaded in Sander was what we call the core operations inference, that if the senior members of management were running the core operations of the company and the issue in question involves the core operations, then you must assume that they had Sander. Here's the problem with that. Ninth Circuit cases, South Ferry in particular, or Metzler as well, say you need, and this is quoting from South Ferry, additional detailed allegations about defendant's exposure to information. Metzler says specific information conveyed to management related to fraud. What you would have to show here is that management knew not only that the sales reps might have been trying to extort advertising, and there's no evidence that they knew that, and secondly, that it was of such a significance as to threaten the financial performance of the company, because knowing that there are just a few rogue salesmen in any company is not enough to threaten your financial performance and make it material. So I think they failed on alleging the proper elements of the core operations inference. And the second area that we talked about were the stock sales. I mean, the stock sales were nonsense in the beginning because they couldn't have anticipated the release of the FOIA data, but even secondly, the sales of stock, and we presented charts and graphs to the lower court, were very consistent. Sales before the class period, sales during the class period, sales after the class period, and very substantial holdings. At the end, this is not a case where people dumped most of their holdings of stock and sold a lot of stock and it was worth a lot of money, but they still had very, very substantial holdings of Yelp stock at the end of the class period. Thank you, counsel. You've seen your argument. Let's give counsel two minutes for a moment. A lot to cover there, but I'd like to go back, Judge Wofford, to, I thought you were onto something when you were talking about confidential witnesses and the analogy there, because regardless of whether the FTC investigates, comes to some conclusion, what we have is a majority of 2,000 essentially witness statements, and the opposing counsel said these are all anonymous. They were not anonymous. As we plead in these areas, 37 and 38 of the complaints, these were identifying information. They talked about the time they were made, the location, who they were, over 200 of them named Yelp representative. As I said before, counsel concedes that 30 of them included admissions from Yelp representatives themselves to act as a connection. And so I think his point was just that these are customers of the company, not company representatives, right? So, yeah, you're right. The customers are reporting. And so the facts should be accepted as true and being at the inferences. And these aren't customers. They're more potential clients. But if these were considered as if they were... In the normal lawsuit, if it wasn't a fraud lawsuit or a securities lawsuit, you'd be on very strong ground with an allegation that you'd get a chance to discovery out. But in a securities fraud context, you have to satisfy both 9B of the federal rules for specificity and also the PSLRA, which requires some strong showing of center. And I guess I have a question of how you can do that from complaints to the FTC if you don't have evidence of some manager who's blowing the whistle on the company or something comparable to that. So there's specificity, there's possibility. And so, yes, under 9B of the PSLRA, we require specificity, which I believe we have done by telling the witness complaints. And although we don't have admission from Yelp itself, there are actually two declarations or statements from a Yelp reviewer, a lawyer who was a reviewer of the Yelp, who also described this quote-unquote between advertising and reviews. And so if you put all those together, we get the inference that it's possible that what the defendants were saying was not true. And there's loose, but then there's CBB. And so there was a subpoena that disclosed the risk, and it was much later there was confirmation that that was actually a problem. And here what we have are the complaints, which disclose the risk. And then the next day or the evening, as these complaints were reviewed, showed confirmed speculation. And as in the most recent case of Culler v. Kirchner, sort of logicalization says the court looks whether a misstatement caused loss by asking whether the subsequent public disclosures were revealed or if he suggested the truth. And that's what we have here. We don't have the proof of truth at this stage of the proceeding, that this wide array of complaints suggested the truth, that Yelp, in fact, was manipulating reviews and the FTC complaints. I think there's also an inference that just threw the iceberg. Those are people who actually went through this formal process of identifying themselves, and then have this detailed application, which suggests it's a much wider group of people who have these complaints on their heart that are involved in the process. Okay? Thank you very much. And the case is just starting to stand some ground. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A31619. Public record. Not legal advice.