The U.S. Supreme Court's historic ruling in Dukes v. Wal-Mart
this year has been the subject of intense debate in labor and employment law firms. Speculation that class action claims would drop off precipitously has not borne out, and instead filings have increased by 10 percent since the ruling. Plaintiffs that once may have banded together in a nationwide class action may now file regionally-based complaints, presenting something of a Faustian bargain for the defense: it will be much more cumbersome to settle serial class actions.
Our panel of experts discusses these issues as well as the NLRB's memorandum on social media and the application of AT&T Mobility v. Concepcion.
They are Dan McCoy and Victor Schachter of Fenwick & West; David Borgen of Goldstein, Demchak, Baller, Borgan & Dardarian; Jody Landry and Garry Mathiason of Littler Mendelson; Kerry Freeman of Miller Law Group; and Mark Budensiek of Rutan & Tucker. The roundtable was moderated by California Lawyer and reported by Krishanna DeRita of Barkley Court Reporters.
Moderator: What impact will the recent decisions from the U.S. Supreme Court (AT&T Mobility v. Concepcion, 131 S.Ct. 1740 (2011)) and the California Court of Appeal (Brown v. Ralph's Grocery Co., 197 Cal.App. 4th 489 (2011)) have on arbitration agreements?
With these decisions arbitration has become a very important vehicle in enforcing class action waivers. However, some caveats are important. The courts will continue to look at whether an arbitration agreement with a class action waiver is too one sided. In Kanbar v. O'Melveny
(2011 WL 2940690 at *6 (N.D. Cal.)), the federal court ruled that mandatory "arbitration agreements are still subject to unconscionability analysis." On the other hand, the possibility that class action waivers might prevent consumers from vindicating their statutory rights will not necessarily preclude enforcement of the waiver. (Kaltwasser v. AT&T Mobility
, 2011 WL 4381748 (N.D. Cal.)). While I believe state courts will respect the thrust of the AT&T
decision, and class action waivers will now be presumptively upheld, issues remain. For example in Brown v. Ralph's Grocery
the arbitration agreement was denied enforcement because PAGA claims were included in the waiver. Though I doubt this decision will be upheld on appeal, the law in this area is clearly evolving.
is a watershed decision confirming that under the Supremacy Clause the Federal Arbitration Act takes priority over contrary state statutes. Class action waivers will be presumptively enforced as part of the agreement to arbitrate. Concepcion
is a commercial transaction case, but the holding should apply with equal force to employment agreements. The test for unconscionability applied by the California Supreme Court in the Discover Bank
case (Discover Bank v. Superior Court
, 36 Cal.4th 148 (2005)) was rejected as being contrary to the FAA's policy favoring arbitration. Across all 50 states most employment arbitration agreements will be enforced, including the waiver of class actions. While it is agreed that one-sided, unbalanced agreements are still challengeable as unconscionable, this is now the same as for any other written contract. Also under Rent-A-Center, West, Inc. v. Jackson
(130 S.Ct. 2772 (2010)), the parties can even agree to have unconscionability decided by the arbitrator rather than the court.
Applying the FAA and the Supremacy Clause, arbitration agreements should be able to waive state PAGA claims in favor of individual arbitrations, the contrary Brown
decision notwithstanding. While the PAGA issue works its way through other California and federal courts, the practical impact appears minimal. With 75 percent of the recovery going to the state, this is not a lawsuit of choice for plaintiffs counsel and their clients.
Prior to 2010, many of our clients were skeptical of arbitration for fear that even with a class action waiver they could be forced into a class arbitration. The U.S. Supreme Court ended that concern with its Stolt-Nielsen
decision (Stolt-Neilsen S.A. v. Animalfeeds Int'l Corp.
, 130 S.Ct. 1758 (2010)) that precluded class arbitration without specific written agreement. While some courts have twisted the terms of an arbitration agreement to allow class arbitrations, a written waiver is almost universally enforceable. The remaining question was then answered in Concepcion
: Class actions could be waived both in arbitration and civil litigation by agreement of the parties.
With the Supreme Court enforcing arbitration agreements, the battleground has shifted to the NLRB. Section 7 of the NLRA protects an employee's right to engage in concerted activities regarding working conditions. It is argued that class action waivers violate that right. Prior to Concepcion
, the NLRB's then-general counsel concluded that an employee could not be disciplined for bringing a class action covering working conditions, but the court would determine the enforceability of the class action waiver, balancing the interests of the FAA with those of the NLRA. Now the NLRB has taken briefing on this issue (D.L. Horton, Inc.
(NLRB Case No. 12-CA-25764)) and the acting general counsel is potentially reconsidering the issue.
If the NLRB determines that an employee cannot be forced to waive bringing a class action as a condition of employment, the federal appellate courts will then determine whether class actions are a procedural or substantive right. With the strong federal policy favoring arbitration, it is likely that the waiver would be found enforceable under the FAA rather than prohibited by the NLRA.
For those who want more certainty, a potential silver bullet exits for defeating attacks on the class action waiver. Consider giving the employee 30 days to opt out of the arbitration agreement after it is first accepted. In Horton
and most of the cases arguing unconscionability, the arbitration agreement with a waiver is a required condition of employment. The opt-out provides more support for the argument that the agreement is voluntary especially when jobs are scarce. All employees have the right to opt out of a class action under Rule 23. Accordingly, opting out or accepting the offer of arbitration with a class action waiver in advance of litigation should almost certainly not violate the NLRA.
On Section 7 of the NLRA, the beauty of that argument is that it gets around the Supremacy Clause, which was the basis of Concepcion
. You don't have preemption of the FAA as opposed to another federal statute like the NLRA. Class actions can be concerted activity. It goes to the argument of whether a court will enforce an arbitration agreement if it's an unlawful agreement because it violates the NLRA.
In any event, trial courts will be reluctant to acquiesce to the overreaching by the very conservative Scalia Supreme Court. There will be a fair amount of resistance to application of Concepcion
. At the outset one can predict greater scrutiny as to whether the agreement is valid in the first place. We also have a whole series of questions of whether certain precedents are affected by Concepcion
Is Gentry v. Superior Court
(42 Cal. 4th 443 (2007)), still good law? Concepcion
. It doesn't say anything about overruling it. It says, "We are overruling Discover Bank.
" The courts that have looked at this matter so far - including Plows v. Rockwell Collins, Inc.
(2011 WL 3501872 (C.D. Cal.)) and others - say Gentry
is still good law.
We also hear a lot of noise about whether Armendariz v. Foundation Health Psychcare Services
(24 Cal. 4th 83 (2000)) is still good law. But Armendariz
just gives guidance as to the unconscionability analysis under California law. We also have the issue of the representative action, which is a different animal than the class action. There is also a precedent that injunctive relief is not appropriate for this kind of class action waiver. (See In re DirectTV Early Cancellation Fee Marketing and Sales Practices Litigation
, 2011 WL 4090774 (C.D. Cal.).)
Finally, assuming the plaintiffs were not subject to removal under CAFA, the California Arbitration Act governs and gives the courts discretion to stay the arbitration and keep the case in the trial court to try the PAGA and/or the injunctive relief action.
Shortly after Concepcion
was decided, the Arbitration Fairness Act of 2011 was introduced in Congress to exclude employment and consumer matters from coverage under the FAA, so guessing where the courts will take this issue may not give us the entire picture if the legislature acts. The employer community was perhaps quick to do a "victory dance" after Concepcion
- in part because of the possibility that the unconscionability barriers raised by Armendariz
may have fallen away, but I believe it is going to take three to five years to work this out. There have been a couple of district court cases now opining that Armendariz
was abrogated in part by Concepcion
(Oguejiofor v. Serramonte Nissan, 2011 WL 3879482 (N.D. Cal.) and Ruhe v. Masimo Corp.
, 2011 WL4442790 (C.D. Cal.)), but neither of those courts has clearly said in what part Armendariz
was abrogated by Concepcion
This area is now so murky in terms of PAGA and Ralph's
, such that the decision to install these class action waivers is enormously more complex. Ralph's strikes me as wrongly decided in terms of the court's attempt to distinguish class actions from PAGA claims. I was not moved by that distinction and hope that the California Supreme Court revisits the case. That said, the supreme court's decisions on arbitration clauses in the last five to seven years are certainly not employer-friendly, and indicate a strong inclination to scrutinize them much more heavily than back in the '90s.
There may have been a premature victory dance by the defense bar and a premature funeral by the plaintiffs bar. As a practical matter, while these decisions are something that we take note of when we advise our clients, the issues of enforceability are still far from well-settled and moving to enforce these agreements adds another step to the litigation that our clients are often not willing to fight over. They'd rather fight battles they feel confident they'll win than test the boundaries of the law.
Picking up on the issue of unconscionability. One additional idea to fight this argument is to provide an opportunity for employees to opt out of the arbitration agreement early in their employment. That allows more time for the employees to make a decision as to whether or not they want to participate. Initially some companies are resistant to this approach. But employers who have allowed for this have found that less than 3 percent of their employees actually opt out. So there's still a good opportunity for arbitration of most claims and the employer is going to have an increased chance to succeed on its motion to compel arbitration in the event it faces an argument that the agreement is unconscionable.
Moderator: How does the NLRB's position on employee social media communications impact your practice?
Employers that have social media policies need to review them to determine whether or not they need to revise the policies in light of the memorandum put out by the general counsel of the NLRB. The NLRB decisions are finding many of these policies to be overbroad and in violation of the NLRA. For that reason employers need to look at what they are trying to prevent and ask what is the legitimate business purpose of each provision in the policy? It can then tailor its policy to meet that business goal.
If an employer adds some disclaimer language at the end of the policy that specifically provides that none of the rules or restrictions is meant to discourage or take away the employee's Section 7 rights, that will hopefully act as a savings clause to uphold the validity of the policy.
The NLRB memo and decisions addressing the use of social media do not provide a bright line test as to what is and what is not acceptable. For that reason it is important to proceed cautiously in imposing discipline or terminating an employee in connection with a communication on social media. The cases where employees posted something to social media on their own time, on their own social media account, and the communication arguably was an attempt to talk to their fellow employees about a work issue, were found to be engaged in protected activity. Thus terminations due to such conduct violated the NLRA.
In another case an employee of a nonprofit facility for homeless people went on Facebook, when he was on the clock, and made fun of the mentally disabled clients to a friend. This was discovered and he was terminated. The NLRB found that the termination did not violate the employee's Section 7 rights because there was no concerted activity; this was clear misconduct while the employee was supposed to be working.
The bottom line is that employers have to develop thicker skins regarding certain things. Employees are likely going to be allowed to say things through social media channels about their employer/managers that employers might not find very flattering and that conduct will be protected if the employees are trying to rally fellow workers to action over a policy they perceive to be unfair.
The general counsel's advice memoranda seem to draw a pretty fine line. You can say, "This is the stupidest employment situation I've ever been in," and it won't be found to be concerted activity absent some call to action like "everybody go down to the SEIU Local Union Hall."
The NLRB has taken a restrained view of what social media communications will be considered concerted activity. The larger question is how have social media invaded our lives as employment lawyers in litigation generally? It's a sea change.
Anyone interested in this area should review the August 18 report on social networking cases issued by the NLRB general counsel. The outcome of these issues is very fact specific. These rulings on protected activity are a very mixed bag, creating chaos for employers trying to establish a professional environment in their workplaces. Our clients seek to maintain a respectful atmosphere, which prohibits bullying or harassment. When the board starts looking at what people do unrestrained on their social networks, and rules that defamatory and disrespectful conduct might be protected under Section 7, even if it violates standards of professionalism, it creates a morass for employers. However, there are a few clear messages: Individual griping through social media is not going to be protected; calls to group action are likely to be protected unless it is maliciously defamatory or false; social network policies with broad prohibitions against disrespectful or unprofessional behavior are going to be subject to close scrutiny. Employers trying to maintain an environment free of bullying and harassment may find that their policies may run afoul of protected Section 7 activity.
Having reviewed the NLRB's position, and surveyed cases regarding when employers can rely on information they've learned from monitoring Internet activity, it is interesting the attention placed on where this activity occurs - that is, is the employee on Facebook at work through the employer's systems or are they doing it away from work on their own time and on their own computer. It's also interesting to see the importance placed on what the employer's policies permit. A lot of employers recognize it's unrealistic to tell employees they can never go on Facebook, and they are reluctant to do so. However, if an employer forbids all personal use of the Internet on the company's computers, then courts generally allow them more leeway to monitor employees' personal usage and factor what they learn into employment decisions.
Meanwhile, there are risks to having policies that allow some personal Internet use (such as policies that say, "while on your lunch break, you can send personal emails"), yet provide that all Internet use will be monitored. Employees may argue that where personal use is permitted, just because something was typed on a company computer does not necessarily mean an employer can monitor and use that information. Also, there may be information regarding an applicant or employee's protected classification that the employer doesn't want to know and should not know but nonetheless learns, or "captures" in its monitoring and thus may arguably be deemed to know, giving life to claims that would go nowhere if the employer could just claim ignorance.
From an employer's perspective, the NLRB's test is unhelpful because in focusing on concerted activity, the board is creating a test that can't be applied until after the communication has been made. It's very difficult as an employer trying to write a policy to know whether a particular posting on a social media site might involve concerted activity until other people come into the conversation, and by that time the thing that you wanted your policy to protect the company against, is already out the door.
At a deeper level, the board's treatment of this issue reveals a culture clash, because the NLRB has always viewed electronic media as an extension of the shop floor. The trouble is that the kind of speech the NLRB considers appropriate in its world of union organizing campaigns and picket lines, includes a lot of things that many of our clients don't think are okay in their offices. For this reason, it is unfortunate that law on social media is being developed by the NLRB rather than by legislatures or courts. Another issue coming with the development of social media is the employer's use of social media in the hiring decision. Many employers now at least think about Googling a job applicant's name before making a hiring decision. In Germany, where privacy laws have developed further than ours, employers are prohibited from considering information on social media websites in hiring decisions.
In Turkey it can be a crime to use information collected through social media as part of the hiring process. Global employers need to consider local requirements. Meanwhile the NLRB is struggling to be relevant in the 21st Century with 92.5 percent of private sector employees being non-union. Since Section 7 recognizes the right of employees to engage in protected concerted activities even without a union, social media communications about working conditions has become a new frontier for the NLRB.
In this effort I see the NLRB overreaching for much the same reason Mark [Budensiek] explained. Protecting "shoptalk" in the 1940s is very different from broadcasting company-disparaging remarks to fellow employees and the entire social-media community, causing multi-million dollar reputation damage. Immediate solutions for employers: Scrutinize social media and electronic communication polices for Section 7 compliance, include the disclaimer described by Jody [Landry], and narrow the policy to cover only what is business justified. Thereafter anticipate that the law in this area will mature as cases reach the full board and enforcement efforts result in appellate court guidance. The NLRB is down to three members out of five and Senate confirmation of new appointees before the next election is unlikely. In 2013 as many as four new board members could be selected, completely changing the direction of the board. NLRB regulations and decisions are already becoming part of the U.S. presidential debate.
There are three key issues for social media. The most prominent issue in the last six months is concerted activity. The second is hiring, and the impact of social media on hiring decisions and discrimination law, and specifically, employers becoming pregnant with knowledge about applicants' protected characteristics by simply going to Facebook, LinkedIn, or some other social-media platform.
The third issue, confidentiality, has received little treatment to date in the case law but it represents fertile ground for claims. Misusing a company's confidential information on and over social media sites, including posting proprietary business information on LinkedIn profiles (typically in a non-malicious manner), and otherwise using those tools to post information that your employer doesn't want you to disclose about the company's secret sauce, constitutes an area of growing risk for employers.
When very small businesses want to put together an employee handbook and ask, "I've got 20 employees and I'm not covered by the FMLA, tell me what are the four or five key policies I need?" social media is now on that list. I feel very strongly about that now as compared to a year ago.
Moderator: How does the U.S. Supreme Court ruling in Dukes v. Wal-Mart Stores, Inc. (131 S.Ct. 2541 (2011) decision below, 603 F. 3d 571 (9th Cir 2010)) change your litigation strategies in class action cases?
Unfortunately, there's now a Dukes
motion in every case we have. The defendant is filing a motion to decertify, reconsider, or never certify, and the underlying case may have nothing at all to do with Dukes
other than that it's an employment class action. The employer won't make a lot of headway on Dukes
, and it's a waste of everyone's time and precious judicial resources.
That said, the plaintiffs bar will now look at cases that we might have once alleged nationwide class action allegations, and really try to pinpoint and tailor the class action allegations and definitions so that we have a narrow class that won't look or smell anything like Dukes
It should not impact our carefully tailored cases on other issues in cases that are not similar to Dukes
. For example, the Dukes
opinion went out of its way to carve out cases of discriminatory testing programs. Dukes
should also have limited impact on wage-and-hour litigation because these are relatively straightforward cases that have nothing to do with subjective decision-making. It should have limited impact on collective action litigation under the Fair Labor Standards Act Section 216(b), which is a completely different animal than Rule 23.
Under California law we have California Code of Civil Procedure 382 that does not build in the structure of Rule 23 and some of the criticisms that the Dukes
court had with regard to the use of (b)(2) certification. So that's not applicable under California state law, and we are encouraged since Sav-on Drug Stores, Inc. v. Superior Court
(34 Cal. 4th 319 (2004)) makes cases certifiable for class purposes where there are important public policy interests like labor law enforcement, which the Dukes
case was not facing.
Furthermore, the old line between what is merits discovery and what is class action or class certification discovery, has been effectively obliterated. Dukes
says we are going to look at the merits and therefore will have broader discovery at the class certification stage. Courts are going to do the more rigorous analysis under Falcon (Gen. Tel. Co. of Southwest v. Falcon
, 457 U.S. 147, 156 (1982); see Dukes
, 131 S.Ct. at 25502557)) and that's going to include a merit-based analysis of whether there's evidence supporting the class action theory that can be adjudicated on a class basis. That's going to bleed into practices on the uses of experts at the class certification stage. Dukes
can be fairly read to say that you can or must have a full blown Daubert
analysis (Daubert v. Merrell Dow Pharmaceuticals, Inc.
509 U.S. 579 (1993)) at the class action stage and then go beyond the admissibility of the expert's opinion to determine whether the expert's analysis should be given sufficient weight as opposed to other available evidence or other experts.
In the Ninth Circuit, we have one decision in Ellis v. Costco Wholesale Corp.
(2011 WL 4336668 (9th Cir.)), that sets ground rules in the Ninth Circuit following Dukes
. It was a gender discrimination case that's going back to the Northern District of California for further consideration under Dukes
going to diminish class action litigation? Since Dukes
was decided there's been about a 10 percent increase in class actions filed. Rather than diminish class action litigation, what's far more likely are much narrower cases, more fact-intensive, regionally based and possibly limited to a single operating unit of a company.
The courts will be open to additional discovery since Dukes
requires a showing of "glue" connecting individual claims not just to establish commonality of questions of law, but commonality of solutions or remedies.
One of the most profound impacts will be on the development of trial plans and affirmative defenses. One of the key defenses to certification will be the need for individualized remedies taking into account mitigation. Even if some of the legal issues are common, the inefficiency of needing to individualize claims may be sufficient to defeat certification. While defense counsel is normally arguing against certification, schizophrenia is likely to set in when a settlement is being negotiated. All of a sudden defense counsel will be generally supportive of a broader class trying to extinguish the maximum number of potential claims. It will be interesting to see whether courts effectively create a dual standard - one for a contested class certification and the other for settlement. My guess is settlements will be a little harder to get approved that are too broad, and we can anticipate that this could create some additional litigation. If one operating entity out of twenty is initially covered by a class action, settlement of that unit could ignite similar litigation elsewhere. I have no doubt however that creative solutions will be found.
A point about trial plans. Every case needs a budget these days. The effects of this decision will make budgeting and estimating costs to oppose class certification much more challenging. Everyone will need to revisit their budgets.
Paradoxically, litigation budgeting might become somewhat easier since we can better predict the outcome of challenging class certifications. The grounds provided by the Supreme Court to attack class certification are considerably clearer. The lessons from Dukes
are: Attack the class action early for not having the "glue" of commonality; attack expert and anecdotal evidence; and, if we sense anything requesting back-pay relief, draw upon the Supreme Court's ruling in Dukes
that such a remedy is inappropriate for class certification because of the individualized determinations required. Plaintiffs lawyers will find it difficult to sustain a class where there is a substantial back-pay component rather than injunctive relief.
We've already seen an impact in a positive way for employers who have solid wage-and-hour policies. For example, we've seen Dukes
have an impact in mediations where employers have good policies about how to handle the meal break periods and wage-and-hour issues, and train their supervisors to follow those policies. Employers can argue that the case is about bad decisions being made by renegade supervisors, something that does not fit into the class action model.
It's been my frustration over the years that some courts do not spend enough or any time focusing on the superiority element at the class certification hearing. A plaintiff alleges that "we can get an expert that will do X, Y, Z" and somehow come up with the trial plan in the future and some courts have accepted that assertion and certified a class. I'm hopeful that the Dukes
decision will force all courts to carefully consider whether or not the proposed class is appropriate for certification and really make sure there is a trial plan in place that would allow the case to be tried efficiently, that certifying the class is superior to allowing individual actions, and the certification of the class does not violate the due process rights of the employer.
Even though Dukes
was not a wage-and-hour case, exempt misclassification cases involve the same kind of localized decision making as the facts in both Dukes
and Ellis v. Costco
, so Dukes
may well have some effect beyond discrimination cases. I also believe that Dukes
will make it more complicated to settle large class actions because the holding creates an incentive to bring serial cases rather than the one big case that won't hold together. If you're the general counsel of a national employer trying to get your litigation budget put together for the next fiscal year and you are looking at a potential row of dominos falling, it will be more difficult to put that first case to bed.