Opinion
Case No: 13 C 3587
10-23-2015
THOMAS MERVYN, individually and on behalf of all others similarly situated, Plaintiff, v. ATLAS VAN LINES, INC., and ACE WORLD WIDE MOVING & STORAGE CO., INC., individually and on behalf of all others similarly situated, Defendants.
District Judge Ronald A. Guzman ORDER
For the reasons provided, Plaintiff's Motion to Compel Defendant to Produce Unredacted Owner-Operator Agreements and Shipment Data is granted [165]. Defendant must produce the documents and data sought in Request Nos. 6 and 7 of the Plaintiff's Second Set of Document Requests on or before November 23, 2015.
BACKGROUND
Defendants in this case, Atlas Van Lines, Inc. ("Atlas") and its owners, make up a cooperative of over 400 independently owned moving and storage businesses across the country. Plaintiff, Thomas Mervyn, is an independent owner of moving trucks, and leases his trucks to entities like defendants. Plaintiff claims that defendants have intentionally miscalculated the compensation owed him, and other owner-operators like him, by reclassifying revenue received to account for discounts. This reclassification reduces the amounts used to calculate owner-operator compensation. For example, plaintiff explains that defendants promise to pay drivers a fixed percentage of revenue received on a shipment and 100% of the fuel surcharge that governs each shipment. Defendants, however, pass on "freebies" to their customers without documenting those discounts. This reduces defendants' revenue which, in turn, reduces the owner-operators' fixed percentage. This, plaintiff claims, is a breach of defendants' payment obligations and violates the Truth-in-Leasing regulations. 49 U.S.C § 14101 et seq.; 49 C.F.R. §376.1 et seq.. Specifically, plaintiff implicates the provision that requires carriers to ensure that owner-operators "receive all the rights and benefits due an owner under the leasing regulation . . . ." 49 C.F.R. §376.12(m).
This Court issued a previous report and recommendation on Plaintiff's Motion for Class Certification, recommending that the motion be denied with leave to renew that motion upon completion of class discovery. (Dkt. 135.) Both parties filed objections, and Judge Guzman adopted this Court's recommendation to the extent it denied class certifications pending further discovery. Following Judge Guzman's order, the parties have engaged in some limited class discovery. On July 1, 2015, Plaintiff served Defendants with Plaintiff's Second Set of Document Requests seeking, inter alia, all owner-operator lease agreements ("PVOs") that identify Atlas as the carrier from May 14, 2009 to May 14, 2013 ("Request 6"), and specific data relating to shipments during the same time period in "Usable Electronic Format" ("Request 7"). On August 12, 2015, pursuant to this Court's order on Plaintiff's prior motion to compel, the parties engaged in an Electronically Stored Information ("ESI") liaison meeting to attempt to resolve some of the issues and conflicts that were arising during class discovery.
To date, Atlas has produced 18 PVOs, with the names of the owner-operators redacted. According to Atlas, this sample of PVOs demonstrates that the contracts all differ greatly, and, therefore, there cannot be sufficient commonality or typicality among the owner-operators to adequately define a workable class. Atlas has refused to produce unredacted versions of the PVOs, arguing that the names of the owner operators are not relevant, and refuses to produce the full catalog of redacted PVOs because the burden of redaction is too great. Atlas further contends that this Court should not compel production of the PVOs because Plaintiff is likely to misuse the information it gains through discovery to bring additional claims or lawsuits against Atlas, in violation of the Protective Order previously entered in this case. (Dkt. 47 at 3-4.)
Regarding the information sought in Request 7, Atlas states that sifting through its database to provide Plaintiff with the specific information he seeks would require Atlas to write a "script," and then run that script to cull the large amount of data it possesses. According to Atlas, writing this script would take approximately one week, it would take the script another week to run the script, and then it would take another week to review the data pulled by the script for quality control purposes (165-5 at 2). Atlas maintains that this request is beyond the scope of discovery as envisioned by the Federal Rules of Civil Procedure because it requires them to create a new document, not simply produce existing documents already in its possession, custody, or control.
Plaintiff filed the instant motion to compel on September 4, 2015, seeking an order compelling Atlas to produce unredacted versions of the PVOs and the data relevant to Request 7. On September 16, 2015, this Court held a hearing on that motion and set a briefing schedule. The motion is now fully briefed and ripe for disposition.
The parties also briefed whether documents in Atlas's agents' possession were within Atlas's "control," such that Atlas would be obligated to produce those documents. However, while this motion was pending, the parties contacted the Court and stated that they had resolved that issue on their own; this Court need not reach a decision on that issue.
DISCUSSION
I. DOCUMENTS SOUGHT IN REQUEST #6
In response to Plaintiff's motion to compel the unredacted PVOs, Atlas makes several arguments against production. First, Atlas argues that the names of the individual owner-operators must be redacted before production because they are irrelevant and prone to misuse by Plaintiff, and that redaction is a time-consuming process that creates an undue burden on Atlas. Second, Atlas asserts that the 18 redacted sample PVOs it has already chosen to produce are sufficient to demonstrate that there is not sufficient commonality or typicality to support class certification, and, therefore, Atlas should not be required to produce additional PVOs. Finally, Atlas contends that, even if it is forced to produce unredacted versions of the PVOs, identifying which PVOs are relevant, responsive, and not duplicative would require review of several thousand documents, which constitutes an undue burden. For the reasons discussed more fully below, this Court rejects all of these arguments, grants Plaintiff's motion, and orders that Atlas produce all documents in its possession, custody, and control that are responsive to Request 6.
A. Atlas Has Not Demonstrated a Basis for Redacting the PVOs.
Atlas has stated that it will need to redact the names of the owner-operators in every contract before it can produce those documents to prevent misuse of the information by Plaintiff and his counsel, and because the names of the owner-operators are irrelevant. Moreover, these redactions are exceedingly time consuming, making production very burdensome on Atlas. Atlas has failed to show any threat of misuse and is incorrect that the names are irrelevant, thereby mooting its argument that redaction would be unduly burdensome.
1. Atlas's Claims Regarding Misuse Are Baseless.
Atlas argues that it has reason to believe that Plaintiff will misuse information it receives through discovery because Plaintiff previously filed a similar lawsuit against Atlas, and that Plaintiff's counsel previously sought to add additional parties in this case through a Petition to Intervene. It is unclear why simply filing two law suits against the same defendant is indicative of wrongdoing on the part of Plaintiff, and Atlas cites no authority to support its suggestion that it is. There is also no evidence that the erstwhile intervenor in this case was the result of any misconduct; Plaintiff states that the intervenor contacted Plaintiff's counsel upon learning of the case on his own, not due to any proactive measures on the part of Plaintiff's counsel, and Atlas has provided no basis to refute this. Ultimately, Atlas has not provided any evidence that Plaintiff has misused information at any time in this or any related cases, and this Court will not base its ruling on unfounded innuendo.
Moreover, the Protective Order entered in this case prevents information learned in this case from being used for any purposes other than this litigation, which sufficiently prevents Plaintiff from misusing any information it learns during the course of class discovery. Furthermore, Atlas's argument that the Protective Order will not prevent the Plaintiff from adding parties or claims in this action is not a valid reason for redaction. The discovery of additional causes of action is one of the purposes of discovery, not an unhappy byproduct. It is unclear how or why this use of information gleaned from discovery would be improper or a reason to require redaction of the owner-operators' names. In fact, some cases have noted that using class discovery to identify and interview putative class members is an acceptable use of such information. See, e.g., Drake v. Aerotek, Inc., 2014 WL 7408715, at *2 (W.D. Wis. Dec. 30, 2014) ("In order to show that this case meets the requirements for certification under Fed. R. Civ. P. 23, and particularly the requirements of typicality and commonality, plaintiff must contact individual class members"). Therefore, this Court rejects Atlas's argument that it should not produce documents due to a threat of misuse on the part of Plaintiff.
2. The Owner-Operator Names Are Relevant.
Atlas further argues that it is necessary to redact the owner-operators' names from the PVOs because of the "'default' rule . . . that names of putative class members are not discoverable prior to certification, unless Plaintiff makes an affirmative showing of relevance." (Dkt. 182 at 6 (emphasis in original).) However, the cases that Atlas cites are distinguishable from the instant suit, and a review of the case law reveals that no such "default rule" exists.
In Oppenheimer Fund, Inc. v. Sanders, 437 U,S, 340, 353-54 (1978), cited by Atlas to support its position, the Supreme Court considered whether it was appropriate to order the defendants to assist the plaintiff in compiling a list of the names and addresses of class members, after class certification had already taken place, in order to provide individual notice pursuant to Rule 23(c)(2). Thus, Oppenheimer cannot stand for a default rule on relevance in document production prior to class certification. Moreover, Oppenheimer specifically disclaimed a holding that there was a default rule against production of class members' identities, noting that "[w]e do not hold that class members' names and addresses never can be obtained under the discovery rules." Id. at 354 n.20.
Similarly, AF Holdings v. Does 1-1058, 752 F.3d 990, 995-996 (D.C. Cir. 2014) - also relied on by Atlas - is inapposite. The plaintiff in that case was an entity formed by "attorneys with shattered law practices," who were "seeking easy money" by acquiring "several copyrights to pornographic movies" and then initiating massive John Doe copyright infringement lawsuits. Id. at 992 (internal citations and quotations omitted). The plaintiff then issued several third-party subpoenas on internet service providers in an attempt to discover the identities and personal information of the John Doe defendants, and the providers refused to comply with the subpoenas. Id. The district court rejected the providers' arguments and ordered them to produce the information; on appeal, the appellate court vacated the district court's opinion, holding that the plaintiff lacked a good faith belief that the court had jurisdiction over the John Doe defendants, that venue was proper, or that joinder of the defendants was permissible. Id. at 995-999. Again, this case has very little bearing on the factual or procedural issues at play in this case, as it was not a class action suit.
Finally, Godson v. Eltman, Eltman & Cooper, 2013 WL 4832715, at *2 (W.D.N.Y. Sept. 11, 2013), also fails to support Atlas's assertions. While the court in Godson denied the plaintiff's motion to compel disclosure of the names and addresses of putative class members, it did so because the identities had no bearing on the class certification arguments raised by the defendant in that case - namely, whether the named plaintiff was an adequate representative and whether the defendant's net worth made a class action a superior vehicle for obtaining recovery. Id. Here, Plaintiff seeks the unredacted owner-operator agreements "for purposes of conducting an analysis as to how putative class members were actually compensated and to what extent such compensation may have varied among the contracting parties, including to what extent compensation terms varied by Atlas's agents." (Dkt. 183 at 9.) Some courts have recognized that the identities of putative class members may be relevant in investigating and analyzing issues related to commonality. See, e.g., Drake, 2014 WL 7408715, at *2 (W.D. Wis. Dec. 30, 2014). In fact, the Supreme Court recognized in Oppenheimer that "discovery often has been used to illuminate issues upon which a district court must pass in deciding whether a suit should proceed as a class action under Rule 23, such as numerosity, common questions, and adequacy of representation." Oppenheimer, 437 U.S. at 351 n.13. Because the Plaintiff has articulated a need for the class members' names to determine issues related to commonality, this information clearly fall within the broad scope of relevance as envisioned by the Federal Rules of Civil Procedure, and this Court rejects Atlas's argument that it must redact the names of the owner- operators.
Because this Court has ruled that redaction is unnecessary, it need not discuss the burden caused by any such redaction. However, this Court notes that any such burden was, at least in part, Atlas's doing. Atlas was aware that some discovery of the PVOs would be necessary by June 2015 at the latest, when Judge Guzman ruled on this Court's Report and Recommendation; Atlas was also aware of the looming discovery deadlines in this matter. However, rather than proactively undertake the redaction that Atlas felt was necessary in order to provide discovery to the Plaintiff, Atlas unilaterally decided that it was simply too burdensome and chose to do virtually nothing until Plaintiff had filed two motions to compel in this matter. Atlas had months to redact and produce the documents sought by Plaintiff, but did not do so, and now argues that it would be too burdensome to do so before the clock runs out on discovery. As this Court stated at the hearing on this motion to compel, Atlas's actions strike this Court as "foot dragging." (Dkt 177 at 9:8-10.)
B. The Samples Provided By Atlas Are Insufficient To Determine Issues Related to Class Certification.
Atlas claims that the sample of 18 contracts it provided to Plaintiff were sufficient to demonstrate a lack of commonality, pointing to the differences among those contracts as proof. While Atlas has certainly demonstrated that those specific contracts lack elements of commonality, it is entirely possible that those contracts are atypical of the greater universe of contracts, and were chosen by Atlas as the sample precisely because they lack commonality. Plaintiff certainly requires a wider, less "cherry-picked" swath of documents in order to determine whether there is sufficient commonality among the owner-operators based on the terms and conditions found in the PVOs. Additionally, even if there are some differences in various classes of contracts, a robust production is necessary to allow Plaintiff an opportunity to accurately define the class in a way that might satisfy the requirements of Rule 23.
Atlas hones in on this Court's language in the Report and Recommendation stating that the "key" to class certification "is that the leases were all the same." (Dkt. 135 at 6.) Apparently, Atlas's position is that the previous ruling held that class certification would be inappropriate unless every PVO was identical, or at least nearly identical. As this Court noted in the hearing on the motion presently before the Court, that argument "oversimplif[ies] our decisions." (Dkt. 177 at 4:9-10.) The issue is not whether the PVOs are identical; the issue is whether there is a subset of PVOs that are sufficiently comparable to Plaintiff's to satisfy the commonality and typicality requirements of Rule 23, such that a class action of similarly situated individuals can appropriately be pursued. Without a wide review of the PVOs possessed by Atlas, neither the Plaintiff nor this Court can make an informed decision on that issue.
Frankly, this Court is confounded by the Defendant's reluctance to provide a larger production of the PVOs. If, as Defendant claims, the contracts are all truly unique, providing several thousand contracts with distinct terms would almost certainly be the most surefire way to defeat class certification. While it is not this Court's place to question Atlas's strategic decisions, the record before this Court clearly demonstrates that the small sample of contracts that Atlas picked to produce is not sufficient to allow Plaintiff to make a reasoned decision regarding commonality or class definition. Therefore, Atlas's argument that it has fulfilled its discovery obligations by providing 18 PVOs that Atlas picked out is also rejected by this Court, and Atlas must produce unredacted versions of all PVOs.
In its brief, Atlas further suggests that discovery in this suit should be limited as it was in the related suit before Judge Young B. Kim, and that this Court should only require a sample of the PVOs. This Court rejects Atlas's arguments. First, the issues before Judge Kim were distinct from this suit, and this Court is not bound by Judge Kim's rulings on the facts and issues before him. Second, this suit has a discovery deadline fast approaching. Creating and executing a sampling protocol would simply slow the discovery process down and create more opportunities for Atlas to complain about the scope of discovery that Plaintiff seeks. This Court's review of the record suggest that Atlas is primarily responsible for the speed, or lack thereof, with which discovery is proceeding in this matter, and believes that a ruling requiring sampling will create more problems that it would solve.
C. Producing All PVOs is not Unduly Burdensome.
Finally, Atlas argues that it would be unduly burdensome to produce the documents that Plaintiff seeks. Rule 26(b)(2)(C)(iii) states that a court must limit discovery where "the burden or expense of the proposed discovery outweighs its likely benefit . . . ." According to Atlas, there are approximately 5,300 electronic documents on two different servers and 16 file cabinets of hard-copy contracts that may house responsive documents. Some of these locations contain documents outside the relevant time period, as well as duplicate and partial contracts (because some of the documents consist of only the first page and signature page of the contract). As such, Atlas claims "it will be necessary to manually review each and every contract to determine which ones" are relevant and responsive, and this creates an undue burden that outweighs the likely benefit of the discovery sought. (Dkt. 182 at 9.)
Atlas has failed to show why production in this matter creates an undue burden. Atlas has not provided any cost or time estimates, and without more, the idea that several thousand documents may have to be reviewed before the appropriate production set is identified does not strike this Court as beyond the pale, particularly given the size and complexity of this suit. Moreover, the PVOs are crucially important to the class certification issues that must be decided before this suit can move forward. Without those documents, neither Plaintiff nor the Court can adequately determine whether a class action is appropriate in this case. Thus, even if the burden were extreme - which this Court does not believe - the relative importance of these documents may very well be strong enough to require production anyway. Atlas's argument that producing all relevant PVOs would create an undue burden is also rejected.
II. DATA SOUGHT IN REQUEST #7
Plaintiff also moves to compel Atlas to produce certain shipping data that Plaintiff claims Atlas maintains in its database. According to Atlas, it should not have to produce this data because it "does not maintain the data in the format requested by Plaintiff," and "would have to write new script - new computer code - to create an entirely new report that retrieves and re-arranges the data into the format that Plaintiff requests." (Dkt. 182 at 11.) Atlas asserts that this would "constitute a requirement to create a new document solely for the purposes of discovery," which is beyond the scope of Rule 34 of the Federal Rules of Civil Procedure. (Id.)
This Court finds the case cited by Plaintiff, Gonzalez v. Google, Inc., 234 F.R.D. 674 (N.D. Cal. 2006), very instructive. In that case, the United States government was seeking, through a third-party subpoena pursuant to Rule 45, "a massive amount of information from Google's search index, and to turn over a significant number of search queries entered by Google users." Id. at 678. Google objected arguing, among other things, that it was unduly burdensome to produce the data the government was seeking "because [Google] does not maintain search query or URL information in the ordinary course of business in the format requested by the government." Id. at 683. While recognizing the "general rule" that "non-parties are not required to create documents that do not exist, simply for the purposes of discovery," the court held that it was not an undue burden on Google simply because "Google contends that it must create new code to format and extract query and URL data from many computer banks, in total requiring up to eight full time days of engineering time." Id. The court concluded that the technical burden of production did not excuse Google from complying with the subpoena. Id.
FN re: Rule 45 burden/fee shifting etc. --------
Another case in the Northern District of California, Apple Inc. v. Samsung Electronics Co. Ltd., 2013 WL 4426512 (N.D. Cal. Aug. 14, 2013), also touches on several issues raised by the parties in the instant motion. In that case, Samsung requested information from Apple regarding its revenues, profit margins, and expenses; when Apple responded do the requests, the information was "compiled at the 'product line level' (e.g., iPhone and iPad) rather than the 'model level' (e.g., iPhone 4S, iPhone 5)," and "Apple refused to present more 'granular financial data,' claiming that it does not maintain reports of such data in its 'ordinary course of business." Id. at *2. After discussing the parties' arguments, the court stated:
"[w]hat emerges from the parties' arguments is that Apple does have financial databases that it could query to generate at least some of the reports sought by Samsung. While this court has held that a party should not be required to create completely new documents, that is not the same as requiring a party to query an existing dynamic database for relevant information. Courts regularly require parties to produce reports from dynamic databases, holding 'the technical burden . . . of creating a new dataset for the instant litigation does not excuse compelling production.' Compelling production here would therefore not violate any established discovery principles."Id. at *3 (internal citations omitted). Ultimately, the court expressed disbelief that the requests would be especially burdensome, but held that Apple did not have to produce the reports because they were of little to no benefit to Samsung, who had already submitted its damages expert's report.
In this case, Atlas argues repeatedly that Plaintiff's request requires it to produce a new document solely for the purposes of discovery, and that doing so is outside the duties imposed on parties by Rule 34. Notably, Atlas has not argued that the data sought in Request 7 is irrelevant to the issue of class certification, or that Plaintiff does not need it to adequately define the putative class. Nor does Atlas provide any alternative avenue for Plaintiff to receive this data, such as allowing Plaintiff to inspect the full database on Atlas's computers to run scripts and pull the relevant data, or running scripts created by the Plaintiff on the database to generate the relevant reports. Instead, Atlas simply says it should not be required to produce it because it is outside the scope of Rule 34, and would be too difficult to do. Neither of these arguments is persuasive.
First, as the court noted in the Apple case, requiring a party to query an existing database to produce reports for opposing parties is not the same as requiring the creation of a new document. Thus, the argument that Plaintiff's request is outside the scope of Rule 34 is unfounded. Second, requiring the creation of new code does not necessarily create an undue technical burden. In Gonzalez, Google claimed that it would take eight full days of engineering time to produce the data the government was seeking. See Gonzalez, 234 F.R.D. at 683. Here, Atlas says it will take a week to create the script, a week to run the script, and a week to check the results. Although this is technically longer than the time cited by Google in Gonzalez, there are critical distinctions between that case and the instant suit. For example, the burden expected of third parties in the context of a Rule 45 subpoena (as Google was in Gonzalez) is lighter than that of a party to litigation. See, e.g., Guy Chemical Co., Inc. v. Romaco AG, 243 F.R.D. 310, 313 (N.D. Ind. 2007). Additionally, one of the weeks cited by Atlas would be passive, as Atlas would apparently simply be waiting for the script to run on its database. Therefore, the actual burden on Atlas would be closer to two weeks of work, which is closer to the amount of time that the court accepted in the Gonzalez case. Simply put, two weeks of active work to provide relevant data that Atlas is required to produce under the federal rules of evidence is not unduly burdensome.
This is also not a case, like Apple, where the benefit of the discovery is negligible. Unlike the plaintiff in Apple, who was able to produce expert reports on damages without the financial data at issue, Mervyn needs this data for Plaintiff's expert to determine whether, and to what extent, the owner-operators' in the putative class were damaged by the terms in the PVOs. This data is of great utility to Plaintiff, and the balancing of the benefit weighs more strongly in favor of the Plaintiff than the burden of producing it weighs in favor Atlas. Therefore, Plaintiff's motion to compel is granted as to the data sought in Request 7.
CONCLUSION
For the reasons discussed above, Plaintiff's Motion to Compel Defendant to Produce Unredacted Owner-Operator Agreements and Shipment Data is granted [165]. Defendant must produce the documents and data sought by Request Nos. 6 and 7 of the Plaintiff's Second Set of Document Requests on or before November 23, 2015. Date: October 23, 2015
ENTERED:
/s/_________
U.S. Magistrate Judge