SHEET METAL WORKERS LOCAL 19 PENSION FUND, individually and on behalf of all others similarly situated, Plaintiffs, v. PROASSURANCE CORPORATION, et al., Defendants Civil Action Number 2:20-CV-00856-AKK United States District Court, N.D. Alabama, Southern Division Filed June 23, 2022 Counsel David J. Guin, Dawn Stith Evans, Tammy McClendon Stokes, Guin Stokes & Evans LLC, Birmingham, AL, Joseph E. White, Pro Hac Vice, Lester R. Hooker, Saxena White, PA, Boca Raton, FL, Bailie L. Heikkinen, Pro Hac Vice, Mason G. Roth, Pro Hac Vice, Reginald E. Janvier, Pro Hac Vice, Robbins Geller Rudman & Dowd LLP, Boca Raton, FL, Jonathan Dov Lamet, Pro Hac Vice, Saxena White, P.A., Boca Raton, FL, Steven B. Singer, Pro Hac Vice, Craig Corey Maider, Pro Hac Vice, Kyla Grant, Pro Hac Vice, Sara M. DiLeo, Pro Hac Vice, Saxena White, P.A., White Plains, NY, James Bringhurst Eubank, Wilson Daniel Miles, III, Beasley Allen Crow Methvin Portis & Miles PC, Montgomery, AL, John T. Long, Pro Hac Vice, Cavanagh & O'Hara, LLP, Springfield, IL, Nathan R. Lindell, Pro Hac Vice, Robbins Geller Rudman & Dowd LLP, San Diego, CA, Roger H. Bedford, Jr., Roger Bedford, Attorney at Law, LLC, Russellville, AL, for Plaintiff. Janet A. Gochman, Pro Hac Vice, Jonathan K. Youngwood, Pro Hac Vice, Carl Jacob Lundqvist, Pro Hac Vice, Simpson Thacher & Bartlett LLP, New York, NY, Walter W. Bates, Cole Robinson Gresham, Jay M. Ezelle, Michael R. Lasserre, Starnes Davis Florie LLP, Birmingham, AL, for Defendants. Kallon, Abdul K., United States District Judge ORDER *1 This putative class action came before the court for a conference call to resolve three discovery disputes. First, whether the time period(s) for the defendants' production of certain documents should cover the alleged Class Period, a shorter period within that Class Period, or a longer period pre- and post-dating that Class Period. Second, whether the defendants must produce documents related to the tail coverage option in the TeamHealth insurance policy that ProAssurance Corporation issued, notwithstanding the court's prior dismissal of certain claims, see docs. 61; 62. And third, to what extent the defendants must produce documents related to ProAssurance's document retention and deletion policies. At the conference, the court instructed counsel to continue conferring as to the third issue, opting to table the dispute until counsel could reach an agreement or narrow it in scope. This order therefore addresses the relevant time period(s) for certain production and the necessity of producing documents related to TeamHealth's tail coverage purchase and related nonrenewal decision. As explained herein, the defendants must produce relevant discovery for the applicable categories of documents for the time period of July 1, 2016, through November 1, 2020. I. By way of background, the plaintiffs—led by the Central Laborers' Pension Fund and the Plymouth County Retirement System—sued ProAssurance and a group of its executives for alleged securities fraud in June 2020. See docs. 1; 44. The plaintiffs claim that for all of the defendants' emphasis on ProAssurance's “conservatism” in its underwriting and reserves practices to investors, the defendants negotiated, signed, and concealed aspects of an extremely risky and unique insurance policy with a physician staffing firm called TeamHealth that led to significant losses. See doc. 44. According to the plaintiffs, the defendants misstated the data on which they relied to set ProAssurance's reserves for its accounts, including for TeamHealth, and hid the truth about the magnitude of the losses ProAssurance could and would consequently experience until it made several disclosures in 2020. See id. The plaintiffs propose a Class Period of August 8, 2018, through May 7, 2020, inclusive, and claim that until the defendants made a public disclosure on May 7, 2020, the defendants concealed that they knew TeamHealth would not renew its policy and would exercise tail coverage, events that would cause further losses.[1] Id. ¶¶ 3, 11–12. *2 In resolving the defendants' motion to dismiss, the court separated the plaintiffs' claims into three categories: the defendants' purported (1) representations “that ProAssurance's loss reserves were adequate and predicated upon certain data,” (2) failure “to disclose TeamHealth's nonrenewal and purchase of tail coverage prior to May 2020,” and (3) continued emphasis on “ProAssurance's commitment to conservative underwriting and reserves practices throughout the Class Period.” See doc. 61 (the MTD Opinion). Ultimately, the court dismissed most of the claims predicated on ProAssurance's loss reserves and TeamHealth's nonrenewal and tail coverage because the plaintiffs failed to sufficiently plead scienter. See id. at 57. However, the MTD Opinion preserved the claims against ProAssurance and two executives seemingly involved in the TeamHealth transaction—Howard Friedman and Michael Boguski—based on alleged misstatements regarding the (1) “observed frequency data underlying the company's reserve estimates” and (2) ProAssurance's “commitment to conservative underwriting and reserve-setting.” Id. Thereafter, the parties engaged in discovery, and in early 2022, they hit their first roadblock.[2] As salient to this order, after conferring, counsel continue to disagree as to whether the defendants have to produce (1) documents for the proposed time period of January 1, 2016, through November 7, 2020, and (2) documents related to the tail coverage option. Because these issues necessarily relate to one another, especially with respect to whether the dismissal of certain claims curtailed the relevant discovery period, the court addresses them together. II. “Parties may obtain discovery regarding any nonprivileged matter that is relevant to any party's claim or defense and proportional to the needs of the case,” subject to factors like the burden of the proposed discovery, its likely benefit, and its importance in resolving the issues. FED. R. CIV. P. 26(b)(1). Although Rule 26 promotes a “traditionally liberal scope of discovery,” Azzia v. Royal Caribbean Cruises Ltd., No. 15-24776-CV-KING/TORRES, 2018 WL 11233847, at *1 (S.D. Fla. Feb. 12, 2018), litigants “may only obtain discovery regarding issues properly presented and pending in the lawsuit,” In re Merck & Co., Inc. Sec., Derivative & ERISA Litig., 2012 WL 4764589, at *7 (D.N.J. Oct. 5, 2012). It follows logically that “rulings granting dispositive motions also define, and narrow, the scope of permissible discovery.” Merck, 2012 WL 4764589, at *7 (quoting Breslin v. Dickinson Tp., No. 09–1396, 2011 WL 3292924, at *7 (M.D. Pa. Aug. 1, 2011)). Thus, to frame a discovery dispute, the court must pinpoint the live claims and defenses and “determine what the purpose of the discovery is.” See McArdle v. City of Ocala, Fla., 451 F. Supp. 3d 1304, 1308 (M.D. Fla. 2020); Luczak v. Nat'l Beverage Corp., No. 18-61631-CIV-MOORE/SNOW, 2020 WL 10505310, at *2 (S.D. Fla. Nov. 2, 2020). “Indeed, as the commentary to Rule 26 informs us, ‘[a] party claiming that a request is important to resolve the issues should be able to explain the ways in which the underlying information bears on the issues as that party understands them.’ ” McArdle, 451 F. Supp. 3d at 1308. Still, it is “the objecting party [who] bears the burden in showing that a discovery request lacks relevance to the claims or defenses raised by either party before an objection is sustained” by demonstrating that a request “is of such marginal relevance” that the potential harm outweighs the “ordinary presumption in favor of broad disclosure.” Azzia, 2018 WL 11233847, at *2; see Luczak, 2020 WL 10505310, at *2. A. *3 Given these principles, it is true that the MTD Opinion may have narrowed the scope of permissible discovery by eliminating or limiting certain claims, as the defendants suggest. But the pertinent claims the court dismissed—those predicated on the failure to disclose TeamHealth's nonrenewal and tail coverage and certain of the claims based on the misstatements about ProAssurance's data and conservative reserves—overlap with the remaining claims such that they implicate similar discoverable documents. In that respect, the court's finding that the plaintiffs failed to plead scienter as to the alleged tail coverage-related omissions does not render irrelevant those documents concerning the tail coverage—which may speak to the falsity or scienter elements of the alleged misstatements about ProAssurance's commitment to conservatism that would help insulate it from losses. Documents from the period when the defendants publicly disclosed information about TeamHealth, including its tail coverage purchase, are also relevant to these live issues. The MTD Opinion preserved the claims against ProAssurance, Friedman, and Boguski as to their alleged misstatements about the company's conservative practices and the data underlying the company's calculations with respect to the TeamHealth account. Because the structure of the tail coverage option relates to these allegations, documents about the tail coverage option—even if they no longer support an independent claim—are relevant to whether the defendants had knowledge of the purported falsity of their statements about ProAssurance's conservatism in underwriting and reserves and the accuracy of their data.[3] In sum, while the defendants accurately observe that rulings on dispositive motions may constrain the substance of and time period for discovery, in this case, the MTD Opinion does not exclude the tail coverage-related documents or necessarily shorten the time period given the overlap between the pending and dismissed claims. B. As to the specific time period at issue, the parties sharply contest whether certain documents from before 2018 or after January 2020 bear relevance to the plaintiffs' claims. “The relevant time period for discovery must be determined on a case-by-case basis. There is no rule that discovery be constrained to a particular time period beyond the general standard under Rule 26(b)(1) requiring that discovery be relevant and proportionate.” In re BofI Holding, Inc. Sec. Litig., No. 3:15-cv-02324-GPC-KSC, 2021 WL 1812822, at *5 (S.D. Cal. May 6, 2021). Thus, in the context of securities claims, some courts “have determined that documents created outside the class period may be relevant.” See id. (citing examples). In other words, an alleged class period does not itself determine the period for discovery; this determination rests on case-specific facts and allegations as defined by the claims in the lawsuit. See id. Here, the plaintiffs assert a Class Period from August 2018 to May 2020, see doc. 44, and request certain categories of information from a period of January 1, 2016, to November 7, 2020. In the defendants' view, the plaintiffs “seek to expand discovery well beyond the narrow categories of alleged misstatements that remain at issue in this Action,” see tabs 7 at 3; 12 at 3, and the defendants counteroffer a period of June 30, 2018, to January 22, 2020. 1. In support of their January 1, 2016, start date, the plaintiffs assert that the following categories of documents contain “highly relevant documents [that] were necessarily created well before the Class Period”: (1) the TeamHealth policy signed in early-2016, including the negotiation, underwriting, signing, monitoring, reserving for, and renewal negotiations related to that policy; (2) issues related to claim frequency and/or severity with regard to the TeamHealth policy, ProAssurance's HCPL book of business, or the HCPL industry in general; and (3) ProAssurance's purportedly ‘conservative’ underwriting and reserve-setting practices and policies, and any potential deviations from such practices and policies. *4 Tab 9 at 2. The defendants, for their part, argue that the plaintiffs' “only purported justification for seeking discovery beginning on January 1, 2016 is that the TeamHealth policy was signed in ‘early 2016’ ” and that the complaint is otherwise “devoid of allegations that frequency, severity, or reserves with respect to the TeamHealth policy were an issue anywhere close to the start of the discovery period Plaintiffs seek.”[4] Tab 12 at 3. The defendants largely propose June 30, 2018, as the start date for the relevant discovery period for these documents and their search terms, see tab 30 at 1, apparently based on the plaintiffs' allegation that the defendants were “on notice of concerns related to the TeamHealth account by no later than ... June 30, 2018,” see tab 12 at 3 (quoting doc. 44). To be sure, the defendants' proposed date precedes the alleged Class Period by about one month, which would help locate relevant information immediately predating some of the losses the plaintiffs allegedly suffered as a result of the purported fraud. But the court sees no basis in the pleadings—and, more specifically, in the pleaded claims still pending—for relegating these categories of discovery to a start date of June 30, 2018. Instead, the court easily discerns from the amended complaint and the MTD Opinion, docs. 44; 61, that the plaintiffs allege the occurrence of certain events dating back to 2016 and 2017 that bear on the pending claims. For example, the plaintiffs assert the following: • In 2015 or 2016, ProAssurance created its “National Healthcare Team,” a group of underwriters “specifically tasked with selling [healthcare professional liability] policies to large physician groups.” See doc. 44 at ¶¶ 5, 47, 51. Friedman led the Team, which underwrote the TeamHealth policy during the second quarter of 2016. See id. ¶¶ 51–52. • In an August 4, 2016, call with investors, Friedman disclosed—without identifying TeamHealth—that a new, large “multistate account” included “the single largest premium” the company had “ever written” and “represent[ed] solid evidence of [ProAssurance's] ability to succeed in an evolving world where physician consolidation [was] continuing ....” Id. ¶ 53. The plaintiffs cite this as evidence that the defendants, and Friedman in particular, knew the nuts and bolts of the TeamHealth policy (and, thus, how unique and risky it was to issue). See id. ¶¶ 53–54, 56. • In a November 3, 2016, earnings call, the defendants, including Friedman, allegedly underscored the risk that firms like TeamHealth present to underwriting and noted that ProAssurance placed “a great deal of emphasis on evaluating the internal processes” of these businesses when structuring their policies. Id. ¶ 60. • In late 2017, ProAssurance “began experiencing a dramatic increase in the number of outstanding claims being reported under its TeamHealth policy,” which “dramatically worsened as the Class Period progressed.” Id. ¶ 76; see id. ¶¶ 115, 120. This spike in claims led to “mounting undisclosed losses” also beginning in late 2017. Id. ¶ 89. • In Forms 10-K for the years ending in 2017, 2018, and 2019, ProAssurance “repeatedly represented that it established loss reserves based upon, among other things, historical experience and claims frequency and severity,” id. ¶ 212, all while TeamHealth claims skyrocketed and general severity in the industry increased. The defendants allegedly “ignored” these critical factors despite the “red flags relating to frequency, severity, and unique risk in setting loss reserves for TeamHealth.” Id. ¶ 214. *5 All in all, these example allegations serve to demonstrate that limiting the time period for the three categories of documents—documents related to the negotiations and monitoring for the TeamHealth policy, the claim frequency and severity for TeamHealth and other healthcare professional liability policies, and ProAssurance's underwriting and reserves practices—to start on June 30, 2018, would severely curtail the discovery of relevant information. And on the flip side, the plaintiffs provide specific allegations of conduct in 2016 and 2017 that bear on the defendants' purported knowledge of the TeamHealth policy, the calculations made for the policy and the company's reserves, and the data supporting those calculations. However, setting aside the one search string for which the defendants have agreed to use a start date of January 1, 2016, see tab 30 at 1, the court ultimately disagrees with the plaintiffs that their allegations require a discovery period beginning in January 2016. As to the other search terms, the court deems appropriate a start date of July 1, 2016. This date would facilitate the discovery of information related to the defendants' knowledge—especially Friedman's—of TeamHealth's allegedly unique business structure and the challenges that posed to underwriting, as well as their potential knowledge about TeamHealth's and the industry's claim severity and frequency starting in late 2016 (when ProAssurance apparently signed the TeamHealth deal) through late 2017 (when TeamHealth claims allegedly began to skyrocket).[5] Providing discovery for these categories dating back to the latter half of 2016 will further enable the plaintiffs to compare data and documents about that data to determine if their claims that frequency and severity rose in 2017 are, in fact, true. Thus, excluding the search string for which the defendants have agreed to a January 1, 2016, start date, the defendants must run the search terms for the remaining categories with a start date of July 1, 2016. 2. The plaintiffs also assert that the relevant time period should conclude on November 7, 2020, six months after the alleged Class Period. See tab 9 at 7. They maintain that (1) the discovery period must extend at least until the last day of the proposed Class Period on May 7, 2020, when ProAssurance made its second relevant corrective disclosure; (2) ProAssurance's “expected losses announced on May 7, 2020 were not officially recorded by ProAssurance until August 10, 2020”; and (3) “relevant information concerning the causes and details of those losses will clearly exist at least through [August 10, 2020], and likely for some period thereafter.” Id. The defendants protest that “the only alleged category of misstatements that were purportedly ‘corrected’ by ProAssurance's disclosures on May 7–8, 2020 relate to TeamHealth's election not to renew its policy and purchase tail coverage—the exact category of alleged misstatements this Court has already dismissed,” and so allowing discovery of materials from May 2020 or later would permit discovery on matters irrelevant to the pending claims. Tabs 7 at 2 (emphasis omitted); 12 at 3. Because the first alleged corrective disclosure occurred on January 22, 2020, the defendants contend that the relevant time period for discovery should conclude on that date. Tab 12 at 3. *6 As an initial matter, for the reasons outlined in Section II.A, the court sees no basis for cutting off discovery on January 22, 2020, by virtue of the dismissed claims alone. The MTD Opinion did not express any findings as to the proposed Class Period. See generally doc. 61. More significantly, the alleged disclosure on May 7, 2020, in which the defendants “disclosed that the unnamed ‘large national account' ... would exercise its option to purchase ‘tail coverage’ ” and expose ProAssurance to additional losses, doc. 44 at ¶ 12, does not bear only on the dismissed tail coverage-related claims. On the contrary, the extent of the defendants' knowledge about TeamHealth's decisions and ProAssurance's expected losses between the January 2020 and May 2020 disclosures is relevant to the pending claims, i.e., whether the defendants knowingly made material misstatements about “observed frequency data underlying the company's reserve estimates and the company's commitment to conservative underwriting and reserve-setting.” See doc. 61 at 58. Further, the court agrees with the plaintiffs that the relevant period should “extend beyond the last day of the Class Period .... because the expected losses announced on May 7, 2020 were not officially recorded by ProAssurance until August 10, 2020.” Tab 9 at 7. As the plaintiffs maintain, “relevant information concerning the causes and details of those losses will clearly exist at least through that date, and likely for some period thereafter.” Id. Extending the discovery timeframe to November 7, 2020—six months after the end of the proposed Class Period—would permit the discovery of relevant information without imposing disproportionate burdens on the defendants when viewed in context with the factors needed to prove securities fraud, if fraud did occur. And “documents created after the alleged fraud was revealed may be relevant to scienter and falsity,” two essential components of the plaintiffs' claims. See BofI Holding, 2021 WL 1812822, at *6. Although November 7, 2020, may appear “somewhat arbitrary, it is no more arbitrary than the ... date proposed by [the] [d]efendants,” given that the court has not, at least at this juncture, determined that January 22, 2020, is the end of the relevant Class Period or dismissed all of the claims on which 2020 disclosures and losses bear relevance. See id. Moreover, “[i]t is not unreasonable to presume that relevant communications continued to occur several months after the filing of the [c]omplaint” and during the same year as the disclosures at issue, making it sensible to extend the discovery period several months after August 2020. See id. Accordingly, the defendants must run the search terms for the documents at issue with an end date of November 7, 2020. III. In summary, the court ORDERS the defendants to run the agreed-upon search terms for the three disputed categories of documents using a time period of July 1, 2016, through November 7, 2020, except the search string for which the defendants have already assented to a January 1, 2016, start date. The court OVERRULES the defendants' objections to producing documents related to TeamHealth's tail coverage option (and the related nonrenewal decision) because these documents bear relevance to pending claims. Counsel should confer as to the dispute concerning document retention and deletion and may subsequently invoke the court's discovery dispute procedure as needed. DONE the 23rd day of June, 2022. Footnotes [1] Specifically, the plaintiffs allege the following: Defendants also concealed that they knew by at least the start of 2020 that TeamHealth was highly unlikely to renew its policy and would instead exercise its option to purchase ‘tail coverage,’ which would (and ultimately did) expose ProAssurance to substantial additional losses. The truth about Defendants' material misstatements and omissions was ultimately revealed through two corrective disclosures, the last of which occurred on May 7, 2020 .... Doc. 44 at ¶ 3. The first corrective disclosure allegedly occurred on January 22, 2020. Id. ¶¶ 11–12. [2] As part of the court's discovery dispute procedure, counsel supplied a compilation of source material to provide context for their disagreements. This material, which the court relies on to frame the disputes and the parties' views, includes requests for production, proposed search terms and parameters, and letters between counsel outlining objections, responses, and counterproposals. Consistent with its standard practice, the court will not make these materials part of the record. Instead, the court will leave it to the parties to preserve the materials they submitted to the court and to make the submission a part of the record if and when it becomes necessary to do so later. [3] For this reason, the court will overrule the defendants' objections to the plaintiffs' requests for documents concerning the tail coverage option in TeamHealth's policy. The court addresses the relevant time period(s) for these requests in the sections that follow. [4] The defendants have agreed to run one particular TeamHealth-related search string with a time period of January 1, 2016, to January 22, 2020, see tab 30 at 1, but as counsel asserted at the conference, the plaintiffs maintain that the defendants must perform specific additional searches using a start date of January 1, 2016. [5] The court finds persuasive the following reasoning from BofI Holding: As to justification for the discovery of documents prepared or created five months prior to the start of the proposed class period, Plaintiff notes that a Confidential Witness (‘CW’) alleged deficiencies in BofI's internal controls stemming from conduct observed no later than May 2013, when the CW left BofI. At minimum, the CW allegations relate to whether BofI's internal controls were in fact deficient and whether Defendants knew of those deficiencies when they made the alleged misstatements. Such information would be relevant as to the falsity and scienter of the alleged misstatements. Plaintiff also notes that the first alleged misrepresentation, BofI's 2013 Form 10-K, covered the fiscal year ending June 30, 2013. Although discovery into time periods further removed from the subject matter of the alleged misstatements would be increasingly less relevant, the Court does not find that the Magistrate Judge erred in permitting discovery to extend back to April of 2013 given that information and communications relating to BofI's practices from that time may well inform whether statements made regarding that time period were knowingly false. 2021 WL 1812822, at *6 (internal citations omitted).