Pension Trust Fund for Operati v. Kohl's Corporaton

• Court: Court of Appeals for the Seventh Circuit
• Decided: July 12, 2018
• Docket: 17-2697
• Status: Published

Opinion

In the

United States Court of Appeals For the Seventh Circuit ____________________ No. 17-2697 PENSION TRUST FUND FOR OPERATING ENGINEERS, et al., Plaintiffs-Appellants,

v.

KOHL’S CORPORATION, et al., Defendants-Appellees. ____________________

Appeal from the United States District Court for the Eastern District of Wisconsin. No. 13-CV-1159 — J. P. Stadtmueller, Judge. ____________________

ARGUED JANUARY 16, 2018 — DECIDED JULY 12, 2018 ____________________

Before WOOD, Chief Judge, and ROVNER and HAMILTON, Circuit Judges. WOOD, Chief Judge. In September 2011, Kohl’s Corporation announced that it was correcting several years of its financial filings because of multiple lease accounting errors. Hard on the heels of that announcement came a putative class action complaint. The plaintiffs, led by the Pension Trust Fund for Operating Engineers, allege that Kohl’s and two of its execu- tives defrauded investors by publishing false and misleading 2 No. 17-2697

information in the lead-up to the corrections. (For ease of ex- position, we refer to the putative class as the Pension Fund.) The Pension Fund took the position that one can infer that the defendants knew that these statements were false or reck- lessly disregarded that possibility at the time they were made, because Kohl’s recently had made similar lease accounting er- rors. Despite those earlier errors, it was pursuing aggressive investments in its leased properties, and at the same time, company insiders sold considerable amounts of stock. The district court dismissed the complaint for failure to meet the enhanced pleading requirements for scienter im- posed by the Private Securities Litigation Reform Act (PSLRA). The court entered that dismissal with prejudice, de- clining to give the Pension Fund even one opportunity to amend to cure the defects. The Pension Fund now appeals both the dismissal of the complaint and the district court’s de- cision to enter it with prejudice. Because the first complaint fell short and the Pension Fund has not been able to suggest how an amendment might help, we affirm. I Kohl’s runs over one thousand department stores across the United States. About 65 percent of those stores are leased—a fact that makes lease obligations a significant com- ponent of Kohl’s financial picture. The treatment of those leases has caused Kohl’s accountants and external auditors some trouble in recent years. The company was forced to ad- just its accounting practices three times—in 2005, 2010, and 2011—to bring its books in line with generally accepted ac- counting principles (“GAAP”). The first and third of these corrections were material and required the restatement of sev- No. 17-2697 3

eral years’ worth of financial statements. The second was com- paratively minor and required an adjustment to income in one quarter. The Pension Fund asserts that these recurring lease accounting errors show that Kohl’s, its CEO Kevin Man- sell, and its CFO Wesley McDonald were at least reckless in overseeing the company’s lease accounting practices by the time of the second and third corrections. Specifically, the Pen- sion Fund contends that purchasers of Kohl’s stock from Feb- ruary 26, 2009, to September 13, 2011 (the “class period”), were defrauded by knowing or reckless false statements in Kohl’s financial reports. The Pension Fund advanced two theories of liability in the district court: securities fraud in violation of section 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b), and SEC Rule 10b-5, 17 C.F.R. § 240.10b-5, against all defendants, and “controlling person” liability under section 20(a) of the Securities Exchange Act, 15 U.S.C. § 78t(a), against Mansell and McDonald. We can limit our discussion to section 10(b) and Rule 10b-5, because a violation of those provisions is nec- essary to support a violation of section 20(a). Pugh v. Tribune Co., 521 F.3d 686, 693 (7th Cir. 2008). To state a claim under section 10(b), a plaintiff must plead “(1) a material misrepresentation or omission by the defend- ant; (2) scienter; (3) a connection between the misrepresenta- tion or omission and the purchase or sale of a security; (4) re- liance upon the misrepresentation or omission; (5) economic loss; and (6) loss causation.” Id. We can narrow our focus even further, for the scienter element is the only point of dispute between the parties. We review the sufficiency of scienter pleadings de novo. Id. at 692. 4 No. 17-2697

Scienter pleadings in securities fraud class actions must satisfy a heightened standard of plausibility. Through the PSLRA, Congress requires that plaintiffs “state with particu- larity facts giving rise to a strong inference that the defendant acted with the required state of mind.” 15 U.S.C. § 78u-4(b)(2)(A) (emphasis added). For a case under section 10(b), that state of mind is “an intent to deceive, demonstrated by knowledge of the statement’s falsity or reckless disregard of a substantial risk that the statement is false.” Higginbotham v. Baxter Int’l, Inc., 495 F.3d 753, 756 (7th Cir. 2007). The Supreme Court has told us that a complaint gives rise to a strong inference of scienter “only if a reasonable person would deem the inference of scienter cogent and at least as compelling as any opposing inference one could draw from the facts alleged.” Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 324 (2007). In making this determination, the allega- tions in the complaint “are accepted as true and taken collec- tively.” Id. at 326. We must consider the relative probability of whether, taken as a whole, the false statements alleged here were “the result of merely careless mistakes at the manage- ment level based on false information fed it from below” or reflect “an intent to deceive or a reckless indifference to whether the statements were misleading.” Makor Issues & Rights, Ltd. v. Tellabs Inc., 513 F.3d 702, 709 (7th Cir. 2008). If the latter inference is not at least as compelling as the former, dismissal is appropriate. II Most of the Pension Fund’s complaint recounts the details of the accounting errors and Kohl’s financial restatements, but both sides argue that we need not wade too deeply into those No. 17-2697 5

details. The Pension Fund insists that because Kohl’s repeat- edly made lease accounting errors, something is up—where there’s smoke, there’s fire. But this inference depends on how (dis)similar the errors are. Kohl’s counters that technical ac- counting errors such as these are well below the pay grade of its executives. But leases are a significant part of Kohl’s finan- cial picture that cannot be expected to evade executive knowledge altogether. See S. Ferry LP, No. 2 v. Killinger, 542 F.3d 776, 784 (9th Cir. 2008) (concluding that a “core-opera- tions inference” can support scienter after Tellabs). We decline to take either simplistic approach. Tellabs’s repeated emphasis on looking at the facts “holistically” tells us that we must do more. 551 U.S. at 326. To apply the PSLRA meaningfully, we must dig deeper into the accounting and other allegations the Pension Fund has raised. After we have done this, we step back to look at what inferences can be drawn from the evi- dence as a whole. A As detailed in the complaint, all three sets of errors were announced through SEC filings accompanied by press re- leases and on at least one occasion, an investor conference call. The first restatement came on February 22, 2005. Kohl’s an- nounced that it was adjusting the period over which its lease obligations were reported. GAAP does not allow firms simply to record lease obligations when they are paid; rather, firms must record start and end dates that reflect the economic re- ality of the lease. As part of the restatement, Kohl’s adjusted how it calculated both the start and the end of lease terms. Previously, Kohl’s had fixed the start of each lease term as the date when it began making payments; as revised, it would set 6 No. 17-2697

the start as the earlier of the date of first payment or first pos- session of the building. Similarly, Kohl’s formerly set the end of the term at the conclusion of the initial non-cancelable lease term; as revised, it would recognize the lease through the ex- pected term, including some cancelable option periods. These changes required Kohl’s to restate its financial statements from 1998 through the third quarter of 2004. Kohl’s next set of accounting adjustments came in the fall of 2010. The company first identified the errors in November, before publicizing its final adjustments in December. These adjustments concerned (again) the start dates of the lease terms. It seems that Kohl’s may have overcorrected in 2005. Kohl’s had used the date of first possession as the start date for some leases even though the obligation to pay rent began earlier, contrary to its 2005 disclosures. Additionally, Kohl’s adjusted depreciation expenses across the terms of some leases and corrected miscategorized incentive payments from landlords. Together, these changes were not material to past financial statements, but they resulted in a $50 million adjust- ment to income in the third quarter of 2010. Finally, in August 2011 Kohl’s announced that it had dis- covered another round of accounting errors. These errors were of a different type. This time, Kohl’s had failed to reclas- sify many of its operating leases as capital leases after making significant investments in the affected stores. Operating leases have no impact on the balance sheet. Rental payments are ex- pensed, the rented property is not counted as an asset, and future rent payments are not recognized as liability. By con- trast, capital leases have a significant effect on the balance sheet. The leased property is recognized as an asset and future rent obligations as liabilities. Rental payments are treated not No. 17-2697 7

as a rental expense, but instead as a combination of deprecia- tion expense and interest expense. These changes were significant—indeed, firms will often go to great lengths to keep their financial obligations off the balance sheet. See Paul B.W. Miller & Paul R. Bahnson, Off- Balance-Sheet Financing: Holy Grail or Holey Pail?, ACCT. TODAY (Oct. 11, 2010), https://www.accountingto- day.com/news/off-balance-sheet-financing-holy-grail-or- holey-pail-AT55794 (“Managers strive after [off-balance-sheet financing] like the Holy Grail … .”); but see Tom Petruno, Why Corporate Leasing Practices Deserve More Respect, UCLA ANDERSON REV. (Apr. 4, 2018), https://www.ander- son.ucla.edu/faculty-and-research/anderson-review/leasing (arguing that reforms to operating lease rules “may be target- ing an accounting abuse that is more imagined than real”). Whatever the firm’s preference, GAAP requires leases to be categorized as capital when the economic reality of the ar- rangement makes the lessee more like the owner. Simplifying the requirements somewhat, capital-lease treatment is required if ownership transfers to the tenant at the end of the term, if the tenant has the right to purchase the property well below its value, or if the term of the lease or lease payments amount to a significant portion of the prop- erty’s value. In Kohl’s case, McDonald suggested that Kohl’s “strategies in negotiating leases and in renovating and con- structing stores” created “ongoing financial interest[s]” in the leased buildings that warranted capital-lease treatment. “Ma- terial weaknesses” in its financial reporting “controls and pro- cedures,” however, allowed these misclassifications to go un- noticed. The next month, Kohl’s restated its financial disclo- sures from 2006 through the second quarter of 2011, with 8 No. 17-2697

large effects on Kohl’s balance sheet, but relatively minor ef- fects elsewhere. According to the Pension Fund, Kohl’s had understated its liabilities from about 26 to 39 percent annually and its assets from about 9 to 12 percent annually as a result of these errors. The complaint supplements this chronology of accounting mistakes and corrections with some additional allegations supporting scienter. First, it alleges that Kohl’s leasing strategies should have put its executives on alert for potential lease accounting issues. By aggressively “renovating and constructing stores,” Kohl’s should have known that capital- lease treatment was appropriate earlier on. Second, the Pension Fund finds highly suspicious a number of stock sales by Mansell, McDonald, and other company insiders. Mansell sold 138,000 shares for $7,676,400 in September 2009. McDonald sold 7,000 shares for $412,000 in September and October 2009, and 2,000 shares for $112,500 in November 2010. Seven other insiders also sold significant numbers of shares during the class period. The Pension Fund argues that these sales underscore that Mansell and McDonald knew that Kohl’s financial statements were false or misleading when they were published. B Taking these facts together, the Pension Fund has made a strong case that many of Kohl’s disclosures regarding its lease accounting practices turned out to be false. But that is not enough. The facts must also give rise to a strong inference of scienter. The complaint fails in this regard if it is more likely that the errors resulted from “careless mistakes at the man- agement level” than from “an intent to deceive or a reckless No. 17-2697 9

indifference to whether the statements were misleading.” Ma- kor Issues & Rights, 513 F.3d at 709. In contrast with the com- plaint’s exhaustive account of the facts of Kohl’s accounting mishaps, the Pension Fund gives us very few facts that would point either toward or away from scienter. This lack of connec- tive tissue is determinative in this case. See Tellabs, 551 U.S. at 326 (noting that “omissions and ambiguities count against in- ferring scienter”). The Pension Fund argues that its strongest evidence of sci- enter is that Kohl’s made similar and significant accounting errors in 2005, 2010, and 2011 related to a core part of its busi- ness. But these errors are not as similar as the Pension Fund suggests. True, one error from 2005 recurred in 2010 (misstat- ing the start date of the lease), but that error led to a relatively minor restatement. The errors leading to major restatements in 2011 were wholly unrelated to the problems of 2005. The classification of leases and the length of lease terms implicate different lease accounting rules and affect firms’ financial statements in very different ways. Shifting start or end dates moves expenses from one period to another, affecting net in- come across periods. The classification of leases, meanwhile, has its primary effect on the balance sheet. The impact and considerations are quite different, even if both involve leases. The Pension Fund tries to overcome the differences be- tween the 2005 and 2011 restatements by arguing that the 2010 and 2011 restatements should be taken as one. The represen- tations in October and November 2010 that the changes would not be material are false, they say, because the 2011 problems were already known. The problem with this theory is that not only is the complaint devoid of evidence to support it—there is actually evidence in the complaint undermining 10 No. 17-2697

it. On June 28, 2011, just over a month before the 2011 lease accounting errors were discovered, Kohl’s announced that it had secured a $1 billion credit agreement requiring a compre- hensive review of its books. Without allegations of facts sug- gesting otherwise, the temporal proximity of these events suggests that an innocent explanation is more likely: the ac- counting errors were discovered during the comprehensive review mandated by contract. To the extent that making the same error again and again suggests recklessness, rather than negligence, the Pension Fund has failed to tell us why these errors are so alike as to make the recklessness inference at least as compelling as any other. That the defendants were employing aggressive invest- ment strategies in their leased properties is similarly of no help to the Pension Fund. Perhaps a reasonable person should have realized that the number of capital leases on Kohl’s bal- ance sheet should have increased as these investments were made. But the allegations do nothing to show why it was reck- less, rather than just negligent, that Kohl’s executives did not realize that something was amiss. Perhaps the executives had a motive to pretend nothing was amiss (though even that does not seem beyond dispute, as they might equally have wanted the most accurate financial picture possible), but a general- ized motive common to all corporate executives is not enough to establish scienter. Otherwise, “virtually every company in the United States that experiences a downturn in stock price could be forced to defend securities fraud actions.” Zucco Partners, LLC v. Digimarc Corp., 552 F.3d 981, 1005 (9th Cir. 2009) (quoting Lipton v. Pathogenesis Corp., 284 F.3d 1027, 1038 (9th Cir. 2002)). It is quite possible that Kohl’s accountants or external auditors knew they were pushing the boundaries of GAAP to keep leases off the balance sheet, but their No. 17-2697 11

knowledge is immaterial to the scienter of those making the statements. See Makor Issues & Rights, 513 F.3d at 708–09. Without more, we cannot say that Kohl’s pursuit of aggressive leasehold improvements counsels for or against scienter. Perhaps suspicious stock sales could tip the balance, but the insider trading allegations in this case do not. “[B]ecause executives sell stock all the time, stock sales must generally be unusual or suspicious to constitute circumstantial evidence of scienter.” Pugh, 521 F.3d at 695. The plaintiffs argue that the sales in this case are suspicious because Mansell and McDon- ald made no sales at all in the year before the class period or in 2011. But that the individual defendants made sales in 2009 and 2010 but not in 2008 or 2011 is not enough to render the sales unusual. See Teachers’ Ret. Sys. of La. v. Hunter, 477 F.3d 162, 185 (4th Cir. 2007) (“[T]he complaint does not provide de- fendants’ trading patterns outside the class period to permit comparison with their trades within the class period.”); Ron- coni v. Larkin, 253 F.3d 423, 435 (9th Cir. 2001) (finding graphs showing trading seven months before and twelve months af- ter the class period insufficient to show trades were suspi- cious or unusual). Once again, the Pension Fund has given us little to go on. The complaint tells us the date of sale, number of shares, and sale price for each trade, but nothing else. We do not know whether these sales were a high percentage of the individual defendants’ holding; we do not know whether the individual defendants sold more shares than they typi- cally would; we do not know if they bought more shares to offset their sales; we have no sense of the typical trading vol- ume of Kohl’s shares; and we do not know how Kohl’s stock price fluctuated around these sales. 12 No. 17-2697

Perhaps we could overlook the complaint’s lack of context if the stock sales resembled a smoking gun, but the probative value of stock sales depends greatly on timing. The most sig- nificant insider sales in this case were made in September 2009, 14 months before the 2010 corrections were announced and 23 months before the 2011 corrections were announced. These periods are more than long enough for any inference of suspicion to dissipate, at least in the absence of concrete facts suggesting otherwise. See In re Harley-Davidson, Inc. Sec. Litig., 660 F. Supp. 2d 969, 1002 (E.D. Wis. 2009); In re Party City Sec. Litig., 147 F. Supp. 2d 282, 313 (D.N.J. 2001) (“A broad tem- poral distance between stock sales and a disclosure of bad news defeats any inference of scienter.”). With nothing to in- dicate that these stock sales were unusual or suspicious, they cannot support an inference of scienter. We have addressed these issues with the complaint one at a time, but we recognize that we need to look at the allega- tions as a whole. Unfortunately for the Pension Fund, this does not help. Each allegation in the complaint is advanced without any sense of how the dots connect. Tellabs requires that a complaint give rise to a “cogent and compelling” infer- ence of scienter. 551 U.S. at 324. The Pension Fund tells us that Kohl’s made similar, but not identical, lease accounting errors; that it did so while management was pursuing an aggressive store-improvement strategy; and that insiders sold stock dur- ing the same period. This could suggest wrongdoing, but it more plausibly suggests negligent oversight of overzealous accounting staff or some other breakdown lower in the corpo- rate hierarchy. The Pension Fund has not taken the extra step to show why these allegations give rise to a strong inference of scienter, even considered collectively. No. 17-2697 13

III Although we agree with the district court that the com- plaint fell short of the PSLRA’s requirements, that court was so unimpressed that it entered a dismissal with prejudice without further ado and refused to entertain an amended complaint. We repeatedly have said that “a plaintiff whose original complaint has been dismissed under Rule 12(b)(6) should be given at least one opportunity to try to amend her complaint before the entire action is dismissed.” Runnion ex rel. Runnion v. Girl Scouts of Greater Chi. & Nw. Ind., 786 F.3d 510, 519 (7th Cir. 2015). This admonition carries special weight in securities fraud cases because “[i]n this technical and de- manding corner of the law, the drafting of a cognizable com- plaint can be a matter of trial and error.” Eminence Capital, LLC v. Aspeon, Inc., 316 F.3d 1048, 1052 (9th Cir. 2003). Our final task is to determine whether the district court abused its dis- cretion through its unusual departure from the standard pro- cedure. Foster v. DeLuca, 545 F.3d 582, 583 (7th Cir. 2008). The district court justified dismissal with prejudice be- cause it thought that the court’s prior rulings “put the plain- tiffs on notice of weaknesses in the amended complaint … .” The district court was right that its prior rulings (which were issued by a different presiding judge) identified weaknesses with the complaint, but those weaknesses were unrelated to the reasons for which the complaint was later dismissed. In its order denying the defendants’ first motion to dismiss with- out prejudice for relying too heavily on exhibits, the district court noted “ongoing concerns about the prolixity of the Amended Complaint—sixty-one pages, with 173 numbered paragraphs.” Whatever the merits of the district court’s criti- 14 No. 17-2697

cism, concerns about the complaint’s length could not possi- bly alert the plaintiffs to problems with their scienter allega- tions. If anything, they reasonably might have thought that more length was necessary to meet the PSLRA’s demanding standards for pleading scienter. The district court’s earlier crit- icism thus does not help support the abrupt end of the case. The defendants argue that the Pension Fund could not have been taken by surprise, because defendants had alerted the plaintiffs to the weaknesses of the complaint. This argu- ment is a non-starter. If briefing in opposition to a motion to dismiss were sufficient basis to deny leave to amend after that motion were granted, there would be little left to the general rule we have just discussed. The only case the defendants cite to the contrary involved denial of leave to amend for the fifth time, when the defects had been identified by the motion to dismiss the second amended complaint. Huon v. Denton, 841 F.3d 733, 745–46 (7th Cir. 2016). In other words, the plaintiffs in Huon had already amended twice with full knowledge of what the defendants would argue. In the usual case, we look only to decisions of the court to determine whether the plain- tiffs knew of faults with their complaint. See Gonzalez-Koeneke v. West, 791 F.3d 801, 806 (7th Cir. 2015) (pointing to “the defi- ciencies identified in the court’s order granting the motion to dismiss”); Bausch v. Stryker Corp., 630 F.3d 546, 562 (7th Cir. 2010) (“But a formal motion for leave to amend was not nec- essary at the Rule 12(b)(6) stage, and the plaintiff was entitled to wait and see if any pleading problems the court might find could be corrected.”). A litigant need not take the opposing side’s legal position as gospel; indeed, it frequently would be unwise to do so. No. 17-2697 15

Although there are problems with the district court’s deci- sion, and better practice might have been to allow one amend- ment, we find no reversible error here. At bottom, the district court was concerned that amendment would be futile, and the plaintiffs have done nothing before this court to dispel that notion. “[A] district court does not abuse its discretion by denying a motion for leave to amend when the plaintiff fails to establish that the proposed amendment would cure the de- ficiencies identified in the earlier complaint.” Gonzalez- Koeneke, 791 F.3d at 807. While the plaintiffs did not have the opportunity to show what they would add before the district court dismissed with prejudice, they have had several oppor- tunities since. They could have moved under Federal Rule of Civil Procedure 59(e) or 60(b) for another opportunity in the district court, see Runnion, 786 F.3d at 521, or they could have told us what more they would plead in their briefing. They took neither step. We asked at oral argument what the plain- tiffs hoped to add if given the opportunity. Again, we were given no indication of what new material the plaintiffs could provide. Reversal is inappropriate if the plaintiff cannot iden- tify how it would cure defects in its complaint. Arlin-Golf, LLC v. Vill. of Arlington Heights, 631 F.3d 818, 823 (7th Cir. 2011). The Pension Fund made no such showing in the district court or on appeal and is not entitled to another chance to do so. IV The Pension Fund failed adequately to plead scienter and has not suggested how it would amend its pleadings to cure this defect. As a result, the judgment of the district court is AFFIRMED.