Fraud at ALC and Broadwind, two Grant Thornton clientsIn 2015, the Securities and Exchange Commission issued an administrative and cease-and desist order against Grant Thornton and two partners as a r

UNITED STATES OF AMERICA Before the SECURITIES AND EXCHANGE COMMISSION SECURITIES EXCHANGE ACT OF 1934 Release No. 76536 / December 2, 2015 ACCOUNTING AND AUDITING ENFORCEMENT Release No. 3718 / December 2, 2015 ADMINISTRATIVE PROCEEDING File No. 3-16976 In the Matter of Grant Thornton , LLP , Respondent. ORDER INSTITUTING PUBLIC ADMINISTRATIVE AND CEASE -AND - DESIST PROCEEDINGS PURSUANT TO SECTION S 4C AND 21C OF THE SECURITIES EXCHANGE ACT OF 1934, AND RULE 102(e) OF THE COMMISSION’S RULES OF PRACTICE , MAKING FINDIk dS, AND IMPOSING REMEDIAL SANCTIONS AND A CEASE - ANa -DESIST ORDER I. The Securities and Exchange Commission (“Commission”) deems it appropriate that public administrative and cease -and -desist proceedings be, and hereby are, instituted pursuant to Sections 4C 1 and 21C of the Securities Exchange Act of 1934 (“Exchange Act”) and Rule 102(e)(1)(ii) of the Commission’s Rules of Practice 2 against Grant Thornton , LLP (“Grant Thornton ” or “Respondent ”). 1 Section 4C provides, in relevant part, that: The Commission may censure any person, or deny, temporarily or permanently, to any person the privilege of appearing or practicing before the Commission in any way, if that person is found . . . (2) . . . to have engaged in unethical or improper prof essional conduct . 2 Rule 102(e)(1) provides, in relevant part: The Commission may censure a person . . . who is found . . . * * * (ii) to have engaged in unethical or improper professional conduct. 2 II. In anticipation of the institution of these proceedings, Respondent has submitted an Offer of Settlement (“Offer”) that the Commission has determined to accept. Respondent admits the facts set forth in Section s III .B, C, and E through F below, acknowledges that its conduct in the course of its Assisted Living Concepts, Inc. and Broadwind E nergy, Inc. engagements violated the federal securities laws, admits the Commission’s jurisdiction over it and the subject matter of these proceedings, and consents to the entry of this Order Instituting Public Administrative and Cease -and -Desist Proceedings Pursuant to Sections 4C and 21C of the Securities Exchange Act of 1934 and Rule 102(e) of the Commission’s Rules of Practice , Making Findings, and Imposing Remedial Sanctions and a Cease -and -Desist Order (the “Order”), as set forth below. III. On the basis of this Order and Respondent’s Offer, the Commission finds 3 that: A. SUMMARY 1. This matter involves impr oper professional conduct by Grant Thornton while serving as the auditor of two clients of Grant Thornton’s Wisconsin practice: Assisted Living Concepts, Inc. (“ALC”), a publicly traded senior living company, and Broadwind Energy, Inc. (“Broadwind”) , a publicly traded alternative energy company. Duri ng the course of its engagements, Grant Thornton repeatedly violated professional standards while ignoring repeated red flags and fraud risks that allowed ALC and Broadwind to file numerous reports with the Commission that were materially false and misleading. * * * (iv) with respect to persons licensed to practice as accountants, “improper professional conduct” under Rule 102(e)(1)(ii) means: * * * (B) either of the following two types of negligent conduct: (1) A single instance of highly unreasonable conduct that results in a violation of applicable professional standards in circumstances in which an accountant knows, or should know, that heightened scrutiny is warranted . (2) Repeated instances of unreasonable conduct, each resulting in a violation of applicable professional standards, that indicate a lack of competence to practice before the Commission. 3 The findings herein are made pursuant to Respondent’s Offer of Settlement and are not binding on any other person or entity in this or any other proceeding . 3 2. Fo r the ALC engagement, for more than three years and under the leadership of two engagement partners , Grant Thornton failed to identify a fraud perpetrated by ALC’s CEO and CFO. That long -running fraud was designed to mask ALC’s defaults on certain occupancy and revenue covenants that had significant financial consequences for ALC in the event of non - complia nce. As a result of the fraud and Grant Thornton’s failed audits, for three years ALC falsely represented to its investors that it was meeting the covenants and avoiding the se rious ramifications of the defaults . 3. For the Broadwind engagement, Grant Thornton ’s failure to exercise due professional care and skepticism contributed to Broadwind improperly omit ting from its financial statemen ts that it had sustained a $58 million impa irment charge caused by the severe deterioration of customer relationships for two of Broadwind’s most important customers. Grant Thornton ’s failures also contributed to Broadwind conduct ing a public offeri ng for its stock which concealed this impairment charge from investors. Grant Thornton ’s negligence fur ther contributed to Broadwind filing multiple financial statements which materially overstated revenue to Broadwind’s investors. 4. Grant Thornton’s failed ALC and Broadwind engagem ents were indicativ e of system ic quality issues and failures to adhere to professional standards on engagements of clients of Grant Thornton’s Wisconsin practice. In particular, Grant Thornton had received numerou s warnings of quality issues involving the managing partner o f the Wisconsin practice. Despite these warnings, Grant Thornton allowed that managing partner to continue to audit public companies, including ALC and Broadwind , and failed to take the appropriate remedial steps that could have stopped ALC’s and Broadwin d’s repeated false and misleading statements to their investors. B. RESPONDENT 5. Grant Thornton is an Illinois limited liability partnership and a PCAOB -registered public accounting firm with its headquarters in Chicago, Illinois. The conduct at issue occurred in the course of audits and reviews of clients of G rant Thornton’s Wisconsin practice . C. RELEVANT GRANT THORNTON PROFESSIONALS 6. Melissa K. Koeppel (“Koeppel”), age 54 , is a Certified Public Accountant (“CPA”) license d to practice in Wisconsin . Koeppel s erved a s the managing partner of Grant Thorn ton’s Wisconsin practice from 2008 through April 2011. Koeppel served as the Grant Thornton engagement partner on , and had final audit responsibility ove r, the ALC engagements from 2006 through 2010 and the Broadwind engagements from 2007 through the second quarter of 2010 . Since 2012 , Koeppel has been employed by Grant Thornton as a managing director , outside the audit -services practice. 7. Jeffrey J. Robinson (“Robinson ”), age 63 , is a CPA licensed to pr actice in Illinois and Wisconsin . Robinson served a s the managing partner of Grant Thornton’s Wisconsin practice from April 2011 through July 2015, when he retired. Robinson served as the Grant Thornton engagement partner on, and had final audit responsibility over, the ALC engagements from 20 11 4 through the first quarter of 2013, when ALC terminated its relationship with Grant Thornton upon ALC’s acquisition by another company . D. OTHER RELEVANT ENTITIES 8. ALC was a Nevada corporation with its prin cipal place of business in Menomonee Falls, Wisconsin. Between November 2006 and July 2013, ALC’s common stock was registered with the Commission pursuant to Section 12(b) of the Exchange Act and traded on the New York Stock Exchange. In February 2013, A LC agreed to be sold to a global private equity firm. In July 2013, when the sale was completed, ALC’s stock ceased trading on the NYSE. 9. Broadwind is an alternative energy company incorporated in Delaware and headquartered in Cicero, Illinois. In Octob er 2007, Broadwind purchased Brad Foote Gear Works, Inc. (“Brad Foote”) to provide gear systems for the wind turbine and other energy industries. Broadwind’s common stock was quoted on the OTC Bulletin Board until April 9, 2009, when its common stock bega n trading on the NASDAQ Global Select Market. Broadwind’s common stock is registered with the Commission pursuant to Section 12(b) of the Exchange Act. On February 5, 2015, the Commission filed a settled action in the Northern District of Illinois ag ains t Broadwind and its former CEO and CFO for Broadwind’s failure to record and disclose a $58 million impairment charge prior to a public offering in January 2010. The SEC also charged Broadwind and its officers with violations arising from accelerated reve nue recognition practices and inadequate disclosures ahead of the offering. Broadwind consented to a judgment enjoining it from violating Section 17(a)(2) of the Securities Act and Sections 13(a), 13(b)(2)(A), and 13(b)(2)(B) of the Exchange Act and Rules 12b -20, 13a -1, and 13a -13 thereunder and imposing a civil penalty of $1 million. The former CEO and CFO also consented to a judgement that enjoined them from future securities laws violations, and imposed disgorgement, prejudgment interest, and civil pen alties. The district court entered the proposed judgments associated with the settlement on February 11, 2015. See SEC v. Broadwind Energy, Inc., et al. , Case No. 15 -cv -1142 (N.D. Ill.). E. GRANT THORNTON’S AUDITS AND REVIEWS OF ALC 1. ALC and the Ventas Lease 10. During the relevant time period, ALC operated more than 200 senior living residences in the United States, totaling more than 9,000 units. On January 1, 2008, ALC purchased the operations of eight assisted living facilities for a total of 540 units in Alabama, Florida, Georgia, and South Carolina (the “Ventas facilities”) and simultaneously entered into a lease with Ventas, Inc. (“Ventas”) , a publicly traded real estate investment trust (“REIT”) and the owner of the facilities, to operate the facili ties ( the “ Ventas lease”). 11. The Ventas lease contained financial covenants (the “financial covenants”), which required that ALC maintain certain quarterly and trailing twelve -month occupancy percentages and coverage ratios, both at each facility and at the portfolio level. The lease defined “coverage ratio” as cash flow divided by rent payments. The Ventas lease required ALC to demonstrate its compliance with the financial covenants on a quarterly basis by providing Ventas, within 45 days of the end of eac h quarter: (1) facility financial statements prepared in accordance with general 5 accepted accounting principles (“ GAAP ”); (2) schedules documenting compliance with the financial covenants; and (3) an officer’s certificate, signed by an ALC executive , attes ting to the completeness and accuracy of such information. 12. The lease provide d that if ALC violated any of the financial covenants, Ventas could: (1) terminate the lease in its entirety; (2) evict ALC from all eight facilities; and (3) require ALC to pay accelerated rent equal to the net present value of the unpaid rent for the remaining term of t he lease . ALC disclosed the net present value of its unpaid rent, as of its 2009, 2010, and 2011 fiscal year -end , to have been approximately $24.9 million, $20.9 million and $16.7 million respectively. The lease also provided that it could only be modified by a writing signed by authorized representatives of both ALC and Ventas and that all “notices, demands, requests, consents, approvals and other communications ” under the lease were to be in writing with a copy to Ventas’s general counsel. Other provisions of the lease required ALC to use the Ventas facilities solely for their primary intended use and in a manner consistent with their operation as healthcare fa cilities. 2. ALC’s Fraudulent Scheme to Hide the Covenant Defaults 13. Beginning in 2008, s hortly after ALC entered the Ventas lease, occupancy at the Ventas facilities declined sharply. As a result of the occupancy d eclines, from at least the first quarter of 2009 throu gh the fourth quarter of 2011 , ALC failed, by a significant margin, many of the occupancy and coverage ratio covenants contained in the Ventas lease. Nevertheless, in each Form 10 -K and 10 -Q ALC filed during that period, ALC falsely represented that it was in compliance with the Ventas lease financial covenants. ALC also disclosed that non -compliance with the financial covenants could result in a “material adverse impact” on ALC. Moreover, beginning with the second quarter of 2011 , ALC falsely represented in its 2011 Forms 10 -K and 10 -Q that “it did not believe that there is a reasonably likely degree o f risk of breach of the [Ventas] covenants.” ALC included this additional disclosure in response to a comment letter received from t he Commission’s Division of Corporation Finance. 14. To hide ALC’s failure to comply with the covenants from Ventas, ALC’s CEO, Laurie Bebo, and CFO, John Buono, directed ALC accounting personnel to include in the covenant calculations between 49 and 103 fabr icated occupants for every day from July 2009 through December 2011. 4 To establish the number of fabricated occupants to be included in the covenant 4 On December 3, 2014, the Commission instituted pu blic administrative and cease and desist proceedings against Bebo and Buono, alleging violations of the antifraud, books and records, internal controls, reporting, and other provisions of the Exchange Act. In the Matter of Laurie Bebo and John Buono, CPA , Exchange Act. Release No. 73722. On January 29, 2015, the Commission entered an Order accepting Buono’s offer of settlement, finding that he violated each securities law provision alleged against him, imposing a $100,000 civil penalty, and barring Buono from practicing before the Commission as an accountant or serving as a n officer or director of a public company. Exchange Act Release No. 74177. On October 2, 2015, following a four -week evidentiary hearing, an Administrative Law Judge issued an Initial Decision making findings against Bebo. Initial Decision Release No. 893. 6 calculations, ALC accounting personnel, at Bebo’s and Buono’s direction, reverse -engineered th e requisite number of additional occupants needed to meet the covenants each quarter. ALC accounting personnel also prepared monthly journal entries to record revenue associated with the fabricated occupants which: (1) credited the fabricated occupant revenue to the individual Ventas facilities; and (2) debited revenue in the same amount in a corporate revenue account. 5 Shortly after the end of each quarter, ALC provided Ventas with covenant calculations which included the fabricated occupants in the occupancy covenant calculations and the revenue associated with the fabricated occupants in the coverage ratio calculations and thus falsely showed that ALC was meeting the covenants. 15. Bebo and Buono told Grant Thornton that Ventas had agreed that ALC could include in the covenant calcul ations ALC employees who travelled to and stayed at the Ventas facilities for business purpos es.6 However, in actuality, no such agreement existed and Ventas was never told that any ALC employees were being included in the covenant calculations. Even if Ventas had agreed that ALC could include in the covenant calculations employees who actually stayed at t he Ventas facilities, given that only a small number of ALC employees actually did so, ALC would still have mis sed the covenants by significant margin s. 16. In the third quarter of 2009, Grant Thornton asked ALC to identify the employees included in the covenant calculations. In response, Bebo created and provided Grant Thornton with a list identifying the employees and their associated lengths of stay at the Ventas facilities. Bebo would subsequently prepare and/or approve such a list for Grant Thornton for every quarter through the fourth quarter of 2011. However, give n the small number of ALC employees that actually stayed at the Ventas facilities and the large number of fabricated occupants necessary to meet the covenants, Bebo chose to inclu de on the list: (1) her family members and friends; (2) family members (including the seven -year old nephew) of another ALC executive; (3) employees who did not travel to, let alone stay at, the facilities; (4) employees of the Ventas facilities, who lived nearby and did not stay overnight at those facilities; (5) employees who had been terminated by ALC or employees who ALC anticipated hiring but who had not yet started; (6) various ALC employees as occupants of multiple Ventas facilities for the same time period; and (7) other individuals who were neither ALC employees nor residents of the Ventas facilities. ALC did not disclose any of this to Ventas. 17. The fraudulent scheme unraveled in the spring of 2012. In April 2012, Ventas, which was still unaware o f ALC’s use of employees in the covenant calculations, filed a lawsuit against ALC resulting from ALC’s unrelated failure to meet state regulatory requirements. The 5 As a result, ALC eliminated in consolidation the revenue associated with the fabricated occupants , and such revenue was not reported in its Commission filings. In addition, ALC did not include the fabricated occupants in the occupancy numbers reported in its Commission filings. 6 ALC’s practice of including in the covenant calculations employees or other non -residents is sometimes referred to hereafter as “the employee adjustment.” 7 following day, Bebo sent Ventas a settlement proposal pursuant to which Ventas would relea se ALC from liability for all claims including those arising from ALC including employees in the covenant calculations. The settlement proposal was the first time Ventas learned that ALC was including employees in the covenant calculations. On May 9, 201 2, Ventas issued a notice of default in which it accused ALC of fraud based on the employee adjustment. 18. In the meantime, on May 2, 2012, one of ALC’s accounting personnel filed a whistleblower complaint with the audit committee of ALC’s Board of Directors. The complaint described the employee adjustment as a “sham” and disclosed that ALC had included in the covenant calculations: (1) employees who did not tr avel to the Ventas facilities; (2) certain employees at multiple facilities on the same day; and (3) Bebo’s parents, husband, and a family friend. ALC immediately initiated an internal investigation, and Bebo was terminated shortly thereafter, purportedly for reasons unrelated to the employee adjustment. 19. In June 2012, ALC and Ventas settled the ir lawsui t. As part of the settlement, ALC purchased the Ventas facilities and certain other facilities from Ventas for an amount far greater than the appraised value of the facilities. ALC paid approximately $100 million to settle the litigation and purchase the facilities, even though independent appraisals only valued the purchased facilities at $62.8 million. Thus, in its second quarter 2012 interim financial statements, ALC included as an expense $37.2 million for “lease termination and settlement” and also wrote off the entirety of the remaining operating lease intangible assets associated with the Ventas facilities, which totaled approximately $8.9 6 million. 20. Grant Thornton issued audit reports containing unqualified opinions on ALC’s 2009, 2010 and 2011 financial statements. Each of those financial statements falsely disclosed that ALC was in compliance with the financial covenants. Those audit r eports and financial statements were included in ALC’s Form 10 -K Commission filings. 3. In Connection with the 2009, 2010, and 2011 ALC Engagements, Grant Thornton Was Aware of Numerous Risks and Red Flags Related to ALC’s Covenant Calculation Practices 21. ALC was one of the larger clients for Grant Thornton’s Wisconsin practice. a. The 2009 Engagement 22. By the beginning of Grant Thornton’s 2009 engagement, Koeppel and other members of the engagement team were aware that occupancy at the Ventas facilities had d eclined and that ALC was close to defaulting on the financial covenants. For this reason, in connection with its planning meeting for the first quarter 2009 review, the ALC engagement team focused on the financial covenants. The planning meeting agenda, which Koeppel reviewed, indicated that the Ventas covenant calculations were an area to which the engagement team should “devote special attention” and that the engagement team should “look closely ” at ALC’s covenant calculations. 23. Three weeks later, i n lat e April 2009, ALC accounting personnel sent a Grant Thornton junior engagement team member ALC’s calculations of the occupancy and coverage 8 ratio covenants for the quarter and underlying support for the numbers used in the covenant calculations. Shortly thereafter, the engagement team member noticed discrepancies for six of the eight Ventas facilities between the occupancy figures used in the covenant calculations and the underlying support. As a result, he sent an email to ALC accounting personnel asking for an explanation for the discrepancies. 24. In response, ALC account ing personnel emailed Grant Thornton a spreadsheet known as an “occupancy recon” which listed for each of the Ventas facilities for each month during the quarter: (1) the actual occupancy of the facilities; and (2) the number of ALC “employees” which were added to the actual occupancy numbers for purposes of the covenant calculations. The occupancy recon revealed that ALC wa s adding a total of 24 “employees” into the occupancy calculations fo r every day in the quarter. A review of the information contained therei n would have show n that ALC fa iled certain occupancy covenants without the inclusion of employees. 25. ALC accounting personnel al so provided the junior engagement team member with purported “support” for the e mployee adjustment, which was a February 4, 2009 email from Bebo to a Ventas emp loyee (the “February 4 email”). However, the February 4 email made no mention of the financial covenants, and merely mentioned that ALC may rent rooms at the Ventas facilities to certain of its employees “in the ordinary course of business.” The junior engagement team member elevated the issue to Grant Thornton ’s engagement manager, ye t Grant Thornton did not receive an y additional support from ALC. The engagement team told Koeppel about the employee adjustment and described it as unusual, but told Koeppel that the team had reviewed documentation which the team believed evidenced an agreement with Ventas. However, Koep pel never reviewed that documentation nor attempted to learn other details about it. 26. Grant Thornton did not perform additional procedures, such as seeking confirmation from Ventas that it had agreed to the use of employees in the covenant calculations.

In fact, G rant Thornton would never seek confirmation from Ventas that it had agreed to the use of employees in the covenant calculations. Had Grant Thornton done so, it would have confirmed that Ventas never agreed to the employee adjustment . 27. Grant Thorn ton’s workpapers for the first quarter of 2009 do not contain the February 4 em ail or refe rence ALC’s inclusion of employe es in the covenant calculations. The workpapers do reflect that Grant Thornton determined that ALC passed the covenant calculations based on the adjusted numbers contained in the occupancy recons. 28. ALC’s management representation letter to Grant Thornton for the first quarter of 2009 and every q uarterly review and audit in fiscal year 2009, 2010 and 2011 contained a representation that ALC had complied with all aspects of contractual agreements that would have a material effect on the financial statements in the event of noncompliance . 29. In the course of Grant Thornton’s second quarter 2009 review, a Grant Thornton summer intern was t aske d with performing the procedures with respect to ALC’s compliance with the Ventas lease covenants under the supervision of an audit senior . The intern reviewed an occupancy recon showing that ALC was including approximately 23 employees in the covenant 9 ca lculations for every day of the quarter . The intern then prepared a note to the workpapers referencing the emplo yee adjustment, which included language from the February 4 email, and stated: The employee adjustment relates to extra rooms at each facility not currently occupied by ALC residents. The rooms are subleased through ALC to improve the overal l performance of each facility…S ince the units are subleased, an adjustment is needed to show ALC occupancy and for Ventas testing. Koeppel signed off on thi s workpaper , even though she knew there were no formal sublease agreements for any employee -occupied rooms. 30. For the third quarter of 2009, Grant Thornton reviewed occupancy recons showing that ALC had included 45, 65 and 75 employees for each day of July, August and September respectively. At Koeppel’s request, the occupancy recons now included a list showing the names of the ALC employees purportedly staying at each Ventas facility, and the attendant length of their purported stay. This information would be contained in each subsequent occupancy recon , which showed employees staying at the facilities for every day of each month they were listed as an occupant . Koeppel never instructed the engagement team to perform additional procedures related to the li st of names to determine whether the list of e mployees, or the length of the employees’ purported stays, was accurate or appropriate. Moreover, Koeppel never asked the engagement team what procedures , if any, the team was performing with respect to the li st of names. Had Koeppel, or later Robinson, asked the engagement team to perform substantive procedures on the list of names , Grant Thornton would have discovered that the list was fraudulent. 31. As part of its third quarter review, Grant Thornton, requested that ALC include in its representation letter the representation that ALC had “ calculated the [Ventas] lease covenants in accordance with the corresponding lease agreement and the lessor’s instructions.” The engagement team wanted this rep resent ation included in part because it was concerned about the employee adjustment. In response, ALC modified the requested representation, and included the following in its representation letter: “We have calculated the [Ventas] lease covenants in accordance with the corresponding lease agreement and as understood by us after conferring with the lessor .” (Emphasis added). Grant Thornton’s workpapers do not contain an explanation for the modification in the language. 32. Grant Thornton’s workpapers for the thir d quarter 2009 review observed that one of the Ventas facilities had unexpected or unusual revenue trends in that changes in revenue could not be explained by corresponding changes in occupancy. The workpapers addressing these trends noted t hat a Grant Th ornton junior engagement team member had spoken with ALC personnel who explained that : (a) ALC employees were staying at the Ventas facilities per an agreement with Ventas; and (b) that such employees “pay rent and increase revenue,” but are not counted toward ALC’s company -wide occupancy numbers. This notation in the workpapers, which Koeppel reviewed, conflicted with her understanding that ALC employees included in the covenant calculations did not pay rent. 10 33. In the course of Grant Thornton’s 2009 year e nd audit, Koeppel reviewed risk assessment workpapers that she understood referred to risks associated with the Ventas financial covenants. 34. In connection with that audit, and in the subsequent 2010 and 2011 audits, Grant Thornton documented that ALC was using “unusual” journal entries in connection with the employee adjustment . Those journal entries record ed “negative revenue” in a corp orate revenue account to offset $1.2 million in non -GAAP revenue on the financial statements of the Ventas facilities associated with the employees included in the covenant calculations. The result of the offsetting revenue meant that the employee revenue reported to Ventas was eliminated from ALC’s consolidate d financial statements. Koeppel, and later Robinson, was aware of these unusual entries yet never instructed the engagement team to perform addition al testing on the employee adjustment. 35. Also in connection with the 2009 year end audit, the Grant Thornto n engagement team planned on conducting site visits to five Ventas facilities in Georgia and one in South Carolina to, among other things, verify facility occupancy figures, physically inspect the houses and the assets therein, make fraud inquiries with fa cility employees and review documentation that was maintained at the facilities. However, ALC requested that the engagement team ch oose other site visit locations in lieu of the Georgia facilities, and Koeppel acquiesced to ALC’s request. 36. Koeppel also revi ewed a workpaper detailing ALC’s monthly revenues for each of its facilities. That workpaper also indicated the occupancy rates for each facility. The workpaper showed that the occupancy rate for seven of the eight Ventas facilities fell far short of the rates required by the lease covenants. 37. In the course of the 2009 year end audit, Grant Thornton reviewed occupancy recons showing that ALC was including 103 employees in the covenant calculations for each day of the fourth quarter , yet failed to mention the employee adjustment in the covenant calculation workpapers . Nevertheless, Koeppel signed off on those workpapers. In addition, Koeppel determined that a representation specific to the use of employees in the covenant calculations, such as the one inc luded in the third quarter 2009 letter, was no longer necessary for ALC’s representation letter. No such representation was included in the representation letters for any subsequent ALC audits and reviews in the relevant period . b. The 2010 Engagem ent 38. In connection with its planning meeting for the first quarter 2010 review, Grant Thornton again focused on the financial covenants. The planning meeting agenda, which Koeppel reviewed, noted that the Ventas covenant calculations were an area to which the engagement team should “devote special attention” and that the engagement team should “ma ke sure we are testing the [Ventas] lease covenant calculations.” 39. For the first and second quarter 2010 reviews, Grant Thornton reviewed occupancy recons showing that ALC included 103 employees in the covenant calculations each day from January through M ay, and included 90 employees each day in June. 11 40. By the time Grant Thornton was conducting field work for the third quarter 2010 review, Grant Thornton national professional standards and risk management personnel ha d become aware of negative quality indica tors with respect to Koeppel, who was placed on a November 2010 monitoring list for pa rtners with such negative indicators . Grant Thornton placed Koeppel o n this list b ecause, among other things, her audit clients had restated their financial statements or interim financial information four times in the preceding two years . T he Public Company Accounting Oversight Board (“PCAOB”) , in its 2008 inspection report, had also found deficiencies on one of Koeppel’s engagements because Grant Thornton had failed to gather sufficient audit evidence. 41. One of the restatements that led to Koeppel’s inclusion on the partner m onitoring list involved Grant Thornton’s audit client, Koss Corporation (“Koss”). In June 2010, Koss restated its financial statements for the prece ding two fiscal years because one of its vice presidents had embezzled $31.5 million from 2005 through 2009 and would ple ad guilty to criminal charges based on the misconduct. Koeppel was the engagement partner for the Koss audit for three of the four years of the embezzlement. In July 2012, Grant Thornton , without admitting any liability, paid $8.5 million to settle a malpractice case f iled by Koss. 42. By November of 2010, Grant Thornton’s National Professional Practice Director for the Midwest region (“ NPPD”) became aware of negative audit quality indicators with respect to Koeppel as a result of an ongoing PCAOB inspe ction . Although the NPPD did not know that Koeppel had been placed on the partner monitoring list, he was informed by others in the firm that Koeppel had a number of negative audit quality indicators. The NPPD also became aware of observations by other Grant Thornton partners that the Wisconsin practice, for which Koeppel was the managing partner, had “gotten off the tracks from a methodol ogy perspective .” 43. By the fall of 2010, Grant Thornton had removed Koeppel f rom all of her other public company engagements ,7 but decided to allow her to continue as th e engagement partner for the 2010 ALC audit. 44. Following the completion of the second quarter 2010 review, the audit m anager assigned to the ALC engagement team resigned . During the planning stages for the third quarter review, Grant Thornton assigned a new engagement manager (the “Engagement Manager”) to be co -engagement manager on the AL C engagement. During the third quarter 2010 review, Koeppel became aware that the other Grant Thornton auditor assigned to be the co -manager of the ALC enga gement was removed from the engagement. Koeppel knew that t his left the Engagement Manager as t he only manager on the ALC engagement . 45. The Engagement Manager ’s assignment as the sole engagement manager for the ALC engagement contravened Grant Thornton policy, which required that lead managers and partners on public company audit engagements be “SEC de signated” to ensure that those 7 Grant Thornton removed Koeppel from the remainder of her audit and review engagements, including private company engagements , after completing her work on financial statements or interim financial info rmation for years or quarters ending December 31, 2010. 12 individuals had the appropriate level of experience and understanding about Commission requirements. To obtain such a designation, an auditor was required to spend 200 hours on public company engagements in a prescribed time period and complete certain continuing professional education courses. 46. Koeppel knew that the Engagement Manager was not SEC designated due to the Engagement Manager having spent insufficient time on publi c company engagements. Koeppel also knew that the other Grant Thornton auditor assigned to be the co -manager of the ALC engagement, who was SEC designated, was removed from the engagement. Koeppel knew that this left the Engagement Manager , despite her lack of SEC designation, as t he only manager on the ALC engagement . 47. On October 22, 2010, s hortly after the Engagement Manager started working on the ALC engagement, one of her subordinates brought the issue of the inclusion of employees in the covenant calculation s to the Engagement Manager ’s attenti on. The email the Engagement Manager received stated: You will also see that the only way some of the houses are passing [the occupancy] covenant is by having these [employee] adjustments. In prior quarters, as well as again this quarter, we have asked f or support for thes e [employee] adjustments but they haven’t [been] able to provide any. In prior quarters I have also asked if they have any support (letters, emails , etc. from [Ventas]) to show that [Ventas] is aware that ALC is adding in [employees] and is okay with it . They have nothing. [ the Engagement Manager ’s predecessor manager] said that all we can do then is rep it. . . Since you are new to the job, I wante d to bring this to your attention and make sure you are comfortable with this as well. Is there anything else we should do with this? 48. A fe w days later, an engagement team member emailed the Engagement Manager a copy of the Ventas lease and noted that she could not locate any provisions tha t defined “occupancy.” The Engagement Manager responded that she also could not find any such provisions and intended to contact ALC accounting personnel who prepared the covenant calculations for a dditional information . The Engagement Manager believed it was necessary to speak to ALC accounting personnel because ALC’s inclusion of employees did not comport with the traditional definition of “occupancy” for assisted living facilities. 49. The Engagement Manager then spoke with the ALC accountant who prepared the occupancy recons. Based on this co nversation, the Engagement Manager believe d that ALC was possibly providing Ventas with the occupancy recons and th at this allowed Ventas to discern that ALC was including employee s in the covenant calculations. In an email documenting her conversation with the ALC accountant, the Engagement Manager wrote that the accountant “did say he gives [Ventas] the more detailed spreadsheet that we have, so he thinks if they had a problem with it, they would have said something.” She added: If I can get John [Buono] to confirm this, then I think we can document that 13 they receive [the occupancy recon], have opportunity to disagree, etc. I’d still like it in the rep letter too then. But at least I feel more comfortable that they see the detail. 50. The Engagement Manager then attempted to contact Buono , who was unavailable. The Engagement Manager left Buono a voicemail inquiring about ALC’s use of employees in the covenant calculatio ns. In t hat voicemail, the Engagement Manager noted that based on her conversation with the ALC accountant, she believed that “maybe” Ventas received the occupancy recons showing ALC’s inclusion of employees. The Engagement Manager did not ask anyone at Grant Tho rnton to review the materials ALC sent to Ventas, which would have shown that Ventas was not receiving information showing ALC’s use of employees. Grant Thornton would not perform such a review until the year end 2011 audit. 51. After leaving Buono the voicem ail, the Engagement Manager spoke with Koeppel . Koeppel indicated to the Engagement Manager that Koeppel was aware that ALC had been including employees in the covenant calculations for the previous six quarters and that Koeppel was comfortable with the p ractice. Following this conversation, the Engagement Manager emailed Buono and told him to ignore the voicemail. 52. The Engagement Manager also suggested to Koeppel that ALC include in its representation letter a representation relating to ALC’s use of emp loyees in the covenant calculations, but Koeppel determined that such a representation was unnecessary. 53. Thus, the Engagement Manager was dissuaded by Koeppel from follow ing either of the Engagement Manager ’s recommendations – speaking with Buono and obta ining additional representations – for obtaining support for ALC’s use of employees in the covenant calculations. 54. In the course of the third quarter 2010 review, Grant Thornton for the first time included copies of the occupancy recons (without the employee names) in its workpapers. Those mater ials, which Koeppel and the Engagement Manager reviewed, showed that ALC was including between 68 and 72 employees in the Venta s covenant calculations for each day of the quarter. 55. Grant Thornton also included two notes in the workpapers regarding ALC’s use of employees in the covenant calculation. In the first note, Grant Thornton indicated that , as a result of the employee adjustment, the occupancy figures used in the covenant cal culations differed from the figures provided in ALC’s internal occupancy reports . In the second note, Grant Thornton explained that the difference was , in part, the result of the employee adjustment and wrote that: “The employee adjustment relates to extr a rooms at each facility that are not currently occupied by ALC residents, but are set aside for ALC employees to improve the overall performance of each facility.” The note also contained the incorrect observation that Ventas was receiving the occupancy recons showing the number of included employees, and had the opportunity to disagree with ALC’s practices. 56. In addition, Grant Thornton included the Ventas lease covenants in its Summary of Significant Matters (“the S SM”) for the third quarter 2010 review. Grant Thornton prepared the 14 SSM to highlight to en gagement partners and engagement quality reviewers, among others, important issues and questions encountered during t he audit or review . In the SSM, Grant Thornton mistakenly noted that it had “confirmed ” with Buono that Ventas received the occupancy recon. The SSM also noted that Grant Thornton was not able to “specifically test” ALC’s use of employees in the covenant calculations, that Grant Thornton accordingly needed to rely on ALC’s representations a bout its covenant practices, and that such a representation was included in ALC’s representation letter. 57. Both Koeppel and the Engagement Manager reviewed the SSM. The Engagement Manager knew or should have known that the SSM incorrectly noted that Buono h ad represented that Ventas received the occupancy recons . 58. In advance of Grant Thornton’s third quarter 2010 meeting with ALC’s audit committee, the Engagement Manager added the Ventas covenants as a written agenda item for Grant Thornton’s presentation to the committee. However, Koeppel removed the agenda item from Grant Thornton’s presentation. 59. In planning for the year end 2010 audit, Grant Thornton scheduled site visits to the Ventas facil ities after Koep pel and the Engagement Manager determined that those facilities were at higher risk for inaccurate occupancy reporting because of the Ventas lease covenant requirements and because ALC was barely passing the covenants. However, Grant Thornton’s workpaper s documenting its site visits to the Ventas facility do not reference any steps to validate the inclusion of employees in the covenant calculations. In fa ct, the Grant Thornton engagement team member who performed the site visit testing was unaware of the employee adjustment. Had Grant Thornton asked, employees who worked at the Ventas facilities would likely have told the Grant Thornton engagement team that only a handful of employees were staying at the facilities at any given time and that no rooms wer e being set aside or reserved at their facilities for employee use. Despite identifying the Ventas facilities as posing a highe r risk, Koeppel failed to sign off on the results of the site visit procedures performed. Koeppel also failed to direct the site visit team to obtain support for ALC’s use of employees in the covenant calculations, and failed to follow -up with the team to see if any such support had been obtained. 60. In the course of the year end 2010 audit , the Engagement Manager received a spreadshe et which showed the number of employees ALC included in the covenant calculations for each month during 2009 and 2010. On e hour after receiving this information, the Engagement Manager sent an email to he r subordinate, in which the Engagement Manager wrot e: 2009 is quite odd -how do they suddenly have 70 employees staying here one month? Did we discuss this with them last year (i.e. how were they tracking these before)? Also, they go from 70, then down to 24, then back up. This seems very odd. . . . Not sure how accurate this is. Nevertheless, the Engagement Manager failed to make any inquiries to ALC regarding her concerns or document her concerns in the workpapers . 15 61. Grant Thornton’ s workpapers for the 2010 year end audit included the occupancy recons, which showed that ALC included 61 employees in the cove nant calculations for each day of the fourth quarter. The workpapers also noted that ALC would fail the occupancy covenants without the inclusion of emplo yees. Both Koeppel and the Engagement Manager reviewed these workpapers. 62. For the 2010 year end audit, a Grant Thornton engagement team member suggested to Koeppel and the Engagement Manager that ALC’s representation letter should contain a specific representation relating to ALC’s use of employees in the covenant calculations. Koeppel responded , consistent with her view the prior quarter , by determining that the suggested representation was not necessary. c. The 2011 Engagement 63. Robinson replaced Koeppel as the ALC engagement partner beginning with the first quarter 2011 review. With the exception of the third quarter 2011 review, Grant Thor nton continued to staff the Engagement Manager as the manager on the ALC engagement. Prio r to the 2011 year end audit, Grant Th ornton approved the Engagement Manager ’s application for SEC designation, even though the Engagement Manager had still not completed the amount of public company engagement hours required under firm pol icy. In approvi ng the application, Grant Thornton demanded that the NPPD continue to perform a detailed review in connection with the year end audit. 64. When Robinson assumed responsibility for the ALC engagement, the engagement team advised Robinson that AL C was includ ing employees in its covenant calculation s. After this was brought to Robinson’s attention, he discussed the issue with Buono during an introductory lunch. Buono told Robinson that there was “an exchange of letters” that allowed ALC to use employees to m eet the covenants. However, Robinson never obtain ed any such documentation. 65. During the first quarter 2011 ALC r eview, the Engagement Manager , as she had done with Robinson, brought the Ventas covenants to the attention of Grant Thornton’s engagement quali ty reviewer (the “EQR”) , who had joined the ALC engagement as the engagement quality reviewer on the 2010 audit . In the course of doing so, she sent an email informing the EQR that ALC was very close to missing the covenants. 66. In the course of the first qu arter 2011 review, Grant Thornton learned that ALC had provided information showing that it failed one of the Ventas lease occupancy covenants. In reality, ALC had mistakenly included an insufficient number of employees to meet the covenant at issue. Aft er Grant Thornton raised the failed covenant with ALC, ALC provided Grant Thornton with revised covenant calculations in which ALC added employees in order to pass all of the covenants. 67. In the following days, a Grant Thornton engagement team member emailed the Engagement Manager to app rise the Engagement Manager of her concerns that ALC had added employees after init ially failing a covenant. The engagement team member also proposed that the Engagement Manager request from Buono a letter confirming Ventas’s agreement to allow 16 employees to be included in the covenant calculations. When the engagement team member spoke with the Engagement Manager, the Engagement Manager shared her assessment that ALC’s use of new employees to remedy a covenant failure was “od d.” However, Grant Thornton did not request additional support from ALC. When a second junior engagement team member learned that Grant Thornton had not followed up with Buono on obtaining a letter concerning the employee adjustment , he w rote an email to the first engagement team member observing that: “I don’t think we’re ever going to get this mystery letter.” 68. ALC’s workpapers for the first quarter 2011 review included the occupancy recons, which showed that ALC included between 60 and 65 employees in the cove nant calculations for each day of the quarter. The workpapers also noted that ALC would fail the occupancy covenants without the inclu sion of employees. Both Robinson and the Engagement Manager reviewed these workpapers. I n a separate section o f the workpaper, Grant Thornton noted that ALC had failed the occupancy covenant for one of the facilities in the first draft of the covenant calculations but explained that Buono had “found” additional employees which allowed ALC to pass . 69. The SSM for the first quarter 2011 review again identified the Ventas covenants as a significant matter. The relevant paragraph was substantially similar to the paragraph in the third quarter 2010 SSM except that it omitted the reference to Buono’s representation that Ventas was receiving occupancy recons. Robinson, the EQR, and the Engagement Manager each reviewed this SSM. 70. At Robinson’s request, Grant Thornton sent ALC a proposed representation letter for the first quarter 2011 review which contained a specifi c representation relating to the inclusion of employees in the covenant calculations. However, when Buono requested that Grant Thornton remove this repre sentation, Robinson and the Engagement Manager acquiesced. 71. In the course of its second quarter 2011 review, Grant Thornton learned that ALC had received a comment let ter from the Commission’s Division of Corporation Finance relating to ALC’s 2010 Form 10 -K, which requested , among other things, that ALC disclose its performance relative to the Ventas cove nants if it remained at “risk of non -compliance.” Grant Thornton reviewed ALC’s proposed response to the comment letter, in which ALC wrote that it did “not believe that it has a reasonably likely degree of risk of breach of the [Ventas] covenants” and th at it would add such a disclosure to its subsequent Commission filings. ALC in fact added such a disclosure to its Commission filings beginning with its second quarter 2011 Form 10 -Q. Grant Thornton did not suggest any revisions to ALC’s final response l etter or to ALC’s additional disclosure in its Commission filings . 72. ALC’s workpapers for the second and third quarter 2011 reviews included the occupancy recons, which showed that ALC included between 73 and 88 employees in the cove nant calculations for eac h day of the quarter. The workpapers again noted that ALC would fail the occupancy covenants without the inclusion of employees. 73. The SSM for the second and third quarter 2011 reviews contained identical paragraphs as the first quarter 2011 SSM with respect to the Ventas covenants. Robinson, the Engagement Manager , and the EQR reviewed both SSM s. 17 74. In preparation f or the 2011 year end aud it, Grant Thornton again recognized the risk associated with the financial covenants. The planning meeting agenda for the audit included an item identifying areas where “ALC management could perpetrate and conceal fraudulent financial reporting.” The ag enda identified “occupancy (affecting covenants and bonuses)” as one such area for po tential fraud. Robinson, the EQR, and the Engagement Manager reviewed this agenda. 75. During the 2011 audit, Grant Thornton for the first time requested a copy of the quarte rly covenant calculations that ALC sent to Ventas. Grant Thornton made this request after the Engagement Manager reviewed the fourth quarter 2011 occupancy recon and saw that 50 percent of the available units at one of t he Ventas facilities and 25 percent of the available units at another facility were purportedly occupied by ALC employees. 76. In response to Grant Thornton’s request, ALC provided copies of the quarterly materials it sent to Ventas. Those materials, like all of the earlier covenant informat ion ALC had provided to Ventas, did not contain any information indicating that employees were being included in the covenant calculations or the number of such employees. 77. A Grant Thornton engagement team member emailed the materials to the Engagement M anager and wrote: “The excel document [sent to Ventas] is exactly what we receive [in the occupancy recons] , except they exclude the tab where they add employees ( as we had kind of expected )” (emphasis added). The Engagement Manager responded to the emai l by writing that “I just don’t know h ow comfortable I am with this.” the Engagement Manager wrote that she called Buono to seek an explanation, but left a voicemail when Buono did n ot answer. The engagement team member then forwarded the Engagement Mana ger ’s email to a junior team member , but not the Engagement Manager , writing: “ I wish I could be on this call . . .” T he junior engagement team member responded: “Holly s --- .” 78. Buono did not return the Engagement Manager ’s call. In the meantime , the engagement team continued to be concerned about the employee adjustment. The day after leaving the voicemail for Buono, the Engagement Manager emailed Robinson and noted that the Ventas covenant calculations were still an “open item” because: (1) AL C wasn’t sending Ventas the employee occupancy information that Grant Thornton had been reviewing; and (2) she wanted to question Buo no on why so many of the occupants at two Ventas facilities were ALC employees. 79. Eight days after the Engagement Manager left the voicemail, Buono still had not returned her call. The Engagement Manager then left another voicemail for Buono in which she said that she wanted to discuss the Ventas covenants because the numbers for the q uarter looked “odd.” Four days later, when Buono still had not returned her calls, the Engagement Manager emailed Robinson to ask whether they should speak with the Chairman of ALC’s Audit Committee because Buono “still hasn’t responded to my questions with my concerns about how they are ‘meeting’ these covenants.” 80. In the mean time, both Robinson and the Engagement Manager again suggested adding a representation to ALC’s representation letter that spe cifically addressed the Ventas covenants, and Grant Thornton included such a representation in the draft letter it sent to ALC . 18 81. Eleven days after the Engagement Manager ’s initial voicemail to Buono, the two final ly spoke. On the call, the Engagement Manager failed to raise her concerns that ALC employees represented 50% of the occupancy of one Ventas facility and 25% of the occupancy of another facility. Rather, Buono requested that Grant Thornton remove the Ventas covenant representation contained i n the draft representation letter Grant Thornton had sent to ALC . 82. In response, the Engagement Manager asked Buono to provide some evidence that Ventas knew that ALC was including employees in the covenant calculations. Buono then described to the Engagement Manager : (1) a conference call with Ventas in 2009 during which Buono claimed that Ventas was notified about the inclusion of employees in the covenant calculations; and (2) an email from Bebo to Ventas confirming such conversation. The convers ation ended with the Engagement Manager requesting a copy of the email. 83. Shortly after the c onversation, Buono sent the Engagement Manager via email a co py of the February 4 email, which contained no mention of the Ventas covenants. T he Engagement Manage r also provided a copy of the February 4 email to Robinson. One minute after receiving the February 4 email from Buono, the Engagement Manager responded to Buono that she would remove the Ventas covenant specific representation from the final r epresentati on letter. The Engagement Manager did not consult with Robinson before removing the representation. However, in a subsequent conv ersation with Robinson, the Engagement Manager described the sequence of events and Robinson indicated his agreement with h er decisions. 84. Grant Thornton’s 2011 audit workpapers included the occupancy recons, which showed that ALC included 92 employees in the cove nant calculations for each day of the fourth quarter. The workpapers also contained a notation that the revenue associated with the employee occupants allowed ALC to satisfy the coverage ratio covenants. Other workpap ers, which Robinson and the Engagement Manager reviewed, also contained the incorrect observation contained in prior workpapers that Ventas received q uarterly information showing the number of employees ALC included in the covenant calculations. 85. As was the case for 2009 and 2010, Grant Thornton issued an audit report that included an unqualified opinion on ALC’s 2011 financial statements. 4. The 2012 Engagement 86. During the reviews of ALC’ s interim financial statements for the first and second qu arters of 2012, Robinson, the EQR , the Engagement Manager and the NPPD received additional information which indicated the impropriety of ALC’s inclusion o f employees in the covenant calculations. Grant Thornton did not withdraw its audit opinions on ALC’s prior period financial statements . 87. In the course of its first quarter 2012 review, Grant Thornton learned that Ventas had issued notices of default to ALC as a result of ALC receiving license revocations at certain of the Ventas faciliti es. On April 24, 2012, the Engagement Manager emailed Robinson to express her concerns regarding ALC’s efforts to resolve its disputes with Ventas, which included ALC offering Ventas $15 million t o terminate the lease. The Engagement Manager wrote: 19 Ventas may not realize the full extent to which occupancy has dropped. Although they are getting close to a deal on the properties now, what happens when Ventas finds out occ upancy is actually much lower than expected as ALC has been filling with employees. This is going to happen when they enter the properties. I wonder if Ventas might ask for more then because they may see it as the properties having depreciated more. 88. On April 26, 2012, when the efforts to settle the license revocation issues collapsed, Ventas sued ALC in federal court. ALC and Ventas would ultimately settle that litigation, after Ventas learned that ALC was including employees in the covenant calculation s, with ALC purchasing the Ventas facilities for $37 million more than the facilities’ appraised value. 89. In May 2012, Robinson, the EQR , the Engagement Manager and the NPPD learned that a whistleblower had submitted a complaint to ALC’s Audit Committee rela ting to the inclusion of employees in the covenant calculations and that ALC ’s Audit Committee had retained a prominent law firm to conduct an internal investigation of the allegations. While Grant Thornton was not provided with a copy of the whistleblowe r complaint until December 2012, ALC’s law firm described the whistleblower’s allegations to Robin son and the Engagement Manager during a meeting on May 14, 2012. In its workpapers, Grant Thornton noted that the whistleblower had alleged that the employee s included in the covenant calculations were not performing any services at the Ventas facilities or even leaving ALC headquarters. The workpapers also noted the whistleblower’s disclosure that the employee numbers were reverse engineered to meet the cove nants . Robinson, the EQR , and the Engagement Manager prepared or reviewed these workpapers. 90. On May 11, 2012, Robinson received an email from the Chairman of ALC’s Audit Committee attaching notices of default which ALC had received from Ventas. In one o f those notices, Ventas asserted that ALC had committed fraud by including employees in the covenant calculations. In the cover email, ALC’s Audit Committee chair indicated that ALC management was not authorized to speak with Grant Thornton about the mat ters being investigated by the law firm retained by ALC’s Audit Committee . 91. In the course of Grant Thornton’s second quarter 2012 review, the Engagement Manager prepared workpapers to explain the accounting for ALC’s purchase of the Ventas facilities, which indicated that ALC had paid Ventas $37 million in excess of the appraised value of those facilities “for damages as a result of occupancy rates falling signif icantly below required covenant occ upancy rates.” Robinson, the EQR , the NPPD and other members of Grant Thornton ’s professional standards group reviewed these workpapers. F. GRANT THORNTON’S AUDIT AND REVIEW OF BROADWIND 1. Background on the Recording of Intangible Assets 92. GAAP typically do es not permit the recognition of intangible assets, such as customer relationshi ps, as independent assets on a company’s balance sheet. An exception to this general rule is intangible assets purchased in connection w ith a business combination. In that 20 context, GAAP requires the consideration for an acquisition to be allocated across the tangible and intangible assets, with the remainder recorded as goodwill. 93. In connection with the Brad Foote acquisition in October 2007, Broadwind recorded amortizable intangible assets of $76 million and goodwill of $26 million. Nearly the entire $76 million intangible asset related to Brad Foote’s contracts with its two most significant customers, Customer 1 and Customer 2 , which w ere recorded at $62 million and $13 million, respectively. To establish the intangible asset value, management relied on a valuation conducted by an appraisal firm, Appraisal Firm 1 . Appraisal Firm 1’s valuation depended in substantial part on the forecasted net cash flows derived from the Customer 1 and 2 contracts over ten - and nine - year periods , respectively . Appraisal Firm 1 calculated those net cash flows from forecasts and estimated growth rates provided by senior managers at Broadwind . The net sales forecasts reflected management’s anticipation of aggressive growth. 94. Once established, an amortizable intangible asset is subject to periodic impairment testing. According to Accounting Standards Codification (“ASC”) 360, originally promulgated as Financial Accounting Standard (“FAS”) 144, an intangible asset is impaired when the carrying amount of the asset exceeds its fair value. A company is required to make this determination “whenever events or changes in circumstance indicate that its carrying amount may not be recoverable.” One such “triggering event” is “a current -period operating or cash flow loss combined with a history of operating or cash flow losses or a projection or forecast that de monstrates continuing losses associated with the use of a long -lived asset (or asset group).” Other e xamples of such triggering events include “a significant adverse change in the extent or manner in which a long -lived asset (or asset group) is being used or in its physical condition” or “a significant adverse change in legal factors or in the business climate that could affect the value of a long -lived asset (or asset group), including an adverse action or assessment by a regulator.” Broadwind purported to follow the accounting principles established by FAS 144. 2. The Decline of Broadwind’s Customer Relationships 95. Beginning in late 2008, Customers 1 and 2 significantly reduced actual and forecasted orders, causing substantial declines in Brad Foote ’s projected revenue associated with those relationships. Broadwind reacted to the downturn by planning or implementing numerous initiatives to rationalize the business, including decreasing headcount, returning machines to equipment suppliers, altering p roduction schedules, and withholding investments in additional capacity. 96. Throughout the first quarter of 2009, the declines in Customer 1 and 2’s revenues and forecasts worsened , falling more than 66% and 71% from the original forecast used to value the customer relationships . In response, Brad Foote took steps to restructure its workforce and operations, laying off more than 200 employees and senior staff through at least three workforce reductions, and reducing its material orders from suppliers. Brad Foote aggressively sought business from other non -wind customers to replace the significant declines and correct the underutilization of its capacity. 21 97. In early 2009, a s its financial condition worsened, Broadwind received early indications of potential impairment while preparing its 2008 annual report . During its 2008 financial statement audit, Broadwind retained Appraisal Firm 1 to test its goodwill and intangible assets for potential impairment. In March 2009, Appraisal Firm 1 informed Broadwind that it had calculated a $15 million impairment charge associated with the Customer 1 contract. Appraisal Firm 1 subsequently modified its calculations, which resulted in no impairment. Although the Company ultimately did not disclose a charge in its 2008 Fo rm 10 -K, following consultation with Grant Thornton , Appraisal Firm 1’s preliminary result placed Broadwind’s management on further notice of the potential for impairment. 98. During the second quarter of 2009, the Customer 1 and 2 relationships deteriorated more precipitously. At a special board meeting convened on June 9, 2009, Brad Foote management reported that Brad Foote would produce no more than $85 million in sales for 2009, compared to $120 million in previously anticipate d sales. Management also reported that 2009 revenue forecasts for Customer 1 had decreased to $30 million and that the 2009 revenue forecasts for Customer 2 had declined to $15 million. In connection with a “massive and sudden schedule” reduction imposed by Customer 1 and other developments, Broadwind and Brad Foote began developing a “life without Customer 1” business plan, and several members of management expressed the desire to exit the relationship. 99. For Brad Foote as a whole, second quarter 2009 rev enue fell 31% against the second quarter of 2008 and 24% compared to the first quarter of 2009, and its gross margin turned negative. As in previous months, Brad Foote continued its efforts to redirect sales from Customers 1 and 2 to new customers and imp lemented additional workforce reductions. 100. In the third quarter of 2009, Brad Foote’s actual and forecasted revenue continued to decline. Specifically, year -to-date revenues from Customer 1 and Customer 2 through September 30, 2009 declined 42% and 25%, re spectively, compared to the same period ending September 30, 2008. 101. On July 29, 2009, Broadwind’s CEO provided the Board a “rationalization update” that described the precipitous decline and the restructuring that management was taking to respond to weak pe rformance. During the meeting, Brad Foote management presented a financial forecast for Brad Foote showing that the expectation for Brad Foote sales for 2009 had declined by $49 million, or 41% to a newly revised forecast of $71 million. Brad Foote manag ement also projected net income of negative $16 million for the year. During a portion of the presentation discussing “risks to [the] income statement,” Broadwind management identified “continued loss of volume” and “impairment at subsidiaries.” 102. The decli nes in actual and forecasted results were not limited to 2009. Revenue forecasts for future years also declined. For instance, on July 30, 2009, C ustomer 1 submitted a 2010 forecast that was only 13% of the $125 million 2010 projection used to establish the acquisition value of the Customer 1 contract. 103. Around this same time, Brad Foote management informed the Board that Brad Foote would generate revenues of $30 million from Customer 2 in 2010, $41 million in 2011, and 22 $54 million in 2012. On average, tho se forecasts represented a 70% decline from the comparable period forecasts originally used to value the Customer 2 contract. 104. A b oard meeting was scheduled for October 1, 2009 to determine whether the Board should sever its relationship with Customer 1. A t that meeting, the Board decided not to terminate Broadwind’s relationship with Customer 1. However, Broadwind continued to budget for and anticipate impairment in several documents and communications through December 2009. 3. Broadwind’s Misrepresenta tions and Omissions in Its Commission Filings 105. As a result of the se developments, by the third quarter of 2009, the intangible assets associated with the Customer 1 and 2 relationships met the triggering event tests established by FAS 144 , and these assets were impaired. Consequently, Broadwind materially overstated its intangible assets and understated a material impairment charge in its Form 10 -Q for the third quarter of 2009, which was filed on November 2, 2009. 106. Given the declines in its business and ot her developments, Broadwind resolved in October 2009 to raise additional capital through a public offering of its stock. In anticipation of the offering that ultimately proceeded in January 2010, Broadwind filed a registration statement with the Commissio n. The registration statement was initially filed on October 30, 2009 and was amended various times between November 6, 2009 and January 14, 2010. The registration statement ultimately went effective on January 14, 2010. All versions of Broadwind’s regi stration statement included the interim financial statements that had been included in the Forms 10 -Q filed through the end of the third quarter, September 30, 2009, and expressly incorporated prior reports by reference. 107. As was the case with the third -qu arter Form 10 -Q, Broadwind’s registration statement was incomplete and misleading. By incorporating third -quarter interim financial statements that did not report the impairment charge, Broadwind failed to disclose that its intangible assets already had b een substantially impaired. Failing to disclose the impairment allowed Broadwind to proceed with an offering that was critical to its financial survival and to give the misleading impression that its business was stronger than actual and predicted results established. 108. Less than two weeks after the completion of the offering, on February 2, 2010, Appraisal Firm 2 informed Broadwind’s management of its preliminary finding of impairment. 109. Approximately one month later, Broadwind disclosed the impairment in its 2009 Form 10 -K and earnings release, filed on March 12, 2010. Broadwind disclosed a $58 million charge to intangible assets and full impairment of its goodwill related to Brad Foote in the amount of $24 million. Described by the Board as “significant,” the charge reduced the value assigned to customer contracts by 94%. Of the $58 million intangible impairment charge, $56 million directly related to the declining value of the Custome r 1 and 2 relationships. Largely as a result of the charge, Broadwind’s operating loss for the year increased from $28 million to $110 million on reported revenues of $198 million. 23 110. Following the revelation of the charge, Broadwind’s stock declined 21%, fr om a closing price of $5.68 on March 11, 2010 to a closing price of $4.47 on March 12, 2010, on increased volume. On the next trading day, March 15, 2010, the price fell another 8% from the March 11, 2010 price to $4.11, for a total decline of 29%. In co ntrast, the broader market, as reflected by the Nasdaq Composite Index, was essentially unchanged for these two days . The charge also was significant because it signaled ser ious weaknesses in Broadwind’s long -term prospects, particularly with two of the i ndustry’s largest players. 111. Because Broadwind did not recognize impairment as a third -quarter event, its Form 10 -K for 2009 misstated third and fourth -quarter results. In the Form 10 -K, Broadwind inaccurately attributed the impairment charge to fourth -quar ter events when the events that resulted in the impairment were present during the third quarter . 4. Grant Thornton’s Early Warnings of Potential Impairment 112. In inspection reports issued begi nning in 2007, PCAOB notified Grant Thornton of several impairment audit deficiencies in othe r engagements. Further, in its 2008 and 2009 inspection report s, PCAOB notified Grant Thornton of issues regarding the effectiveness of Grant Thornton ’s firmwide quality controls with respect to (1) profess ional skepticism, (2) engagement supervision, and (3) impairment procedures. 113. As Broadwind’s auditor, Grant Thornton learned of the decline in sales and volume that began in late 2008, as well as the restructuring steps undertaken by Broadwind to address the declines, through its audit work. For instance, Grant Thornton reviewed Broadwind’s board and committee materials in connection with each of its reviews and audits. Grant Thornton’s work papers also noted various negative developments through its rev iew of financial metrics and conversation with accounting personnel and others at Broadwind. 114. In addition, because it reviewed and analyzed Appraisal Firm 1’s work during the 2008 annual audit in early 2009, Grant Thornton also received the early indication s of potential impairment identified by Appraisal Firm 1 . Although Broadwind ultimately did not disclose an impairment charge in its 2008 Form 10 -K, the preliminary result placed Broadwind and Grant Thornton on notice of the potential for impairment. 115. Duri ng the quarterly review for the period ending March 31, 2009, Grant Thornton asked management to prepare a memorandum documenting its consideration of triggering events that could indicate impairment of the intangible assets associated with the Customer 1 and 2 relationships. Management’s memorandu m to Grant Thornton recited the standard for impairment, noted the recent downturns, but summarily concluded that no triggering events had occurred. 116. Grant Thornton also requested a similar assessment for the seco nd quarter of 2009. Management’s memorandum again noted key customers’ intentions to scale back orders, but summarily concluded these declines simply represented a delay in the timing of revenues and associated cash flows. With respect to the second quart er 2009 assessment, the senior manager for the Grant Thornton engagement team wrote the following to Koeppel: 24 The problem is that the ability to forecast has been suspect with this group and whether or not the sales are suspended or eliminated is the quest ion . . . their position is suspended which is how they get to no impairment. I think this meets adequate documentation at an interim date and not sure we are going to get a better product. Please let me know if you disagree and whether or not we should have more expansive forecasting to illustrate the expected revenues from these customers that have pushed out. 117. Despite these concerns, Koeppel failed to direct the engagement team to obtain more e xpansive forecasting from Broadwind , or to perform any a ddit ional procedures , during the second -quarter review. The Grant Thornton review file similarly included no documentation of any a dditional forecasting by Broadwind or additional procedures performed . 5. Grant Thornton Learns of Broadwind’s Expectation of Impairment 118. In August 2009, after the continued deterioration of the Customer 1 and 2 customer relationships, Broadwind management shared with Grant Thornton its expectation that its intangible assets would be substantially impaired. Specifically, Grant Thornton received internal budgeted balance sheets and income statements for 2010 and other documents reflecting management’s assumption that the entire Customer 1 contract intangible asset would be impaired by December 31, 2009. Several of these document s specifically identified an expected impairment charge of $48 million to be recorded in the fourth quarter. 119. After being informed on September 9, 2009 of management’s expectation, Koeppel wrote to the senior manager : “Guess they see the writing on the wa ll for [Brad Foote] — seems like we need a triggering event?!?!?” One day later, the senior manager summarized for Koeppel a discussion with Broadwind’s Chief Accounting Officer (“CAO”) and Controller: The impairment has not been booked or determined, but t hey believe a Customer 1 triggering event is a week or two away and is based on a customer retention decision at the board level . . . [the CAO] feels they are slanting the board materials to keep Customer 1 , but are not incorporating the true costs of the customer relationship which would potentially paint a different light. 120. Grant Thornton incorporated management’s expectation in its planning for the third -quarter review and year -end audit of Br oadwind’s 2009 financial statements . On September 29, 2009, B roadwind and Grant Thornton met to plan the upcoming annual audit. Agenda and notes drafted by Grant Thornton in preparation for the meeting reflect “expected impairment,” “loss of Customer 1 . . . FAS 144 writeoff,” and the need “to get started as soon a s possible.” Relatedly, a client meeting agenda listed “impairment analysis” as a topic of discussion. Attachments to the agenda included other references to impairment and an estimated $48 million charge. A few days after the audit planning meeting, on October 4, 2009, Grant Thornton forwarded Broadwind’s Controller guidance on impairment disclosures “as a follow up to [their] discussion regarding potential impairment.” 25 6. Grant Thornton’ s Failed Review of Broadwind’s Third -Quarter Impairment Assessment 121. Broadwind’s third quarter ended on September 30, 2009. As discussed above, i n early October 2009, Broadwind decided to raise additional capital through a follow -on offering of its stock, referred to as a “re -IPO,” and began to prepare a registration state ment. The Grant Thornton engagement team was aware that Broadwind was embarking upon an offering and understood that the offering was critical to Broadwind’s financial survival. Grant Thornton also understood that Broadwind’s CEO planned to participate i n the offering by selling shares that he personally owned . Broadwind originally planned to complete the offering in late November 2009. 122. Grant Thornton’s quarterly review of Broadwind’s third -quarter interim financial statements began on October 21, 2009. On or about October 27, 2009, immediately before a planned filing of a registration statement, Grant Thornton learned that Broadwind’s management had decided not to file the registration statement due to the withdrawal of the lead underwriter. After being informed by Broadwind’s CFO that there were “underwriting issues at the last minute,” Koeppel conveyed her understanding to the senior manager and manager that “impairment may be an issue/concern — they want to highlight it may occur — [the CAO’s] and my concern is it starts to look like you should have already recorded it then — details, details!” 123. In connection with the third -quarter review, on or about October 29, 2009, despite prior communications of an expect ed impairment, Broadwind management once again prepared an impairment assessment for Grant Thornton that concluded that no triggering event had occurred. In the assessment, Broadwind incorporated a comparative table of year -to-date revenues by customer th at reflected substantial revenue declines. Management summarily characterized these declines as temporary and asserted that long -term volumes had not been materially changed. However, management offered no specific evidence to support its view that order s would return to the volumes forecasted in 2007, and other facts contradicted its assertion. Based on this unsupported and incorrect conclusion, Broadwind failed to disclose a significant impairment charge in its Form 10 -Q filed on November 2, 2009, opti ng instead for a generalized risk disclosure of the possibility of such a charge. Management’s assessment was incorporated into a Grant Thornton working paper that was reviewed and approved by Koeppel and others on the engagement team. 124. Notwithstanding th e awareness of a significant likelihood that the Customer 1 and 2 contracts were impaired, and despite lingering uncertainty about management’s forecasting ability, Koeppel and other s on the engagement team accepted management’s unsupported assertion that the reduced orders only represented a temporary deferral of sales that would be recovered over t he life of the supply agreement. Grant Thornton noted in its workpapers , which Koeppel revi ewed : GT notes that although volumes are below those projected by mg mt during the 2008 integrated audit, management does not believe that this is a triggering event due to the following. (i) BFGW continues to have long - term contracts with [Customer 1 and 2 ]. For example, the Customer 1 contract contains automatic renewals , termination provisions, etc. Current discussions with Customer 1 indicate increased future activity. (ii) These are 26 long -term relationships with an estimated useful life of 9 [ Customer 2 ] and 10 [ Customer 1 ] years. Thus, it is reasonable to assume that short -term reductions in volume will be made up over the longer term. Management does not believe, and GT concurs, that a short -term decrease from budgeted activity qualifies as a triggering event and such analysis does not seem to be supported by FAS 14 4. 125. Even though multiple facts known to Koeppel and the engagement team did not align with management’s conclusion, and Grant Thornton knew that Broadwind wa s pursuing an offering, Grant Thornton did not perform any procedures to determine whether the order reductions were other than tempor ary. On the contrary, the review documentation noted that Grant Thornton simply “assume[d]” that these reductions were temporary. Among o ther things, Grant Thornton did not a sk Broadwind to provide detailed information a bout Customer 1 and 2 volume expectations beyond 2009, which was available to management at the time. Grant Thornton’s files do not include any documentation indicating how Koeppel or the engagement team determined the reasonablene ss and consistency of Br oadwind ’s response in light of the results of other review procedures and their knowledge of the business. 126. In fact, Broadwind ’s response conflicted with longer -term forecasts presented to the Board as early as July 2009 , which were available to Grant Tho rnton. Koeppel’s experience with Broadwind since 2007 provided her with an understanding of Brad Foote’s operations to recognize that Brad Foote required longer -term forecasts from Customer 1 and 2 to manage its material requirements and production schedu le. Consequently, she should have known that 2010 volume forecasts were available at the time of the third -quarter triggering event assessment. 127. Further, during Grant Thornton’s review of Broadwind ’s third -quarter impairment assessment memorandum, Grant T hornton’s professional standards partner (the “PSP”) assigned to the Broadwind engagement provided comments to Koeppel and the manager . The PSP had been assigned by Grant Thornton to monitor the Broadwind engagement as a result of Grant Thornton’s designation of the engagement as high risk. In a communication to the engagement team about Broadwind ’s impairment assessment, the PSP wrote the following: “Are the revenues consistent wi th the projections used for their last impairment analysis? If they are significantly declined than that, then you would have a trigger you may need to revisit. They don’t address this here. Also need GT conclusion & view on impairment in the files.” Alth ough the team’s own third -quarter revenue analytics working papers demonstrated that the revenues were not consistent with the projections used for Broadwind’s last impairment analysis, the engagement team did not revisit the triggers for an impairment ana lysis. The manager responded to the PSP simply that management viewed the contracts as long term, such that a short -term decrease from budget would not qualify as a triggering event. 7. Grant Thornton Defers to Broadwind in Its Comfort Letter 128. Broadwind’s third -quarter 2009 interim financial statements were incorporated into its registration statement for the public offering in January 2010. In connection with the offering, Grant Thornton was retained to provide a comfort letter to investment bankers and reviewed draft registration statements and impairment disclosures. As Grant Thornton performed its procedures, 27 Koeppel and the engagement team learned that the Commission’s Division of Corporation Finance had questioned Broadwind’s impairment disclosures. As part of its review of Broadwind’s registration statement, the Division of Corporation Finance issued Broadwind a comment letter in late November 2009. In response to comments questioning the Company’s impairment disclosures, Broadwind added more deta il to its description of its significant accounting policies in its MD&A and the notes to consolidated interim financial statements. However, the additional language simply provided more detail about the testing process and did not alter the substance of Broadwind’s disclosures regarding impairment or the risk of impairment. 129. Following this expanded process disclosure, Koeppel and the senior manager defended the quality of Broadwind’s disclosures surrounding impairment in email exchanges with the PSP. The PSP voiced serious reservations to the engagement team about the impairment disclosures and emphasized the company’s obligation to provide an early warning to investors if it believed an impairment charge was reasonably possible. The PSP specifically caution ed Koeppel that “[w]e can’t just turn a blind eye if we believe there is a good possibility they will have impairment.” Koeppel and the senior manager addressed the PSP’s concerns by obtaining a representation from management that impairment testin g was in process but had not yet been completed. Grant Thornton’s comfort letter expressly cautioned that it had not performed any procedures surrounding impairment of intangible assets with respect to the period from October to November 2009, deferring i nstead to its “inquiry of management.” 8. Grant Thornton Negligently Audits Management’s Allocation of Impairment Charge to the Fourth Quarter 130. In the course of its year end audit of Broadwind, Grant Thornton’s workpapers failed to differentiate the fourth -quarter declines from comparable declines that occurred in prior periods. At the urging of the PSP just days prior to audit signoff, the engagement team considered whether the impairment charge should be recorded in the fourth quarter of 2009 or some earlier period. The resulting audit file memo randum , which Koeppel reviewed , relied on selected sales metrics to demonstrate a purported deterio ration in the fourth quarter of 2009. In the memorandum, Grant Thornton reasone d that: (1) the fourth quarter 2009 revenue decline from fourth quarter 2008 (65%) and from third qua rter 2009 (32%) was significant; and (2) prior to the fourth quarter, actual and forecasted revenue amounts related to the reporting unit were consistent with previous expectations. 131. Grant Thornton ’s observation that, prior to the fourth quarter, actual revenues were consistent with the prior year’s forecasted revenues was incorrect and was contradicted by the engagement team’s own revenue analytics documen tation in each of the first three quarters of 2009. The team’s third -qu arter review impairment work papers also acknowledged this underperformance by stating “ that , although volumes a re below those projected by [management] during the 2008 integrated audit , management does not believe that this is a trigg ering event.” In fact, Brad Foote’s actual quarterly revenues in 2009 failed to meet the forecasted revenues used in the fiscal year 2008 impairment analysis by significant margins. Moreover, Grant Thornt on’s reasoning applied with equal force to prior quarters in 2009 , and these declines were documented by Grant Thornton each quarter in its revenue analytics working papers . In the second and third quarters, Brad Foote realized comparable declines in reve nue. Second quarter 2009 revenue fell 28 31% against second quarter 2008 and 24% compared to first quarter 2009. Third quarter 2009 revenue fell 49% compared to third qua rter 2008 and 19% compared to second quarter 2009. 132. The engagement team’s memorandum on this issue failed to discuss whether or how it considered the significant sales declines that had occurred in the third quarter in concluding the company was correct in recording the impairment charge as a fourth -quarter event. Further, the tea m’s incl usion of the inaccurate statement that actual revenues were consistent with prior -year projections reinforced the company’s incorrect allocation of the impairment charge to the fourth quarter. Koeppel and Grant Thornton sought only evidence to corroborate management’s conclu sion while disregarding evidence from their own prior work that contradicted management’s conclusion. 133. Around the same time that the PSP had asked the engagement team to evaluate the timing of the impairment charge, Koeppel provided Br oadwind’s CFO with comments on Broadwind’s draft earnings release to be issued in connection with the filing of Broadwind’s 2009 Form 10 -K. The CFO’s draft had attributed the impairment charge to “reduced wind gearing purchases under key customer contract s beginning in late 2008 .” (emphasis added). Koeppel urged the CFO to reevaluate the language in the release . Koeppel wrote : “We view the following reference to 2008 as problematic as it may suggest that you should have taken the impairment charge earli er . . . Suggest that you expand the sentence to focus on Q4 events which drove this ass essment.” Koeppel proposed these revisions to the press release prior to the engagement team’s drafting of the memo randum purporting to document its consideration of t he charge’s timing. 9. Grant Thornton’s Failure to Detect Broadwind’s Overstatement of Revenue a. Grant Thornton Identifies Revenue Recognition Risks 134. In the course of the 2009 year end audit, Koeppel and Grant Thornton identified the risk of material misstatement due to fraud in the area of revenue recognition at Broadwind’s most significant subsidiary, Brad Foote , as a specific risk. Audit planning documentation identified the risk that “sales include fr audulent transactions” as “high” and “reasonably possible.” Planning documentation further noted “possible incentive to play with earnings especially at the BFGW level (there are monthly sales targets included in their loan covenants).” More generally, G rant Thornton had identified the Broadwind engagement as a high -risk audit. This conclusion was influenced in part by Broadwind’s prior disclosure s of material weaknesses in controls over revenue recogni tion and other controls in earlier periods . b. Overstatement of Revenue by Broadwind 135. This risk in fact materialized. The deterioration in customer relationships that produced the impairment charge also compromised Brad Foote’s ability to meet monthly debt covenants associated with its primary credit f acility. To avoid default and other negative consequences, Brad Foote personnel accelerated revenue to meet its covenants until Broadwind could raise funds to retire the credit facility through the offering in Janua ry 2010. Broadwind failed to disclose t his practice and its effect on future revenue in the registration statement used in 29 the offering. In addition, as a result of the transactions, Broadwind reported $4 million of improperly recognized revenue for the third and fourth quarters of 2009, inclu ding certain bill -and - hold transactions entered with Customer 2. This revenue was material to Broadwind’s financia l results. 136. The Customer 2 bill -and -hold transactions had their genesis in a broader “pull - ahead agreement” between Customer 2 and Brad Fo ote. In response to forecast reductions in early 2009, Brad Foote personnel approached Customer 2 about pulling $6 million of orders , consisting of 150 sets of gear boxes, from 2010 into 2009 “to ensure [Brad Foote’s] future compliance with debt covenants ” and its ability to continue supplying gearboxes to Customer 2. Brad Foote’s proposal was not requested by Customer 2 or tied to any commercial need on the part of Customer 2 beyond the survival of a critical supplier. The 150 sets were to be pulled fro m requirements that were scheduled to ship in 2010 and would not be consumed until the first half of 2010. Because Customer 2 had no need for the sets and would carry the 150 sets as excess inventory, Brad Foote proposed “to cover Customer 2 ’s carrying an d storage costs through deflation in 2010.” In addition, because the long -term agreement with Customer 2 provided for an annual reduction in prices paid by Customer 2 , Brad Foote agreed to accept 2010 prices for the parts. Brad Foote committed that it wo uld not ship the products to Customer 2 if it were able to identify new business from other customers. After initially refusing the request , Customer 2 agreed to provide $3 million of support . Customer 2 scheduled the 75 sets to be delivered from late Au gust 2009 through November 2009. Brad Foote’s delivery of these sets caused significant disruption at Customer 2 , given its lack of need for the parts until 2010 . Brad Foote paid Customer 2 the carrying cost and the price reduction through a 1.5% discount that was spread over sh ipments that occurred in 2010. 137. As a result of the pull -ahead agreement, by October 2009, Customer 2 exceeded its ability to store the excess inventory. Consequently, Customer 2 approached Brad Foote about storing the remain ing gear sets through a bill -and -hold arrangement. On October 31, 2009, Brad Foote agreed and entered into a bill -and -hold arrangement with respect to 30 gear sets totaling $1,247,160. The gears were not shipped to Customer 2, but instead were supposed t o be segregated and maintained onsite at Brad Foote. Customer 2 required Brad Foote to segregate the product and provide evidence of compl etion and periodically inspect the product on site. Notwithstanding these efforts, Brad Foote failed to segregate th e product consistently. The transaction failed to meet the criteria to recognize revenue under a bill -and -hold arrangement. See In the Matter of Stewart Parness , Accounting and Auditing Enforcement Rel. No. 108 (Aug. 5, 1986), Staff Accounting Bulletin 1 04, Revenue Recognition . 138. On November 30, 2009, Brad Foote placed 30 sets into a bill -and -hold arrangement similar to the one entered with Customer 2 in October 2009. Fifteen of these 30 sets completed the pull -ahead arrangement with Customer 2 , and Br ad Foote documented Customer 2’ s request through the same email authorizing delivery in October 2009. The other half corresponded to an additional 15 sets not formally ordered by Customer 2 until December 2009. In total, Brad Foote recognized $1,194,471 of revenue associated with these 30 sets. As with the October 2009 shipment, Broadwind failed to produce evidence of Customer 2’s substantial business purpose for the arrangement and failed to comply with other requirements of a proper bill -and -hold arran gement. 30 c. Grant Thornton’s Audit of the Bill -and -Hold Arrangement 139. Grant Thornton never learned the details of the bill -and -hold arrangement. During the 2009 year end audit, Grant Thornton identified the October 2009 transaction as a transaction for further review. However, the engagement team’s testing of this transaction was limited to obtaining a summary of its terms from management. The summary failed to identify several aspects of the transaction critical to its proper acco unting, including, among other things, the price discount and the inventory carrying cost. For example, i nterviews with Brad Foote personnel , review of documents associated with the transaction, or confirming the terms directly with Customer 2 could have revealed that the bill -and -hold arrangement was the product of a large -scale pull -ahead agreement and had no independent business purpose. 140. More broadly, prior to the conclusion of the audit, management identified multiple problematic transactions at Brad Foote designed to meet monthly revenue covenants during a pending offering. These transactions included improperly recorded bill -and -hold transactions that should have raised additional questions about the October 2009 transaction with Customer 2. The engagement team did not document how it altered the nature, timing, or extent of its audit procedures in a manner that addressed this risk. 141. Koeppel participate d in the audit planning process and fraud brainstorming discussion, signed off on the audit workin g papers in which the identified risks were documented, and signed off on working papers concluding that the Customer 2 bill -and -hold transaction met applicable revenue recognition criteria. d. Retention of Supporting Documentation in the Audit File 142. As discussed above, Broadwind identified and corrected certain transactions in which it had improperly recognized revenue through an internal review conducted by management from late 2009 to early 2010. A report of the internal review provided to Grant Th ornton and referenced in its working papers identified multiple instances of improper revenue recognition, which included backdated letters, side agreements, unauthorized bill -and -hold transactions, and a directive not to record a credit memo. These trans actions contributed to an overstatement of revenue that enabled Brad Foote to barely meet its debt covenants in the months leading to a critical stock offering. However, the engagement team did not retain a copy of the report or supporting documentation f or the questionable transactions in the audit file. G. VIOLATIONS RULE 102( e) AND SECTION 4C OF THE EXCHANGE ACT 143. Grant Thornton’s 2009, 2010 , and 2011 audits of ALC, and its 2009 third quarter review and year end audit of Broadwind were deficient and not performed in accordance with PCAOB s tandards .8 Section 4C(b) and Rule 102(e)(1)(iv) define improper professional conduct 8 References to auditing standards in this Order are to PCAOB standards in effect at the time the audit work was performed. For example, the PCAOB risk assessment standards (AS 8 -15) 31 with respect to persons licensed to practice as accountants. Pursuant to these provisions, “improper professio nal conduct” includes two types of negligent conduct: (1) a single instance of highly unreasonable conduct that results in a violation of professional standards in circumstances in which an accountant knows, or should know, that heightened scrutiny is war ranted; or (2) repeated instances of unreasonable conduct, each resulting in violations of professional standards, that indicate a lack of competence. 144. As set forth above, Grant Thornton knew, or should have known, that the Ventas financial covenant ca lculations were an area in which heightened scrutiny was warr anted in connection with the 2009, 2010 and 2011 audits of ALC . Moreover, Grant Thornton knew or should have known facts tha t called into question ALC’s claims that it was meeting the Ventas lea se covenants by virtue of an agreement with Ventas to include employees and other non - residents in the covenant calculations. Throughout the 2009, 2010, and 2011 audits and r eviews, Grant Thornton was aware of repeated red flags surrounding ALC’s practice of meeting the covenants by treating employees and other non -residents as occupants of the Ventas facilities. In light of these red flags, Grant Thornton failed to take reasonable steps to verify that an agreement with Ventas existed or that the employee s ALC claimed to be occupants of the facilities were in fact staying there. Had Grant Thornton taken such measures, it could have exposed and put an end to ALC’s fraud. Instead, Grant Thornton issued audit reports in 2009, 2010 , and 2011 containing unqualified opinions that were filed with ALC ’s financial statements in the Form 10 -Ks. In those reports , Grant Thornton inaccurately stated that the audit had been conducted in accordance w ith PCAOB standards and that ALC ’s financial statements presented fairly, in all material respects, the company’s position and results in conformity with accounting principles generally accepted in the United States of America . 145. Further, i n the course of its 2009 third quarter review of Broadwind, Grant Thornton knew, or should have known, that Broadwind’s financial statements omitted a $58 million impairment charge associated with its deteriorated customer relationships for Brad Foote’s two most important customers. Additionally, in the course of its 2009 year end au dit of Broadwind, Grant Thornton knew or should have known that Broadwind’s impairment charge was not a fourth -quarter event and that Broadwind’s third and fourth quarter financial statements materially overstated revenue. Failure to Properly Plan the Aud it (AU § § 311 and 312 , AS 8, AS 9) 146. PCAOB standards require an auditor to consider the nature, extent and timing of work to be performed in planning the audit and prepare a written audit program which sets forth in reasonable detail the audit procedures necessary to accomplish the audit objectives. (AU § 311.05). Auditors must also consider audit risk and materiality in planning the audit and designing audit procedures. (AU § 312.12). Auditors additionally must plan the audit so that audit risk will be limited to a low level appropriate for expressing an opinion on the finan cial statements. (AU § became effective for audits of fiscal years beginning on or after December 15, 2010 (i.e. , the 2011 aud it of ALC) and superseded AU §§ 311, 312, 326, among others. 32 312.13) . Auditors are also required to consider significant risk of material misstatement of the financial statements in: (1) determining the nature, timing or extent of procedures; (2) assigning staff; and (3) requiring appropriat e levels of supervision. ( AU § 312.17). In planning an audit, auditors must also design procedures to obtain reasonable assurance of detecting mis statements that the auditor believes could be material. (AU § 312.25 , AS 8. 3). 147. PCAOB standards also require auditors to properly plan their audits and develop and document an audit plan that includes a description of, among other things: (1) the planned nature, timing and extent of substantive procedures; and (2) other planned audit procedures required to be performed so that the engagement compli es with PCAOB standards. (AS 9.4, AS 9.10). 148. As a result of Grant Thornton’s conduct described above, Grant Thornton failed to properly plan its 2009 , 2010 and 2011 audits of ALC . Failure to exercise due profe ssional care and professional skepticism (AU §§ 230 , 316 , 722 , and AS 13 ) 149. PCAOB standards require auditors to exercise due professional care in the planning and performance of the audit and the preparation of the report. (AU § 230.01) . Auditors must maintain an attitude of professional skepticism, which includes “a questioning mind and a critical assessment of audit evidence.” (AU § 230.07 , AS 13. 7). In addition, the auditor should “consider the competency and sufficiency of the evid ence. Since evidence is gathered and evaluated throughout the audit, professional skepticism should be exercised throughout the audit process.” (AU § 230.08) . In exercising profess ional skepticism, an auditor should not be satisfied with less than persu asive evidence because of a belief that management is honest. (AU §§ 230.09 and 316.13). Further, auditors should : (1) perform an ongoing questioning of whether the information and evidence obtained suggests that a material misstatement due to fraud has occurred; and (2) conduct the engagement with a mindset that recognizes that a material misstatement due to fraud could be present, regardless of past experience with the entity and the auditors’ belief about management’s honesty and integrity. (AU § 316 .13) . Auditors should also exercise due professional care and professional skepticism in the course of reviews of interim financial information. (AU § 722.01 ). 150. As a result of Grant Thornton ’s conduct described above , Grant Thornton failed to exercise due professional care and professional skepticism in its 2009, 2010 and 2011 audits of ALC and its 2009 third quarter review and year end audit of Broadwind . Failure to obtain sufficient evidence (AU §§ 326 and 333, AS 13, 14 and 15 ) 151. PCAOB s tandards require d auditors to obtain sufficient competent evidential matter (for the 2009 Broadwind audit and the 2009 and 2010 ALC audits) and sufficient appropriate audit evidence (for the 2011 ALC audit) to afford a reasonable basis for an opinion with respect to the financial statements under audit . (AU § 326 .22 , AS 15. 4). Auditors must be thorough in their search for evidential matter and unbiased in its evaluation and consider relevant evidential matter 33 regardless of whether it corroborates or contradicts assertions in the financial statements. (AU § 326.25 , AS 15.2 and 15.29 ). 152. PCAOB s tandards also provide that management representations “are not a substitute for the application of th[e] auditing procedures necessary to afford a reasonable basis for an opinion regarding the financial statements under audit,” that “the auditor obtains written representations from management to complement other auditing procedures,” and that “[i]n exercising professional skepticism, the auditor should not be satisfied with less than persuasive evidence because of a belie f that management is honest. ” (AU §§ 333.02, 333.03 , 230.09) . Auditors must also: (1) obtain corroboration for management’s explanation regarding significant unusual or unexpected transactions, events, amounts or relationships; and (2) perform procedures if management’s responses to the auditor’s inquiries appear to be implausible, inconsistent with other audit evidence, imprecise or not at a sufficient level o f detail to be useful. (AS 14.8 ). 153. Auditors should also design and perform audit procedures in a manner that addresses the assessed risks of material misstatement for each relevant assertion of each significant account and disclosure. (AS 13. 8). In designing such audit proce dures, auditors should obtain more persuasive audit evidence the higher the audito r’s assessment of risk. (AS 13. 9). Auditors should also perform substantive procedures, including tests of details, for significant risks, and the evidence auditors obtain from substantive procedures should increase as the assessed risk of material misstatement increases. (AS 13.11, AS 13.3 7). 154. As a result of Grant Thornton ’s conduct described above , Grant Thornton failed to obtain sufficient evidence supporting assertion s in ALC’s 2009, 2010 and 2011 Form 10 -K financial statements that ALC was in compliance with the Ventas lease covenants . In the course of its 2009 year end audit of Broadwind , Grant Thornton also failed to obtain sufficient evidence supporting the conclusion that the impairment was a fourth -quarter event and that the Customer 2 bill and hold transaction met the applicable revenue recognition criteria. Failure to Properly Supervise the Engagement Team (AU § 311 and AS 10 ) 155. PCAOB standards note that audit “assistants,” including firm personnel other than the auditor with final responsibi lity for the audit, are to be “properly supervised.” (AU §§ 311.01 and 311.02). Those standards further require that assistants be informed of their responsibilities and the objectives of procedures assigned to them, and that the work of assistants be reviewed to determine whether it was adequate ly performed. (AU §§ 311.12, 311.13 , and AS 10.5 ). 156. As a result of Grant Thornton’s conduct described above, Grant Thornt on failed to properly supervise the engagement team on its 2009 , 2010 and 2011 ALC engagements . Failure to Make Additional Inquires or Perform Additional Procedures in the Course of Reviewing Interim Financial Information (AU § 722) 157. PCAOB standards provide: 34 If, in performing a review of interim financial information, the accountant becomes aware of information that leads him or her to believe that the interim financial information may not be in conformity with generally accepted accounting principles in all material respects, the accountant should make additional inquiries or perform other procedures that the accountant considers appropriate to provide a basis for communicating whether he or she is aware of any material modifications that sh ould be made to the interim financial information. (AU § 722.22). 158. As a result of Grant Thornton’s conduct described above, Grant Thornton violated AU § 722.22 when it failed to make appropriate additional inquires or perform other procedures in the course of its third quarter 2009 review of Broadwind’s impairment assessment. Failure to Prepare Required Documentation (AS 3) 159. PCAOB standards mandate that an audit or’s documentation contain sufficient information to enable an experienced auditor, having no pre vious connection to the engagement to: (1) understand the nature, timing, extent and results of the procedures performed, evidence obtained and conclusions reached; and (2) determine who performed the work and the date such work was completed as well as th e person who reviewed the work and the date of such review. (AS 3.1, 3. 6). Auditors are also required to document significant findings and issues, including the actions taken to address them and the basis for the conclusions reached. (AS 3.12). 160. As a re sult of Grant Thornton’s conduct described above, Grant Thornton failed to obtain required audit documentation on its 20 09, 2010 and 2011 audits of ALC and it s 2009 year end audit of Broadwind. Failure to Perform Adequate Personnel Management (QC 20 and 40) 161. PCAOB Quality Control Standards require an auditing firm to establish policies and procedures which provide the firm with reasonable assurance that (a) those hired possess the appropriate characteristics to enable them to perform competently and (b) w ork is assigned to personnel having the degree of technical training and proficiency required in the circumstances.

(QC 20.13 and QC 40.2) . 162. As a result of Grant Thornton’s conduct described above, Grant Thornton violated QC 20.13 and QC 40.2 in staffing the ALC engagements. Finding 163. As a result of the conduct described above, the Commission finds that Grant Thornton engaged in improper professional conduct within the meaning of Section s 4C(a)(2) and 4C(b)( 2) of the Exchange Act and Rule s 102(e)(1)(ii) and 102(e)(1)(iv)(B) of the Commission’s Rules of Practice. Grant Thornton’s conduct in the 2009, 2010, and 2011 audits of ALC, and its 35 2009 third quarter review and year end audit of Broadwind involved repeated instances of unreasonable conduct, each res ulting in violations of PCAOB standards and indicating a lack of competence, and also satisf ies the standard of highly unreasonable conduct resulting in violations of PCAOB standards in circumstances in which heightened scrutiny was warranted. GRANT THOR NTON WAS A CAUSE OF VIOLATION S OF SECTION 13(a) OF THE EXCHANGE ACT AND RULE S 13a -1 AND 13a -13 THEREUNDER 164. Section 13(a) of the Exchange Act and Rule s 13a -1 and 13a -13 thereunder require that every issuer of a security registered pursuant to Section 12 of the Exchange Act file with the Commission annual and quarterly reports (i.e., Form s 10 -K and 10 -Q) as the Commis sion may require . The obligation to file such reports e mbodies the requirement that they be true and correct. 165. ALC’s annual reports on Form 10 -K for fiscal years 2009, 2010 and 2011 included audit reports from Grant Thornton that stated its audits of ALC’s financial statements were conducted “in accordance wit h the standards of the Public Company Accounting Oversight Board ” and that ALC ’s financial statements presented fairly, in all material respects, the company’s position and results. Broadwind’s 2009 Form 10 -K contained a similar representation by Grant Thornton. The se statement s were materially misleading. As a result of Grant Thornton ’s above -described conduct, Grant Thornton’s 2009, 2010 and 2011 audits of ALC and 2009 audit of Br oadwind were not conducted in accordance with PCAOB standards and t he financia l statements included in ALC’s 2009, 2010 and 2011 Form s 10 -K were materially misstated because, among other things, they incorrectly represented that ALC was in compliance with the Ventas lease financial covenants. As for Broadwind, its Form 10 -Q for the third quarter of 2009 materially overstated its intangible assets and understated a material impairment charge, and its 2009 10 -K misstated revenue an d incorrectly described its $58 million impairment charge as a fourth quarter event. At a minimum, Grant Thornton knew or should have known that its unreasonable conduct would contribute to ALC ’s filing of inaccurate 2009, 2010 and 2011 Form s 10 -K and Broadwind’s filing of an inacc urate third quarter 2009 Form 10 -Q and 2009 Form 10 -K. 166. As a result of the conduct described abov e, the Commission finds that Grant Thornton was a cause of ALC ’s violations of Section 13(a) of the Exchange Act and Rule 13a -1 thereunder and Broadwind’s violations of Section 13(a) of the Exchange Act and Rule s 13a -1 and 13a -13 thereunder. H. UNDERTAKINGS 167. Grant Thornton’s Review. Within 120 days after the entry of this Order, Grant Thornton shall perform and complete a review and evaluation (“Grant Thornton’s Review”) of the sufficiency and adequacy of Grant Thornton’s quality controls, including its policies and procedures for a udits and interim reviews regarding the following (hereinafter referred to as “Grant Thornton’s Policies”): 36 a) the exercise of due professional care and professional skepticism (as set forth in AU 230); b) auditing estimates (as set forth in AU 342 and AS 14) , including, but not limited to: (i) considering the relevance, reliability, and sufficiency of the factors and data used in forming the assumptions underlying estimates; and (ii) evaluating the results of procedures performed, including whether the evidence obtai ned supports or contradicts the estimates included in the financial statements; c) appropriately responding to the risk of material misstatement related to potential impairment, including through the performance of procedures to resolve inconsistencies in the evidence obtained (as set forth in AS 13 and AS 15); d) assessing the risk of fraud, including the risk of fraud involving management override of controls, and appropriately responding to identified fraud risks (as set forth in AU 316, AS 12, AS 13 and AS 1 4); e) evaluation of evidence obtained concerning matters that are the subject of written representations from management in order to consider the reliability of the representation made and whether reliance on management’s representations is appropriate and justified (as set forth in AU 333) ; f) obtaining sufficient appropriate audit evidence (as set forth in AS 15); g) effective supervision by engagement partners and managers, including review of work performed by the engagement team (as set forth in AS 10); h) consultations with local, regional or national office technical oversight professionals ; i) appropriate resol ution of concerns or disagreements in views among an engagement team regarding conclusions (i.e., accounting and auditing) reached in connection with an audit (as set forth in AS 10); j) audit documentation , including risk assessment procedures and responses , significant findings and resulting actions, work paper sign -off , archiving, and dating (as set forth in AS 3); k) compliance with the requirements for engagement quality reviewers (as set forth in AS 7); l) obtaining reasonable assurance (as set forth in QC 2 0 and QC 40) that 37 i) the engagement partner and other individuals assisting the engagement partner in supervising the engagement possess the competencies that are necessary and appropriate in the individual circumstances , and ii) work is assigned to personnel having the degree of technical training and proficienc y required in the circumstances ; m) the identification, monitoring and remediation of audit partners with negative quality indicators, including the development and implementation of corrective actions ( as set forth in QC 30); and n) reporting, evaluation, and compensation of professional practice personnel (as set forth in QC 20 and QC 40). Grant Thornton’s Review shall assess the forgoing areas to determine whether Grant Thornton ’s Policies are adequate an d sufficient to provide reasonable assurance of compliance with all relevant Commission regulations and PCAOB standards and rules. 168. Grant Thornton Report. Within 60 days of completing the Grant Thornton Review , Grant Thornton shall deliver to the Commission staff a detailed written report (“Grant Thornton Report”) summarizing its review and changes to Grant Thornton’s Policies, if any, to provide reasonable assurance of compliance with all relevant Commission regulations and PCAOB standards and rul es. The Grant Thornton Report shall identify the undertaking, provide written evidence of compliance in the form of a narrative, and be supported by exhibits sufficient to demonstrate compliance. The Commission staff may make reasonable requests for furt her evidence of compliance, and Grant Thornton agrees to provide such evidence. 169. Independent Consultant’s Review. Grant Thornton has undertaken to retain, within 180 days after the entry of this Order, an independent consultant (“Independent Consultant”), not un acceptable to the Commission staff. Grant Thornton shall provide to the Commission staff a copy of the engagement letter detailing the scope of the Independent Consultant’s responsibilities. The Independent Consultant’s compensation and expenses sh all be borne exclusively by Grant Thornton . Grant Thornton shall deliver to the Independent Consultant the Grant Thornton Report at the same time as Grant Thornton provides such report to the Commission s taff as specified in paragraph 168 above. Grant Th ornton shall require that the Independent Consultant perform a review ( the “IC Review”) of Grant Thornton ’s Policies to determine whether Grant Thornton’s Policies are adequate and sufficient to provide reasonable assurance of compliance with all relevant Commission regulations and PCAOB standards and rules. Grant Thornton shall cooperate fully with the Independent Consultant and shall provide reasonable access to firm personnel, information, and records as the Independent Consultant may reasonably request for the IC Review (including training materials pertaining to the undertaking in paragraph 173), subject to Grant Thornton’s right to withhold from disclosure any information or records protected by any applicable protection or privilege such as the attor ney -client privilege or the attorney work product doctrine. 38 170. Independent Consultant’s Report. After the IC Review is completed, but no later than ninety days after receiving the Grant Thornton Report, the Independent Consultant shall issue a detailed wri tten report ( the “IC Report”) to Grant Thornton : (a) summarizing the IC Review; and (b) making recommendations, where appropriate, reasonably designed to ensure that Grant Thornton’s Policies are adequate and sufficient to provide reasonable assurance of compliance with all relevant Commission regulations and PCAOB standards and rules. Grant Thornton shall require the Independent Consultant to provide a copy of the IC Report to the Commi ssion staff and the PCAOB staff when the IC Report is issued. 171. Grant Thornton shall adopt, as soon as practicable, all recommendations of the Independent Consultant in the IC Report. Provided, however, that within thirty days of issuance of the IC Report, Grant Thornton may advise the Independent Consultant in writing of a ny recommendation that it considers to be unnecessary, outside the scope of this Order, undu ly burdensome, or impractical. Grant Thornton need not adopt any such recommendation at that time, but instead may propose in writing to the Independent Cons ultant and the Commission s taff an alternative policy or procedure designed to achieve the same objective or purpose. Grant Thornton and the Independent Consultant shall engage in good -faith negotiations in an effort to reach agreement on any recommendations ob jected to by Grant Thornton . In the event that the Independent Consultant and Grant Thornton are unable to agree on an alternative proposal within sixty days, Grant Thornton shall abide by the determinations of the Independent Consultant . 172. Certification by Grant Thornton ’s CEO . Within sixty days of issuance of the IC Report, but not sooner than thirty days after a copy of the IC Report is provided to the Commission staff, Grant Thornton’s chief executive officer (“CEO”) must certify to the Commission sta ff in writing that (i) Grant Thornton has adopted and has implemented or will implement all recommendations of the Independent Consultant , if any; and (ii) the Independent Consultant agrees with Grant Thornton’s adoption and implementation of the recommend ations . To the extent that Grant Thornton has not implemented all recommendations of the Independent Consultant within sixty days of issuance of the IC Report, Grant Thornton’s CEO must certify to the Commission staff in writing, thirty days after their i mplementation, that (i) Grant Thornton has adopted and has implemented all recommendations of the Independent Consultant; and (ii) the Independent Consultant agrees that the recommendations have been adequately adopted and implemented by Grant Thornton. The certifications by Grant Thornton’s CEO shall identify the undertakings, provide written evidence of compliance in the form of a narrative, and be supported by exhibits sufficient to demonstrate compliance. The Commission staff may make reasonable reque sts for further evidence of compliance, and Grant Thornton agrees to provide such evidence. 173. Training . Prior to November 30, 2016, Grant Thornton shall require each audit professional serving public company audits to complete successfully a minimum of 32 h ours of audit -related training. The audit -related training shall cover the topics specified in Paragraph 1 67 . However, the following topics should be accorded the training hours noted below: a) At least 8 hours shall be devoted to the importance of exercising due care and professional skepticism in evaluating audit evidence; 39 b) At least 4 hours shall be devoted to procedures and techniques used in auditing estimates; c) At least 4 hours shall be devoted to the auditor’s responsibilities with respect to reviews of interim financial information, specifically focusing on circumstances where extended procedures are appropriate (a s discussed in AU § 722.22); d) At least 8 hours shall be devoted to fraud -detection training. The training shall include techniques i n detecting and responding to possible fraud in the course of public company audits by audit clients or by employees, officers or directors of audit clients. Particular attention should be focused on the auditor’s consideration of the impact of audit find ings on the assessment of fraud risks as discussed in AS 14 (Par. 28 -29), as well as the documentation requirements outlined in AU § 316.83. Such training will also include the auditor’s responsibilities under Section 10A of the Exchange Act; and e) At least 4 hours shall be devoted to appropriate reliance on management representations. 174. To ensure the independence of the Independent Consultant, Grant Thornton : (1) shall not have the authority to terminate the Independent Consultant or substitute another indepe ndent compliance consultant for the initial Independent Consultant, without the prior written approval of the Commission staff; and (2) shall compensate the Independent Consultant and persons engaged to assis t the Independent Consultant for services render ed pursuant to this Order at their reasonable and customary rates. 175. Grant Thornton shall require the Independent Consultant to enter into an agreement that provides that, for the period of engagement and for a period of two years from completion of the enga gement, the Independent Consultant shall not enter into any employment, consultant, attorney -client, auditing or other pr ofessional relationship with Grant Thornton , or any of its present or former affiliates, directors, officers, employees, or agents acti ng in their capacity as such. The agreement shall also provide that the Independent Consultant will require that any firm with which he/she is affiliated or of which he/she is a member, and any person engaged to assist the Independent Consultant in perfor mance of his/her duties under this Order shall not, without prior written consent of the Division of Enforcement, enter into any employment, consultant, attorney -client, auditing or other professional relationship with Grant Thornton , or any of its present or former affiliates, directors, officers, employees, or agents acting in their capacity as such for the period of the engagement and for a period of two years after the engagement. 176. Grant Thornton shall not be in, and shall not have an attorney -client rel ationship with the Independent Consultant and shall not seek to invoke the attorney -client privilege or any other doctrine or privilege to prevent the Independent Consultant from transmitting any information, reports, or documents to Commission staff . 177. Grant Thornton shall inform its audit professionals of the terms of the Order within ten business days after entry of the Order. 40 178. By December 31, 2016, Grant Thornton’s CEO shall certify, in writing, compliance with the undertakings set forth in paragraphs 173 and 177 . The certification shall identify the undertakings, provide written evidence of compliance in the form of a narrative, and be supported by exhibits sufficient to demonstrate compliance. The Commission staff may make reasonable requests for fu rther evidence of compliance, and Grant Thornton agrees to provide such evidence. 179. Annual Certifications . With respect to each of the calendar year periods 2017 and 2018, Grant Thornton ’s National Managing Partner, Audit Risk Management (“Managing Partner ”) shall certif y that Grant Thornton has assessed whether Grant Thornton’ s Policies are adequate and sufficient to provide reasonable assurance of compliance with all relevant Commission regulations and PCAOB standards and rules by, among other things, tes ting the firm’s implementation of Grant Thornton ’s Policies during the twelve (12) months preceding the certification (“Annual Certification”). The Annual Certification shall describe the nature and scope of Grant Thornton ’s testing. The Annual Certifica tion shall represent that the Managing Partner has reviewed and evaluated the firm’s assessment and testing process and that, based on belief and after reasonable in quiry, the Managing Partner believes that Grant Thornton ’s Policies are adequate and suffic ient to provide reasonable assurance of compliance with all relevant Commission regulations and PCAOB standards and rules. If t he Managing Partner cannot represent that Grant Thornton ’s Policies are adequ ate and sufficient, then the Managing Partner shall describe in reasonable detail the reasons for the inability to so certify. The Managing Partner shall provide the Annual Certifications to the Commission’s staff within sixty days of the end of the annual period. Grant Thornton shall preserve and retain all documentation regarding the Managing Partner’ s Annual Certification for seven (7) years and will make it available to the staffs of the Commission or the PCAOB upon request. 180. All reports and certifications mentioned in these undertakings shall be submi tted to Robert Burson, Associate Regional Director, Division of Enforcement, Securities and Exchange Commission, 175 W. Jackson Blvd., Suite 900, Chicago, IL 60604, with a copy to the Office of Chief Counsel of the Enforcement Division. 181. For good cause show n, the Commission staff may extend any of the procedural dates relating to the undertakings. Deadlines for procedural dates shall be counted in calendar days, except that if the last day falls on a weekend or federal holiday, the next business day shall b e considered to be the last day. 182. In determining whether to accept Grant Thornton’s Offer, the Commission has considered these undertakings. Grant Thornton agrees that if the Division of Enforcement believes that Grant Thornton has not satisfied these unde rtakings, it may petition the Commission to reopen the matter to determine whether additional sanctions are appropriate . 41 IV. In view of the foregoing, the Commission deems it appropriate and in the public interest to impose the sanctions agreed to in the Respondent’s Offer. Accordingly, it is hereby ORDERED that: A. Pursuant to Section 21C of the Exchange Act, Grant Thornton shall cease and desist from committing or causing any violations and any future violations of Se ction 13(a) of the Exchange Act and Rule s 13a -1and 13a -13 thereunder . B. Grant Thornton is censured. C. Grant Thornton shall comply with the undertakings enumerated in Section H above. D. Grant Thornton shall within 14 days of the entry of this Order, pay disgorgement of $1,305,396 , which represents profits gained as a result of the conduct described herein and prejudgment interest of $ 231,174.19 , to the Commission for transfer to the general fund of the United States Treasury, subject to Exchange Act Section 21F(g)(3) . If timely p ayment is not made, additional interest shall accrue pursuant to SEC Rule of Practice 600. E. Grant Thornton shall pay a civil penalty of $3 million to the Commission within 14 days of the entry of this Order for transfer to the general fund of the United States Treasury, subject to Exchange Act Section 21F(g)(3) . If timely payment is not made additional interest shall accrue pursuant to 31 U.S.C. 3717 . F. Payments ordered in paragraphs D and E above must be made in one of the following ways: (1) Respondent may transmit payment electronically to the Commission, which will provide detailed ACH transfer/Fedwire instructions upon request; (2) Respondent may make direct payment from a bank account via Pay.gov through the SEC website at http://www.sec.gov/about/offices/ofm.htm ; or (3) Respondent may pay by certified check, bank cashier’s check, or United States postal money order, made payable to the Se curities and Exchange Commission and hand -delivered or mailed to: Enterprise Services Center Accounts Receivable Branch HQ Bldg., Room 181, AMZ -341 6500 South MacArthur Boulevard Oklahoma City, OK 73169 42 Payments by check or money order must be accompanied by a cover letter identifying Grant Thornton as a Respondent in these proceedings, and the file number of these proceedings; a copy of the cover letter and check or money order must be sent to Robert Bur son , Chicago Regional Office, Securities and Exchange Commission, 175 West Jackson , Suite 900, Chicago, IL 60604 . By the Commission. Brent J. Fields Secretary