The Granston memo: two years later

In January 2018, we wrote an article in the Daily Journal about a recently released internal Department of Justice memo. The memo -- referred to as the Granston memo for its author Michael Granston, the Director of the DOJ Civil Fraud Section -- encouraged government attorneys to be more aggressive in dismissing False Claims Act cases that do not serve government interests.

The Granston memo marked a policy change regarding the DOJ's handling of FCA investigations and litigation. Until the Granston memo was released, the DOJ had never provided any public guidance about its approach to dismissing whistleblower claims. In practice, the government would perform an investigation and then decide whether to intervene -- a critical decision with significant consequences for the expected value of a case. But when the government declined to intervene, it rarely moved to dismiss. The Granston memo described an apparent shift from this traditional, more hands-off approach. The Granston memo identified seven "non-exhaustive" grounds that may justify the government exercising its right to dismiss a qui tam action filed on its behalf under the FCA, 31 U.S. Section 3730. These grounds include: (1) curbing facially meritless actions; (2) preventing "parasitic or opportunistic" actions that add nothing to ongoing government investigations; (3) preventing interference with government policies and programs; (4) protecting the DOJ's litigation prerogatives; (5) protecting classified information and national security; (6) preserving government resources; and (7) addressing egregious procedural errors.

Shortly after the memo's release, we wrote that the Granston memo signaled a welcome policy change, especially for qui tam defendants. We predicted that the memo would lead to more DOJ dismissals, place more weight on the government pre-intervention investigation, and change the FCA landscape in other ways. Two years later, how have our assessments and predictions fared? Pretty well. It is safe to say that the current presidential administration has not grown any fonder of or kinder to whistleblowers.

If anyone doubted that the Granston memo reflected a shift in official DOJ policy, that doubt was put to rest in February 2019 when the DOJ incorporated the policy into its newly updated Justice Manual (formerly called the United States Attorneys' Manual). The Justice Manual tracks language from the Granston memo nearly verbatim, reciting each of the seven factors that may justify dismissal. See section 4-4.111. The Justice Manual also embraces the core policy of the memo, observing that "[w]hile it is important to be judicious in in utilizing [dismissals under] § 3730(c)(2)(A), such dismissals also provide an important tool to advance the government's interests, preserve limited resources, and avoid adverse precedent." In other words, the DOJ not only supports the Granston memo, but requires that assistant U.S. attorneys put its policy into practice.

As predicted, the government has accelerated its dismissals of FCA cases since the Granston memo was released. The DOJ's annual report on FCA recoveries for the fiscal year 2019, released last month, acknowledges that the government "made increasing use" of its "authority ... to dismiss [qui tam] cases that do not advance the goal of fraud prevention." The DOJ's report also invokes the language of the Granston memo, claiming that such dismissals "help prioritize and protect the expenditure of government resources." DOJ included similar language in its 2018 report, claiming that it "made increasing use" of its dismissal authority in 2018, and that such dismissals "help prioritize the use of government resources."

Deputy Associate Attorney General Stephen Cox said in a January 2019 speech that the DOJ used to invoke dismissal "sparingly," dismissing just "a few cases" in a given year. Since the memo, he explained, the DOJ has dismissed dozens of cases.

At the same time, the DOJ has signaled some limits to its newly aggressive policy. Cox noted in his speech that the government's "exercise of [dismissal] authority will remain judicious." The eponymous Michael Granston has also tamped down expectations of large-scale policy changes. In a March 2019 keynote speech at the Federal Bar Association's FCA Conference, Granston cautioned that "dismissal will remain the exception rather than the rule." He also warned that the government could not be baited into dismissing FCA cases, such as by defendants who take excessive discovery to inflate the government's litigation cost and burden.

Despite DOJ's warning, the Granston memo policy creates opportunities for defense lawyers. Long before the Granston memo was issued, circuits split on the scope of the government's dismissal authority under Section 3730(c)(2)(A). The U.S. Court of Appeals for the D.C. Circuit holds that the DOJ's right to dismiss is "unfettered." Swift v. United States, 318 F.3d 250, 252, 253 (D.C. Cir. 2003). The 9th Circuit holds that the government must show (1) a valid government purpose for dismissal and (2) "a rational relation between dismissal and accomplishment of the purpose." United States ex rel. Sequoia Orange Co. v. Baird-Neece Packing Corp., 151 F.3d 1139 (9th Cir. 1998). If the government meets both prongs, the burden shifts to the relator to show that dismissal is "fraudulent, arbitrary and capricious, or illegal." Id. Although this circuit split raises interesting questions about the appropriate level of deference, the split has for years been mostly academic, with the DOJ prevailing under either standard. Swift's "unfettered right" test is toothless by design, and Sequoia Orange's rational basis test is toothless in effect. We are not aware of any circuit court ever sustaining a relator's objection to a government dismissal.

But perhaps this tradition of toothless review tracked the government's tradition of using dismissal "sparingly." Could the government's new, aggressive policy lead courts to scrutinize Section 3730(c)(2)(A) dismissals more carefully, making this circuit split meaningful?

Two recent district court decisions support that view. In United States v. Academy Mortgage Corp., the Northern District of California rejected the government's dismissal request under the Sequoia Orange standard "because the complaint ha[d] not been fully investigated," and the amended complaint had not been investigated at all. 16-CV-02120-EMC, 2018 WL 3208157, at *3 (N.D. Cal. June 29, 2018). In a subsequent order, the court explained that its job was not to "simply rubber-stamp" the government's basis for dismissal. United States v. Acad. Mortg. Corp., 16-CV-02120-EMC, 2018 WL 4794231, at *3 (N.D. Cal. Oct. 3, 2018). In United States ex rel. CIMZNHCA, LLC v. UCB, INC., likewise, the Southern District of Illinois denied a motion to dismiss under the Sequoia Orange standard where the government had failed to perform a "minimally adequate investigation," its purported basis for dismissal may have been "pretextual," and its "true motivation [wa]s animus toward the relator." 17-CV-765-SMY-MAB, 2019 WL 1598109, at *3-4 (S.D. Ill. Apr. 15, 2019).

Even if Academy Mortgage and CIMZNHCA can withstand appellate review, they appear to be the exceptions that prove the rule. These cases place significant stock in the government's pre-intervention investigation, a change that we argued was long overdue. Had the government performed adequate investigations in those cases -- and in the case of CIMZNHCA, not disparaged the "professional relator" who brought the action -- their dismissal requests would almost surely have been granted.

In fact, the Granston memo makes it even easier for the government to justify dismissal under any appellate standard. So long as the government performs a facially adequate investigation, the DOJ can identify a valid government purpose for dismissal merely by citing to one or more of the seven factors set forth in the memo -- and now in the Justice Manual. As we noted in our January 2018 article, most of these seven factors do not concern the merits of the underlying claim and could favor dismissal even of a meritorious claim.

For these reasons, we encouraged defendants to think creatively about how the non-substantive factors may apply to their case. A review of cases from the past two years shows that this approach has been effective. The government always recites at least one of the Granston memo factors as a basis for dismissal, and many dismissals have little or nothing to do with the substantive merit. The most evergreen justification is factor six, the government's interest in preserving resources. Every case consumes resources. Yet with the notable exceptions of Academy Mortgage and CIMZNHCA, the government can generally obtain a dismissal on this factor alone. See United States v. Bayer HealthCare Pharm., Inc., CV 14-031 WES, 2019 WL 5310209, at *2 (D.R.I. Oct. 21, 2019) ("the Government has shown at least one "valid government purpose" for dismissing this action -- the burden this continuing litigation would place on the Government's resources"). In Gilead Sciences, Inc. v. U.S. ex rel. Campie, the government dismissed a qui tam action in part because "burdensome discovery" would "distract from the [FDA's] public-health responsibilities." United States v. Gilead Scis., Inc., 11-CV-00941-EMC, 2019 WL 5722618, at *3 (N.D. Cal. Nov. 5, 2019).

The Gilead Sciences dismissal came after the 9th Circuit had reversed an earlier dismissal and remanded to the district court. Id. at *2. In Polansky v. Exec. Health Res., Inc., 12-CV-4239, 2019 WL 5790061 (E.D. Pa. Nov. 5, 2019), dismissal was granted on the basis of preserving government resources, even though the government had previously declined to dismiss, and reversed its own position apparently only to avoid complying with certain unfavorable discovery rulings. Notably, the courts in Gilead Sciences and Polansky each cited directly to the Granston memo in recognizing the valid governmental purpose supporting dismissal. See Polansky at *5, n. 13.

Courts also routinely grant dismissal for other non-merits-based factors, such the government's interest in safeguarding classified information and national security interests. See United States ex rel. Johnson v. Raytheon Co., 395 F. Supp. 3d 791, 795 (N.D. Tex. 2019) (government request for dismissal granted where discovery would risk disclosure of classified information and consume government resources).

Finally, we predicted that the DOJ's shift to more aggressive dismissals might lead courts and advocates to interpret a government decision not to dismiss as a tacit endorsement of a qui tam action. It is probably too early to measure, but at least one recent circuit opinion -- though later vacated -- supports this prediction. In September 2019, the 3rd Circuit reversed an order granting a qui tam defendant's motion to dismiss, suggesting that the bulwark against frivolous qui tam lawsuits is not the judiciary, but the DOJ. United States ex rel. Bookwalter v. UPMC, 938 F.3d 397, 417 (3d Cir.), reh'g granted, judgment vacated, 944 F.3d 965 (3d Cir. 2019), and on reh'g, 946 F.3d 162 (3d Cir. 2019). The opinion pointed to the Granston memo in holding that "[f]ederal courts are not the first line of defense against abusive suits; the Justice Department is." Id. This no doubt overstates the aggressiveness of the DOJ's policy, but it signals a serious rethinking of government review of FCA cases. No pre-Granston memo court would have contended that the courts should not worry too much about abusive qui tam lawsuits, because the government will be ruthless in dismissing them. Times have changed. 