Barba v. Bonta CA4/1

• Court: California Court of Appeal
• Decided: November 17, 2023
• Docket: D081194
• Status: Unpublished

Opinion

Filed 11/17/23 Barba v. Bonta CA4/1 NOT TO BE PUBLISHED IN OFFICIAL REPORTS California Rules of Court, rule 8.1115(a), prohibits courts and parties from citing or relying on opinions not certified for publication or ordered published, except as specified by rule 8.1115(b). This opinion has not been certified for publication or ordered published for purposes of rule 8.1115.

COURT OF APPEAL, FOURTH APPELLATE DISTRICT

DIVISION ONE

STATE OF CALIFORNIA

ASHLEYMARIE BARBA et al., D081194

Plaintiffs and Respondents,

v. (Super. Ct. No. 37-2022- ROB BONTA, as Attorney General, 00003676-CU-CR-CTL) etc.,

Defendant and Appellant.

APPEAL from an order of the Superior Court of San Diego County, Katherine A. Bacal, Judge. Reversed and remanded. Rob Bonta, Attorney General, Thomas S. Patterson, Assistant Attorney General, R. Matthew Wise, Ryan R. Davis, Paul E. Stein and Sebastian Brady, Deputy Attorneys General, for Defendant and Appellant. Benbrook Law Group, Bradley A. Benbrook and Stephen M. Duvernay for Plaintiffs and Respondents. I. INTRODUCTION California state law has required the California Department of Justice (the DOJ) to maintain records of essentially all handgun transfers occurring in the state of California since at least the 1950’s, and has further mandated the inclusion of long gun transfers since 2014. (Pen. Code § 11106, originally

enacted by Stats. 1953, ch. 1385, § 1)1 In 2015, the California electorate passed a voter initiative that requires the DOJ to maintain similar records regarding any sale or transfer of ammunition. (Prop. 63, as approved by voters, Gen. Elec. (Nov. 8, 2016); § 30352.) This data provides a unique opportunity for research not available anywhere else. Since at least 1989, researchers at the University of California, Davis have been utilizing that data in research studies aimed at understanding and preventing various forms of firearm violence. In 2016, the Legislature directed the Regents of the University of California to establish a Firearm Violence Research Center (the Center) with the goals of overseeing interdisciplinary research addressing the nature and consequences of firearm violence, and working with policymakers “to identify, implement, and evaluate innovative firearm violence prevention policies and programs.” (Pen. Code §§ 14230, 14231.) To aid in those goals, the Legislature mandated that several state agencies, including the DOJ, provide the Center with “the data necessary for [it] to conduct its research.” (§ 14231, former subd. (c), current subd. (c)(2).) However, the DOJ began restricting the Center’s access to data in the following years. In response, the Legislature passed Assembly Bill No. 173 (2021–2022 Reg. Sess.) (Assem. Bill 173), which amended several statutory provisions to make clear the Legislature’s intent that the DOJ provide all necessary data, including personally identifying data, to the Center upon proper request. Plaintiffs filed suit, challenging the constitutionality of the amendments. Among other claims, they assert that the data sharing prescribed by the amendments violates their right to privacy under the

1 All further unspecified statutory references are to the Penal Code.

2 California Constitution. The Attorney General demurred, and Plaintiffs filed a motion for preliminary injunction. Following a combined hearing in October 2022, the trial court overruled the demurrer as to the privacy claim and issued a preliminary injunction enjoining the DOJ from transferring any additional personal identifying information from the firearm and ammunition databases to researchers until further notice. On appeal, the Attorney General asserts the trial court erred by conflating the legal standards for the co-pending demurrer and preliminary injunction motions and that, under the proper standard, it was an abuse of discretion to grant the injunction. We agree. Even assuming that Plaintiffs have met the threshold inquiries to establish a privacy claim, the Attorney General presented a legitimate countervailing interest and presented evidence explaining why Plaintiffs’ proposed alternatives are not adequate or sufficient. Having failed to rebut that evidence, Plaintiffs cannot establish a probability of success on the merits as a matter of law, and we therefore reverse the trial court’s order granting the preliminary injunction and remand the matter to the trial court with instructions to enter a new order denying the motion for a preliminary injunction. II. FACTUAL AND PROCEDURAL BACKGROUND A. California’s Long History of Retaining Firearm Transfer Data California has required dealers to retain records regarding firearm transfers since at least 1917. (See Stats. 1917, ch.145, § 7, pp. 222–223.) Under the original statute, “[e]very person in the business of selling, leasing, or otherwise transferring a pistol, revolver or other firearm, of a size capable of being concealed upon the person,” was to “keep a register in which [they] entered the time of sale, the date of sale, the name of the salesman making the sale, the place where sold, the make, model, manufacturer’s number,

3 caliber or other marks of identification on such pistol, revolver or other firearm.” (Ibid.) The law required the purchaser to sign the form and affix their address, and required the dealer to mail a duplicate of the form to the municipal police department or county clerk where the firearm was sold. (Ibid.) Failure to comply with those requirements or the use of a fictitious name or address on the form gave rise to a misdemeanor offense. (Ibid.) This same basic system for recording firearm transfers persists today but has now been expanded to include all firearms—not just handguns or those capable of being concealed upon the person—and to provide for digital transfer of the data to the DOJ. (See Assembly Bill 809, Stats. 2011, ch. 745, § 2.5, operative Jan. 1, 2014 [conforming the law so that transfers of handguns and firearms other than handguns (i.e., long guns) are treated the same]; § 28205 [addressing electronic transfers of firearm purchaser information to the DOJ].) With some limited exceptions, every sale or transfer of a firearm in California must now go through a licensed dealer, under the Dealer’s Record of Sale (DROS) process. (§§ 26500, 28050.) Under that process, an individual wishing to obtain a firearm must first fill out a DROS form, provided by the dealer. (§§ 28100, 28160.) The contents of the DROS form are statutorily mandated, and include information regarding the firearm, such as the make, model, and serial number; information regarding the dealer; and information about the purchaser, including their full name, date of birth, address, telephone number, occupation, gender, and a physical description. (§ 28160.) The form also contains mandatory questions regarding the purchaser’s criminal and mental health history and requires an imprint of the purchaser’s right thumbprint. (§ 28160, subds. (a)(29), (b).) The dealer must submit the information on the DROS form to the DOJ, which allows the DOJ to conduct

4 a background check to ensure the purchaser is not precluded from possessing a firearm. (§§ 28205, 28220.) If the DOJ approves the transaction, the dealer may deliver the firearm to the purchaser following a 10-day waiting period. (Cal. Code Regs., tit. 11, § 4230, subd. (a).) Beyond enabling background checks, the DROS data has long served as the basis for a firearm registration system maintained by the DOJ. As originally enacted in 1953, section 11106 required the DOJ to “properly file a complete record of all copies of fingerprints[ and] duplicate carbon copies of applications for licenses to carry concealed weapons and dealers’ records of sales of deadly weapons,” among other items, “to assist in the investigation of crime, the arrest and prosecution of criminals and the recovery of lost, stolen or found property.” (§ 11106, as originally enacted, Stats.1953, ch. 1385, § 1, p. 2966.) The DOJ has maintained these records in a repository known as the Automated Firearms System (the AFS) for all known handgun acquisitions since approximately 1980, and for all new legally acquired

firearms, including both handguns and long guns, since January 1, 2014.2 (Code Regs., tit. 11, §§ 4350, 4281, subds. (a), (d) [defining the “Automated Firearm System”]; Cal. Attorney General, “APPS Database” https://oag.ca.gov/ogvp/apps-database [last visited, Aug. 25, 2023].) Section 11106 has expanded over time, and now requires the Attorney General to maintain records of numerous other statutorily specified items. (§ 11106, subd. (a)(1)(A)–(I).) But the primary purpose has not changed. As the statute specifies, these records are retained “to assist in the investigation

2 The AFS “is populated by way of firearm purchases or transfers at a California licensed firearm dealer, registration of assault weapons, an individual’s report of firearm ownership to the Department, Carry Concealed Weapons Permit records, or records entered by law enforcement agencies.” (Cal. Code Regs., tit. 11, § 4350.)

5 of crime, the prosecution of civil actions by city attorneys pursuant to paragraph (3) of subdivision (b), the arrest and prosecution of criminals, and

the recovery of lost, stolen, or found property.”3 (§ 11106, subd. (a)(1).) Consistent with that goal, section 11106, subdivisions (a)(2) and (b)(3) require the Attorney General to furnish the information, upon proper application, “to the officers referred to in Section 11105,” which similarly requires the DOJ to maintain and provide “state summary criminal history information” (§ 11105). Section 11105, subdivision (b), in turn, delineates numerous circumstances under which the Attorney General must provide information to courts, peace officers, city attorneys, and other public agencies. B. California’s More Recent Retention of Ammunition Transfer Data More recently, the Legislature enacted similar laws requiring vendors to keep records of firearm ammunition transfers. (See § 12061, enacted by Stats. 2009, ch. 628, § 2 (Assem. Bill 962); renumbered without substantive change to § 30352, et seq. by Stats. 2010, ch. 711, § 6 (Sen. Bill 1080).) Like the DROS system, the ammunition transfer data includes information regarding the type of ammunition and the purchaser’s name, address, date of birth, state issued identification number, and an image of the purchaser’s right thumbprint. (Former § 30352, subd. (a)(3)(A)–(H), eff. Jan. 1, 2012 to Nov. 8, 2016.) Vendors must maintain these records for at least five years and make them available to any peace officer who is “conducting an investigation where access to those records is or may be relevant, is seeking

3 Section 11106, subdivision (b)(3) provides further: “Information in the registry referred to in this subdivision shall, upon proper application therefor, be furnished to the officers referred to in Section 11105, to a city attorney prosecuting a civil action, solely for use in prosecuting that civil action and not for any other purpose, or to the person listed in the registry as the owner or person who is listed as being loaned the particular firearm.”

6 information about persons prohibited from owning a firearm or ammunition, or is engaged in ensuring compliance with the Dangerous Weapons Control Law, as defined in Section 23500, or any other laws pertaining to firearms or ammunition.” (§ 30357, subd. (a).) In 2016, the California electorate voted to approve Proposition 63, commonly known as “The Safety for All Act of 2016” (Prop. 63). The proposition amended various provisions of the Penal Code to increase oversight of ammunition sales in the state and to enhance the record keeping associated with ammunition transfers. (Prop. 63, supra.) In doing so, the people made the following findings and declarations: “1. Gun violence destroys lives, families and communities. From 2002 to 2013, California lost 38,576 individuals to gun violence. . . .

“2. In 2013, guns were used to kill 2,900 Californians, including 251 children and teens. That year, at least 6,035 others were hospitalized or treated in emergency rooms for non-fatal gunshot wounds, including 1,275 children and teens.

[¶] . . . [¶]

“4. This tragic violence imposes significant economic burdens on our society. Researchers conservatively estimate that gun violence costs the economy at least $229 billion every year, or more than $700 per American per year. In 2013 alone, California gun deaths and injuries imposed $83 million in medical costs and $4.24 billion in lost productivity.

(Ibid.)

Among other provisions, Proposition 63 amended section 30352 to require the DOJ to create and maintain a database like the AFS to track and record ammunition sales. Under the amended statute, vendors submit the ammunition transfer data they were previously required to obtain to the DOJ and the DOJ retains the information in a database called the Ammunition

7 Purchase Records File (APRF). (Prop. 63, supra; § 30352, former subd. (b), current subd. (b)(1).) The amended section 30352 specifies that the APRF data “shall remain confidential and may be used by the department and those entities specified in, and pursuant to, subdivision (b) or (c) of Section 11105, through the California Law Enforcement Telecommunications System [CLETS], only for law enforcement purposes.” (§ 30352, former subd. (b), current subd. (b)(1).) C. California’s History of Utilizing DROS Data in Research Aimed at Informing Policies to Reduce Firearm Violence

The DROS system and the associated AFS and APRF databases create a unique data set regarding gun and ammunition ownership not available anywhere else. Researchers in California have used this data to conduct

empirical research4 regarding firearm-related violence for some time. The DOJ first provided DROS records to researchers at the University of California, Davis (UC Davis), in approximately 1989. The researchers used the data to conduct “research on risk factors for criminal activity among legal purchasers of firearms that was supported by the Centers for Disease Control and Prevention” (the CDC). At the time, the CDC viewed gun violence as a public health issue and was funding studies aimed at reducing injuries and deaths resulting from such violence. (See Rostron, Allen, The Dickey Amendment on Federal Funding for Research on Gun Violence: A Legal Dissection (2018) Am. J. of Public Health, Vol. 108, No. 7, pp. 865–867.) Not long after, Congress passed an amendment to a CDC appropriations bill,

4 “An ‘empirical study’ is commonly understood to mean a research study that relies on empirical evidence; it is designed to test a theory or hypotheses by collecting independently verifiable data or information and making conclusions based on that information.” (Wendz v. State Dept. of Education (2023) 93 Cal.App.5th 607, 652.)

8 commonly referred to as the “Dickey Amendment,” that essentially ended CDC funding for research concerning firearm violence. (Ibid.) Nevertheless, researchers at UC Davis continued to study firearm violence using data supplied by the DOJ. 1. The Creation of the Firearm Violence Research Center

In 2016, apparently sharing many of the same concerns as the electorate, the Legislature declared its intent that the Regents of the University of California establish and administer a Firearm Violence Research Center to expand and continue this valuable research. (Assembly Bill No. 1602 (2015–2016 Reg. Sess.), ch. 24, § 30 (Assem. Bill 1602).) In newly enacted section 14230, the Legislature made the following findings: “(a) Firearm violence is a significant public health and public safety problem in California and nationwide. Nationally, rates of fatal firearm violence have remained essentially unchanged for more than a decade, as declines in homicide have been offset by increases in suicide.

“(c) Nationwide, the annual societal cost of firearm violence was estimated at $229,000,000,000 in 2012. A significant share of this burden falls on California. In 2013, the Office of Statewide Health Planning and Development noted that government-sponsored insurance programs covered nearly two-thirds of the costs of hospitalizations for firearm assaults in California, and about one-half of the costs of hospitalizations for unintentional injuries or those resulting from deliberate self-harm.

“(d) California has been a leader in responding to this continuing crisis. However, although rates of fatal firearm violence in California are well below average for the 50 states, they are not low enough.

9 “(e) Too little is known about firearm violence and its prevention. This is in substantial part because too little research has been done. The need for more research and more sophisticated research has repeatedly been emphasized. . . . Because there has been so little support for research, only a small number of trained investigators are available.

“(f) When confronted by other major health and social problems, California and the nation have mounted effective responses, coupling an expanded research effort with policy reform in the public’s interest. Motor vehicle accidents, cancer, heart disease, and tobacco use are all examples of the benefits of this approach.

“(g) Federal funding for firearm violence research through the federal Centers for Disease Control and Prevention has been virtually eliminated by Congress since 1996, leaving a major gap that must be filled by other sources.”

Based on these findings, the Legislature enacted section 14231, in which it requested that the University of California establish and administer the Firearm Violence Research Center. (§ 14231, subds. (a), (e).) Statutorily enumerated goals and principals provide that the Center is to conduct interdisciplinary research addressing: “(A) The nature of firearm violence, including individual and societal determinants of risk for involvement in firearm violence, whether as a victim or a perpetrator.

“(B) The individual, community, and societal consequences of firearm violence.

“(C) Prevention and treatment of firearm violence at the individual, community, and societal levels.”

(Id. at subd. (a)(1)(A)–(C).)

In addition, the Center is to “work on a continuing basis with policymakers in the Legislature and state agencies to identify, implement,

10 and evaluate innovative firearm violence prevention policies and programs.” (§ 14231, subd. (a)(3).) To achieve those goals, the Legislature mandated further that, “[s]ubject to the conditions and requirements established elsewhere in statute, state agencies, including, but not limited to, the Department of Justice, the State Department of Public Health, the State Department of Health Care Services, the Office of Statewide Health Planning and Development, and the Department of Motor Vehicles, shall provide to the [C]enter, upon proper request, . . . the data necessary for the [C]enter to conduct its research.” (Id., current subd. (c)(2), former subd. (c).) 2. The DOJ’s Restriction of the Center’s Access to Data, and the Legislature’s Response

Despite the Legislature’s emphasis on the importance of the Center’s research, the DOJ stopped providing data to the Center sometime around 2017 and, in 2021, then Attorney General Becerra announced that the DOJ was considering a rule change based on privacy concerns that would restrict the release of certain personal identifying information to the Center. In a statement issued at the time, the DOJ said that it valued the Center’s research and policy impacts, but also took seriously its “ ‘duty to protect Californians’ sensitive personally identifying information.’ ” The Legislature passed Assembly Bill No. 173 (2021–2022 Reg. Sess.) (Assem. Bill 173) in response, making several amendments to the law, solidifying the DOJ’s ability, and obligation, to provide data to the Center. First, the Legislature amended the findings enumerated in section 14230 to add: “California’s uniquely rich data related to firearm violence have made possible important, timely, policy-relevant research that cannot be conducted elsewhere.” (§14230, subd. (e), as amended by Assem. Bill 173, 2021 Stats., ch. 253, § 4.)

11 Second, the Legislature amended sections 14231, 11106, and 30352 to clarify the data sharing requirement for both the AFS and APRS. Section 14231, subdivision (c)(1) now clarifies that “[i]t is the intent of the Legislature that the [C]enter be provided with access to data kept by state agencies that is necessary for the conduct of its research.” Section 14231, subdivision (c)(2) maintains the previous language regarding requests for data, and newly added subdivision (c)(3) now states, more specifically: “Material identifying individuals shall only be provided for research or statistical activities and shall not be transferred, revealed, or used for purposes other than research or statistical activities, and reports or publications derived therefrom shall not identify specific individuals. Recognizing the time-sensitive nature of the [C]enter’s research, data shall be provided in a timely manner. . . . If a request for data or letter of support for research using the data is denied, the state agency shall provide a written statement of the specific reasons for the denial.”

That same language is repeated in newly added section 11106, subdivision (d) and 30352, subdivision (b)(2). In addition, section 11106, subdivision (d) provides: “All information collected pursuant to this section shall be maintained by the department and shall be available to researchers affiliated with the California Firearm Violence Research Center at UC Davis for academic and policy research purposes upon proper request and following approval by the [C]enter’s governing institutional review board when required. At the department’s discretion, and subject to Section 14240, information collected pursuant to this section may be provided to any other nonprofit bona fide research institution accredited by the United States Department of Education or the Council for Higher Education Accreditation for the study of the prevention of violence and following approval by the institution’s governing institutional review board or human subjects committee when required.”

12 And section 30352, subdivision (b)(2), provides:

“The information collected by the department as provided in paragraph (1) shall be available to researchers affiliated with the California Firearm Violence Research Center at UC Davis following approval by the institution’s governing institutional review board, when required. At the department’s discretion, and subject to Section 14240, the data may be provided to any other nonprofit bona fide research institution accredited by the United States Department of Education or the Council for Higher Education Accreditation for the study of the prevention of violence, following approval by the institution’s governing institutional review board or human subjects committee, when required, for academic and policy research purposes.”

The foregoing provisions went into effect on September 23, 2021. On November 29, 2021, the DOJ approved the Center’s most recent application and authorized 19 users to access data from the DROS and AFS system. The data that the DOJ allowed the researchers to access included all available personal identifying information. On July 1, 2021, the DOJ approved an application from Stanford University and authorized similar access for an additional 8 researchers and, on November 15, 2021, the DOJ provided DROS data extracts to Stanford. As of April 2022, there were no further applications for AFS or DROS data pending before the DOJ. As of that same date, neither the Center nor any other research institution had submitted a request for APRS data. D. Plaintiffs Challenge to the Sharing of Firearm and Ammunition Data and the Associated Preliminary Injunction

In early 2022, Plaintiffs Ashleymarie Barba (an individual gun owner), Firearms Policy Coalition, Inc., Second Amendment Foundation, California Gun Rights Foundation, San Diego County Gun Owners PAC, Orange County Gun Owners PAC, and the Inland Empire Gun Owners PAC (collectively,

13 Plaintiffs) filed a complaint against Rob Bonta, in his official capacity as Attorney General of California, challenging the constitutionality of the amendments enacted by Assem. Bill 173. In the operative first amended complaint, Plaintiffs assert that the amendments are invalid for three reasons: 1) the disclosure of personal identifying information permitted by the amended statutes violates their right to privacy under Article I, section I of the California Constitution; 2) Assem. Bill 173 improperly amends Proposition 63, a voter initiative; and 3) the amendments violate the Second Amendment of the United States Constitution. Based on these claims, Plaintiffs seek declaratory and injunctive relief precluding the DOJ from sharing personal identifying information from the AFS or APRS databases. On March 3, 2022, Plaintiffs filed a motion for a preliminary injunction, asking the trial court to enjoin the DOJ from sharing personal identifying information collected pursuant to sections 11106 and 30352. In addition, they asked the court for a mandatory injunction requiring the DOJ to retrieve all previously transferred personal identifying information. Although they had raised other arguments in the complaint, Plaintiffs premised their motion for preliminary injunction solely on their right to privacy under Article I, Section 1 of the California Constitution. Later that same month, the Attorney General filed a demurrer, asserting that all three of Plaintiffs’ claims fail as a matter of law. At the parties’ request, the trial court stayed the demurrer as to the third claim, based on the Second Amendment, pending resolution of the same issue in a different matter pending in the federal

courts.5

5 On January 12, 2023, while the present appeal was pending, the federal district court granted the Attorney General’s request to dismiss the complaint in the federal matter. (See Doe v. Bonta (S.D.Cal. 2023) 650 F.Supp.3d 1062.)

14 The trial court heard both motions at a consolidated hearing in October 2022. The court began with the demurrer. After hearing argument, the court indicated it was inclined to confirm its tentative on the demurrer, overruling the demurrer as to the first cause of action alleging violations of the right to privacy under the California Constitution, and sustaining it as to the second cause of action alleging that Assem. Bill 173 improperly amends a voter initiative. The court then turned its attention to the preliminary injunction motion. The Attorney General began by pointing out that the injunction was based on a different standard than the demurrer. He argued that Plaintiffs had the burden of establishing a likelihood of success on the merits to obtain an injunction and that, because they had brought only a facial challenge to the amended statutes, they first had to overcome the inherent presumption that the statutes are constitutional. In response, the court pointed out that the standard for a preliminary injunction was a balancing test that did not necessarily rest primarily on the likelihood of success on the merits. The Attorney General went on to address the state’s countervailing interest in conducting research aimed at preventing or reducing firearm violence and the imminent harms resulting from precluding or delaying that research. He concluded: “This court has to look beyond whether we have raised factual issues in the pleadings and must address whether the plaintiffs have shown they are likely to succeed on the merits on their facial challenge to [Assem. Bill] 173. And because they have not and because the balance of harms cuts against denying Californians the benefit of critical research into an extraordinarily pressing problem, the injunction should be denied.” After hearing argument from Plaintiffs, the trial court confirmed its tentative decision to grant the preliminary injunction, in part. In a

15 subsequent written order, the court found that “the balancing of plaintiffs’ privacy interest against the state’s interest in reducing firearms violence, is not an appropriate inquiry at the demurrer stage.” It then incorporated that ruling into its discussion of the preliminary injunction ruling, and stated: “Just as plaintiffs’ cause of action for violation of privacy under the California Constitution survived defendant’s demurrer, for the same reasons plaintiffs have also shown a likelihood of success on the merits to satisfy the factor of the preliminary injunction inquiry.” Turning to the balance of harms, the court weighed the fact that much of the information had already been shared with researchers, against the alleged harms associated with additional transfers, and concluded the balance of harms weighed in favor of issuing a prohibitory, but not mandatory, injunction. Based on that ruling, the court issued the following preliminary injunction: “The California Department of Justice is enjoined from transferring to researchers (1) personal identifying information collected in the Automated Firearms System pursuant to Penal Code section 11106(d) and (2) personal identifying information collected in the Ammunition Purchase Records File pursuant to Penal Code section 30352(b)(2), until further notice and order by the Court.” The Attorney General filed a timely notice of appeal. III. DISCUSSION The Attorney General asserts the trial court erred by misapplying the demurrer standard to its ruling on the preliminary injunction, and that under the correct legal standard, the trial court’s issuance of the preliminary injunction was an abuse of its discretion. Plaintiffs respond that the trial court was not required to make express findings as to each element of the preliminary injunction test, that the trial court’s balance of harms analysis

16 shows that it did consider the appropriate legal standard, and that the trial

court did not err by issuing the injunction.6 A. Relevant Legal Principals Regarding Preliminary Injunctions

“[T]he question whether a preliminary injunction should be granted involves two interrelated factors: (1) the likelihood that the plaintiff will prevail on the merits, and (2) the relative balance of harms that is likely to result from the granting or denial of interim injunctive relief.” (White v. Davis (2003) 30 Cal.4th 528, 554.) “ ‘The ultimate goal . . . is to minimize the harm which an erroneous interim decision may cause.’ ” (Ibid., italics omitted.) The party seeking injunctive relief has the burden to show all elements necessary to support the issuance of a preliminary injunction. (O’Connell v. Superior Court (2006) 141 Cal.App.4th 1452, 1481 (O’Connell).) We review the trial court’s ruling on a motion for a preliminary injunction for an abuse of discretion. (See Cohen v. Board of Supervisors (1985) 40 Cal.3d 277, 286.) “Generally, the ruling on an application for a preliminary injunction rests in the sound discretion of the trial court. The exercise of that discretion will not be disturbed on appeal absent a showing that it has been abused.” (Ibid.) However, “[t]he abuse of discretion standard is not a unified standard; the deference it calls for varies according to the aspect of a trial court’s ruling under review.” (Haraguchi v. Superior Court (2008) 43 Cal.4th 706, 711.) “For instance, the superior court’s express and implied findings of fact are accepted by appellate courts if supported by substantial evidence, and the superior court’s conclusions on issues of pure law are subject to independent review.” (Smith v. Adventist Health System/West (2010) 182 Cal.App.4th 729, 739 (Smith).) In addition, “ ‘ “[a]

6 Neither party disputes the trial court’s ruling on the Attorney General’s demurrer.

17 trial court abuses its discretion when it applies the wrong legal standards applicable to the issue at hand.” ’ ” (Zurich American Ins. Co. v. Superior Ct. (2007) 155 Cal.App.4th 1485, 1493.) Because the preliminary injunction at issue here arises from Plaintiffs’ claim that the disclosure of personal identifying information permitted by amended sections 11106 and 30352 violates their right to privacy under Article I, section I of the California Constitution, we must consider the trial court’s ruling in the context of the law surrounding the right to privacy under the California Constitution. B. Relevant Legal Principals Regarding the Right to Privacy Established by Article I, Section I of the California Constitution

In 1972, the People of the State of California voted to approve a ballot initiative commonly referred to as “The Privacy Initiative,” which amended Article I, Section I of the California Constitution to include a universal right to privacy. (See Lewis v. Superior Court (2017 3 Cal.5th 561, 569 (Lewis).) As our high court has explained, “The Privacy Initiative addressed the ‘accelerating encroachment on personal freedom and security caused by increased surveillance and data collection activity in contemporary society.’ ” (Ibid.) The initiative addressed three primary concerns: “ ‘(1) “government snooping” and the secret gathering of personal information; (2) the overbroad collection and retention of unnecessary personal information by government and business interests; (3) the improper use of information properly obtained for a specific purpose, for example, the use of it for another purpose or the disclosure of it to some third party; and (4) the lack of a reasonable check on the accuracy of existing records.’ ” (Ibid.) The ballot materials suggest that proponents of the amendment were primarily concerned with the government compiling secret dossiers, or “ ‘cradle to grave’ profiles,” on citizens, without affording individual citizens

18 any right or ability to review the information contained therein for accuracy. (Voter Initiative Guide for 1972 General Election (1972), pp. 26–27.) The proponents asserted the right to privacy “should be abridged only when there is a compelling public need. Some information may remain as designated public records but only when the availability of such information is clearly in the public interest.” (Id. at p. 27.) Plaintiffs now assert that the sharing of firearm and ammunition data with researchers in accordance with amended sections 11106 and 30352 is not in the public interest and violates their right to privacy under the California Constitution. The California Supreme Court considered an analogous question in Lewis. The issue there was whether the patients’ right to privacy was violated when the Medical Board of California accessed data from a database used to monitor prescription drugs during an investigation of the patients’ physician. (Lewis, supra, 3 Cal.5th at p. 565.) In evaluating the merits of that claim, the Court applied the two-part inquiry for determining whether there has been a violation of the right to privacy under the California Constitution, previously articulated by the Court in its decision in Hill v. National Collegiate Athletic Assn. (1994) 7 Cal.4th 1 (Hill). (Lewis, supra, 3 Cal.5th at p. 571.) “First, the complaining party must meet three ‘ “threshold elements” . . . utilized to screen out claims that do not involve a significant intrusion on a privacy interest protected by the state constitutional privacy provision.’ ” (Lewis, supra, 3 Cal.5th at p. 571.) They “must demonstrate: ‘(1) a legally protected privacy interest; (2) a reasonable expectation of privacy in the circumstances; and (3) conduct by defendant constituting a serious invasion of privacy.’ ” (Ibid., quoting Hill, supra, 7 Cal.4th at pp. 39–40.) “Whether a legally recognized privacy interest is present in a given case is a question of

19 law to be decided by the court. [Citations.] Whether plaintiff has a reasonable expectation of privacy in the circumstances and whether defendant’s conduct constitutes a serious invasion of privacy are mixed questions of law and fact. If the undisputed material facts show no reasonable expectation of privacy or an insubstantial impact on privacy interests, the question of invasion may be adjudicated as a matter of law.” (Hill, at p. 40.) Second, any such privacy interest “must be balanced against other important [or countervailing] interests.” (Hill, supra, 7 Cal.4th at p. 37; accord Lewis, supra, 3 Cal.5th at p. 572.) Thus, even if a plaintiff is able to establish a serious invasion of privacy under the threshold inquiry, the defendant may still prevail by “ ‘proving, as an affirmative defense, that the invasion of privacy is justified because it substantively furthers one or more countervailing interests.’ ” (Lewis, at p. 572.) “ ‘The plaintiff, in turn, may rebut a defendant’s assertion of countervailing interests by showing there are feasible and effective alternatives to defendant’s conduct which have a lesser impact on privacy interests.’ ” (Ibid., quoting Hill, at p. 40.) “The standard that a defendant’s proffered countervailing interests must satisfy varies based on the privacy interest asserted: ‘Where the case involves an obvious invasion of an interest fundamental to personal autonomy, e.g., freedom from involuntary sterilization or the freedom to pursue consensual familial relationships, a “compelling interest” must be present to overcome the vital privacy interest. If in contrast, the privacy interest is less central, or in bona fide dispute, general balancing tests are employed.’ ” (Lewis, supra, 3 Cal.5th at p. 572, quoting Hill, supra, 7 Cal.4th at p. 34.) “ ‘The existence of a sufficient countervailing interest or an alternative course of conduct present[s] threshold questions of law for the

20 court. The relative strength of countervailing interests and the feasibility of alternatives present mixed questions of law and fact. . . . [I]n cases where material facts are undisputed, adjudication as a matter of law may be appropriate.’ ” (Lewis, at p. 572, quoting Hill, at p. 40.) C. The Trial Court Abused Its Discretion by Granting the Injunction

Here, by relying on its analysis on the demurrer to resolve the motion for preliminary injunction, it appears that the trial court failed to consider the second step of the Hill analysis. Applying the full analysis compelled by Hill, we conclude that the Attorney General established a legitimate countervailing interest and that, although Plaintiffs proposed some alternatives, they did not rebut the Attorney General’s evidence regarding the lack of feasibility of those alternatives. Thus, on the record before us, Plaintiffs cannot establish a likelihood of success on the merits as a matter of law, and the preliminary injunction must be reversed. 1. The Trial Court Did Not Apply the Correct Legal Standard

The trial court began by considering the constitutional privacy claim in Plaintiffs’ first cause of action in the context of the Attorney General’s demurrer. As the trial court explained, the standard for reviewing the sufficiency of a complaint against a general demurrer is well settled. (Blank v. Kirwan (1985) 39 Cal.3d 311, 318.) The reviewing court is to treat the demurrer as accepting all material facts properly pleaded and, doing so, must determine simply whether the complaint states facts, if taken as true, sufficient to constitute a cause of action. (Ibid.) Addressing the demurrer here, the trial court acknowledged the threshold requirements to state a claim for an invasion of privacy set forth in Hill and concluded that Plaintiffs had alleged facts sufficient to satisfy each requirement.

21 In overruling the demurrer as to Plaintiffs’ first cause of action, the trial court concluded that Plaintiffs adequately alleged (1) that they had a legally protected privacy interest in the personal information collected in conjunction with firearm and ammunition transactions; (2) that they had a “reasonable expectation of privacy in this information not being disclosed for reasons other than law enforcement and not being disclosed to third parties who are hostile to their interests”; and (3) that the transfer and use of the data constituted a serious invasion of privacy insofar as it was done without their knowledge or consent. The court went on to explain that issues such as the reasonableness of Plaintiffs’ expectations of privacy and the seriousness of the alleged invasion were questions of fact not suitable for resolution at the demurrer stage. Similarly, the court expressly stated that “the balancing of plaintiffs’ privacy interest against the state’s interest in reducing firearms violence, is not an appropriate inquiry at the demurrer stage.” (Italics added.) The trial court then incorporated this same analysis into its decision on the preliminary injunction motion. Addressing Plaintiffs’ likelihood of success on the merits, the trial court stated: “Just as plaintiffs’ cause of action for violation of privacy under the California Constitution survived defendant’s demurrer, for the same reasons plaintiffs have also shown a likelihood of success on the merits to satisfy the factor of the preliminary injunction inquiry. Defendant’s arguments do not compel a different outcome.” (Italics added.) The Attorney General asserts that this was error, and argues that the trial court had to go beyond the adequacy of the facially accepted allegations in the complaint to adequately consider whether Plaintiffs had established a likelihood of success on the merits of their claim. In other words, the trial court needed to consider those issues that it had deemed not appropriate for

22 consideration at the demurrer stage and, in doing so, needed to consider the supporting materials submitted by the parties. Because the trial court declined to conduct that analysis, it abused its discretion. Plaintiffs do not dispute that the standard for overcoming a demurrer differs from the standard for obtaining a preliminary injunction. Rather, they contend that we must presume that the court applied the correct analysis, even if it did not make express findings as to each element of the preliminary injunction test, particularly where, as here, the trial court referenced the correct standard in its ruling. As Plaintiffs explain, further, “[r]ecognition of, and deference to, implied findings is derived from the principle that an appellate court must interpret the facts in the light most favorable to the prevailing party and indulge all reasonable inferences in support of the trial court’s decision regarding the preliminary injunction.” (Smith, supra, 182 Cal.App.4th at p. 739, italics added.) The fact that the trial court relied, to some extent, on its reasoning in ruling on the demurrer in ruling on the preliminary injunction motion is not in and of itself sufficient to establish an abuse of discretion. As the trial court pointed out, the decision whether to grant a preliminary injunction is “ ‘guided by a “mix” of the potential-merit and interim-harm factors; the greater the plaintiff’s showing on one, the less must be shown on the other to support an injunction,’ ” and, thus, the court only had to conclude that Plaintiffs established at least a minimal probability of success on the merits. (See Butt v. State of California (1992) 4 Cal.4th 668, 678 (Butt).) However, a “trial court may not grant a preliminary injunction, regardless of the balance of interim harm, unless there is some possibility that the plaintiff would ultimately prevail on the merits of the claim.” (Ibid.; Law School Admission Counsel, Inc. v. State of California (2014) 222 Cal.App.4th 1265, 1280.)

23 Because Plaintiffs, as the moving party, had the burden to show all elements necessary to support the issuance of the injunction, they had to establish at least some likelihood of success on the merits of their privacy claim to prevail on their motion for a preliminary injunction. (See O’Connell, supra, 141 Cal.App.4th at p. 1481.) The trial court could not determine whether Plaintiffs had done so without applying the full two-step analysis set forth in Hill. In ruling on the demurrer, the trial court explicitly declined to balance Plaintiffs’ privacy interest against the state’s alleged countervailing interest in conducting empirical research aimed at reducing firearm violence. By subsequently incorporating that analysis into the ruling on the preliminary injunction motion, it appears the trial court overlooked the second step of the two-party inquiry set forth in Hill. Indeed, the trial court does not mention the Attorney General’s asserted countervailing interest or Plaintiffs’ alleged feasible alternatives at all in its ruling. Thus, it appears that the trial court did not apply the full, or correct, legal standard in determining whether Plaintiffs had established a likelihood of success on the merits. This was error. And, although we generally presume the trial court did apply the correct analysis, as Plaintiffs assert, “[w]here the record demonstrates the trial judge did not weigh the evidence, the presumption of correctness is overcome.” (Kemp Bros. Construction, Inc. v. Titan Electric Corp. (2007) 146 Cal.App.4th 1474, 1477.) As we explain next, under the correct legal standard, the trial court’s granting of the preliminary injunction was likewise an abuse of its discretion. Assuming that Plaintiffs met their burden under the threshold inquiry, the Attorney General presented a legitimate countervailing interest—promoting research informing policy decisions aimed at preventing or reducing firearm

24 violence—and presented detailed declarations from several individual researchers explaining the importance of and need to use personally identifiable data to conduct that research. Although Plaintiffs offered some argument as to feasible alternatives, the declarations submitted by the Attorney General explained why those alternatives are not adequate. Plaintiffs did not seriously dispute that evidence or provide any evidence of their own establishing a material dispute of fact as to either issue. Thus, although we indulge all reasonable inferences in support of the trial court’s decision, we conclude, on the record before us, that Plaintiffs cannot establish a likelihood of success on the merits as a matter of law. (See Lewis, supra, 3 Cal.5th at p. 572, quoting Hill, supra, 7 Cal.4th at p. 40 [“ ‘The relative strength of countervailing interests and the feasibility of alternatives present mixed questions of law and fact. . . . [I]n cases where material facts are undisputed, adjudication as a matter of law may be appropriate’ ”].) 2. Under the Correct Legal Standard, Plaintiffs Have Not Established a Likelihood of Success on the Merits

a. Threshold Inquiry We begin with the three-part threshold inquiry “ ‘utilized to screen out claims that do not involve a significant intrusion on a privacy interest protected by the state constitutional privacy provision.’ ” (See Lewis, supra, 3 Cal.5th at p. 571; Hill, supra, 7 Cal.4th at p. 34.) The trial court found that Plaintiffs’ allegations as to each of the three threshold factors were adequate to survive a demurrer, and that Plaintiffs established a likelihood of success on the merits “for the same reasons.” In doing so, the court expressly declined to consider whether it was reasonable for Plaintiffs to expect the state to refrain from sharing DROS data with researchers despite the history of the DOJ doing so, or whether the narrow scope of the disclosure prevented it from being a serious invasion of privacy.

25 The parties argue both points on appeal, but we need not, and expressly do not, decide these issues here. Assuming without deciding that Plaintiffs are able to satisfy each prong of the threshold inquiry, we nevertheless conclude that they cannot establish a likelihood of success on the merits because they do not rebut the Attorney General’s evidence supporting its claims that the statutory amendments substantively further a countervailing interest and that there are no feasible alternatives. b. The Attorney General Presented Uncontested Evidence Establishing That the Statutory Amendments Substantively Further One or More Countervailing Interests

Setting aside the threshold inquiry, “[a] defendant may prevail in a state constitutional privacy case by . . . pleading and proving, as an affirmative defense, that the invasion of privacy is justified because it substantively furthers one or more countervailing interests. Plaintiff, in turn, may rebut a defendant’s assertion of countervailing interests by showing there are feasible and effective alternatives to defendant’s conduct which have a lesser impact on privacy interests.” (Hill, supra, 7 Cal.4th at p. 40.) “The existence of a sufficient countervailing interest or an alternative course of conduct present threshold questions of law for the court. The relative strength of countervailing interests and the feasibility of alternatives present mixed questions of law and fact. Again, in cases where material facts are undisputed, adjudication as a matter of law may be appropriate.” (Ibid.) Even a serious “[i]nvasion of a privacy interest is not a violation of the state constitutional right to privacy if the invasion is justified by a competing interest. Legitimate interests derive from the legally authorized and socially beneficial activities of government and private entities. Their relative importance is determined by their proximity to the central functions of a

26 particular public or private enterprise. Conduct alleged to be an invasion of privacy is to be evaluated based on the extent to which it furthers legitimate and important competing interests.” (Hill, supra, 7 Cal.4th at p. 38.) Before turning to the Attorney General’s proffered countervailing interest, we note that “[l]egally recognized privacy interests are generally of two classes: (1) interests in precluding the dissemination or misuse of sensitive and confidential information (‘informational privacy’); and (2) interests in making intimate personal decisions or conducting personal activities without observation, intrusion, or interference (‘autonomy privacy’).” (Hill, supra, 7 Cal.4th at p. 35.) In Hill, the Court explained, “[w]here the case involves an obvious invasion of an interest fundamental to personal autonomy, e.g., freedom from involuntary sterilization or the freedom to pursue consensual familial relationships, a ‘compelling interest’ must be present to overcome the vital privacy interest. If, in contrast, the privacy interest is less central, or in bona fide dispute, general balancing tests are employed.” (Id. at p. 34.) More recently, in Lewis, the California Supreme Court noted that it had applied the general balancing test “without requiring the asserted countervailing interest to be compelling” in all but one case since Hill. (Lewis, supra, 3 Cal.5th at p. 573.) That one case, American Academy of Pediatrics v. Lungren (1997) 16 Cal.4th 307 (Lungren), plainly fell into the second category of privacy interests as it “involved a challenge to a statute requiring a pregnant minor to obtain parental consent or judicial authorization before having an abortion, an issue that ‘unquestionably impinges upon “an interest fundamental to personal autonomy.” ’ ” (Lewis, at p. 573.) By contrast, the class of privacy interest at issue here is informational; it involves the dissemination of sensitive information and not

27 an obvious invasion of an interest fundamental to personal autonomy like the one at issue in Lungren. Plaintiffs summarily assert that the compelling interest standard applied in Lungren applies here, but they offer no explanation or argument to support that conclusion. In our view, under Hill and Lewis, the general balancing test applies in this case. (Hill, supra, 7 Cal.4th at pp. 34–35.) The People of California, the Legislature, and the Attorney General have identified the following countervailing interest justifying any invasion of privacy resulting from the DOJ transferring firearm and ammunition data containing personally identifying information for approved research projects: empirical research supporting informed policymaking aimed at reducing and preventing firearm violence. In passing Proposition 63 in 2016, the electorate acknowledged both the large number of lives lost, and the high costs associated with medical care and lost productivity resulting from various forms of firearm violence in recent years. (Prop. 63, supra.) The Legislature made similar findings when enacting section 14230, also in 2016 (§ 14230, subd. (b)) and found that “[t]oo little is known about firearm violence and its prevention,” because “too little research has been done. The need for more research and more sophisticated research has been repeatedly emphasized.” (Id., subd. (e), italics added.) The Legislature noted further that California has a history of mounting effective responses to similar concerns, by “coupling an expanded research effort with policy reform in the public’s interest.” (§ 14230, subd. (f).) And California is even more uniquely situated in its ability to analyze and develop novel policy initiatives reducing the impacts of firearm violence. As the Legislature added in 2021, “California’s uniquely rich data related to firearm

28 violence have made possible important, timely, policy-relevant research that cannot be conducted elsewhere.” (§ 14230, subd. (e).) Plaintiffs rely on Lungren, supra, 16 Cal.4th 307, to assert that the courts should not give deference to these legislative findings. But Lungren is not directly on point. As noted, in Lungren, the Court addressed the validity of a statutory provision that required pregnant individuals under the age of 18 to secure parental consent or judicial authorization before obtaining an abortion. (Id. at p. 313.) Thus, “the case involve[d] an obvious invasion of an interest fundamental to personal autonomy,” that could only be overcome by a compelling interest. (Id. at p. 329.) In that context, the Court addressed certain legislative findings supporting the asserted countervailing interests of protecting “the physical, emotional, and psychological health of minors,” and preserving and promoting “the parent-child relationship.” (Lungren, supra, 16 Cal.4th at pp. 325, fn. 12, 348.) The Court agreed that these were, on their face, “ ‘compelling interests,’ ” but went on to find that the statute did not actually further those interests. (Ibid.) In doing so, the Court acknowledged that courts generally do give deference to legislative findings but concluded that legislative findings regarding “the need for, or probable effect of, [a] statutory provision” are not “determinative for constitutional purposes.” (Id. at pp. 349–350, italics added.) By contrast, the privacy interest at issue here is informational, such that the general balancing tests applies, and the legislative findings that the Attorney General relies on do not concern such private and constitutionally protected matters as the parent-child relationship. Nor are those findings limited to the Legislature. Rather, both the electorate and the Legislature found that firearm violence is a public health and safety concern requiring the attention of policymakers. And the Legislature then determined that

29 they, as policymakers, would benefit from “timely, policy-relevant research.” Plaintiffs do not seriously dispute the statistics regarding the impact of firearm violence set forth in the Proposition 63 materials or the findings set forth in section 14230. Nor do they present any evidence contradicting the Legislature’s findings that there is a dearth of available research concerning firearm violence and that California is uniquely situated to fill that gap by conducting “important, timely, policy-relevant research that cannot be conducted elsewhere.” (See § 14230, subd. (e).) In any event, the Attorney General did not rely solely on the legislative findings in the trial court. Rather, in opposition to the preliminary injunction motion, the Attorney General presented evidentiary support detailing the importance of the asserted countervailing interest. Professor Garen J. Wintemute, the director of the Center, presented a detailed declaration concerning the importance of the research at issue. As he explained, “The lack of basic information on the epidemiology of firearm violence; personal, community, and societal risk factors for that violence; its personal, community, and societal consequences; and optimal measures for addressing it has led to widespread misunderstanding of the problem and has impeded prevention efforts. Evidence of the effects of state policies and programs for reducing firearm violence as well as basic information on benefits, risks, and prevalence of firearm ownership in California are also lacking.” The Attorney General also submitted a declaration from David M. Studdert, a professor at Stanford University who has also utilized firearm data provided by the DOJ. Professor Studdert explains: “Gun violence is an important public health problem. Nearly 40,000 Americans die each year from firearm-related suicides, homicides, and accidents. These deaths occur in all segments of the community—among men and women, among the young

30 and the old, among all racial and ethnic groups, among city-dwellers and people who live in rural areas, and among gun owners and non-owners.” Accordingly, a major focus of his team’s “research is to advance understanding of the nature of the relationship between access to firearms and risks of firearm-related mortality, including suicide, homicide, and accidental deaths.” In 2016, his team “launched the Longitudinal Study of Handgun Ownership and Transfer (LongSHOT),” “a large, population-level cohort study . . . examining the mortality risks and benefits associated with access to firearms.” The LongSHOT study is “one of the largest studies of firearm injury ever undertaken.” The researchers have published six peer- reviewed manuscripts based on the study—two addressing risk factors associated with firearm suicides—and expect to produce at least as many more. Finally, Professor Daniel Webster, an independent expert on evidence- based firearm policy, submitted a declaration in which he opined that “[t]here have been many important research questions answered because of researchers’ access to information in [California] DOJ’s records, specifically the DROS System and AFS, that are relevant to efforts to reduce homicide, violent crime, suicide, and accidents.” These include, among others, how the risk of use of a firearm in a suicide is distributed over time; which prior criminal offenses are predictive of future risk of criminal activity involving a firearm; and whether denial of a handgun purchase application after a background check reduces risks of subsequent commission of crime. “For example, Wintemute and colleagues reported that in the first week after the purchase of a handgun, the rate of suicide by means of firearms among purchasers was 57 times as high as the adjusted rate in the general population,” suggesting that suicidal ideation drives gun acquisition in at

31 least some cases. As Professor Webster explained further: “Each of these finding are incredibly important for shaping fair and effective gun policy and even for personal decision-making for those considering whether to purchase a handgun.” Plaintiffs offer no evidence of their own to negate the existence, or extent, of the asserted impacts of firearm violence or the importance of empirical research in developing and evaluating policies to reduce such violence. Rather, they present several arguments as to why the state does not have a legitimate countervailing interest in promoting this vital research, none of which are availing. First, Plaintiffs assert that research is not a constitutionally identified function of any branch of the state government. Plaintiffs adopt too narrow a view. As the Legislature itself has explained, the primary focus of the research is to inform policymaking, which is a central function of the legislative branch. (See Carmel Valley Fire Protection Dist. v. California (2001) 25 Cal.4th 287, 299 [“ ‘Essentials of the legislative function include the determination and formulation of legislative policy.’ ”].) The Legislature is in the best position to determine what materials or resources are helpful to it in performing that function. While there may be a wide range of opinions surrounding the appropriate policy responses to firearm violence, it is undoubtedly a topic that the Legislature must consider. Plaintiffs do not dispute the general notion that empirical research may be valuable to policymakers. Instead, they dispute the value of the research at issue here and assert that it will be used only to “promote reduced access to firearms” and “justify limitations on firearm rights.” Again, the Attorney General presented evidence to refute this claim. As Professor Wintemute explains, the researchers “seek to contribute to a clear and objective

32 understanding of firearm violence and its prevention, and in that effort, we go where the data take us.” Moreover, the research covers a broad range of topics; it has been used to inform policies aimed at preventing unintentional and intentionally inflicted self-injury and has, in some cases, questioned the effectiveness of firearm control policies. Plaintiffs have not directly addressed or disputed these specific claims, nor have they provided any evidence to the contrary. Regardless, the potential, yet to be determined, outcomes of the research should not be used to evaluate its inherent value. Finally, Plaintiffs assert that there is no evidence the research has enhanced public policy or resulted in any “tangible public policy change[s].” To the contrary, the Attorney General presented evidence that the findings made by researchers using the DROS and AFS data—including, for example, the elevated risks for firearm violence associated with prior convictions for alcohol related offenses like driving under the influence—“are incredibly important for shaping fair and effective gun policy and even for personal decision-making for those considering whether to purchase a handgun.” Plaintiffs provide no authority or evidence of their own suggesting that the state does not have a countervailing interest in the research absent a direct link between a specific research project and a specific policy change. Rather, much of the research is expressly aimed at assessing the effectiveness of firearm control policies that are put into place by the Legislature. For example, Professor Wintemute explained that, “In 1991, California prohibited the purchase or possession of firearms by persons convicted of violent misdemeanor crimes for 10 years. [His team] conducted a study, funded by the National Institute of Justice, that used [linkage procedures] to perform a rigorous evaluation of the law’s effectiveness.” Likewise, Professor Wintemute’s team has conducted similar studies to

33 assess the effectiveness of California’s unique Armed and Prohibited Persons System (APPS), and is currently studying the effectiveness of California’s extreme risk protection order (ERPO) law. Empirical studies like these “provide shape and guidance to California’s understanding of violence and its prevention and facilitate the development of evidence-based policies and programs.” Moreover, as Professor Wintemute explained, “Another study in progress, funded by the National Institute of Justice, shows great promise in offering new methods for assessing threats of violence, including mass shootings,” and could lead to such policies in the future. These types of studies are consistent with the Legislature’s stated intent that the Center “work on a continuing basis with policymakers in the Legislature and state agencies to identify, implement, and evaluate innovative firearm violence prevention policies and programs.” (§ 14231, subd. (a)(3).) Plaintiffs did not present any evidence to the contrary. c. Plaintiffs Did Not Present Evidence Sufficient to Establish the Existence of Any Feasible Alternatives

Beyond disputing the importance of the research itself, Plaintiffs seek to rebut the Attorney General’s assertion that it constitutes an important countervailing interest by asserting that there are at least two feasible and effective alternatives that would reduce or avoid the privacy intrusion:

34 1) individuals could be given the option of opting out of having their

information shared; and 2) the data could be de-identified.7 Plaintiffs presented these proposed alternatives in their preliminary injunction motion. They also provided a “Compendium of Evidentiary Documents” in support of their motion. The compendium is comprised of several news articles concerning the DOJ’s approach to data sharing, information from the DOJ’s website about the firearm and ammunition data collection process and the AFS and DROS databases, scientific papers published by the researchers using the AFS and DROS data, an article from the United States Department of Commerce entitled “De-Identification of Personal Information,” and several federal court opinions. In response, the Attorney General asserted, and maintains, that the proposed alternatives are neither feasible nor effective to serve the asserted countervailing interest. The Attorney General argued, and now maintains, further, that the evidence Plaintiffs provided is not sufficient to establish the effectiveness or feasibility of the proposed alternatives. By contrast, the Attorney General points to its own expert declarations, which include detailed explanations as to why the alternatives are not feasible, as well as additional declarations detailing the safeguards surrounding the information sharing at issue.

7 Plaintiffs also presented a third alternative, that the DOJ hire its own researchers to conduct the same studies in house. Although they claim that this option would be “less harmful to plaintiffs’ privacy interests,” they do not concede that it would eliminate the privacy concerns altogether. Regardless, Plaintiffs do not explain how this would minimize their privacy concerns, and we perceive no material difference between the DOJ providing data to researchers at accredited research institutions subject to an extensive application process and hiring independent researchers itself to conduct the same research using the same data.

35 As to the first proposal—giving individuals the option to opt out of having their data shared—the researchers explained in their declarations that it was not feasible because it would create selection bias that would undermine the results of the studies. Professor Wintemute stated, “I believe this suggestion violates widely-accepted standards for the conduct of scientific research. Records-based research such as ours is regularly conducted on other major health problems, such as heart disease and cancer, without such an opt-out mechanism.” In these types of studies, “where individual risk factors are very important, identified individual-level data linked across multiple datasets are frequently essential.” “These projects proceed only with Institutional Review Board approval and are subject to strict identity protection requirements, as is the case with our research. The individuals whose records provide the data for the research are not aware of the existence of the research.” Beyond the one article generally discussing various de-identification methods, which we discuss further, post, Plaintiffs do not dispute Professor Wintemute’s explanation of the research norms or offer any evidence to the contrary. Consistent with Professor Wintemute’s explanation, the statutory amendments effectuated by Assem. Bill 173 expressly require that the DOJ make the data available to researchers at the Center—and permit the DOJ to make it available to researchers at other bona fide and accredited nonprofit research institutions—only “upon proper request and following approval by the [C]enter’s governing institutional review board [or human subjects committee] when required.” (§§ 11106, subd. (d); 30352, subd. (b)(2).) Professor Webster provided a detailed description of the Institutional Review Board (IRB) and data protection protocols required to access firearm and ammunition data containing personally identifying information, and both

36 Professors Wintemute and Studdert averred that the studies they conducted were in fact subject to IRB approval and review. The Attorney General also provided declarations from a research data supervisor and an information security officer detailing the strict security protocols associated with any data transfers. They explained that the DOJ shares data only after an extensive application process, during which the researchers must identify each individual that will have access to the data. Moreover, the security requirements vary with the type of data request, becoming increasingly stricter when more sensitive data is requested. The researchers must agree to notify the DOJ of any data breaches and, to date, none of the researchers have had a breach, or otherwise inadvertently disclosed personally identifying information to any unauthorized person. This evidence establishes that the DOJ is already minimizing the privacy intrusion to the extent feasible while still permitting the important research to proceed, and Plaintiffs provide no evidence establishing these procedures are inconsistent with typical human subject research protocols or otherwise insufficient. (See Lewis, supra, 3 Cal.5th at p. 576 [“ ‘[p]rotective measures, safeguards and other alternatives may minimize the privacy intrusion’ and thus should be considered when balancing a plaintiff’s privacy interest against a defendant’s countervailing interests”].) In addition, Professor Webster explained the important differences between the empirical research that is currently being conducted using individually identifiable DROS and AFS data and the prior research, which necessarily relied on “temporal and spatial associations between the adoption of policies and various [measurable] outcomes.” As he explained, the use of individually identifiable data cures many of the limitations of the prior studies, and “[t]here is simply no way to accurately assess risk” in the way

37 these studies have been able to do without the use of personal identifiers. “If those who opt out differ in important ways with respect to risks for future involvement in violence or suicide from those who do not opt out, the findings from the research will be biased and, therefore, of much less value.” And, he opined further that “[d]esigning and implementing an opt-out process would be extremely expensive, time-intensive, and almost likely infeasible.” Professor Wintemute likewise explained, “[a]ccurate research results depend on having a population of research subjects that represents (i.e., closely resembles) the larger population from which it is derived. An opt-out approach would introduce important bias into research on firearm violence and reduce the validity of its findings.” Professor Studdert agreed, stating that “[f]or these and other reasons, large-scale, population-level epidemiological studies that use administrative data, like LongSHOT, rarely if ever use the kind of consent and notification procedures that are common in other types of research.” Thus, all three professors agreed that requiring individual consent “as a pre-requisite for data use would, for practical purposes, render the research undoable,” both because of the cost and because it would “leave a non-representative group of firearm purchasers.” And, as Professor Wintemute concluded: “This is not a hypothetical matter. When access to federal firearm tracing data for research was lost under the terms of the Tiahrt amendment (a change in federal law that limited use of tracing data in ATF’s possession), research on the structure and function of criminal firearms markets was severely impaired.” Plaintiffs do not dispute these specific explanations or offer any evidence to the contrary. Plaintiffs simply assert, in conclusory fashion, that these declarations do not adequately explain why an opt-out mechanism would not be feasible. They offer no argument or evidence rebutting the

38 specific explanations given as to why personal identifying information allowing for linkage is essential to the research being conducted, nor do they provide any affirmative evidence suggesting that an opt-out mechanism would be feasible or effective. As to the second proposed alternative—having the DOJ de-identify the data before providing it to researchers—Plaintiffs do provide some evidence regarding the various methods that can be employed to de-identify personal information from a given data set. In his declaration, Professor Studdert candidly concedes that “[f]or some studies, deidentifying data before sharing it with researchers is entirely appropriate—indeed, good research practice dictates that it should be done. A core principle in the responsible conduct of research involving sensitive data is the ‘minimum necessary’ principle: researchers should receive/collect and use the least amount of personal information needed to perform the study,” and that “[s]ome types of gun violence research can be conducted without any significant compromise in their quality without personally-identifying information.” However, for many of the same reasons already explained in the context of the proposed opt-out procedure, the DOJ’s data transfer policies already follow this core principle and provide personally identifying data only in cases where it is necessary to conduct the research. Professor Studdert averred that “LongSHOT falls into this latter category; it is a type of study that cannot be done without the use of personal identifiers” for the reasons already explained. And Professor Wintemute similarly averred that many of the key studies he and his team published concerning firearm violence “could not have been conducted without identified individual-level data. In each case, the studies assess the relationship, at the individual level, between behavior in the past and events in the future. This is simply not possible

39 with aggregate, population-level data, or without the ability to link records for identified individuals across multiple datasets.” As he explained further, this proposed alternative “misapprehends the research process because [the researchers] do not request identified individual-level information unless it is necessary to our research. Before obtaining that type of data, we must obtain UC Davis IRB approval. The IRB’s mandate is to ensure that research projects are designed and executed in such a way that subject’s personal identifying information is given a very high degree of protection. It reviews a detailed protocol describing the study’s rationale, design, execution, and security procedures.” As with the first proposed alternative, Plaintiffs generally dispute these assertions, and claim that “public policy researchers will always want to gather as much data as possible to conduct ever more research.” But Plaintiffs do not provide any argument or evidence of their own in response to the declarations presented by the Attorney General, and do not explain why de-identification would be both feasible and effective in these specific types of studies. Without doing so, they have not carried their burden to establish the existence of a feasible and effective alternative. (See Lewis, supra, 3 Cal.5th at p. 575 [plaintiffs have the burden of demonstrating the existence of a feasible and effective alternative by substantial evidence].) In sum, the Attorney General presented evidence in the trial court sufficient to establish a countervailing interest that outweighed any associated invasion of privacy and Plaintiffs did not provide evidence or argument sufficient to establish the existence of a feasible and effective alternative in response. (See Lewis, supra, 3 Cal.5th at p. 572.) However, by incorporating its analysis on the demurrer into its analysis on the preliminary injunction motion, it appears the trial court did not fully consider

40 the arguments and evidence concerning the state’s countervailing interest in determining whether Plaintiffs had established a likelihood of success on the merits. Because Plaintiffs failed to rebut the Attorney General’s evidence establishing a legitimate countervailing interest or establish a feasible alternative, Plaintiffs could not establish the requisite likelihood of success on the merits as a matter of law, and the trial court could not properly grant

the preliminary injunction.8 (See Butt, supra, 4 Cal.4th at p. 678; see also Lewis, supra, 3 Cal.5th at p. 572, quoting Hill, supra, 7 Cal.4th at p. 40 [“ ‘The relative strength of countervailing interests and the feasibility of alternatives present mixed questions of law and fact. . . . [I]n cases where material facts are undisputed, adjudication as a matter of law may be appropriate’ ”].)

8 Having concluded that Plaintiffs cannot establish a probability of success on the merits as a matter of law, we need not consider the relative balance of potential harms.

41 IV. DISPOSITION The order granting the preliminary injunction is reversed and the matter is remanded to the superior court with directions to enter a new order denying Plaintiffs’ preliminary injunction motion. Appellant is awarded costs on appeal.

KELETY, J.

WE CONCUR:

DATO, Acting P. J.

DO, J.