California’s constitutional privacy guarantee needs a reset



California voters passed Proposition 11 in 1972, which amended the state constitution to include a fundamental right to privacy. The ballot arguments expressed a clear voter intent to set a high bar for invaders to justify privacy invasions.[1] Yet the California Supreme Court misinterpreted Proposition 11, and all but abrogated the electorate’s intent when it instead set a low bar to justify privacy invasions. California’s constitutional privacy doctrine needs a reset: Hill v. National Collegiate Athletic Association should be disavowed, and privacy doctrine should return to something closer to what the voters intended with Proposition 11.


How the California Supreme Court rewrote voter intent in Hill

The California constitutional right to privacy is distinct from the federal right. Like its federal counterpart, the state right to privacy extends to both to both informational and autonomy privacy.[2] Yet the federal right is only implied, while the California right is codified in the state constitution. The California Supreme Court has taken this to suggest the state right should be broader than its federal counterpart.[3] As a result, in theory Californians have privacy protections that extend beyond the “penumbral” protections under the federal charter, in both liberty and informational privacy.

The California Supreme Court established the current California privacy doctrine in two key cases: Hill v. NCAA and Loder v. City of Glendale.[4] In the first cases to consider the new constitutional privacy right created by Proposition 11, courts used a compelling interest test to review privacy claims. The California Supreme Court adopted that test in White v. Davis, its first decision interpreting Proposition 11, where the court applied a compelling interest test to a privacy violation by law enforcement surveillance on university campuses.[5] The court relied on the Proposition 11 ballot argument, which states that “[t]he right should only be abridged when there is compelling public need.”[6] The White decision concluded that the ballot arguments were “clear” evidence that privacy infringements must be justified by a compelling interest.[7]

The California Supreme Court abandoned that approach in Hill, where the court considered a privacy claim raised by college athletes who objected to a drug testing program.[8] The court distinguished White as concerning privacy interests under First Amendment free speech and association rights.[9] Yet Proposition 11 showed no voter intent to separate privacy claims into subject-matter categories. The court then applied common law privacy tort doctrine to reject the compelling interest standard of review and re-interpret the state constitutional right to privacy.[10]

The Hill test first requires plaintiffs to establish three elements: conduct violates a legally protected privacy right; there is a reasonable expectation of privacy from the intrusion; and the intrusion is a substantial impact on privacy.[11] If the plaintiff establishes these threshold elements, then the defendant must show that the intrusion was justified by a “legitimate” counter-interest.[12] The California Supreme Court briefly seemed to expand the Hill test in Loder v. City of Glendale when it described the three requirements as “threshold elements” that could be used to screen out claims that do not involve significant intrusion on a privacy interest.[13] But that version of the Hill test does not operate today. In County of Los Angeles v. Los Angeles County Employer Relations Commission, the California Supreme Court reinstated the rigid Hill approach: “In general, the court should not proceed to balancing unless a satisfactory threshold showing is made. A defendant is entitled to prevail if it negates any of the three required elements.”[14]

The upshot is that the three-element Hill test — premised on common law tort doctrine and a misinterpretation of voter intent — governs California constitutional privacy claims today. That test, and the analysis it rests on, is flawed for two reasons: it perverts the electorate’s intent, and it conflates the common law and constitutional remedies.

Hill misinterpreted the intent of Proposition 11

The California Supreme Court in Hill misinterpreted the electorate’s intent for Proposition 11. Although the Hill opinion acknowledged that the proponent’s ballot arguments seemed to impose a compelling need standard, the decision instead relied on one inconsistent reference to “legitimate needs” in the proponent’s rebuttal as evidence of the intent the court wanted to find.[15] The court adopted that legitimate need standard over the compelling need standard the voters intended.[16] On that basis, the court abandoned the compelling need standard it previously adopted in White.

The Hill decision’s reading of voter intent stretches credibility. In the sentence preceding the reference to legitimate need in the proponent’s rebuttal, the proponent reaffirmed the “compelling public need standard” by stating that the right to privacy is limited only by “compelling public necessity and the public’s need to know.”[17] And the rebuttal’s reference to “legitimate need” does not undermine the compelling need test — “compelling need” appears three times in the ballot arguments, while a reference to “legitimate need” appears just once.[18]

The first instance of “compelling need” frames the right: “The right of privacy is the right to be left alone. It is fundamental and compelling.”[19] The second instance frames “compelling need” as a test: “The right should only be abridged when there is compelling public need.”[20] That “compelling interest” test is the same standard used in federal privacy cases.[21] Because this sentence is placed immediately after a sentence that references the federal rights, it is reasonable to infer that the voters intended to adopt the federal compelling interest as a test.

The third instance frames the compelling need standard as a balancing test and includes the only mention of legitimate needs: “The right to privacy will not destroy the welfare nor undermine any important government program. It is limited by ‘compelling public necessity’ and the public’s need to know. Proposition 11 will not prevent the government from collecting any information it legitimately needs.”[22] The quotes on “compelling public necessity” appear in the ballot arguments, suggesting an emphasis on that term. Because the reference exists in the rebuttal to the arguments against Proposition 11, the quote marks reaffirm the earlier references to compelling need as the test. The reference to “a legitimate need” in the following sentence is neither an alternative nor a modifier to the intended compelling need standard — it only explains that the effect of the compelling need standard will not prevent all government data collection.

The way Hill and Loder read Proposition 11 violates the tenets California courts abide by when interpreting initiatives. Ballot measures must be reasonably interpreted, with every word’s ordinary meaning given significance, even when a court disagrees with its outcome.[23] This doctrine applies so long as it does not produce absurd results.[24] Yet this standard was not employed in Hill. As shown above, a reasonable interpretation would result in reading the repeated references to “compelling need” as the test for state privacy right claims; indeed, that was the California Supreme Court’s first reading of Proposition 11. Yet Hill hangs on the single reference to “a legitimate need” — which when read in its context affirms the “compelling need” test — and abandons the court’s prior interpretation of the privacy right. That was error.

Current doctrine conflates the California constitutional privacy right with the common law tort

The California Supreme Court in Hill conflated the constitutional privacy right and the common law tort doctrine, which contravenes the electorate’s intent. Hill stated that its “reference to the common law as background to the California constitutional right to privacy is not intended to suggest that the constitutional right is circumscribed by the common law tort.”[25] Yet in Hernandez v. Hillsides, Inc.[26] the court expressly conflated the two: it said that the two sources of privacy protection “are not unrelated” under California law, and that the California constitutional right to privacy “sets standards similar to the common law tort of intrusion.”[27]

That was error, because there is no evidence that voters intended constitutional privacy to be subsumed under common law tort doctrine. To the contrary, the ballot arguments show a desire to establish the state constitutional doctrine as a novel fundamental right. The ballot argument in support of Proposition 11 is unequivocal: “This measure, if adopted, would revise the language of this section to list the right of privacy as one of the inalienable rights.”[28] The ballot argument stated that “the right of privacy is the right to be left alone. It is a fundamental and compelling interest.”[29] Nothing in the ballot arguments suggests that voters intended to constitutionalize common law privacy or tort law.

The distinction is not an accident: inalienable rights are distinct from tort laws.[30] One right is constitutionally guaranteed, while the others are either statutory claims or common law claims.[31] Conflating a constitutional right with tort remedies undermines privacy’s status as a fundamental right and reduces it to a mere civil wrong. Every law student learns that constitutional rights are superior to statutory and common law claims in the law’s hierarchy.

Reducing constitutional privacy claims to the status of a tort means that their remedies coincide. Tort remedies generally provide compensation, while the default fundamental rights remedy is stopping the intrusive conduct.[32] Providing a single set of remedies for constitutional, common law, and statutory privacy claims gives plaintiffs little incentive to pursue the constitutional claim when the tort is arguably easier to prove — and provides compensation.[33] That disincentive in turn discourages further judicial development of the constitutional right, because the constitutional claims will be brought less often. And (as shown below) because the constitutional claim is routinely rejected, plaintiffs will be further discouraged from making the constitutional claim.

Compelling interest does not necessarily mean strict scrutiny

The departure in Hill from the compelling interest test apparently flowed from the court’s concern that applying strict scrutiny for every asserted privacy interest would create an “impermissible inflexibility” for courts.[34] The court was concerned that private businesses, which need to collect some data to process transactions, would be barred from doing so because their need would never be compelling enough. That amounts to a court substituting its policy judgment for the electorate’s to justify lowering the standard of review. And such concerns are never a basis for rewriting an initiative.[35]

Yet that concern was itself erroneous, because it was based on a misreading of the intended compelling interest test. Justice Mosk’s dissent and then-Justice George’s concurring and dissenting opinion in Hill illustrate this misunderstanding. Both justices noted that in this context a compelling interest is a point on a sliding scale test, where “the greater the intrusiveness of the defendant, the more compelling the interest required in order to justify the intrusion.”[36] The “inflexibility” lies not in the test, rather “the error lies in those courts’ understanding and application of the compelling interest standard itself.”[37] The Hill majority identified the problem but reached the wrong solution.

The Hill majority apparently was concerned that equating a compelling interest in the privacy context to that concept’s accepted meaning in the equal protection context would require applying strict scrutiny.[38] Hill referenced federal cases to that effect.[39] But that concern was unwarranted. California’s privacy right is broader than the federal right.[40] And linking the standard of review for state privacy claims to an unrelated federal equal protection doctrine would undermine the California provision’s independent meaning. Unfortunately, the court’s concern about avoiding strict scrutiny led it to further depart from compelling need, reaching its apex in Williams v. Superior Court — where the court limited the compelling interest test to only “obvious invasion(s) of an interest fundamental to personal autonomy.”[41] For the lesser informational privacy interest a mere general balancing test applies.[42]

Williams reversed the electorate’s intended burden of proof. The Proposition 11 ballot argument is unequivocal about requiring a compelling interest to review privacy claims. The legitimate need standard essentially reverses the burden: it requires a defendant to show a mere legitimate interest, while plaintiffs must show a fundamental interest.[43] That in turn perverts the usual judicial approach to initiatives: rather than liberally interpreting the electorate’s intent to guard its initiative power, the existing privacy analysis negates the electorate’s will.

The irony is that (as shown in the case analysis below) like the strict scrutiny test Hill sought to evade, the existing test is similarly fatal in fact — to plaintiffs. Considering that autonomy claims are in the minority, and that the compelling interest test is only applies to a subset of claims within that minority, the result is that the compelling interest test operates to bar nearly all constitutional privacy claims.

The existing test is a substantive limit on California’s constitutional privacy right

The Hill test effectively bars constitutional privacy claims. Tables 1 and 2 present all published cases in which California constitutional privacy were adjudicated. Table 1 shows the number of state and federal privacy claims that have been upheld or denied since 2009 (when Hernandez v. Hillsides was decided). Table 2 shows the state privacy claims by state and federal court. These results show that courts reject 80% of constitutional privacy claims. That, in turn, shows that the legitimate need required to counterbalance a privacy interest is so trivial that the privacy claim is prohibitively difficult to establish. The threshold elements prevent those claiming the privacy right from proceeding with their case, even if the intruding party has provided no justification for the conduct.[44] Bizarrely, federal courts are applying an analysis similar to compelling interest — and yet the federal courts uphold the privacy claim more often than California courts.[45]

Table 1 showing aggregated state and federal claims for California’s constitutional right of privacy:

State & Federal Court: Public
Informational Privacy Claim Autonomy Privacy Claim
38 17
Privacy Claim Upheld (%) Privacy Claim Denied Privacy Claim Upheld (%) Privacy Claim Denied
8 (21%) 30 (79%) 2 (12%) 15 (88%)
State & Federal Court: Private
Informational Privacy Claim Autonomy Privacy Claim
33 4
Privacy Claim Upheld (%) Privacy Claim Denied Privacy Claim Upheld (%) Privacy Claim Denied
8 (24%) 25 (76%) 1 (25%) 3 (75%)

Table 2 showing disaggregating state and federal claims for California’s constitutional right of privacy:

State Court: Public Federal Court: Public
Informational Privacy Claim Autonomy Privacy Claim Informational Privacy Claim Autonomy Privacy Claim
30 14 8 3
Privacy Claim Upheld (%) Privacy Claim Denied Privacy Claim Upheld (%) Privacy Claim Denied Privacy Claim Upheld (%) Privacy Claim Denied Privacy Claim Upheld (%) Privacy Claim Denied
5 (17%) 25 (83%) 1 (7%) 13 (93%) 3 (38%) 5 (62%) 1 (33%) 2 (77%)
State Court: Private Federal Court: Private
Informational Privacy Claim Autonomy Privacy Claim Informational Privacy Claim Autonomy Privacy Claim
11 2 22 1
Privacy Claim Upheld (%) Privacy Claim Denied Privacy Claim Upheld (%) Privacy Claim Denied Privacy Claim Upheld (%) Privacy Claim Denied Privacy Claim Upheld (%) Privacy Claim Denied
2 (18%) 9 (82%) 1 (50%) 1 (50%) 6 (27%) 16 (72%) 0 (0%) 1 (100%)

Table 1 shows that in 92 state and federal claims, 80% failed.[46] Informational privacy claims make up 77% of the total failed claims. And, consistent with the weaker metric the current analysis applies to invaders, the informational claims fail at higher rates. Indeed, if cases disapproved by Williams for applying a compelling interest test to an informational privacy claims are removed, the rejection rate falls to 76%. The high failure rate suggests that the interest required to counterbalance a privacy interest is so low that the informational privacy claim is fatal in fact against plaintiffs.

For example, the court in People v. Laird upheld a trial court motion denying expungement of DNA data after a felony was reduced.[47] The court reasoned that even with redesignation “to an infraction for all purposes, the state’s legitimate interests in the collection and retention of Laird’s DNA outweighs any privacy interest Laird may have in expungement.”[48] The court repeatedly uses “legitimate interests” to characterize the state’s concern, consistent with the Williams view of informational privacy as a “lesser interest.” This result shows that the legitimate interest test reverses the electorate’s intended standard, because it only requires a defendant to show a lesser, legitimate interest while a plaintiff must show a fundamental interest.

Table 2 shows that federal courts denied 71% of privacy claims, while state courts denied 84% of privacy claims. The variance in results flows from the fact that some federal courts do not apply Hill — instead, they evaluate California constitutional privacy claims using an analysis that is closer to the intended compelling interest test. For example, in Carter v. County of Los Angeles a federal court held that a county violated their workers’ privacy rights by surveilling them with a hidden camera to investigate possible misconduct.[49] That’s very similar to the facts in Hillsides, where the California Supreme Court rejected a privacy claim; in Carter the federal court applied strict scrutiny and upheld a privacy claim.[50] The differing analyses were outcome-determinative: the federal court focused on the egregiousness of the surveillance, and it did not consider whether the actions furthered “legitimate interests” as the California court did in Hillsides.[51] The reverse example is equally probative: when federal courts apply the Hill test, the privacy claim gets denied.[52] The upshot is that when federal courts apply a stricter compelling interest test that better fits Proposition 11, plaintiffs are more likely to prevail.

These data suggest that the Hill test is a substantive limitation on state constitutional privacy claims. The number of claims rejected, the apparent difficulty of the threshold questions, and the particularity of the claims that were approved support this conclusion. The difference in the federal treatment of some privacy claims evidences the preclusive effect Hill has on plaintiffs, and shows how differences in the analysis affect the rejection rates. These data show that the state constitutional privacy claim will be upheld or rejected because of the test, regardless of a claim’s merits. In fact, federal courts have commented that the standard California courts apply to state constitutional privacy claims is “a high bar.”[53] It’s a strange day when federal courts are better at validating a California constitutional right than the state courts are.


The Hill test has scuttled Proposition 11. The electorate’s intent to use a compelling interest test for reviewing privacy claims, regardless of type, is plain. Hill misinterpreted that intent to fashion a new standard of review and instead imposed a pseudo-tort constitutional right of privacy. That error opened a path for courts to further depart from compelling need. As the analysis of modern state constitutional privacy right claims shows, the Hill test and its progeny have largely curtailed California’s constitutional privacy right. It’s time to reset this area of the law: abandon Hill and restore the White compelling interest test.


Rodolfo Rivera Aquino is a research fellow at the California Constitution Center. The center’s executive director contributed to this article.

  1. Ballot Pamp., General Elec. (Nov. 7, 1972) at 27 (arguments in support and opposition to Proposition 11) (hereafter “Ballot Arguments”).
  2. In Griswold v. Connecticut (1965), the U.S. Supreme Court described privacy as a protected interest implied within the “penumbra” of the enumerated, individual fundamental rights. Id. at 483. The U.S. Supreme Court later found a federal implied right to privacy to include informational privacy in Whalen v. Roe (1977) at 600. The California Supreme Court has consistently affirmed the right extends to informational and autonomy privacy. See White v. Davis (1975) at 774; Valley Bank of Nevada v. Superior Court (1975).
  3. City of Santa Barbara v. Adamson (1980) at 130 fn.3.
  4. Hill v. NCAA (1994); Loder v. City of Glendale (1997).
  5. White (1975).
  6. Ballot Arguments at 27.
  7. White at 775 (“[T]he statement makes clear that the amendment does not purport to prohibit all incursion into individual privacy but rather that any such intervention must be justified by a compelling interest.”).
  8. Hill at 10.
  9. Id. at 32.
  10. Id. at 22–28.
  11. Id. at 36–37.
  12. Ibid.
  13. Loder at 893.
  14. County of Los Angeles v. Los Angeles County Employer Relations Commission (2013) at 926.
  15. Hill at 20–22.
  16. Id. at 20.
  17. Ballot Arguments at 28.
  18. Id. at 26–28.
  19. Id. at 27 (emphasis added).
  20. Ibid. (emphasis added).
  21. See Lawrence v. Texas (2003) at 593 (the right to privacy, like other fundamental rights, cannot be infringed “unless the infringement is narrowly tailored to serve a compelling state interest.”).
  22. Ballot Arguments at 28 (emphasis added).
  23. Dempsey v. Market Street Ry. Co. (1943) at 113; see also Amador Valley Joint Union High Sch. Dist. v. State Bd. of Equalization (1978) at 257–58.
  24. Amador Valley at 258.
  25. Hill at 22–27.
  26. Hernandez v. Hillsides, Inc. (2009) at 288 (“In light of the (similarities), we will assess the parties’ claims … under the rubric of both the common law and constitutional tests for establishing a privacy violation . . .”)
  27. Id. at 286–87.
  28. Ballot Arguments at 26.
  29. Id. at 27 (emphasis added).
  30. Hill at 80–86 (dis. opn. of Mosk).
  31. See, e.g., Kizer v. County of San Mateo (1991) at 145 (the Tort Claims Act “is a comprehensive statutory scheme that sets forth the liabilities and immunities of public entities and public employees for torts”); Rowland v. Christian (1968) at 116–20 (providing a historical analysis of common law torts).
  32. Turpin v. Sortini (1982) at 232; Kelso, California’s Constitutional Right to Privacy (1992) 19 Pepperdine L.Rev. 327, 394–97 (violation of a fundamental right limits actions and noting that a violation does not give rise to action for damages absent statutory authorization).
  33. While the privacy intrusion tort shares similar elements to the constitutional claim, the tort claim is easier to prove because it does not require a balancing of interests. See Judicial Council of Cal., Civil Jury Instruction (2020) No. 1800. The privacy intrusion tort also allows for “[d]amages flowing from an invasion of privacy [to] include an award for “mental suffering and anguish.” Ibid. Conversely, infringement of constitutional privacy does not necessarily lead to compensation. Ibid. (“It is an open question whether the constitutional privacy provision can also provide direct and sole support for a damages claim.”).
  34. Hill at 35.
  35. Professional Engineers in Cal. Gov. v. Kempton (2007) at 833 (“Our role as a reviewing court is to simply ascertain and give effect to the electorate’s intent guided by the same well-settled principles we employ to give effect to the Legislature’s intent when we review enactments by that body. We do not, of course, pass upon the wisdom, expediency, or policy of enactments by the voters any more than we would enactments by the Legislature.”) (citations and quotations omitted).
  36. Hill at 65 (conc. and dis. opn. of George); id. at 87–86 (dis. opn. of Mosk) (stating that a plaintiff must prove there was a right of privacy and interference, which plaintiff has to counter-balance).
  37. Id. at 65 (conc. and dis. opn. of George).
  38. Hill at 30–31; Brown v. Superior Court (2016) at 351 (when a word or phrase appearing in a statute has a well-established legal meaning, it will be given that meaning in construing the statute).
  39. Hill at 30–31.
  40. See Am. Academy of Pediatrics v. Lungren (1997) at 326 (holding, “the scope and application of the state constitutional right of privacy is broader and more protective of privacy than the federal right of privacy”).
  41. Williams v. Superior Court (2017) at 556.
  42. Ibid.
  43. Id. at 552–53.
  44. Willard v. AT&T Communications of Cal., Inc. (2012) (finding no need to inquire whether appellants had a privacy interest because they could not establish a reasonable expectation of privacy); Faunce v. Cate (2013) (finding that prisoner failed to show that he had a reasonable expectation of privacy when he met with prison medical staff); In re Luis F. (2009) (finding that a teacher failed to state the elements for an invasion of privacy, so no balancing was required).
  45. See, e.g., In re Vizio, Inc., Consumer Privacy Litigation (2017) at 1233 (“Considering the quantum and nature of the information collected, the purported failure to respect consumers’ privacy choices, and the divergence from the standard industry practice, Plaintiffs plausibly allege Vizio’s collection practices amount to a highly offensive intrusion.”).
  46. The cases represent claims made to California state courts and federal courts. State cases comprise reported and unreported Court of Appeals decisions, including an Appellate Division. It also includes California Supreme Court decisions. Similarly, federal cases comprise reported and unreported cases. Because district court cases are reported, unlike state trial court cases, they are included alongside Court of Appeals decisions. A variety of cases were omitted even when appellants raised a privacy claim because their procedural posture prevented the court from resolving the claim. Duarte Nursery, Inc. v. Cal. Grape Rootstock Improvement Com. (2015) (privacy claim rejected because it was not raised in trial court); Grafilo v. Wolfsohn (2019) (avoiding the privacy right claim by stating that there was not enough evidence for a subpoena); Strawn v. Morris, Polich & Purdy, LLP (2019) (stating that a demurrer on appeal does not provide sufficient factual record to hold on the privacy claim). This analysis is further complicated by the different presumptions a court gives to claims given the procedural stance the case is in. Compare In re Q.R. (2020) (rejecting the privacy claim after reviewing lower courts’ holding on abuse of distraction standard) with Lopez v. Youngblood (2009) (approving the privacy claim on a motion to dismiss standard).
  47. People v. Laird (2018) (review denied, Jan. 2, 2019).
  48. Id. at 473.
  49. Carter v. County of Los Angeles (2011) at 1045–47.
  50. Hernandez at 300.
  51. Carter at 1054.
  52. See, e.g., Cahen v. Toyota Motor Co. (2015) at 973 (“As pleaded, defendants’ tracking of a vehicle’s driving history, performance, or location ‘at various times,’ is not categorically the type of sensitive and confidential information the constitution aims to protect.”).
  53. In re Google Assistant Privacy Litigation (2020) at 830.