California Court of Appeal
Castellanos v. State of California
A1636550 citations·
Summary of the case Castellanos v. State of California
In Castellanos v. State of California, the plaintiffs challenged Proposition 22, arguing it violated the California Constitution. The trial court ruled the proposition invalid for intruding on legislative authority and violating the single-subject rule. On appeal, the court found Proposition 22 did not infringe on legislative authority or the single-subject rule but did violate separation of powers principles regarding amendments. The unconstitutional provisions were severable, leading to a partial affirmation and reversal of the trial court's decision.
Key Issues of the case Castellanos v. State of California
- Validity of Proposition 22 under the California Constitution
- Separation of powers and amendment provisions
Key Facts of the case Castellanos v. State of California
- Proposition 22 was approved by voters in November 2020.
- The trial court ruled Proposition 22 invalid for intruding on legislative authority and violating the single-subject rule.
Decision of the case Castellanos v. State of California
The court affirmed the judgment declaring certain provisions invalid and reversed other parts.
Opinions
Filed 3/13/23
CERTIFIED FOR PUBLICATION
IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA
FIRST APPELLATE DISTRICT
DIVISION FOUR
HECTOR CASTELLANOS
et al.,
Plaintiffs and A163655
Respondents,
(Alameda County
v. Super. Ct. No.
STATE OF CALIFORNIA RG21088725)
et al.,
Defendants and
Appellants,
PROTECT APP-BASED
DRIVERS AND SERVICES
et al.,
Interveners and
Appellants.
In November 2020, the voters approved Proposition 22, the
Protect App-Based Drivers and Services Act (Proposition 22).
(Bus. & Prof. Code, 1 §§ 7448–7467, as added by Prop. 22,
approved by the voters at Gen. Elec. (Nov. 3, 2020).) Shortly
afterwards, Hector Castellanos, Joseph Delgado, Saori Okawa,
Michael Robinson, Service Employees International Union
Undesignated statutory citations are to the Business and
1
Professions Code.
1
California State Council, and Service Employees International
Union (SEIU; collectively, plaintiffs) filed a petition for writ of
mandate seeking a declaration that Proposition 22 is invalid
because it violates the California Constitution. 2 The trial court
granted the petition, ruling that the proposition (1) is invalid in
its entirety because it intrudes on the Legislature’s exclusive
authority to create workers’ compensation laws; (2) is invalid to
the extent that it limits the Legislature’s authority to enact
legislation that would not constitute an amendment to
Proposition 22, and (3) is invalid in its entirety because it violates
the single-subject rule for initiative statutes.
Proposition 22’s proponents and the state appeal, arguing
the trial court was mistaken on all three points. We agree that
Proposition 22 does not intrude on the Legislature’s workers’
compensation authority or violate the single-subject rule, but we
conclude that the initiative’s definition of what constitutes an
amendment violates separation of powers principles. Because the
unconstitutional provisions can be severed from the rest of the
initiative, we affirm the judgment insofar as it declares those
provisions invalid and to the extent the trial court retained
jurisdiction to consider an award of attorney’s fees, and otherwise
reverse.
BACKGROUND
In 2019, the Legislature enacted Assembly Bill No. 5
(2019–2020 Reg. Sess.), which established a new test for
2Undesignated citations to constitutional articles and
sections are to the California Constitution.
2
distinguishing between employees and independent contractors
for the purposes of the Labor Code and Unemployment Insurance
Code. (Stats. 2019, ch. 296; Lab. Code, § 2775, subd. (b)(1);
People v. Uber Technologies, Inc. (2020) 56 Cal.App.5th 266, 274–
277 [describing background of statute].)
In response, Davis White and Keith Yandell, supported by
a group called Protect App-Based Drivers and Services (Protect
Drivers; collectively, interveners), proposed Proposition 22.
(§ 7449, subd. (d).) An “[a]pp-based driver” is a person who works
as a driver or courier for transportation or delivery network
companies, which are businesses that operate transportation or
delivery services using an electronic application or platform to
connect passengers seeking transportation or customers seeking
delivery of goods to drivers or couriers willing to provide those
services with their personal vehicles. (§ 7463, subds. (a), (f), (i),
(q).) Among the supporters of Protect Drivers and Proposition 22
were rideshare and delivery network companies such as Uber
Technologies, Inc., Lyft, Inc., and DoorDash, Inc.
When interveners requested a title and summary of the
measure so they could gather the necessary signatures to qualify
it for the ballot, the Attorney General gave it the title “Changes
Employment Classification Rules for App-Based Transportation
and Delivery Drivers.” The Attorney General later modified the
title for the purposes of the voter information guide, titling it
“Exempts App-Based Transportation and Delivery Companies
from Providing Employee Benefits to Certain Drivers.” (Voter
Information Guide, Gen. Elec. (Nov. 3, 2020) title and summary
3
of Prop. 22, p. 56 (Voter Guide).) White and Protect Drivers filed
a petition for writ of mandate in Sacramento County Superior
Court to compel the Attorney General to revise the title and
summary, but that court denied the petition.
Proposition 22 added sections 7448 to 7467 to the Business
and Professions Code. (Proposition 22, § 1, available at
[as of
Mar. 13, 2023].) Section 7450 states the initiative’s purposes are
to (1) “protect the basic legal right of Californians to choose to
work as independent contractors with rideshare and delivery
network companies”; (2) “protect the individual right of every
app-based rideshare and delivery driver to have the flexibility to
set their own hours for when, where, and how they work”;
(3) “require rideshare and delivery network companies to offer
new protections and benefits for app-based rideshare and
delivery drivers”; and (4) “improve public safety by requiring
criminal background checks, driver safety training, and other
safety provisions to help ensure app-based rideshare and delivery
drivers do not pose a threat to customers or the public.” (§ 7450.)
To achieve these purposes, section 7451, titled “Protecting
Independence,” provides, “Notwithstanding any other provision of
law, including, but not limited to, the Labor Code, the
Unemployment Insurance Code, and any orders, regulations, or
opinions of the Department of Industrial Relations or any board,
division, or commission within the Department of Industrial
Relations, an app-based driver is an independent contractor and
not an employee or agent with respect to the app-based driver’s
4
relationship with a network company” if the company does not
control the drivers in certain specified ways. (§ 7451.) 3
Proposition 22 then details certain benefits to which drivers are
entitled, including a health care subsidy for drivers meeting
certain minimum requirements for hours spent providing services
(as opposed to waiting to provide services); a minimum earnings
guarantee based on time spent providing services; occupational
accident insurance; and contract, anti-discrimination, and
termination rights. (§§ 7452–7456, 7463, subd. (j).) The
initiative also includes various obligations for drivers relating to
safety, such as background check and rest requirements.
(§§ 7458, 7461.)
Section 7465 is the only section contained in article 9,
which is titled “Amendment”; it describes the circumstances in
which the Legislature can amend Proposition 22 without voter
approval. (§ 7465.) As relevant here, section 7465 states that the
3 The specific conditions section 7451 imposes for an app-
based driver to qualify as an independent contractor are:
“(a) The network company does not unilaterally prescribe specific
dates, times of day, or a minimum number of hours during which
the app-based driver must be logged into the network company’s
online-enabled application or platform. [¶] (b) The network
company does not require the app-based driver to accept any
specific rideshare service or delivery service request as a
condition of maintaining access to the network company’s online-
enabled application or platform. [¶] (c) The network company
does not restrict the app-based driver from performing rideshare
services or delivery services through other network companies
except during engaged time. [¶] (d) The network company does
not restrict the app-based driver from working in any other
lawful occupation or business.”
5
Legislature can amend Proposition 22’s provisions with a statute
passed by a seven-eighths majority in both houses, so long as the
statute is “consistent with, and furthers the purpose of,” the
initiative and the Legislature complies with certain procedural
requirements. (§ 7465, subd. (a).) Section 7465, subdivision (c)
(section 7465(c)) then addresses the application of these
requirements. Section 7465(c)(1) states that the initiative’s
purposes are described in sections 7448 to 7450. Section
7465(c)(2) states that any statute that amends the definition of
app-based drivers as independent contractors in section 7451
does not further those purposes, effectively preventing the
Legislature from amending that section without voter approval.
Section 7465(c)(3) declares that any statute that places unequal
regulatory burdens on app-based drivers, such as a rule that
prohibits only app-based drivers from performing particular
services, constitutes an amendment of the initiative. And section
7465(c)(4) declares that a statute constitutes an amendment if it
“authorizes any entity or organization to represent the interests
of app-based drivers in connection with drivers’ contractual
relationships with network companies, or drivers’ compensation,
benefits, or working conditions.”
Proposition 22 contains a severability clause declaring that
if any provision is held to be invalid, the remainder of the
initiative shall remain valid, except that the invalidity of
anything in section 7451—which declares drivers to be
independent contractors and not employees under certain
conditions—would invalidate the entire initiative. (§ 7467.)
6
The voters approved Proposition 22 in November 2020,
with 58.6 percent of voters in favor and 41.4 percent opposed.
In January 2021, plaintiffs filed a petition for writ of
mandate in the California Supreme Court seeking a declaration
that Proposition 22 is invalid. The Supreme Court denied the
petition in February 2021 “without prejudice to refiling in an
appropriate court,” though two justices were of the opinion that
the court should have issued an order to show cause.
(Castellanos v. State of California, S266551, Supreme Ct. Mins.,
Feb. 3, 2021.)
A week later, plaintiffs filed a petition for writ of mandate
in Alameda County Superior Court. Plaintiffs named as
defendants the State of California and Katie Hagen as the
director of the Department of Industrial Relations (defendants).
By stipulation, the trial court granted interveners leave to
intervene to oppose the petition as real parties in interest.
Plaintiffs alleged that Proposition 22 is invalid for four
reasons. First, they argued it improperly limits the Legislature’s
authority in article XIV, section 4 of the California Constitution
to create a workers’ compensation system. 4 Second, plaintiffs
alleged that Proposition 22’s provision defining what types of
4 The provision was originally found in article XX,
section 21. (See Mathews v. Workmen’s Comp. Appeals Bd. (1972)
6 Cal.3d 719, 724, fn. 2, 734 (Mathews).) It was renumbered in a
constitutional reorganization in 1976, without substantive
change. (See Pacific Legal Foundation v. Brown (1981) 29 Cal.3d
168, 184, fn. 8; Ballot Pamp., Prim. Elec. (June 8, 1976)
pp. 58–59.) For simplicity, we refer to this provision using its
current numbering, regardless of the time period at issue.
7
statutes would constitute amendments to the initiative violates
the separation of powers doctrine because it restricts the courts’
authority to interpret the Constitution. Third, they alleged that
the amendment provision violates the separation of powers
because it attempts to prevent the Legislature from enacting
laws on matters not substantively addressed within the measure.
Fourth, plaintiffs alleged that one aspect of the amendment
provision violates the rule in article II, section 8 of the
Constitution limiting initiatives to a single subject because it
imposes restrictions on subjects not substantively addressed in
the initiative and it deceived voters into adopting restrictions
that they did not understand.
In August 2021, the trial court agreed with plaintiffs’ first,
third, and fourth arguments. It issued a judgment in September
2021 declaring Proposition 22 invalid in its entirety and ordering
Hagen, as director of the Department of Industrial Relations, not
to enforce any of Proposition 22’s provisions.
DISCUSSION
I. General Legal Principles and Standard of Review
The trial court’s ruling that Proposition 22 is
unconstitutional turns on the interplay between the language of
Proposition 22 and constitutional provisions governing workers’
compensation law, the initiative power, and the separation of
powers. “We apply similar principles when construing
constitutional provisions and statutes, including those enacted
through voter initiative. [Citation.] Our primary concern is
giving effect to the intended purpose of the provisions at issue.
8
[Citation.] In doing so, we first analyze provisions’ text in their
relevant context, which is typically the best and most reliable
indicator of purpose. [Citations.] We start by ascribing to words
their ordinary meaning, while taking account of related
provisions and the structure of the relevant statutory and
constitutional scheme. [Citations.] If the provisions’ intended
purpose nonetheless remains opaque, we may consider extrinsic
sources, such as an initiative’s ballot materials. [Citation.]
Moreover, when construing initiatives, we generally presume
electors are aware of existing law. [Citation.] Finally, we apply
independent judgment when construing constitutional and
statutory provisions.” (California Cannabis Coalition v. City of
Upland (2017) 3 Cal.5th 924, 933–934.)
“ ‘[T]he Constitution’s initiative and referendum provisions
should be liberally construed to maintain maximum power in the
people.’ ” (Independent Energy Producers Assn. v. McPherson
(2006) 38 Cal.4th 1020, 1032 (McPherson).) The Supreme Court
has declared it the courts’ “ ‘solemn duty to jealously guard the
precious initiative power, and to resolve any reasonable doubts in
favor of its exercise.’ ” (Briggs v. Brown (2017) 3 Cal.5th 808, 827
(Briggs).) “ ‘ “We do not consider or weigh the economic or social
wisdom or general propriety of the initiative. Rather, our sole
function is to evaluate [it] legally in the light of established
constitutional standards.” ’ [Citations.] ‘ “[A]ll presumptions and
intendments favor the validity of a statute and mere doubt does
not afford sufficient reason for a judicial declaration of invalidity.
Statutes must be upheld unless their unconstitutionality clearly,
9
positively, and unmistakably appears.” [Citations.] If the
validity of the measure is “fairly debatable,” it must be
sustained.’ ” (Id. at p. 828.)
“We consider only the objections raised by the [case] before
us. ‘We have no occasion at this time to consider other possible
attacks,’ and ‘except as necessary to resolve the basic questions
before us, we do not consider in this case possible interpretive or
analytical problems’ that might arise from the measure in the
future.” (Briggs, supra, 3 Cal.5th at p. 827.) We review here a
facial challenge to the constitutionality of Proposition 22, and we
express no view on claims that might be asserted in specific
applications of the initiative.
II. Article XIV, Section 4 of the California Constitution
Article XIV, section 4 of the California Constitution begins,
“The Legislature is hereby expressly vested with plenary power,
unlimited by any provision of this Constitution, to create, and
enforce a complete system of workers’ compensation, by
appropriate legislation, and in that behalf to create and enforce a
liability on the part of any or all persons to compensate any or all
of their workers for injury or disability, and their dependents for
death incurred or sustained by the said workers in the course of
their employment, irrespective of the fault of any party.” (Cal.
Const., art. XIV, § 4.) 5
The trial court’s ruling based on this provision is
straightforward. Article II, section 10, subdivision (c) (Cal.
5Additional portions of article XIV, section 4 are quoted in
other parts of this opinion. (See, e.g., fn. 11 and pp. 26–28, post.)
10
Const., art. II, § 10(c)) allows the Legislature to amend an
initiative statute only if the voters approve the amendment,
unless the initiative permits amendment without voters’
approval, in which case the Legislature must comply with any
conditions the voters impose. 6 Because of this authority, the
Legislature cannot amend or repeal section 7451 at all to classify
app-based drivers as employees for the purposes of workers’
compensation or amend the initiative in any other respect
without a seven-eighths majority or the approval of the voters.
The trial court concluded these restrictions on the Legislature are
contrary to article XIV, section 4’s statement that the
Legislature’s power to create a workers’ compensation system is
“plenary” and “unlimited by any provision of this Constitution.”
The trial court pointed out that the voters added the phrase
“unlimited by any provision of this Constitution” to article XIV,
section 4, in a constitutional amendment in 1918, seven years
after they amended the Constitution to allow for voter initiatives.
It concluded the plain meaning of article XIV, section 4 prevailed
over the more general provision in article II, section 10(c).
Finally, because Proposition 22 states that any invalidation of
section 7451 invalidates the entire initiative, the trial court found
the constitutional conflict renders Proposition 22 invalid in its
entirety.
6Article II, section 10(c) states in pertinent part, “The
Legislature may amend or repeal an initiative statute by another
statute that becomes effective only when approved by the electors
unless the initiative statute permits amendment or repeal
without the electors’ approval.”
11
The Supreme Court’s decision in McPherson, supra,
38 Cal.4th 1020, which involved a provision worded similarly to
article XIV, section 4, is contrary to the trial court’s ruling, so we
discuss it in some detail. At issue there was article XII, section 5,
which states, “The Legislature has plenary power, unlimited by
the other provisions of this constitution but consistent with this
article, to confer additional authority and jurisdiction upon the
[Public Utilities Commission (PUC)], to establish the manner and
scope of review of commission action in a court of record, and to
enable it to fix just compensation for utility property taken by
eminent domain.” (McPherson, at p. 1032.) An initiative statute
would have expanded the PUC’s authority over electric service
providers. (Id. at p. 1026.) Opponents of the initiative argued
that the initiative statute violated the Legislature’s exclusive
authority to expand the PUC’s authority. (Id. at p. 1027.)
The Court of Appeal held that this constitutional provision
was clear and unambiguous and dictated that only the
Legislature could confer additional authority on the PUC, not the
voters. (McPherson, supra, 38 Cal.4th at p. 1032.) The Court of
Appeal recognized that courts had previously held that references
in the Constitution to the Legislature having a power did not
deprive the voters of their initiative power. (McPherson, at
pp. 1033–1035.) But the Court of Appeal found those cases
distinguishable because article XII, section 5 gave the Legislature
“plenary” power that was “ ‘unlimited by the other provisions of
this constitution.’ ” (McPherson, at p. 1035.)
12
Before the Supreme Court, the proponents of the initiative
challenged the Court of Appeal’s reasoning by pointing out that a
“plenary power” is complete but is not necessarily exclusive.
(McPherson, supra, 38 Cal.4th at p. 1035.) They also noted that
the Court of Appeal’s broad interpretation of the “unlimited”
clause logically would have meant “that a statute passed by the
Legislature pursuant to article XII, section 5 would not be subject
to any provision of the California Constitution, including, for
example, the provision authorizing the Governor to veto a bill
approved by the Legislature.” (Id. at p. 1036.) The Supreme
Court agreed that these arguments showed that article XII,
section 5 was not unambiguous, “[p]articularly in light of the
numerous past California authorities holding that constitutional
references to the Legislature’s authority to take specified action
generally are not interpreted to limit the initiative power.”
(McPherson, at p. 1036.) The court therefore considered the
origin and background of the constitutional language. (Id. at
pp. 1036–1037.) The voters had added the language to the
Constitution in 1911, the same year as the initiative power.
(McPherson, at pp. 1037–1041.) Because both the initiative
power and the language allowing the Legislature to expand the
authority of the PUC were part of the reform program of the
progressive movement, the court found it “most improbable” that
the voters intended the amendment to article XII, section 5 to
limit the scope of the initiative power they approved
simultaneously, “without any direct or explicit statement to this
effect.” (McPherson, at p. 1042.)
13
McPherson reveals two flaws in the trial court’s ruling.
First, McPherson expressly approved “long-standing California
decisions establishing that references in the California
Constitution to the authority of the Legislature to enact specified
legislation generally are interpreted to include the people’s
reserved right to legislate through the initiative power.”
(McPherson, supra, 38 Cal.4th at p. 1043.) This principle—which
plaintiffs do not dispute—deprives the trial court’s textual
argument of much of its force. Rather than read article XIV,
section 4 as conferring plenary, unlimited power on the
Legislature and only the Legislature, McPherson requires that
we read article XIV, section 4 as though it said, “The Legislature
or the electorate acting through the initiative power are hereby
expressly vested with plenary power, unlimited by any provision
of this constitution, to create, and enforce a complete system of
workers’ compensation . . . .” (See McPherson, supra, 38 Cal.4th
at pp. 1032 [“ ‘the power of the people [to enact statutes] through
the statutory initiative is coextensive with the power of the
Legislature’ ”], 1033, 1042–1043 [reading article XII, section 5 as
though it referred to the electorate’s initiative power].) 7
7 This aspect of McPherson answers the contention
underlying much of the dissenting opinion that article XIV,
section 4 delegates power specifically to the Legislature and not
the voters. (Conc. & dis. opn., post, at p. 30.) The dissenting
opinion’s view rests primarily on the fact that article XIV, section
4 is not self-executing. (Conc. & dis. opn., post, at pp. 30–31.)
There is no logical conflict between article XIV, section 4 needing
implementing legislation and the voters retaining their initiative
power in the same field; both can coexist.
14
Given that article XIV, section 4 must be construed to grant
lawmaking authority to both the Legislature and the electorate,
it is not significant that article XIV, section 4 confers plenary
power, nor that the people may exercise their initiative power in
a way that limits the Legislature’s authority under article XIV,
section 4. Article XIV, section 4 is not concerned with the
allocation of power between the Legislature and the electorate,
but rather with ensuring that the lawmaking bodies jointly and
severally have authority to create a workers’ compensation
system. If the people enact an initiative statute to create or
modify the workers’ compensation system, they have exercised
the plenary, unlimited authority that article XIV, section 4
confers and satisfied that aspect of the Constitution. As courts
must liberally construe the initiative power and resolve doubts in
favor of the use of the initiative wherever reasonable, this is the
interpretation of article XIV, section 4 that we must adopt.
(Briggs, supra, 3 Cal.5th at pp. 827–828.)
Second, McPherson shows that the trial court erred when it
read article XIV, section 4 as a plain statement prevailing over
the initiative power and that plaintiffs likewise err when reading
it as an express repeal of the initiative power. McPherson held
that nearly identical language regarding the Legislature’s power
vis à vis the PUC was “at most ambiguous.” (McPherson, supra,
38 Cal.4th at p. 1025.) It reached that conclusion in part based
on its reasoning that applying the “unlimited” language literally
would mean that the Legislature could enact a law without
having to comply with provisions of the Constitution like the one
15
that gives the Governor the right to veto legislation. (McPherson,
at p. 1036.)
The same rationale applies here. Since article XIV,
section 4’s “unlimited” clause cannot mean that workers’
compensation laws are exempt from every other aspect of the
Constitution, it is ambiguous as to which aspects of the
Constitution continue to apply and which do not. As in
McPherson, this finding of ambiguity would require us to consult
the ballot materials from the election in 1918 at which the voters
added the “unlimited” language to article XIV, section 4, in order
to discern the intent behind it. The initiative power was already
part of the Constitution at that time, as the trial court noted.
But the Supreme Court has already concluded that the history of
article XIV, section 4 shows the provision “was added to the
Constitution and then amended for the sole purpose of removing
all doubts as to the constitutionality of the then existing
workmen’s compensation statutes.” (Mathews, supra, 6 Cal.3d at
pp. 734–735, italics added.) 8 Plaintiffs cite no authority or
8 The dissenting opinion cites Mathews, supra, 6 Cal.3d at
page 735, as supporting its theory that neither the Legislature
nor the electorate can change any of the basic features of the pre-
1918 workers’ compensation system. (Conc. & dis. opn., post, at
p. 6.) The point of Mathews was only that article XIV, section 4
was intended to authorize the workers’ compensation system that
already existed—not that those features became sacrosanct and
untouchable by either the Legislature or the electorate. (City and
County of San Francisco v. Workers’ Comp. Appeals Bd. (1978)
22 Cal.3d 103, 114 [article XIV, § 4’s purpose “was simply to
remove any doubt as to the constitutionality of the existing
workers’ compensation legislation, and not to erect any new
16
evidence indicating to the contrary that article II, sections 8 and
10 and article IV, section 1—relating to the initiative power—
were provisions from which the voters intended to free the
Legislature when enacting workers’ compensation laws. Absent
such evidence, the notion that article XIV, section 4 should be
read as limiting the voters’ initiative power falls apart. 9 To
restrictions on the exercise of legislative power,” (citing Mathews,
at pp. 733–734, fn. 11, italics added)].) Because article XIV,
section 4’s purpose was to ensure that the workers’ compensation
system was “ ‘beyond the possibility of being attacked on
technical grounds or by reason of any questioned want of
constitutional authority,’ ” Mathews rejected a constitutional
challenge to the Legislature’s enactment of an amendment to the
workers’ compensation scheme that expanded on a pre-1918 type
of exclusion from coverage. (Mathews, supra, 6 Cal.3d at p. 735 &
fn. 11, italics omitted.) Nothing in Mathews suggests that article
XIV, section 4 prevents the Legislature from changing workers’
compensation in any ways it sees fit, which is unsurprising given
its grant of plenary authority to create a system requiring “any or
all persons to compensate any or all of their workers.” (Cal.
Const., art. XIV, § 4, italics added.) The dissenting opinion
disregards this language as merely confirming that the
Legislature had not yet covered all potentially eligible workers,
but that is precisely the point: it gives the Legislature the
authority to change coverage if it chooses. (Conc. & dis. opn.,
post, at p. 56.) Notably, even plaintiffs agree that it has this
authority, since they admit the Legislature could have excluded
app-based drivers from workers’ compensation coverage.
9 Interveners ask us to judicially notice two sections from a
treatise describing the history and purpose of article XIV,
section 4 of the Constitution, as well as a 1918 newspaper
editorial discussing the initiative that amended this provision.
Relatedly, amicus curiae California Constitution Center seeks
judicial notice of a host of news articles concerning the 1911 and
1918 propositions that created and amended article XIV, section
4. We deny these requests as unnecessary. (County of San Diego
17
paraphrase McPherson, it is “most improbable” that the voters in
1918—seven years after they “approved a far-reaching measure
incorporating a broad initiative power as part of the California
Constitution”—would have intended, “without any direct or
explicit statement to this effect, to limit the use of the initiative
power by virtue of the language” in article XIV, section 4.
(McPherson, supra, 38 Cal.4th at p. 1042.)
Plaintiffs do not agree that McPherson controls here. Like
the trial court, they rely on footnote 9 of that decision, where the
Supreme Court “emphasize[d]” that its holding was “limited to a
determination that the provisions of article XII, section 5 do not
preclude the use of the initiative process to enact statutes
conferring additional authority upon the PUC.” (McPherson,
supra, 38 Cal.4th at p. 1044, fn. 9.) The court continued, “We
have no occasion in this case to consider whether an initiative
measure relating to the PUC may be challenged on the ground
that it improperly limits the PUC’s authority or improperly
conflicts with the Legislature’s exercise of its authority to expand
the PUC’s jurisdiction or authority. Should these or other issues
arise in the future, they may be resolved through application of
the relevant constitutional provision or provisions to the terms of
the specific legislation at issue.” (Ibid., second italics added.)
Plaintiffs assert that their challenge to Proposition 22
raises the type of conflict that McPherson foresaw and about
which it reserved judgment, since they contend Proposition 22
v. State of California (2008) 164 Cal.App.4th 580, 613, fn. 29
[denying request for judicial notice as unnecessary].)
18
improperly conflicts with the Legislature’s exercise of its
authority to enact workers’ compensation laws. Moreover,
according to plaintiffs, applying McPherson’s logic to decide that
a voter initiative relating to workers’ compensation can conflict
with the Legislature’s authority under article XIV, section 4
would be contrary to McPherson’s instruction that a challenge to
the Legislature’s plenary, unlimited power under article XII,
section 5 should be resolved “through application of the relevant
constitutional provision or provisions to the terms of the specific
legislation at issue.” (McPherson, supra, 38 Cal.4th at p. 1044,
fn. 9.) According to plaintiffs, there would have been no reason
for our Supreme Court to include this footnote if a voter initiative
can always undo what the Legislature does with plenary,
unlimited power, like that which article XIV, section 4 provides.
We do not read McPherson’s footnote 9 as broadly as
plaintiffs or the trial court. That footnote states only that the
court had “no occasion” to address a challenge like the one in this
case, meaning the court was not resolving such challenges either
way. (McPherson, supra, 38 Cal.4th at p. 1044, fn. 9.) The
court’s generalized instruction to resolve such challenges
“through application of the relevant constitutional provision or
provisions to the terms of the specific legislation at issue” did not
imply that the principles it had enunciated and applied in
McPherson would have no bearing on such challenges. (Ibid.)
The McPherson court may have wanted to leave open, as a
precedential matter, the possibility that an argument could be
made that an initiative statute improperly limited the
19
Legislature’s authority in some fashion. But plaintiffs have
failed to offer any argument to overcome the implications of
McPherson’s reasoning on this question, which we do not lightly
cast aside. (Cf. City and County of San Francisco v. All Persons
Interested in Matter of Proposition C (2020) 51 Cal.App.5th 703,
715–717 [interpreting constitutional provision by applying
reasoning and principles from Supreme Court decision
interpreting separate but similar constitutional provision].) We
see no justification for reaching a different interpretation than
McPherson reached with respect to virtually identical language.
Seeking to distinguish McPherson and its point that a
literal reading of “unlimited” would exclude the veto power,
plaintiffs argue that article XIV, section 4, unlike the PUC-
related provision at issue in McPherson, states that the
Legislature’s power to enact workers’ compensation laws must be
exercised “by appropriate legislation.” They argue that
appropriate legislation must be enacted bicamerally and
presented to the Governor for veto.
Framing the inquiry in terms of “appropriate legislation”
does not change the analysis. If we followed plaintiffs’ argument,
we would still have to determine what makes legislation
appropriate or inappropriate, which would entail choosing which
constitutional provisions would apply and which would not. It is
not clear why the veto power and the initiative power would fall
on different sides of that line. Plaintiffs characterize the veto as
part of the “normal legislative process” and article II, section
10(c)’s voter-approval requirement for amendments to initiative
20
statutes as a “special limitation” on the Legislature’s power. But
plaintiffs cite nothing to support this distinction; both such
limitations derive from the Constitution and have equal force.
Plaintiffs’ distinction is also inconsistent with the principle that
“ ‘the Constitution’s initiative and referendum provisions should
be liberally construed to maintain maximum power in the
people.’ ” (McPherson, supra, 38 Cal.4th at p. 1032.)
Rather than take up such free-floating standards, we
adhere instead to Hustedt v. Workers’ Comp. Appeals Bd. (1981)
30 Cal.3d 329. The Supreme Court there affirmed that “the
adoption of article XIV, section 4 ‘effected a repeal pro tanto’ of
any state constitutional provisions which conflicted with that
amendment. [Citations.] A pro tanto repeal of conflicting state
constitutional provisions removes ‘insofar as necessary’ any
restrictions which would prohibit the realization of the objectives
of the new article.” (Id. at p. 343; see id. at pp. 343–344 [article
XIV, section 4 did not effect a pro tanto repeal of the judiciary’s
constitutional authority to discipline attorneys because
permitting the Workers’ Compensation Appeal Board (WCAB) to
discipline attorneys was not necessary to effectuate the objectives
of article XIV, section 4].) Because McPherson teaches that
article XIV, section 4’s objective was not to give the Legislature
exclusive authority over workers’ compensation laws, but rather
to give such authority to the Legislature or the voters, Hustedt
dictates that article XIV, section 4 did not repeal the voters’
21
initiative power to enact legislation concerning workers’
compensation. 10
It is also important to remember that, by its nature, “ ‘the
California Legislature possesses plenary legislative authority
except as specifically limited by the California Constitution.’ ”
(Howard Jarvis Taxpayers Assn. v. Padilla (2016) 62 Cal.4th 486,
498; see Yosemite L. Co. v. Industrial Acc. Com. (1922) 187 Cal.
774, 780 [“Nothing is added to the force of [article XIV, section 4]
by the use of the word ‘plenary.’ If the legislature has power to
do a certain thing, its power to do it is always plenary. It is
merely surplus verbiage”].) Thus, unlike grants of power to
Congress in the United States Constitution, the voters had no
need to give a specific power to the Legislature to allow that body
to legislate on workers’ compensation. (City and County of San
Francisco v. Workers’ Comp. Appeals Bd., supra, 22 Cal.3d at
p. 114 [“[e]ven without such specific authorization [in article XIV,
10 The dissenting opinion cites the statement in Hustedt,
supra, 30 Cal.3d at p. 343, fn. 11, that Mathews interpreted the
paragraph in article XIV, section 4 ratifying and confirming the
creation and existence of the WCAB and all its functions “to bar
only the invalidation of ‘basic features’ of the workers
compensation laws ‘as they have existed since 1911.’ ” (Conc. &
dis. opn., post, at p. 6.) While the language is perhaps
ambiguous, the reference to “invalidation” suggests that Hustedt
read Mathews, as we do, as holding only that article XIV,
section 4 was not intended to render invalid any preexisting
features of the workers’ compensation system. If Hustedt
intended to announce a sweeping holding that the 1918 workers’
compensation system was beyond the Legislature’s ability to
change (which was unnecessary to resolve the dispute before it),
we believe it would have said so in far more direct language.
22
section 4], the Legislature possesses the authority, under the now
firmly established view of the concept of the police power, to
adopt appropriate legislative measures for the protection of
employees and their dependents”].) The only reason to enact
article XIV, section 4 was to create an exemption from provisions
elsewhere in the Constitution that might have been viewed at the
time as preventing the enactment of certain features of such
legislation. (See Western Indemnity Co. v. Pillsbury (1915)
170 Cal. 686, 725 (dis. opn. of Henshaw, J.) [espousing view that
workers’ compensation statute under prior version of art. XIV, § 4
violated 14th Amendment of United States Constitution]; see also
Bixby v. Pierno (1971) 4 Cal.3d 130, 142 [describing courts’ shift
over time to give “less emphasis to outmoded rights of property
and to shibboleths of freedom of contract”]; Proposed
Amendments to the Constitution of the State of California, with
Legislative Reasons for and against Adoption Thereof, Gen. Elec.
(Oct. 10, 1911), argument in favor of Sen. Const. Amdt. No. 32
[prior version of art. XIV, § 4 was enacted to abrogate court
decisions holding that compulsory workers’ compensation law
was a taking of property without due process], available at
[as of
Mar. 13, 2023].) Nothing in the initiative power otherwise
prevented the enactment of workers’ compensation legislation, so
article XIV, section 4 did not repeal any aspect of the initiative
power.
To plaintiffs, Proposition 22 is inconsistent with article
XIV, section 4 because the benefits that Proposition 22 gives app-
23
based drivers do not amount to a “complete system of workers’
compensation” as defined in article XIV, section 4. 11 But article
XIV, section 4 does not require every worker to be covered by
workers’ compensation. (Facundo-Guerrero v. Workers’ Comp.
Appeals Bd. (2008) 163 Cal.App.4th 640, 650 [intent of article
XIV, section 4 “was not to impose a lawmaking mandate upon the
Legislature, but to endow that body expressly with exclusive and
‘plenary’ authority to determine the contours and content of our
state’s workers’ compensation system, including the power to
11 The portion of article XIV, section 4 relevant to plaintiffs’
argument that Proposition 22 does not provide a “complete”
compensation system provides as follows: “A complete system of
workers’ compensation includes adequate provisions for the
comfort, health and safety and general welfare of any and all
workers and those dependent upon them for support to the extent
of relieving from the consequences of any injury or death incurred
or sustained by workers in the course of their employment,
irrespective of the fault of any party; also full provision for
securing safety in places of employment; full provision for such
medical, surgical, hospital and other remedial treatment as is
requisite to cure and relieve from the effects of such injury; full
provision for adequate insurance coverage against liability to pay
or furnish compensation; full provision for regulating such
insurance coverage in all its aspects, including the establishment
and management of a state compensation insurance fund; full
provision for otherwise securing the payment of compensation;
and full provision for vesting power, authority and jurisdiction in
an administrative body with all the requisite governmental
functions to determine any dispute or matter arising under such
legislation, to the end that the administration of such legislation
shall accomplish substantial justice in all cases expeditiously,
inexpensively, and without incumbrance of any character; all of
which matters are expressly declared to be the social public policy
of this State, binding upon all departments of the state
government.” (Cal. Const., art. XIV, § 4.)
24
limit benefits”]; Wal-Mart Stores v. Workers’ Comp. Appeals Bd.
(2003) 112 Cal.App.4th 1435, 1442 [Constitution does not make a
right to workers’ compensation benefits “absolute”; Legislature
can choose to exclude certain workers].) 12 The Legislature or the
voters may exclude app-based drivers entirely from workers’
compensation benefits (cf. Lab. Code, § 3352, subd. (a) [excluding
various types of workers from the definition of “employee” for
12 The dissenting opinion concedes that courts have long
given the Legislature “wide berth” to revise the worker’s
compensation system, but it contends that no case has held that
the Legislature may eliminate basic features of the pre-1918
workers’ compensation system. (Conc. & dis. opn., post, at pp. 58-
59.) Apart from Wal-Mart Stores v. Workers’ Comp. Appeals Bd.,
there is also Stevens v. Workers’ Comp. Appeals Bd. (2015)
241 Cal.App.4th 1074, 1094–1096, which held that article XIV,
section 4 did not prevent the Legislature from taking away the
authority of the WCAB to determine medical necessity of
treatment, which it had held since 1917 (Stats. 1917, ch. 586,
§ 9(a), p. 837), and vesting it instead in an independent medical
review organization whose determinations are effectively final.
(Accord, Ramirez v. Workers’ Comp. Appeals Bd. (2017)
10 Cal.App.5th 205, 226; see also Rio Linda Union School Dist. v.
Workers’ Comp. Appeals Bd. (2005) 131 Cal.App.4th 517, 525–
527, 532 [workers’ compensation rights are wholly statutory, so
Legislature could change existing law (which dated to 1917, see
Stats. 1917, ch. 586, § 3(4), p. 833) and require apportionment of
disability based on causation and pre-existing conditions].) The
dissenting opinion recognizes that the WCAB’s authority to
resolve disputes between employers and employees over medical
necessity of treatment was a basic feature of the pre-1918 law.
(Conc. & dis. opn., post, at p. 46.) By contrast, no case has ever
held that article XIV, section 4 prevented the Legislature from
changing basic features of the pre-1918 workers’ compensation
system, as the dissenting opinion would hold.
25
purposes of workers’ compensation]), so the relative insufficiency
of Proposition 22’s benefits is of no constitutional moment.
In their last argument, plaintiffs contend that article XIV,
section 4’s reference to the Legislature should not be read as
including the initiative power because doing so changes the scope
of the article XIV, section 4 power. They reason that the
Legislature on its own could not restrict its own future power,
while an initiative like Proposition 22 can. (In re Collie (1952)
38 Cal.2d 396, 398.) Initiatives do bind the Legislature by virtue
of article II, section 10(c), discussed further, post, in a way that
the Legislature cannot bind itself. (Rossi v. Brown (1995)
9 Cal.4th 688, 715–716.) But the same argument could be made
about any reference to the Legislature in the Constitution, which
would completely defeat the long-established rule that references
to the Legislature should be read as including the initiative
power. (McPherson, supra, 38 Cal.4th at p. 1043.) We therefore
reject this argument.
We turn finally to what the dissenting opinion deems
article XIV, section 4’s “particularly notable” declaration that the
workers’ compensation scheme shall be “binding upon all
departments of the state government.” (Conc. & dis. opn., post, at
p. 39, quoting Cal. Const., art. XIV, § 4). Based on its
interpretation of that phrase, the dissent asserts that when the
people adopt an initiative statute, they “are encompassed within
the phrase ‘all departments of State Government.’ ” (Conc. & dis.
opn., post, at p. 39.) From that premise, the dissent then
concludes that when there is a conflict between a legislative
26
enactment and an initiative statute relating to workers’
compensation, the electorate is bound by the Legislature’s view of
workers’ compensation policy. (Id. at pp. 40–41, 44.) The
dissenting opinion cites nothing to support its assertion that the
people enacting an initiative constitute a “department of the state
government” or its conclusion that the phrase “binding upon all
departments of the state government” was intended to convey a
limitation on the initiative power. In actuality, this phrase
appears to have been intended only to mean that the workers’
compensation system applies to the state and local governments
as employers. (Bautista v. State of California (2011)
201 Cal.App.4th 716, 726; Sacramento v. Industrial Acc. Com.
(1925) 74 Cal.App. 386, 395.)
A second, more fundamental problem is that construing
“departments of the State government” to include the
electorate—one premise for the dissenting opinion’s conclusion
that article XIV, section 4 limits the initiative power—runs afoul
of California Cannabis Coalition v. City of Upland, supra,
3 Cal.5th 924. There, our Supreme Court rejected the City’s
contention that a constitutional provision limiting the authority
of “local government” with respect to taxes should be construed to
mean that such a limitation applies to the electorate when it
enacts a tax-related initiative statute. (Id. at pp. 930–931
[construing article XIII C].) Not only did the court reason that
construing “local government” to include the electorate was
contrary to the term’s common understanding (California
Cannabis Coalition, at p. 937), the court also repeatedly
27
instructed that there must be some “unambiguous indication that
a provision’s purpose was to constrain the initiative power” before
such a limitation would be imposed. (Id. at p. 945; see id. at p.
946 [“the best way to implement our oft-repeated references to
the importance of the initiative is to avoid presuming that a
provision constrains that power without a clear statement or
equivalent evidence that such was the provision’s intended
purpose” (italics added)], 948 [“Unless a provision explicitly
constrains the initiative power or otherwise provides a similarly
clear indication that its purpose includes constraining the voters’
initiative power, we will not construe provisions as imposing such
limitations” (italics added)].)
Similar to the phrase “local government” in the
constitutional provision at issue in California Cannabis
Coalition, article XIV, section 4’s reference to “departments of the
state government” contains no unambiguous indication that the
phrase was intended to include the electorate and thereby
constrain the people’s initiative power. This does not make the
voters “exogenous” to our plan of government (conc. & dis. opn.,
post, at p. 39); the California Constitution plainly provides for the
initiative power, after all. This merely recognizes that a
reference to “State government” or its departments does not
naturally include the voters.
We therefore conclude that Proposition 22 does not violate
article XIV, section 4.
28
III. Single-subject Rule
In addition to challenging the entirety of Proposition 22
based on article XIV, section 4, plaintiffs argued below that
section 7465(c)(4) violates article II, section 8, subdivision (d) of
the Constitution, which provides that “[a]n initiative measure
embracing more than one subject may not be submitted to the
electors or have any effect.” Plaintiffs asserted that section
7465(c)(4) violates this single-subject rule because it is not
reasonably germane to the purpose of Proposition 22 and imposes
restrictions not substantively addressed in the initiative. While
directed at only section 7465(c)(4), plaintiffs’ challenge still aims
at the complete invalidation of Proposition 22 because “when an
initiative measure violates the single-subject rule, severance is
not an available remedy.” (See Senate of the State of Cal. v. Jones
(1999) 21 Cal.4th 1142, 1168.)
The Supreme Court’s “jurisprudence in this area is well
developed.” (Briggs, supra, 3 Cal.5th at p. 828.) It has “ ‘upheld
a variety of initiative measures in the face of a single-subject
challenge, emphasizing that the initiative process occupies an
important and favored status in the California constitutional
scheme and that the single-subject requirement should not be
interpreted in an unduly narrow or restrictive fashion that would
preclude the use of the initiative process to accomplish
comprehensive, broad-based reform in a particular area of public
concern.’ ” (Ibid.) “ ‘[T]he single-subject provision does not
require that each of the provisions of a measure effectively
interlock in a functional relationship. [Citation.] It is enough
29
that the various provisions are reasonably related to a common
theme or purpose.’ ” (Ibid.) “The ‘reasonably germane’ standard
is applied ‘in an accommodating and lenient manner so as not to
unduly restrict . . . the people’s right to package provisions in a
single bill or initiative.’ ” (Id. at p. 829.)
Section 7465 is the only statutory section in the article
titled “Amendment.” Subdivision (a) of section 7465 allows the
Legislature to amend Proposition 22 under certain conditions,
including that any amendments must be enacted by a seven-
eighths majority and must further the purpose of the initiative.
(§ 7465, subd. (a).) Under the Constitution, the Legislature may
also amend an initiative statute if the voters subsequently
approve it. (Cal. Const., art. II, § 10, subd. (c).)
Section 7465(c)(4) defines a specific type of legislation that
would constitute an amendment of Proposition 22, stating, “Any
statute that authorizes any entity or organization to represent
the interests of app-based drivers in connection with drivers’
contractual relationships with network companies, or drivers’
compensation, benefits, or working conditions, constitutes an
amendment” to the initiative. Section 7465(c)(4)’s language is
broad, but it is undisputed that if the Legislature seeks to enact a
law allowing app-based drivers to collectively bargain, it must
comply with the requirements for amendments to the initiative.
Such a legislative enactment would be necessary for drivers to
collectively bargain because antitrust law prevents independent
contractors from doing so unless they obtain specific state
authorization. (Chamber of Commerce of the USA v. City of
30
Seattle (9th Cir. 2018) 890 F.3d 769, 780–790 (Chamber of
Commerce); 15 U.S.C. § 17.)
Plaintiffs alleged and the trial court agreed that section
7465(c)(4) does not relate to the purposes of Proposition 22, as
specifically set forth in section 7450: protecting app-based
drivers’ rights to work as independent contractors, protecting
their right to have flexibility in their schedules and locations,
offering them new benefits and protections, and improving public
safety relating to app-based drivers. (§ 7450, subds. (a)–(d).) We
conclude to the contrary that section 7465(c)(4) is “ ‘reasonably
related to a common theme or purpose’ ” of the initiative and thus
satisfies the single-subject rule, given the accommodating and
lenient fashion in which the Supreme Court has instructed us to
apply it. (Briggs, supra, 3 Cal.5th at p. 828.)
Proposition 22’s common theme or purpose is, as
interveners argue, the creation of a new balance of benefits and
obligations for app-based drivers in lieu of either traditional
employment or traditional independent contractor status.
Section 7465(c)(4) is reasonably germane to this subject because
it relates to drivers’ ability to change that balance by limiting the
Legislature’s ability to authorize collective bargaining over
drivers’ compensation, benefits, or working conditions. Stated
slightly more generally, Proposition 22’s overarching single
subject is regulation of the relationships between app-based
drivers and network companies, and section 7465(c)(4)’s
restrictions on the Legislature’s ability to allow drivers to
collectively bargain relate to those relationships.
31
The trial court took a different approach to the analysis,
which plaintiffs support on appeal. Rather than identifying a
single purpose for the entire law, plaintiffs recite Proposition 22’s
four declared purposes: classifying drivers as independent
contractors, protecting driver independence, providing new
benefits, and protecting public safety. (§ 7450, subds. (a)–(d).)
Like the trial court, they point out that section 7465(c)(4)’s
restrictions on the enactment of laws allowing collective
bargaining have no relationship to the rest of the initiative’s
sections and are not necessary to achieve its four stated purposes.
Plaintiffs resist interveners’ definition of Proposition 22’s purpose
as comprehensive reform of app-based drivers’ relationships with
network companies, arguing that we should rely only on the
express statements of Proposition 22’s purpose in its title,
findings, and declarations.
Given their argument here, plaintiffs are correct that we
should draw Proposition 22’s theme or purpose from its stated
aims, but plaintiffs’ framing of their single-subject challenge
requires us to summarize and derive from Proposition 22’s
multiple purposes a single, overarching theme or purpose against
which to measure section 7465(c)(4). This is how the Supreme
Court has conducted the single-subject analysis, including in the
cases that plaintiffs cite in support of their argument. (Briggs,
supra, 3 Cal.5th at p. 831 [identifying initiative’s purpose as “an
extensive reform of the entire system of capital punishment to
make it more efficient, less expensive, and more responsive to the
rights of victims”]; Manduley v. Superior Court (2002) 27 Cal.4th
32
537, 576 [“Addressing the problem of juvenile crime and gang-
related crime properly can be considered the common purpose of
Proposition 21”]; Legislature v. Eu (1991) 54 Cal.3d 492, 512
[“The unifying theme or common purpose of Proposition 140 is
incumbency reform”].) League of Women Voters v. Eu (1992)
7 Cal.App.4th 649, 654–655, which plaintiffs also claim supports
their position, compared the process of identifying a measure’s
purpose to the arithmetic involved in adding unlike fractions, in
that courts should “identify the lowest common denominator of
the various provisions of the initiative, i.e., the most narrowly
defined object or purpose which nevertheless is sufficiently broad
to encompass all such provisions.” (Id. at p. 659.) Such an
approach starts with the provisions of an initiative and draws
from them the overall theme or purpose of the entire measure.
Plaintiffs’ piecemeal comparison of section 7465(c)(4) to
each of the initiative’s separate purposes misses the forest for the
trees. Two of Proposition 22’s stated purposes, classification of
drivers as independent contractors and protecting driver
independence, relate to each other, but the other two, driver
benefits and public safety, do not. (§ 7450, subds. (a)–(d).) We
therefore cannot give each of these purposes equal significance,
as plaintiffs urge us to do, because doing so would mean
Proposition 22 has three purposes and therefore three subjects.
Additionally, plaintiffs’ approach would make the single-subject
inquiry unworkable. Initiatives commonly state multiple
purposes or motivating concerns, between their titles, preambles,
findings, declarations, and substantive provisions. Proposition 22
33
is relatively simple in this regard with four purposes. Other
initiatives state many more, like the proposition at issue in
Briggs, which set out 11 findings and declarations. (Briggs,
supra, 3 Cal.5th at p. 823; Voter Information Guide, Gen. Elec.
(Nov. 8, 2016) text of Prop. 66, § 2, pp. 212–213.) Eschewing the
identification of initiatives’ common or dominant purpose and
conducting a multivariate analysis, as plaintiffs advocate, is not
feasible. To assess their claim that section 7465(c)(4) represents
a different subject than the remainder of the initiative, we must
elucidate from the initiative’s stated purposes a single theme or
subject against which we can evaluate section 7465(c)(4).
In a fallback argument, plaintiffs contend that even if
Proposition 22’s theme or subject can be isolated from its stated
purposes, its theme is the classification of app-based drivers for
purposes of employment law and section 7465(c)(4) does not
relate to classification. This description of Proposition 22’s
subject is too narrow. Classification is just one of the initiative’s
stated purposes, and only two of the initiative’s statutory sections
relate to it. (§§ 7451 [reclassifying drivers], 7452.5 [nothing else
in Proposition 22 should be construed as altering the
classification of app-based drivers as independent contractors].)
Most of the rest of the initiative’s statutory sections are devoted
to achieving its other stated purposes by detailing the benefits
that app-based drivers must receive and public safety
requirements with which they must comply. (§§ 7454–7462.)
These sections do not relate to classification. Because these
provisions represent the bulk of the initiative, plaintiffs’
34
statement of the initiative’s overall subject cannot be correct.
Moreover, even if classification could be said to be Proposition
22’s single subject, that determination would not help plaintiffs.
If the initiative’s benefits and public safety requirements sections
relate to classification (perhaps because they provide benefits and
obligations to replace those that app-based drivers would have as
employees), then section 7465(c)(4), too, relates to classification
because it concerns the Legislature’s authority to change the
procedures by which drivers can increase the replacement
benefits.
Plaintiffs’ remaining arguments are unavailing. First, they
argue that section 7465(c)(4) restricts the Legislature’s power to
allow drivers to collectively bargain even though the rest of the
initiative, which plaintiffs describe as its “operative” provisions,
does not mention representation of drivers, either individually or
collectively. We have more to say about the relationship between
section 7465(c)(4) and the rest of Proposition 22, post, but for
purposes of the single-subject rule, plaintiffs’ distinction between
section 7465(c)(4) and the initiative’s operative provisions is
irrelevant. Plaintiffs cite no authority that requires different
provisions of an initiative to cross-reference each other, and the
law is to the contrary. “[A] measure’s separate provisions have
been considered to be reasonably germane to each other within
the meaning of the standard so long as all of the provisions are
reasonably germane to a single common theme, purpose, or
subject.” (Californians for an Open Primary v. McPherson (2006)
38 Cal.4th 735, 764, fn. 29, italics omitted.)
35
Second, plaintiffs contend that collective bargaining is a
separate subject because collective bargaining would not impair
drivers’ ability to set their own hours or work independently, and
collective bargaining for increased benefits would not conflict
with the benefits Proposition 22 provides, which are established
as minimum benefits. (See, e.g., §§ 7454, subd. (a)(1)–(2)
[requiring health care subsidy “greater than or equal to” certain
reference standards], 7455, subd. (a) [setting “minimum”
coverage for insurance].) But the single-subject test does not look
at whether a provision is necessary for the rest of an initiative to
function. Plaintiffs are in effect seeking to impose a requirement
that an initiative’s provisions must all functionally relate to one
another. The Supreme Court has already rejected this argument.
(Briggs, supra, 3 Cal.5th at p. 828 [“ ‘[T]he single-subject
provision does not require that each of the provisions of a
measure effectively interlock in a functional relationship’ ”].)
Third, plaintiffs compare this case to California Trial
Lawyers Assn. v. Eu (1988) 200 Cal.App.3d 351, 358 (Trial
Lawyers), which held that an initiative violated the single-subject
rule. The comparison is not apt. Trial Lawyers involved a
challenge to an initiative titled “Insurance Cost Control Initiative
of 1988,” which consisted of 120 pages covering 67 sections.
(Trial Lawyers, at pp. 354–355.) The initiative would have
created “revolutionary” changes in insurance law, including
creation of no-fault insurance for auto accidents, limiting
recovery for injuries beyond the no-fault limits, reducing
premiums for certain coverages, and limiting future insurance
36
regulation legislation. (Id. at pp. 355–356.) One section of the
initiative, located “inconspicuously” at pages 52 and 53 of the
measure, would have enacted a statute providing that public
officials would not be required to disqualify themselves from any
decisions affecting an insurer’s interests based upon the insurer’s
campaign contributions. (Id. at p. 356.) The court found no
possible connection between avoiding such disqualification and
the general object and purpose of the initiative. (Id. at p. 359.)
The court rejected the initiative supporter’s argument that the
provision was germane because both it and the rest of the
initiative dealt generally with regulation of insurance industry
practices. (Id. at pp. 359–360.) The court first noted that the
express purpose of the initiative was to control insurance costs.
(Id. at p. 360.) The court further concluded that the supporter’s
argument would mean that any two provisions that affected the
business of insurance would comply with the single-subject rule,
which the court found would “render the constitutional single-
subject limitation nugatory.” (Ibid.)
Proposition 22 is not like the measure at issue in Trial
Lawyers. The only discernable relationship between the anti-
disqualification provision of that initiative and the rest of the
measure was that both had some effect on the business of
insurance. (Trial Lawyers, supra, 200 Cal.App.3d at p. 360.) But
as just discussed, section 7465(c)(4) relates to the rest of
Proposition 22 because both concern the subject of app-based
drivers’ benefits, with the bulk of Proposition 22 establishing that
drivers would not receive employee benefits and providing a
37
replacement set of benefits, and section 7465(c)(4) restricting the
Legislature’s ability to allow drivers to negotiate collectively to
increase those benefits. Because of the “undeniably liberal
nature of the standards which have been formulated” in the
Supreme Court’s cases, this connection is sufficient. (Trial
Lawyers, at p. 359.)
Finally, plaintiffs seek to buttress their arguments by
pointing out that the single-subject rule is intended to avoid voter
confusion and the exploitation of the initiative process through
the combination of disparate provisions which might not have
commanded majority support if considered separately, which is
known as logrolling. (Amador Valley Joint Union High Sch. Dist.
v. State Bd. of Equalization (1978) 22 Cal.3d 208, 231–232.)
However, as plaintiffs admit, voter confusion and logrolling are
not standalone bases for invalidating an initiative. “The single-
subject rule is the method by which the state Constitution guards
against” those risks, so we may examine only whether
Proposition 22 satisfies the rule, without also examining whether
it satisfies the purposes behind the rule. (Kennedy Wholesale,
Inc. v. State Bd. of Equalization (1991) 53 Cal.3d 245, 255;
accord, California Gillnetters Assn. v. Department of Fish &
Game (1995) 39 Cal.App.4th 1145, 1162 & fn. 11.) While
concerns about voters’ understanding of section 7465(c)(4) may
bear on our interpretation of the statute for the purposes of
plaintiffs’ separation of powers challenge, as we discuss post,
these concerns are not sufficient to rescue plaintiffs’ otherwise
unpersuasive single-subject rule argument.
38
IV. Separation of Powers
We turn now to plaintiffs’ separation of powers challenge
with respect to sections 7465(c)(3) and (4). We have already
discussed section 7465(c)(4) in detail in relation to the single-
subject rule. Section 7465(c)(3) defines another class of
legislation that constitutes an amendment to Proposition 22,
namely, “[a]ny statute that prohibits app-based drivers from
performing a particular rideshare service or delivery service
while allowing other individuals or entities to perform the same
rideshare service or delivery service, or otherwise imposes
unequal regulatory burdens upon app-based drivers based on
their classification status.” Plaintiffs argued in their petition
that sections 7465(c)(3) and (4) are facially unconstitutional
because they intrude on the judiciary’s power to define what
constitutes an amendment to Proposition 22, and that section
7465(c)(4) on its face unconstitutionally limits the Legislature’s
authority to enact related but distinct legislation. The trial court
found that section 7465(c)(3) passes constitutional muster but
concluded that section 7465(c)(4) is invalid because it violates the
separation of powers by intruding on the Legislature’s powers.
As noted ante, section 7465 represents an exercise of the
voters’ power under article II, section 10(c) “to decide whether or
not the Legislature can amend or repeal initiative statutes. This
power is absolute and includes the power to enable legislative
amendment subject to conditions attached by the voters.”
(California Common Cause v. Fair Political Practices Com. (1990)
221 Cal.App.3d 647, 652, italics omitted.) For lack of a better
39
term, we will refer to the area of law in which an initiative
constrains the Legislature’s legislative actions as the initiative’s
article II, section 10(c) shadow.
A. Facial challenge
Interveners and the state first contend the trial court erred
in holding section 7465(c)(4) invalid because plaintiffs fail to meet
the standard for facial challenges, as they cannot show that
section 7465(c)(4) is unconstitutional in all or almost all of its
applications. We disagree.
“The standard for a facial constitutional challenge to a
statute is exacting. It is also the subject of some uncertainty.”
(Today’s Fresh Start, Inc. v. Los Angeles County Office of
Education (2013) 57 Cal.4th 197, 218.) In the stricter
formulation of the standard, “legislation is invalid only if it
presents a total and fatal conflict with applicable constitutional
prohibitions.” (T-Mobile West LLC v. City and County of San
Francisco (2019) 6 Cal.5th 1107, 1117, fn. 6.) This standard
requires a challenger to “ ‘ “ ‘establish that no set of
circumstances exists under which the [law] would be valid.’ ” ’ ”
(American Civil Rights Foundation v. Berkeley Unified School
Dist. (2009) 172 Cal.App.4th 207, 216.) But other cases have
“ ‘applied a more lenient standard, asking whether the statute is
unconstitutional “in the generality or great majority of cases.” ’ ”
(California School Boards Assn. v. State of California (2019)
8 Cal.5th 713, 724.) “Either way, we consider only the text and
purpose of the statute, and ‘[plaintiffs] cannot prevail by
suggesting that in some future hypothetical situation
40
constitutional problems may possibly arise as to the particular
application of the statute.’ ” (Ibid., italics omitted.)
Interveners press for the application of the stricter
standard and argue it is not met because they can imagine
collective bargaining laws that would constitute amendments to
Proposition 22, demonstrating that section 7465(c)(4) has at least
some constitutional applications. For example, interveners argue
that section 7465(c)(4) would be constitutional as applied to a
statute authorizing mandatory collective bargaining over
minimum hours that app-based drivers must work (see Chamber
of Commerce, supra, 890 F.3d at pp. 777–778), since such a
statute would amend Proposition 22 by taking away “the
individual right of every app-based rideshare and delivery driver
to have the flexibility to set their own hours for when, where, and
how they work.” 13 (§ 7450, subd. (b).)
As they did below, plaintiffs argue that by attempting to
define what constitutes an amendment, sections 7465(c)(3) and
(4) intrude on the judiciary’s power to define Proposition 22’s
article II, section 10(c) shadow. Interveners’ contention that
plaintiffs’ separation of powers challenge is not facial ignores this
13 Plaintiffs also argue that section 7465(c)(4) would
prohibit the Legislature from allowing the Labor Commissioner
or some other public entity to represent individual app-based
drivers as to their contractual relationships with transportation
and delivery network companies. Interveners do not offer
examples of laws authorizing public representation of individual
drivers that section 7465(c)(4) could constitutionally define as
amendments to Proposition 22. Since the parties focus primarily
on section 7465(c)(4) as a measure directed at collective
bargaining, we do the same.
41
aspect of plaintiffs’ argument, which does not depend on the
content of any specific piece of legislation; rather, the mere
existence of sections 7465(c)(3) and (4) completes the alleged
intrusion. In other words, even if sections 7465(c)(3) and (4)
would correctly declare some statutes to be amendments to
Proposition 22, sections 7465(c)(3) and (4) would still violate the
judiciary’s exclusive right to make such determinations.
Plaintiffs’ separation of powers challenge based on this intrusion
on the judiciary’s power is therefore a facial one, even under
interveners’ argument.
As for plaintiffs’ argument that section 7465(c)(4) intrudes
on the Legislature’s power, plaintiffs concede that interveners’
hypothetical statute authorizing mandatory collective bargaining
over minimum hours that app-based drivers must work would
constitute an amendment to Proposition 22. 14 But they argue it
would be an amendment even in the absence of section 7465(c)(4),
so section 7465(c)(4) would be irrelevant in such a scenario. They
urge us to disregard such situations and focus only on scenarios
in which section 7465(c)(4) would be relevant. Thus, plaintiffs’
facial attack on section 7465(c)(4) is more accurately described as
an attack on section 7465(c)(4) in all instances in which it would
declare a collective bargaining statute to be an amendment of
14 In the trial court, plaintiffs contended that section
7465(c)(3) also intrudes on the Legislature’s power, not just
section 7465(c)(4). The trial court rejected this argument, and
plaintiffs do not renew it on appeal. We express no opinion on
this point or on whether such an argument would be cognizable
as a facial challenge in the same manner as plaintiffs’ contention
that section 7465(c)(4) intrudes on the Legislature’s power.
42
Proposition 22 that would not otherwise constitute an
amendment. Or, using the terminology we have adopted here, it
is an attack on section 7465(c)(4) in every instance in which it
affects Proposition 22’s article II, section 10(c) shadow.
The Supreme Court endorsed plaintiffs’ approach in a
slightly different context in American Academy of Pediatrics v.
Lungren (1997) 16 Cal.4th 307 (plur. opn. of George, C. J.).
There, the Supreme Court noted, “ ‘[l]egislation is measured for
consistency with the Constitution by its impact on those whose
conduct it affects. . . . The proper focus of constitutional inquiry is
the group for whom the law is a restriction, not the group for
whom the law is irrelevant.’ ” (Id. at p. 345.)
American Academy of Pediatrics v. Lungren paired this
rationale with its discussion of the more lenient standard for
facial challenges, which some Courts of Appeal view as applicable
only in fundamental rights cases, but the Supreme Court
continues to treat the standard of review for facial challenges as
generally unsettled. (E.g., T-Mobile West LLC v. City and County
of San Francisco (2016) 3 Cal.App.5th 334, 345, fn. 12 [rejecting
application of lenient standard as appropriate only for First
Amendment and abortion rights], affd. (2019) 6 Cal.5th 1107;
T-Mobile West LLC v. City and County of San Francisco, supra, 6
Cal.5th at p. 1117, fn. 6 [declining to settle on a “precise
formulation of the applicable standard”].) Because plaintiffs’
argument fits squarely within the rationale expressed in
American Academy of Pediatrics v. Lungren for measuring the
constitutionality of statutes by looking only at the behavior they
43
restrict, and nothing about that rationale is logically or
inherently related to fundamental rights or the more lenient
facial standard, American Academy of Pediatrics v. Lungren
supports our conclusion that plaintiffs’ challenge to section
7465(c)(4) may proceed as a facial attack. 15 (Cf. San Francisco
Apartment Assn. v. City and County of San Francisco (2016)
3 Cal.App.5th 463, 486–488 [finding that plaintiffs had mounted
a successful facial challenge to a local ordinance on preemption
grounds by focusing on the instances where the local ordinance
would in fact impermissibly impact plaintiffs’ exercise of state
law rights, not the ordinance in isolation].)
Our acceptance of plaintiffs’ challenge as a facial one does
not, as interveners argue, make a facial challenge the easiest
challenge to assert rather than the hardest by ignoring the
constitutional applications of the statute and focusing only on the
unconstitutional ones. We are not ignoring the constitutional
applications of the statute, but rather ignoring the applications
where the statute would be irrelevant or, perhaps more
accurately, immaterial. If every instance in which section
7465(c)(4) has a material impact is also one in which it is
unconstitutional, then the law is facially infirm, even if it could
be constitutionally applied as surplusage in other instances.
15 Although it is plaintiffs’ burden to establish facial
invalidity of section 7465(c)(4), we note that interveners and the
state have not rebutted plaintiffs’ challenge by pointing to any
instance in which section 7465(c)(4) could constitutionally apply
to legislation that would fall outside the “natural” article II,
section 10(c) shadow.
44
B. Ripeness
Interveners and the state next contend plaintiffs’
separation of powers challenge is not ripe because the Legislature
has not enacted any legislation that might constitute an
amendment to Proposition 22. “The ripeness requirement, a
branch of the doctrine of justiciability, prevents courts from
issuing purely advisory opinions. [Citation.] It is rooted in the
fundamental concept that the proper role of the judiciary does not
extend to the resolution of abstract differences of legal opinion. It
is in part designed to regulate the workload of courts by
preventing judicial consideration of lawsuits that seek only to
obtain general guidance, rather than to resolve specific legal
disputes. However, the ripeness doctrine is primarily bottomed
on the recognition that judicial decisionmaking is best conducted
in the context of an actual set of facts so that the issues will be
framed with sufficient definiteness to enable the court to make a
decree finally disposing of the controversy. On the other hand,
the requirement should not prevent courts from resolving
concrete disputes if the consequence of a deferred decision will be
lingering uncertainty in the law, especially when there is
widespread public interest in the answer to a particular legal
question.” (Pacific Legal Foundation v. California Coastal Com.
(1982) 33 Cal.3d 158, 170.)
The Supreme Court has applied a two-prong test for
deciding whether a dispute is ripe, examining “ ‘the fitness of the
issues for judicial decision and the hardship to the parties of
withholding court consideration.’ ” (Pacific Legal Foundation v.
45
California Coastal Com., supra, 33 Cal.3d at pp. 171–174, italics
omitted.) “ ‘Under the first prong, the courts will decline to
adjudicate a dispute if “the abstract posture of [the] proceeding
makes it difficult to evaluate . . . the issues” [citation], if the court
is asked to speculate on the resolution of hypothetical situations
[citation], or if the case presents a “contrived inquiry” [citation].
Under the second prong, the courts will not intervene merely to
settle a difference of opinion; there must be an imminent and
significant hardship inherent in further delay.’ ” (Communities
for a Better Environment v. State Energy Resources Conservation
& Development Com. (2017) 19 Cal.App.5th 725, 733–734.)
The fitness prong of this test indicates plaintiffs’ separation
of powers challenge to the initiative is ripe because the voters
have already approved Proposition 22 and, as discussed ante,
plaintiffs’ challenge to it is a facial one for which concrete facts
are unnecessary. (Alliance for Responsible Planning v. Taylor
(2021) 63 Cal.App.5th 1072, 1082 [“Nothing precludes resolution
of the controversy, as the facial allegation does not depend on the
application of the measure to a particular petitioner or future
County interpretation”].) Interveners emphasize that the
Legislature has not enacted a statute that plausibly amends
Proposition 22, so there is no statutory text to interpret to
determine whether it changes the scope and effect of the
initiative. But plaintiffs contend the definitions of amendments
in sections 7465(c)(3) and (4) violate the separation of powers on
their face, so there is no need to wait for the Legislature to enact
a specific statute to evaluate plaintiffs’ arguments.
46
The hardship prong also militates in favor of finding
ripeness. Sections 7465(c)(3) and (4) are in effect, so they are
already intruding on the judiciary’s powers. Section 7465(c)(4)
can also affect the Legislature’s and stakeholders’ political
calculations, including those of labor organizations like plaintiff
SEIU, about whether to try to pass a law allowing app-based
drivers to collectively bargain or authorizing some other form of
representation of individual drivers. The only three avenues for
such legislation to become effective would be if the Legislature
submits the legislation to the voters for approval, which is a
costly undertaking; if the Legislature approves the legislation by
a seven-eighths majority, which is a bar so high as to be virtually
insurmountable; or if the Legislature passes such legislation by
less than a seven-eighths majority and the courts subsequently
agree that section 7465(c)(4) is unconstitutional, which is both
time-consuming and difficult to predict. Given these problematic
paths to effective legislation, the most likely outcome is that
legislators would not even undertake the effort. This is a
significant hardship justifying judicial resolution at this stage.
Interveners and the state argue that this type of chilling
effect is insignificant because the Legislature regularly enacts
laws even though the courts might find them to be amendments
to initiatives. (See, e.g., Amwest Surety Ins. Co. v. Wilson (1995)
11 Cal.4th 1243, 1261 [rejecting Legislature’s stated claim that a
statute merely clarified the scope of an initiative].) However,
interveners do not cite any law requiring a seven-eighths
majority for passage, which imposes an apparently uniquely high
47
barrier to legislative action. Besides, even if the chilling effect
were not a significant hardship, the concrete nature of the
parties’ arguments here would still make the issue ripe for our
resolution. The Supreme Court has held that “the ripeness
requirement does not prevent [the courts] from resolving a
concrete dispute if the consequence of a deferred decision will be
lingering uncertainty in the law, especially when there is
widespread public interest in the answer to a particular legal
question.” (Hunt v. Superior Court (1999) 21 Cal.4th 984, 998.)
Because the parties’ dispute has enough specificity for judicial
resolution at present, the uncertainty in the law that sections
7465(c)(3) and (4) create, as well as the public interest in the
validity of Proposition 22, counsel in favor of proceeding to the
merits of plaintiffs’ arguments.
C. Merits
Having disposed of interveners’ and the state’s prefatory
arguments, we turn now to the merits of plaintiffs’ separation of
powers challenge to sections 7465(c)(3) and (4).
1. Intrusion on judicial powers
Plaintiffs first argue that sections 7465(c)(3) and (c)(4)’s
definitions of amendments intrude on the judiciary’s power
because only the judiciary has the authority to say what
constitutes an amendment within the meaning of article II,
section 10(c).
“The powers of state government are legislative, executive,
and judicial. Persons charged with the exercise of one power may
not exercise either of the others except as permitted by this
48
Constitution.” (Cal. Const., art. III, § 3.) “ ‘The judicial power is
conferred upon the courts by the Constitution and, in the absence
of a constitutional provision, cannot be exercised by any other
body.’ ” (McClung v. Employment Development Dept. (2004)
34 Cal.4th 467, 472.) “[I]t is well established that it is a judicial
function to interpret the law, including the Constitution.”
(Schabarum v. California Legislature (1998) 60 Cal.App.4th
1205, 1213; accord, Raven v. Deukmejian (1990) 52 Cal.3d 336,
354 [“ ‘The judiciary, from the very nature of its powers and
means given it by the Constitution, must possess the right to
construe the Constitution in the last resort’ ”].)
Because the definitions in sections 7465(c)(3) and (4)
constitute an attempt to define the boundaries of Proposition 22’s
article II, section 10(c) shadow, sections 7465(c)(3) and (4) on
their face intrude on the judiciary’s authority to define the
meaning of “amendment” in that section of the Constitution. The
trial court determined otherwise because it read sections
7465(c)(3) and (4) as defining “amendment” only for purposes of
the optional, seven-eighths majority procedure in section 7465,
subdivision (a), not for article II, section 10(c). The trial court is
correct that sections 7465(c)(3) and (4) only govern which
amendments must comply with the conditions established in
section 7465, subdivision (a). But section 7465, subdivision (a)’s
power to set conditions on the Legislature’s enactment of future
legislation is the authority conferred by article II, section 10(c).
If a statute does not qualify as an amendment of Proposition 22
within the meaning of the Constitution, the Legislature need
49
neither secure the voters’ approval nor comply with the
conditions in section 7465, subdivision (a). Sections 7465(c)(3)
and (4)’s definitions of “amendment” as used in section 7465,
subdivision (a) are thus necessarily also an attempt to define
“amendment” as used in article II, section 10(c), which is
impermissible because such authority rests solely with the
judiciary.
2. Intrusion on legislative powers
Plaintiffs’ argument that section 7465(c)(4) intrudes on the
Legislature’s authority, with which the trial court agreed, is
slightly more intricate but also has merit. In determining
whether the Legislature has intruded on the voters’ initiative
power, courts have devised several different definitions of what
constitutes an amendment of an initiative under article II,
section 10(c). Decisions have defined an amendment variously as
“ ‘a legislative act designed to change an existing initiative
statute by adding or taking from it some particular provision’ ”
(People v. Superior Court (Pearson) (2010) 48 Cal.4th 564, 571),
one that “prohibits what [an] initiative authorizes, or authorizes
what the initiative prohibits” (ibid.), or one that “changes its
scope and effect” (Proposition 103 Enforcement Project v.
Quackenbush (1998) 64 Cal.App.4th 1473, 1486). But the
Supreme Court has cautioned against using too broad of a
formulation because, “despite the strict bar on the Legislature’s
authority to amend initiative statutes, judicial decisions have
observed that this body is not thereby precluded from enacting
laws addressing the general subject matter of an initiative. The
50
Legislature remains free to address a ‘ “related but distinct
area” ’ [citations] or a matter that an initiative measure ‘does not
specifically authorize or prohibit.’ ” (People v. Kelly (2010) 47
Cal.4th 1008, 1025–1026 & fn. 19, italics omitted (Kelly).) This
admonition implicitly recognizes that the definition of an
amendment for the purposes of article II, section 10(c) operates
both to prevent the Legislature from “ ‘ “undoing what the people
have done, without the electorate’s consent” ’ ” (Kelly, at p. 1025)
and to demarcate the boundaries of the restrictions that an
initiative places on the Legislature’s broad authority to legislate.
Collective bargaining legislation would not necessarily
amend Proposition 22 under any of these definitions. Apart from
section 7465(c)(4), no other provision of Proposition 22 directly
concerns the procedures for driver representation or collective
bargaining. The mere classification of drivers as independent
contractors is not determinative of their ability to collectively
bargain, as independent contractors can, in some circumstances,
collectively bargain. (See Chamber of Commerce, supra, 890 F.3d
at pp. 780–790; Welf. & Inst. Code, §§ 10420.5, 10423
[authorizing family child care providers to appoint organizations
to represent them without making such providers employees].)
Likewise, the benefits and obligations that Proposition 22
provides and imposes are established as minimums, not
maximums, so nothing prevents drivers from negotiating for
more, collectively or individually. (See, e.g., §§ 7454, subd. (a)(1)–
(2) [requiring health care subsidy “greater than or equal to”
certain amounts], 7455, subd. (a) [setting “minimum” coverages
51
for insurance], 7458, subd. (e) [nothing in statute “shall be
interpreted to prevent a network company from imposing
additional standards relating to criminal history”].) By extending
Proposition 22’s article II, section 10(c) shadow to bar legislation
on subjects which Proposition 22 does not otherwise directly
address, section 7465(c)(4) intrudes on the Legislature’s authority
to address a “ ‘ “related but distinct area” ’ ” or a matter that
Proposition 22 “ ‘does not specifically authorize or prohibit.’ ”
(Kelly, supra, 47 Cal.4th at pp. 1025–1026, italics omitted; cf.
People v. Nash (2020) 52 Cal.App.5th 1041, 1059–1060
[legislation changing the bases for murder liability did not amend
initiative that mandated increased sentences for murder
convictions].) On its face, section 7465(c)(4) therefore violates the
separation of powers for this reason as well.
3. Interveners’ and the state’s arguments
Interveners do not defend sections 7465(c)(3) and (4) as
written and instead downplay them by construing them as
merely statements of intent, precatory declarations of the voters’
views of what constitutes an amendment. 16 Interveners
analogize sections 7465(c)(3) and (4) to provisions the Legislature
commonly deploys when amending initiative statutes. They cite,
for example, Amwest Surety Ins. Co. v. Wilson, supra, 11 Cal.4th
16 Although plaintiffs renewed their separation of powers
challenge to section 7465(c)(3) in their respondents’ brief,
interveners do not mention section 7465(c)(3) in their reply and
discuss only section 7465(c)(4). But interveners defended section
7465(c)(3) and (4) in the trial court on the same basis on which
they defend section 7465(c)(4) on appeal, so we mention both
section 7465(c)(3) and (4) in our discussion of their argument.
52
at page 1260, which dealt with a statute declaring “that the act
‘furthers the purpose of Proposition 103 by clarifying the
applicability of the proposition to surety insurance.’ ” (See
Proposition 103 Enforcement Project v. Quackenbush, supra,
64 Cal.App.4th at p. 1481 [considering a statute stating, “ ‘The
Legislature finds and declares that this statute furthers the
purpose of Proposition 103,’ ” italics omitted].)
Interveners’ construction of sections 7465(c)(3) and (4) is
not plausible. Questions of purpose are suitable for precatory
declarations, since they turn on questions of legislative intent
and will vary from one initiative to another. The Legislature’s or
the voters’ input could be relevant when discerning such intent.
Thus, there is no apparent issue with section 7465(c)(1)’s
definition of Proposition 22’s purposes and section 7465(c)(2)’s
declaration that a statute that amends section 7451—which
defines app-based drivers as independent contractors—does not
further those purposes.
Unlike sections 7465(c)(1) and (2) or the legislative
statements that interveners cite, however, sections 7465(c)(3) and
(4) do not concern Proposition 22’s purposes or what would
further those purposes. Sections 7465(c)(3) and (4) address the
distinct question of what constitutes an amendment of
Proposition 22. (See O.G. v. Superior Court (2021) 11 Cal.5th 82,
99–100 [issues of whether a statute amends an initiative and
whether amendments further initiative’s purpose are separate].)
That question is governed by standards announced in numerous
judicial decisions over the years. By purporting to declare what
53
types of enactments would constitute amendments subject either
to the constitutional alternative of voter approval or section 7465,
subdivision (a)’s strict seven-eighths majority requirement (and
associated procedural conditions), sections 7465(c)(3) and (4) seek
to bypass this jurisprudence or declare what satisfies it, to the
exclusion of the courts. To the extent that section 7465(c)(4)
seeks to cast a broader article II, section 10(c) shadow than the
otherwise applicable judicial definitions of legislative
amendments, it also intrudes on the Legislature’s authority.
Interveners also contend that plaintiffs’ argument that
section 7465(c)(4) intrudes on the Legislature’s authority is a
single-subject argument in disguise and an attempt to sidestep
the lenient single-subject standard in favor of a “stricter (but
undefined and unprecedented) subject-based limitation on the
initiative power.” They contend that if section 7465(c)(4) satisfies
the single-subject rule, there is no basis to distinguish it from the
rest of the initiative for purposes of the separation of powers
analysis. But the Supreme Court in Kelly, supra, 47 Cal.4th at
pages 1025–1026 and footnote 19, recognized the “related but
distinct” standard and both the Supreme Court and the Courts of
Appeal continue to apply it, so it is hardly undefined or
unprecedented. (E.g., People v. Superior Court (Pearson), supra,
48 Cal.4th at p. 573; People v. Steward (2021) 63 Cal.App.5th
895, 899.) In addition, the single-subject rule and the separation
of powers are analytically distinct and serve different purposes,
so there is no reason to allow the former to swallow the latter.
The single-subject analysis intentionally “is applied ‘in an
54
accommodating and lenient manner so as not to unduly
restrict . . . the people’s right to package provisions in a single bill
or initiative.’ ” (Briggs, supra, 3 Cal.5th at p. 829.) By contrast,
the separation of powers doctrine is intended to protect the
Legislature’s sphere of control, to ensure that it “remains free to
enact laws addressing the general subject matter of an initiative,
or a ‘related but distinct area’ of law that an initiative measure
‘does not specifically authorize or prohibit.’ ” (Kelly, at p. 1026,
fn. 19, italics omitted.) Our agreement with interveners that
Proposition 22 complies with the single-subject rule does not free
us of our obligation to enforce the separation of powers.
Interveners further argue that prohibiting the voters from
expressing their views on the types of legislation that would be
subject to an initiative’s amendment process will discourage
voters from allowing amendments at all in the future, but we are
not convinced. Sections 7465(c)(3) and (4) are apparently unique
in the annals of initiative statutes. Voters chose to permit
amendments of initiatives for decades before Proposition 22. It
seems likely they will continue do so even though we now declare
sections 7465(c)(3) and (4) invalid. We also question whether
voters allow the Legislature to amend their initiative measures
as a sort of gift or a consolation prize, as interveners’ argument
implies. Rather, voters likely permit amendments so that the
Legislature can close loopholes, fix problems, and tweak initiative
statutes to meet unexpected circumstances. Declaring sections
7465(c)(3) and (4) to be unconstitutional attempts to expand
55
Proposition 22’s article II, section 10(c) shadow will not eliminate
or reduce these incentives.
For its part, the state at least reads sections 7465(c)(3)
and (4) as exactly what they purport to be: attempts to define
certain types of legislation that will constitute amendments of
the initiative. But the state strays when it maintains that
sections 7465(c)(3) and (4) are nonetheless proper. The state
contends that legislation can amend an initiative even without
altering its text and Proposition 22 regulates collective
bargaining (albeit without saying so directly), so section
7465(c)(4) appropriately declares that new legislation authorizing
collective bargaining would change the initiative’s effect. 17 The
state believes Proposition 22 regulates app-based drivers’ ability
to collectively bargain by classifying them as independent
contractors, who, as a matter of law, cannot collectively bargain.
The state does not meaningfully respond to plaintiffs’
argument that sections 7465(c)(3) and (c)(4) interfere with the
judiciary’s role to determine what constitutes an amendment
within the meaning of article II, section 10(c). And the state’s
attempt to rebut plaintiffs’ argument regarding section
7465(c)(4)’s infringement on the Legislature’s powers is flawed.
We have no quarrel with the principle that legislation can amend
an initiative without expressly changing its wording, but that
17 While the state treats section 7465(c)(4) as dealing only
with collective bargaining, by its terms the provision would also
apply to legislation authorizing other forms of representation of
app-based drivers, including representation on an individual
basis. (See fn. 13, ante.)
56
principle has no application here. As discussed ante, Proposition
22 does not directly regulate collective bargaining. And as noted,
independent contractors are not entirely barred from collectively
bargaining, so long as the Legislature enacts a law that satisfies
the requirements for state action immunity to antitrust law.
Section 7465(c)(4) therefore extends more broadly than
Proposition 22’s natural article II, section 10(c) shadow.
Additionally, as several election law professors point out in an
amicus brief, the voters would have had little reason to obliquely
call out potential collective bargaining statutes in section
7465(c)(4) as amendments to the initiative if Proposition 22’s
substantive provisions already addressed that issue. If there
were any doubt about whether the initiative’s classification of
app-based drivers as independent contractors foreclosed the
possibility of collective bargaining, the far more direct way to
address the issue would be to add a provision expressly saying so.
The only discernable reason to include section 7465(c)(4) was to
expand the scope of the initiative’s article II, section 10(c) shadow
beyond Proposition 22’s substantive provisions.
In a variation on the state’s argument, interveners contend
in their reply brief that we could interpret section 7465(c)(4) itself
as a form of substantive regulation. In this view, because app-
based drivers cannot collectively bargain in the absence of an
authorizing statute and there is presently no such law, section
7465(c)(4)’s restriction on the enactment of such a law effectively
locks in place the status quo of drivers not being able to
collectively bargain.
57
Treating section 7465(c)(4) as equivalent to a direct
pronouncement that app-based drivers cannot collectively
bargain presents even more difficulties than the other
interpretations. To begin with, this approach is inconsistent with
the text of the initiative. Section 7465(c)(4) is the only section
contained in an article titled “Amendment.” (People v. Garfield
(1985) 40 Cal.3d 192, 199 [titles are “a useful guide in
determining the intended scope of legislation,” but not the only
one].) And by its terms, section 7465(c)(4) discusses only the
content of future legislation. It does not affirmatively state that
app-based drivers cannot collectively bargain or that the
Legislature cannot allow a public entity to represent them on an
individual basis.
Even if section 7465(c)(4)’s text could be read to imply that
drivers may not collectively bargain, there is no indication that
the voters intended section 7465(c)(4) to operate in this fashion.
“ ‘ “[I]n the case of a voters’ initiative statute . . . we may not
properly interpret the measure in a way that the electorate did
not contemplate: the voters should get what they enacted, not
more and not less.” ’ ” (People v. Valencia (2017) 3 Cal.5th 347,
375.) Section 7465(c)(4) does not use the phrase “collective
bargaining,” even though all the parties here acknowledge that
restricting collective bargaining was the provision’s aim. Nor
does such an intent appear in Proposition 22’s prefatory findings
and statements of purpose. The Attorney General did not
identify such a restriction as one of the initiative’s effects, either
in the original circulating title and summary or in the revised
58
title and summary included in the voter guide. (Proposition 22,
at p. 2, available at
[as of
Mar. 13, 2023]; Voter Guide, supra, title and summary of Prop.
22, p. 56.) Nor did the Legislative Analyst. (Voter Guide, supra,
analysis of Prop. 22, pp. 56–57.) Nor did the arguments for and
against the initiative. 18 (Id. at pp. 58–59.)
To paraphrase our Supreme Court’s remarks about a
different initiative, “[w]e recognize that the materials in the
ballot pamphlet may not touch on every aspect of an initiative, no
18 Interveners ask us to judicially notice a tweet from the
campaign against Proposition 22 and a webpage maintained by
that campaign that they claim demonstrate the voters
understood that Proposition 22 would affect app-based drivers’
ability to collectively bargain. Assuming for the sake of
argument that these materials are properly subject to notice,
they do not change our analysis. These materials predicted that
“Proposition 22 would make it almost impossible for workers to
have legal protections if they want to collectively bargain.” But
they did not describe whether this would come from classifying
app-based drivers as independent contractors, thereby requiring
an act by the Legislature to enable collective bargaining, or from
the section 7465(c)(4) limitations on such legislation. The tweet
and webpage therefore do not demonstrate that voters
understood the effect of section 7465(c)(4).
Interveners also ask us to notice a webpage from an online
encyclopedia that they contend demonstrates that Proposition 22
received widespread publicity and public discussion. This
webpage was not submitted to the trial court, so we deny this
request. (Haworth v. Superior Court (2010) 50 Cal.4th 372, 379,
fn. 2 [“ ‘Reviewing courts generally do not take judicial notice of
evidence not presented to the trial court’ absent exceptional
circumstances”].) This webpage also does not demonstrate
anything about which specific aspects of Proposition 22 received
publicity or discussion.
59
matter how minor. ‘A statute, of course, must prevail over any
summary. Were it not so, no statute could ever be enacted whole
and entire. For every summary, by definition, is incomplete.’
[Citation.] When, for example, an initiative contains a clear and
unambiguous provision that, because of its relatively limited
significance, is not mentioned in ballot summary or arguments,
the absence of such a reference will not nullify its effectiveness.
Here, however, the language of Proposition [22] is not free from
ambiguity. And the application of its definition of [an
amendment regulating app-drivers’ ability to collectively bargain]
is a matter of such substantial import that the voters could
reasonably expect that, if Proposition [22] applied [in such a
way], the ballot materials would mention it.” (People v. Valencia,
supra, 3 Cal.5th at p. 364, fn. 6.)
Finally, we are reluctant to accept section 7465(c)(4) as an
indirect form of substantive regulation because of the troubling
implications for the initiative process. Treating section
7465(c)(4)’s ostensibly procedural regulation of the amendment
process as equivalent to a direct statement that app-based
drivers cannot collectively bargain would encourage
gamesmanship and reward initiative proponents for drafting
confusing, or even outright misleading, initiatives. Adhering to
the plain text of initiatives’ enactments will instead help ensure
that proponents draft clear and intelligible proposals for the
voters’ consideration.
The state and interveners emphasize that we must
jealously guard the initiative power. But our duty to guard the
60
initiative power works both ways; “we guard this power with both
sword and shield. We must not only protect against interference
with its proper exercise, but must also strike down efforts to
exploit the power for an improper purpose.” (Widders v.
Furchtenicht (2008) 167 Cal.App.4th 769, 785.) We do the voters
no favors by upholding a provision that either unconstitutionally
intrudes on the powers of the judiciary and the Legislature or
whose effect was unclear and not called to their attention.
The state and interveners also invoke the doctrine of
constitutional avoidance, asking us to resolve all doubts in favor
of the validity of section 7465(c) and to construe it in a way that
avoids constitutional conflicts. (California Redevelopment Assn.
v. Matosantos (2011) 53 Cal.4th 231, 253 [courts avoid
“constitutional conflicts whenever possible by construing
legislative enactments strictly against the impairment of [courts’]
constitutional jurisdiction”]; Briggs, supra, 3 Cal.5th at pp. 857–
859 [construing plainly mandatory provision in initiative as
directive, in part to avoid separation of powers concerns].) But
this does not allow us to simply pretend that Proposition 22
regulates collective bargaining, as the state suggests, when the
text does not reasonably admit of such a construction. The
doctrine of constitutional avoidance does not allow us to clothe
the emperor in such a fashion.
We could perhaps construe section 7465(c)(4) as precatory,
as interveners urge, even though its plain language is to the
contrary. (See Briggs, supra, 3 Cal.5th at pp. 857–859.) But that
would not meaningfully change the result. If section 7465(c)(4)
61
were a mere declaration of the voters’ intent that the initiative
forbids app-based drivers from collectively bargaining, it might
survive constitutional scrutiny on its face. However, any court to
examine the question of whether Proposition 22 actually restricts
the Legislature’s authority to authorize collective bargaining
would very likely disregard such a statement of intent, since
nothing in the text of the rest of the initiative supports it, as
discussed ante. (O.G. v. Superior Court, supra, 11 Cal.5th at
p. 91 [“In discerning the purposes of a proposition, ‘we are guided
by, but are not limited to, the general statement of purpose found
in the initiative’ ”].) We see little reason to uphold section
7465(c)(4) by using the fig leaf of construing it as precatory, when
such a construction would render it just as ineffectual as
declaring it facially invalid.
In sum, we conclude that sections 7465(c)(3) and (4) are
facially invalid on separation of powers grounds because they
intrude on the judiciary’s authority to determine what constitutes
an amendment to Proposition 22, and section 7465(c)(4) fails for
the additional reason that it intrudes on the Legislature’s
authority by artificially expanding Proposition 22’s article II,
section 10(c) shadow. As the trial court ruled and the parties
agree, the proper remedy for the separation of powers violation is
to sever section 7465(c)(3) and (4) and allow the rest of
Proposition 22 to remain in effect, as the voters indicated they
wished. (§ 7467, subd. (a).)
62
DISPOSITION
The judgment is affirmed to the extent it declared sections
7465(c)(3) and (c)(4) invalid and to the extent the trial court
retained jurisdiction to consider a motion for attorney fees under
Code of Civil Procedure section 1021.5. In all other respects, the
judgment is reversed. The matter is remanded to the trial court
with instructions to enter a new judgment not inconsistent with
this opinion. All parties shall bear their own costs on appeal.
BROWN, Acting P. J.
WE CONCUR:
STREETER, J.
POLLAK, J. *
*Retired Presiding Justice of the Court of Appeal, First
Appellate District, Division Four, assigned by the Chief Justice
pursuant to article VI, section 6 of the California Constitution.
63
STREETER, J., Concurring and Dissenting.
I. Introduction
I concur in Justice Brown’s opinion except for one key
aspect of part II of her discussion for the majority addressing
article XIV, section 4 of the California Constitution, which vests
the Legislature with “plenary power, unlimited by any provision
of this Constitution, to create, and enforce a complete system of
workers’ compensation, by appropriate legislation . . . .” 1 My
disagreement on this point leads me to a different overall result.
I would affirm the judgment, but I prefer to go further. I believe
we must invalidate Proposition 22 2 in its entirety.
All legislative power—whether exercised by the
Legislature, or by the voters acting as “electors” when they enact
an initiative statute—must abide by constitutional limits, federal
and state. (In re Marriage Cases (2008) 43 Cal.4th 757, 851.)
In addition to the article III, section 3 separation of powers
problem identified by Justice Brown based on what she terms the
“shadow” effect of the initiative (tying the hands of the
Legislature in the distinct and unrelated area of collective
bargaining representation), Proposition 22 violates the California
Constitution on three other grounds as well.
1 All further undesignated citations to constitutional
articles and sections thereof are to the California Constitution.
2Business & Professions Code, sections 7448–7467, as
added by Proposition 22, approved by the voters, General
Election (Nov. 3, 2020).
1
First, by undermining the “complete system of workers’
compensation” that the people have expressly defined in article
XIV, section 4, Proposition 22 constitutes a sub rosa attempt to
amend the Constitution in the guise of statutory change. Second,
article XIV, section 4, delegates to the Legislature—not the voter
electors—specific lawmaking power to “create” and “enforce” a
“complete system of workers’ compensation.” This case presents
a direct conflict between the voter electors’ power to adopt
initiative statutes and the Legislature’s power under article XIV,
section 4, and on this record we must resolve that conflict in favor
of the Legislature. So long as the Legislature has acted within
the scope of its authority under article XIV, section 4, voter
electors may not countermand a prior determination by the
Legislature about what the Constitution requires. Third,
Proposition 22 usurps a “core” or “essential” governmental
function provided for in article XIV, section 4—judicial power—in
violation of article III, section 3.
II. Proposition 22 Violates Article XIV, Section 4,
in Two Respects
Unlike the petitioners in Independent Energy Producers
Assn. v. McPherson (2006) 38 Cal.4th 1020 (McPherson),
respondents do not argue that a specific constitutional grant of
lawmaking authority to the Legislature categorically precludes
any use of the initiative power to legislate on the subject
addressed by that grant. Their complaint is different. They say
Proposition 22 so fundamentally undermines the workers’
compensation system the Legislature created prior to 1918 that it
2
transgresses article XIV, section 4, and thus could only have been
adopted validly by ballot amendment to the Constitution.
I believe the argument is well taken.
Proposition 22 removes app-based drivers from the
constitutionally mandated workers’ compensation system and
substitutes a private accident insurance mandate. (Bus. & Prof.
Code, §§ 7451, 7455, subd. (a).) The linchpin of Proposition 22 is
the “independent contractor” definition in Business & Professions
Code section 7451. Proposition 22 reversed the Legislature’s
judgment on this issue 3 by redefining the scope of “employment,”
a term that is used in the language of article XIV, section 4. No
one disputes that the effect of the “independent contractor”
definition in Proposition 22 is to expel app-based drivers, as a
class, from the “complete system of workers’ compensation”
established by the Legislature more than a century ago. Because
the basic architecture of that system was literally written into
the Constitution in 1918, statutory changes to it must be closely
scrutinized for compatibility with article XIV, section 4.
Proposition 22 cannot withstand that scrutiny in two
respects, the first detailed in parts II.A.–II.B. and the second in
parts II.C.–II.D.
3 See Assembly Bill No. 5 (2019–2020 Reg. Sess.) (Assembly
Bill 5); former Labor Code § 2750.3, as added by Stats. 2019,
ch. 296, § 2; repealed and transferred to Lab. Code, §§ 2775, 2785.
3
A. Certain “Basic Features” of the “Complete System of
Workers’ Compensation” Mandated by Article XIV,
Section 4, Cannot Be Revised by Statute
Article XIV, section 4, by its express terms, defines in detail
what must be in a “complete system of workers’ compensation,”
and the new privatized scheme for app-based drivers falls far
short. There are no safety and health provisions; no provisions
for vocational training if a driver cannot return to work; no
compensation provisions for permanent disability; and no
provisions for an administrative body exercising judicial power to
resolve disputes expeditiously, with direct review in the Court of
Appeal. At best, app-based drivers are left with a breach of
contract remedy for any deprivation of workplace accident
benefits, and the value of even that is dubious since these
workers (like so many these days) must sign form contracts that
force them into arbitration, without access to judicial review.
Proposition 22 also permanently withdraws the
Legislature’s authority to address these deficiencies by restoring
article XIV, section 4’s protections to this class of workers.
It allows no legislative amendments unless passed by seven-
eighths vote of both houses of the Legislature, and even then any
amendment must be “consistent with, and further[] the purpose
of [Proposition 22].” (Bus. & Prof. Code, § 7465, subd. (a).) It also
provides that “[a]ny statute that amends Section 7451”—the
provision that removes app-based drivers from the workers’
compensation system (as well as from other worker protections)—
“does not further the purposes of ” Proposition 22. (Bus. & Prof.
4
Code, § 7465, subd. (c)(2).) As a result, Proposition 22 not only
adopts an incomplete system of workers’ compensation that fails
to meet article XIV, section 4’s constitutional requirements, it
prevents the Legislature from acting to rectify the situation.
The Constitution, plainly read, prohibits all of this. It is
well established that article XIV, section 4, places substantive
limits on any exercise of legislative power, whether exercised by
initiative statute or by enactment of the Legislature. The
definition of “complete system of workers’ compensation” in
article XIV, section 4, tracks the statutory workers’ compensation
system that existed in 1918. As early as 1922, the Supreme
Court made clear that the Legislature has no power to go beyond
the pre-1918 scheme and create new elements of a “complete
system of workmen’s compensation” not mentioned in the
constitutional language. (Yosemite L. Co. v. Industrial Acc. Com.
(1922) 187 Cal. 774, 780 (Yosemite Lumber) [construing former
article XX, section 21, adopted in November 1918, later
renumbered article XIV, section 4].) 4 That foundational
4 Former article XX, section 21 was repealed and readopted
as article XIV, section 4, without substantive change in 1976.
(City of Los Angeles v. Workers’ Comp. Appeals Bd. (2009)
179 Cal.App.4th 134, 142, fn. 3; see Six Flags, Inc. v. Workers’
Comp. Appeals Bd. (2006) 145 Cal.App.4th 91, 95 (Six Flags)
[“The constitutional enabling provision establishing the workers’
compensation scheme has remained the same since 1918 with
two exceptions: (1) a 1972 amendment adding the State of
California as a beneficiary entitled to workers’ compensation
benefits in some cases; and (2) a 1974 amendment making the
provision gender neutral, changing ‘workmen’ to ‘workers.’ ”].)
5
limitation precludes any legislative attempt to revise the basic
outline of the constitutionally mandated scheme for
compensating workers injured or killed while engaged in
“employment.” 5
Yosemite Lumber and other early cases applying this
principle to what is now article XIV, section 4, have continuing
force today. In the 1970’s and 1980’s the Supreme Court
confirmed that, while legislation consistent with the
constitutionally mandated pre-1918 statutory scheme is
permissible, statutory revisions altering the “basic features” of
the scheme are not. (Mathews v. Workmen’s Comp. Appeals Bd.
(1972) 6 Cal.3d 719, 734–735 (Mathews) [statute conditioning the
right to compensation on the absence of willful misconduct held
to be valid because it was consistent with “basic features” of the
pre-1918 statutory scheme]; see Hustedt v. Workers’ Comp.
5 Pacific G. & E. Co. v. Industrial Acc. Com. (1919) 180 Cal.
497, 500 (PG&E) (“the terms ‘employers,’ ‘employees’ and
‘employment’ as used in [former] section 21, article XX, of the
constitution, as amended in October, 1911, must be construed in
the light of their meaning at the time of the adoption of the
amendment, and cannot be extended by legislative definition, for
such extension would, in effect, be an amendment of the
constitution, if accepted as authoritative”); Employers’ Liability
Assur. Corp. v. Industrial Acc. Com. (1918) 179 Cal. 432, 437
[same]. In parts II.B. and II.D.2.–II.D.3. below, I explain the
history of article XIV, section 4, as it evolved from its original
form in 1911. Though some of the terminology changed in the
course of its evolution, the scheme was always founded on the
existence of an “employment” relationship. Each iteration of the
constitutional language, in 1911 and in 1918, uses the term
“employment” to designate the boundary of who is covered and
who is not.
6
Appeals Bd. (1981) 30 Cal.3d 329, 341–346 & fn. 11 (Hustedt)
[statute conferring power on Workers’ Compensation Appeals
Board to discipline attorneys invalid because it was not a “basic
feature” of the pre-1918 statutory scheme].)
Of course, “ ‘[a] statute cannot trump the Constitution.’ ”
(City of San Diego v. Shapiro (2014) 228 Cal.App.4th 756, 788.)
“[O]ur state Constitution is the highest expression of the will of
the people acting in their sovereign capacity as to matters of state
law. When the Constitution speaks plainly on a particular
matter, it must be given effect as the paramount law of the
state.” (Playboy Enterprises, Inc. v. Superior Court (1984)
154 Cal.App.3d 14, 28, citing Ex Parte Braun (1903) 141 Cal. 204,
211; People v. Parks (1881) 58 Cal. 624, 635; Dye v. Council of the
City of Compton (1947) 80 Cal.App.2d 486, 490; McMillan v.
Siemon (1940) 36 Cal.App.2d 721, 725.) Thus, voter electors
legislating by initiative, and the Legislature enacting statutes,
can only act within the powers afforded to them by the people in
article XIV, section 4. 6
6 To be sure, voter electors can always change the “basic
features” of the pre-1918 system of workers’ compensation by
ballot constitutional amendment. For example, former article
XX, section 21, the precursor to article XIV, section 4, authorized
the Legislature to create a system “to compensate . . . workers for
injury or disability, and their dependents for death incurred or
sustained by the said workers.” (Prop. 23, approved by the
voters, Gen. Elec. (Nov. 5, 1918); Ballot Pamp. text of Prop. 23 at
[as of Mar. 13, 2023].) The
California Supreme Court held in 1922 that these words did not
7
What this means here, as a practical matter, is that there
is a minimum constitutional baseline to our workers’
compensation system no statute can go below. If, after adoption
of article XIV, section 4—which codified in the Constitution itself
what a “complete system of workers’ compensation” meant prior
to 1918—the Legislature had attempted to create a different
statutory scheme by lopping off some of the “basic features” of
that system (Mathews, supra, 6 Cal.3d at p. 735), the statute
would have been invalid. Nothing could be more “basic” to the
system of workers’ compensation than the scope of the
“employment” relationship. Because Proposition 22 attempts to
redefine that term in a manner that is contrary to the laws of
workplace accident protection as “those laws . . . have existed
since 1911” (Mathews, at p. 735), I believe it is unconstitutional.
B. Article XIV, Section 4, Has Continuing Vitality Today
Insisting upon a narrow reading that gives article XIV,
section 4, no contemporary substance at all, the Attorney General
and the interveners (collectively, the Proposition 22 proponents)
authorize payment of compensation benefits to a state fund if the
worker died without dependents (Yosemite Lumber, supra,
187 Cal. at p. 782), a holding that was later extended to
invalidate awards to the estates of deceased workers (see
Commercial Cas. Ins. Co. v. Industrial Acc. Com. (1930) 211 Cal.
210, 215). Many years later, the voters amended article XIV,
section 4, to allow the Legislature to authorize payment of a
compensation award for a worker who dies without dependents
either to the state (Prop. 13, approved by the voters, Gen. Elec.
(Nov. 7, 1972); see Six Flags, supra, 145 Cal.App.4th at p. 97), or
to the worker’s estate (Prop. 14, approved by the voters, Prim.
Elec. (June 8, 1976)).
8
cite Mathews for the idea that the only purpose of article XIV,
section 4, was to provide constitutional armor against judicial
invalidation by Lochner era courts. 7 While there is a grain of
truth to this line of argument, the Proposition 22 proponents take
the idea too far. Our Supreme Court has indeed recognized that
the “sole purpose” of article XIV, section 4, was to put to rest “all
doubts” about the constitutionality of the existing statutory
workers’ compensation scheme in 1918. (Mathews, supra,
6 Cal.3d at p. 734; City and County of San Francisco v. Workers’
Comp. Appeals Bd. (1978) 22 Cal.3d 103, 114.) But the Mathews
court said nothing about Lochner or Lochnerism, and
understandably so, because there is considerably more to the
origin story of article XIV, section 4, than that.
For proper perspective, some historical background is
necessary. By 1918, our Legislature had passed three versions of
a statutory workers’ compensation scheme, each building on the
last one in sequence. These schemes—which were viewed as
“radical,” even “revolutionary,” at the time (Western Indemnity
Co. v. Pillsbury (1915) 170 Cal. 686, 692 (Western Indemnity I ))—
eliminated employers’ common law tort defenses against
employee workplace accident claims and created an
administrative system for adjudicating and enforcing such claims
on a no-fault liability basis. The first, passed in 1911 and known
as the Roseville Act (Stats. 1911, ch. 399, §§ 1–31, pp. 796–806),
7See Lochner v. New York (1905) 198 U.S. 45, disapproved,
West Coast Hotel Co. v. Parrish (1937) 300 U.S. 379.
9
was purely voluntary; the second, passed in 1913 and known as
the Boynton Act (Stats. 1913, ch. 176, §§ 1–92, pp. 279–320), was
compulsory. Refinements were made in the “workmen’s
compensation, insurance and safety act of 1917” (Stats. 1917,
ch. 586, §§ 1–74, pp. 831–879 (the 1917 Act)), which partially
repealed and restated the Boynton Act (1917 Act, at
pp. 831–832). 8 This series of enactments reflected the cumulative
statutory evolution of the pre-1918 workers’ compensation
system. These statutes are what the Mathews court refers to
when it speaks of the “basic features of those laws as they have
existed since 1911.” (Mathews, supra, 6 Cal. 3d at p. 735.) 9
Article XIV, section 4, evolved along a parallel path, also in
two steps. First, by ballot initiative in 1911—in fact, at the same
special election that brought us the powers of initiative,
8 The current workers’ compensation statutory scheme
appears in the Labor Code. (King v. CompPartners, Inc. (2018)
5 Cal.5th 1039, 1046 [“First created more than a century ago,
California’s workers’ compensation system is now governed by
the Workers’ Compensation Act (WCA; Lab. Code, § 3200 et seq.),
‘a comprehensive statutory scheme governing compensation given
to California employees for injuries incurred in the course and
scope of their employment.’ ”].)
9 The majority opinion appears to question whether the
Mathews and Hustedt courts actually meant what they said in
stating that the language of article XIV, section 4, codified the
“basic features” of the pre-1918 statutory scheme. (Maj. opn.,
ante, at p. 22, fn. 10.) But the constitutional language—which
these courts were simply describing when they said article XIV,
section 4, outlines the “basic features” of a “complete system of
workers’ compensation” (Hustedt, supra, 30 Cal.3d at pp. 342–
343, fns. 10 & 11)—speaks for itself.
10
referendum and recall—the voters laid a constitutional
foundation for the Roseville Act by adding former article XX,
section 21, to the Constitution. Framed in abbreviated language
compared to what later became article XIV, section 4, as we now
know it, that amendment gave the Legislature plenary power to
create and enforce a system of liability without fault for workers
injured in the workplace. Second, seven years later, in November
1918, the Legislature by joint resolution proposed the adoption of
an amendment to the 1911 version of former article XX,
section 21 (Mathews, supra, 6 Cal.3d at p. 733), this time adding
much greater specificity and detail. The 1918 amendment
augmenting former article XX, section 21, is in all material
respects what we now have in article XIV, section 4.
The amended language adopted in 1918 delineates with
particularity the Legislature’s authority to make “adequate
provisions” for relieving workers and their families of the
“consequences of any injury or death incurred or sustained by
workmen in the course of their employment, irrespective of the
fault of any party”; “full provision for securing safety in places of
employment”; “full provision for such medical, surgical, hospital
and other remedial treatment as is requisite to cure and relieve
from the effects of such injury”; “full provision for adequate
insurance coverage against liability to pay or furnish
compensation”; “full provision for otherwise securing the payment
of compensation”; and “full provision for vesting power, authority
and jurisdiction in an administrative body with all the requisite
governmental functions to determine any dispute or matter
11
arising under such legislation, . . .” (Former art. XX, § 21,
amended by Prop. 23, Gen. Election (Nov. 5, 1918).)
Of particular significance here—because this reflects the
superior position article XIV, section 4, enjoys relative to other
provisions within the Constitution, as well as the preeminent role
the Legislature has in making “social public policy” pertaining to
workers’ compensation vis-à-vis other governmental actors—the
amended language makes clear for the first time that (1) the
Legislature’s “plenary” power to do all of the above-listed things
is “unlimited by any provision of this constitution” and (2) the
“social public policy” reflected in the system of workers’
compensation established by the Legislature pursuant to this
authority is “binding upon all departments of the state
government.” (Former art. XX, § 21, amended by Prop. 23, Gen.
Election (Nov. 5, 1918).)
Stepping back for a moment to understand why a second
ballot amendment was necessary in 1918, we must look to case
law developments. In 1915, our Supreme Court upheld the
Boynton Act against constitutional attack in Western Indemnity I
over a vigorous dissent by Justice Henshaw, who predicted that
“the decision in this case will doubtless pass under the scrutiny of
the supreme court of the United States” (Western Indemnity I,
supra, 170 Cal. at p. 721 (dis. opn. of Henshaw, J.)). The
Henshaw dissent in Western Indemnity I was wide-ranging.
It attacked the Boynton Act as an unreasonable exercise of the
police power—a classic Lochner-style substantive due process
12
argument 10—but it raised equal protection and takings clause
issues as well. 11 A concurring opinion by Justice Shaw took the
position that the no fault liability provisions of the Boynton Act
were sustainable under the taxing power, but agreed that, in a
case where an employee’s injury was brought about by his own
negligence, the application of the Boynton Act would constitute
“an attempt to take private property from the owner for the
private use of another person . . . without lawful excuse or right
and without compensation.” (Western Indemnity I, supra, at
p. 735 (conc. opn. dubitante of Shaw, J., on rehg. den.)
Perhaps most importantly, the concurrence and the dissent
in Western Indemnity I together cast doubt on whether former
article XX, section 21, provided adequate support for the Boynton
Act as a matter of state law. 12 Justice Henshaw, who saw the
Boynton Act as an exercise in “socialistic paternalism” (Western
10 Western Indemnity I, supra, 170 Cal. at pages 711–716;
see Ives v. South Buffalo Ry. Co. (1911) 201 N.Y. 271 citing and
relying on Lochner v. New York, supra, 198 U.S. 45, disapproved,
West Coast Hotel Co. v. Parrish, supra, 300 U.S. 379, to
invalidate New York’s workers’ compensation statute.
11 Western Indemnity I, supra, 170 Cal. at page 716 (the
statute is “obnoxious to the constitutional provisions
guaranteeing equal protection . . . and forbidding confiscation”).
12 Western Indemnity I, supra, 170 Cal. at page 729
(dis. opn. of Henshaw, J.) (Boynton Act exceeds the Legislature’s
powers under former article XX, section 21, as adopted in 1911);
cf. Western Indemnity I, at page 735 (conc. opn. dubitante of
Shaw, J., on rehg. den.) (combination of taxing power and police
power insufficient to save Boynton Act from constitutional
invalidity where employee fault implicated in his own injury).
13
Indemnity I, supra, 170 Cal. at p. 724), articulated this view most
forcefully. His dissent hammered away at the theme that the
basis for the Boynton Act was not an exercise of the police power
at all, but instead was an unbridled expression of sovereign “fiat”
(Western Indemnity I, at p. 710 (dis. opn. of Henshaw, J.)) that
“places upon the employer, to the peril of his welfare and the loss
of all his property, the performance of a duty resting upon the
state itself.” (Id. at p. 723.) According to Justice Henshaw, the
“[l]egislature mistakenly put on this constitutional grant” a vast
and oppressive power that went well beyond its “declared limits.”
(Id. at p. 729.)
This backdrop in the California Supreme Court is
important here, but just as relevant are developments prior to
1918 in the United States Supreme Court. While concerns about
potential constitutional invalidation as a matter of substantive
due process under the United States Constitution were real in
1915, as Justice Henshaw’s dissent demonstrates, those concerns
had largely evaporated by 1918. The prediction of a grant of
certiorari in Western Indemnity I did not come true, and in a trio
of cases decided in the high court’s 1917 term, that court
addressed—and rejected—Lochner challenges to statutory
workers’ compensation systems of different kinds. (See Mountain
Timber Co. v. State of Washington (1917) 243 U.S. 219 [voluntary
scheme]; Hawkins v. Bleakly (1917) 243 U.S. 210 [voluntary
scheme]; New York Cent. R. Co. v. White (1917) 243 U.S. 188
14
[compulsory scheme].) White, in particular, was a turning point,
since it dealt with a compulsory scheme similar to California’s. 13
At both the federal and state level, what this brief review of
the pertinent case law shows is that, by the time the proposed
amendment to revise former article XX, section 21, came on the
ballot in November 1918, it was not Lochner that cast
constitutional doubt on California’s statutory workers’
compensation scheme. Rather, the source of doubt was the split
of opinion in Western Indemnity I over the adequacy of former
article XX, section 21, as a state constitutional foundation for
California’s workers’ compensation scheme. Given the state of
the law at the time, the 1918 amendment sought to bolster and
enhance the language of former article XX, section 21, as
originally adopted in 1911, in direct response to the Henshaw
view that that grant of lawmaking power was limitless and
unbounded. The objective was to anchor article XIV, section 4,
13 Following the decisions in these cases, in June of 1919
the high court brushed aside another attempt to challenge a state
workers’ compensation statute on substantive due process
grounds with the following words: “In view of our recent
decisions sustaining state laws imposing upon employers in the
hazardous industries responsibility in one form or another for the
consequences of injuries received by employees in the course of
the employment in the absence of fault on the employer’s part
. . . , little need now be said.” (New York Cent. R. Co. v. Bianc
(1919) 250 U.S. 596, 601 [citing New York Cent. R. Co. v. White,
supra, 243 U.S. 188 and Mountain Timber Co. v. State of
Washington, supra, 243 U.S. 219, among other cases].)
15
more strongly in our Constitution, independently of the police
power. 14
Accordingly, while it is true that the purpose of the 1918
amendment was to “remove all doubts as to the constitutionality
of then existing workmen’s compensation laws” (Mathews, supra,
6 Cal.3d at p. 733)—a remark which must be understood against
the backdrop of what the remaining doubts were when voters
went to the polls—it is an overstatement to go a step further and
suggest that the demise of Lochnerism in the mid-1930’s
transformed article XIV, section 4, into a constitutional dead
letter. In service of an argument that article XIV, section 4, may
safely be ignored today, that account of what happened places a
convenient expiration date on this provision of the Constitution.
But it is incomplete and misleading as history.
For proof, we need look no further than the cases
construing article XIV, section 4, to put substantive limitations
14 In the modern terminology of United States Supreme
Court jurisdiction, this would be called an effort to insulate the
decision in Western Indemnity I against federal constitutional
attack—on any ground—by ensuring that the vehicle for such a
future challenge would be a California decision resting on
“adequate and independent state grounds.” (E.g., Republican
Nat. Committee v. Burton (1982) 455 U.S. 1301, 1302.) As a
recipe to ward off federal review, this is not a theory I am reading
into the case law based on hindsight. It is a strategy that
actually played out in a case involving Arizona’s workers’
compensation scheme, where, just as we see with article XIV,
section 4, the challenged statute was based on a specific grant of
lawmaking power in that state’s constitution. (See Arizona
Employers’ Liability Cases (1919) 250 U.S. 400, 417–419.)
16
on legislative power, dating from as early as 1922 (Yosemite
Lumber, supra, 187 Cal. 774) to as recently as 1981 (Hustedt,
supra, 30 Cal.3d 329) and even more recently, 2006 (Six Flags,
supra, 145 Cal.App.4th 91), long after Lochner was gone. If we
take stock of the full history, article XIV, section 4, remains as
vital today as it was when adopted in 1918. It should be treated
with the seriousness and dignity it deserves as one of several
enduring constitutional achievements of the progressive reform
era, along with and adjacent to the powers of direct democracy.
Here, what that means is we must stand ready to strike down
any statute that exceeds the outer limits of legislative power
established by article XIV, section 4. I believe Proposition 22
crosses that line.
C. This Case Presents the Conflict of Legislative Powers
Issue Anticipated in Footnote 9 of the Supreme
Court’s Opinion in Independent Energy Producers
Assn. v. McPherson
1. The Conflict Here Is Between Concurrent Powers,
Not Coextensive Powers
Although the most glaring constitutional violation
presented here is the use of a ballot initiative to effect what
amounts to an amendment of the Constitution, a second, related
article XIV, section 4, violation requires the invalidation of
Proposition 22.
By placing the eligibility of app-based drivers for workers’
compensation benefits off-limits to amendment by the
Legislature, Proposition 22 seeks to override constitutional
responsibility delegated to the Legislature. The Proposition 22
17
proponents contend that, in striking the initiative down on that
basis, the trial court erred by construing the term “plenary” as
“exclusive.” This is not an accurate characterization of the trial
court’s ruling. The trial court saw a conflict between article XIV,
section 4, and article II, section 10(c), and resolved it in favor of
the Legislature. I believe it was correct to do so.
The deficiency I see here is structural. The delegated
power we are dealing with—the power to carry out a specific
constitutional task (i.e., to establish and maintain a “complete
system of workers’ compensation”)—was conferred on “the
Legislature” under article XIV, section 4, not on the voters acting
as “electors.” Because voter “electors” and the “Legislature”
share the police power, they may each legislate on the subject of
workers’ compensation, which is why our Supreme Court has
held that the article II, section 8(a) power to adopt initiative
statutes is “encompass[ed]” within the Legislature’s article XIV,
section 4 power. (McPherson, supra, 38 Cal.4th at p. 1025.)
But as the trial court correctly concluded, when voter
electors exercise the police power in a way that comes into
conflict with the Legislature’s article XIV, section 4, power, we
have the dilemma the Supreme Court flagged in footnote 9 of its
McPherson opinion. (McPherson, supra, 38 Cal.4th at p. 1044,
fn. 9.) In that footnote, the court carefully left open issues that
may be raised in a scenario where an initiative statute
“improperly conflicts with the Legislature’s exercise of its
authority” under a specific delegation of legislative power in the
Constitution. (McPherson, at p. 1044, fn. 9., original italics.)
18
Emphasizing that we must liberally construe the people’s
“initiative power” (Legislature v. Eu (1991) 54 Cal.3d 492, 501)
and “resolve any reasonable doubts in favor” of this right
(Brosnahan v. Brown (1982) 32 Cal.3d 236, 241, italics omitted),
the Proposition 22 proponents tell us there is no genuine conflict
here because the legislative power of the voters is at least
coextensive with and in some respects “greater than that of the
Legislature” (Rossi v. Brown (1995) 9 Cal.4th 688, 704). But
because we are dealing with a ballot statute that treads on a
constitutionally delegated power assigned specifically to the
Legislature, sweeping statements about the people’s “precious
rights” of initiative (Amador Valley Joint Union High Sch. Dist.
v. State Bd. of Equalization (1978) 22 Cal.3d 208, 248) provide
little guidance.
Precision is crucial. Voter electors are not the “people”
when they legislate under article II, section 8(a). They exercise
sovereign power as the “people” only when they approve ballot
constitutional amendments. Although voter electors and the
Legislature may both exercise the police power to pass
legislation, and ordinarily the electors have the upper hand when
both wish to legislate on the same subject (art. II, § 10(c); People
v. Kelly (2010) 47 Cal.4th 1008), this is not an ordinary case. The
power exercised by voter electors under article II, section 8(a),
and the power exercised by the Legislature under article XIV,
section 4, overlap, but these two powers remain distinct, as the
history of article XIV, section 4, recounted above shows. Nothing
in McPherson holds to the contrary. The McPherson court
19
pointed out that the electors’ ability to adopt statutes by
initiative is a “similar power” to that of the Legislature under
article XII, section 5 (McPherson, supra, 38 Cal.4th at p. 1033),
not that the electors are “the Legislature” when they legislate on
workers’ compensation matters, having simply stepped into the
shoes of the Legislature, clothed with article XII, section 5,
authority.
Because only the Legislature has the constitutional power
to act “pursuant to” article XIV, section 4 (Graczyk v. Workers’
Comp. Appeals Bd. (1986) 184 Cal.App.3d 997, 1008), I believe
the most accurate way to describe these two powers when they
conflict is that they are concurrent, not that they are coextensive.
(See Youngstown Sheet & Tube Co. v. Sawyer (1952) 343 U.S.
579, 637–638 (Youngstown Sheet & Tube) (conc. opn. of
Jackson, J.).) The friction generated by a direct clash of two
concurrent legislative powers within our basic plan of
government presents a rare and difficult constitutional question.
One of these two sources of power must give way when they
collide, and as I read it, footnote 9 of McPherson signals that the
usual rules of deference to legislation adopted by ballot initiative
may have to yield in the face of competing constitutional
considerations. In my view, this is such a case.
2. McPherson Provides the Backdrop
Before delving further into the conflict of concurrent
powers issue before us, some background discussion of
McPherson is useful. At issue there was an initiative statute,
20
Proposition 80, that expanded the regulatory jurisdiction of the
Public Utilities Commission (PUC). Article XII, section 5, gives
the Legislature “plenary power” to confer “additional regulatory
authority and jurisdiction” upon the PUC, and the question was
whether that clause prevents the voters from expanding the
PUC’s regulatory jurisdiction by statutory ballot initiative.
(McPherson, supra, 38 Cal.4th at p. 1023.) The Court of Appeal
said yes, holding that legislation by ballot initiative was
preempted on all subject matter covered by article XII, section 5.
(McPherson, at p. 1023.)
The Supreme Court reversed. After canvassing its
precedents, the court concluded, first, “long-standing California
decisions establish[] that references in the California
Constitution to the authority of the Legislature to enact specified
legislation generally are interpreted to include the people’s
reserved right to legislate through the initiative power”
(McPherson, supra, 38 Cal.4th at p. 1043); and second, “in light of
the background and purpose of the relevant language of article
XII, section 5, . . . [the clause] does not preclude the people,
through their exercise of the initiative process, from conferring
additional powers or authority upon the PUC” (id. at pp. 1043–
1044).
The second step of the analysis in McPherson, laying out
the historical backdrop to article XII, section 5, was key. At the
time it was added to the Constitution in 1911, there was great
concern that railroad companies exercised de facto control over
many organs of state government, including the Railroad
21
Commission, the precursor to the PUC. (McPherson, supra,
38 Cal.4th at pp. 1038–1039.) By controlling the membership of
the Railroad Commission, the railroads managed to evade
regulation of their rates. (Ibid.) The motivating purpose of the
ballot initiative that became article XII, section 5, was to take
membership control of the Railroad Commission away from the
railroads and ensure that, going forward, the Railroad
Commission had adequate regulatory authority to control
railroad rates. (McPherson, at p. 1040.) The clause in article XII,
section 5 conferring “plenary power” on the Legislature to grant
“additional authority and jurisdiction” addressed the potential
that, to ensure the Railroad Commission’s continued regulatory
effectiveness, expanded regulatory authority might be needed in
the future.
When McPherson was decided in 2006, the regulatory
target was different, but the overall purpose of article XII,
section 5, was the same. Independent electric service providers
(ESP’s) were then an important new source of electric power to
the public grid, and utilities’ cost of power purchases from them
was increasingly affecting rates to consumers. (McPherson,
supra, 38 Cal.4th at pp. 1025–1027.) But ESP’s emerged in the
1990’s as creatures of the energy deregulation movement and
were unregulated, so the PUC had no rate-setting or enforcement
control over them. (Ibid.) To fill the gap, Proposition 80
extended the agency’s rate-setting and enforcement authority to
ESP’s. (McPherson, at pp. 1025–1027.)
22
The Supreme Court concluded that this new grant of
regulatory authority was consistent with the “origin and purpose”
of article XII, section 5 (McPherson, supra, 38 Cal.4th at p. 1025):
To ensure the continuing effectiveness of the Railroad
Commission, and hence of its regulatory successor, the PUC.
McPherson would have been a very different case—the type of
case we have here—had Proposition 80, for example, removed the
PUC’s statutory jurisdiction to regulate investor-owned utilities
rather than added to it. That is why respondents invoke
footnote 9 of the McPherson opinion. They argue that, while
ballot statutes may be used to build upon and refine our
constitutionally sanctioned “complete system of workers’
compensation,” the power to legislate by initiative may not be
used to undermine that system.
The interveners deride respondents’ reading of McPherson
as an attempt to create a “one-way ratchet” permitting additive
revisions to our statutory workers’ compensation system, but not
subtractions from it. What the interveners overlook, in my view,
is that, structurally, our state Constitution supports respondents’
reading much better than it does theirs. They are asking that we
allow the voter electors to “undo” what the Legislature has done
pursuant to its article XIV, section 4 powers, and at the same
time bar the Legislature from restoring what the electors have
taken away, even if a majority of the Legislature reads the
Constitution to require it.
We are not dealing simply with who gets the legislative last
word. Because the issue of whether an app-based driver is an
23
employee or an independent contractor determines threshold
eligibility for workers’ compensation, it is squarely within the
heartland of lawmaking power conferred solely upon the
Legislature in article XIV, section 4. By seeking to reserve all
statutory lawmaking power for themselves on this issue, the
Proposition 22 voter electors go well beyond a measure in which
voter electors have told the Legislature not to tinker with their
statute. This statutory initiative attempts to seize and
permanently redistribute constitutional power in a manner that
subverts article XIV, section 4, itself.
Let me be concrete about the problem I see here.
If, mindful of the robust support for the Proposition 22 in the
November 2020 election, a majority of elected legislators
understand it to be their solemn constitutional obligation under
article XIV, section 4, to override the definition of “independent
contractor” adopted by voter electors—none of whom took any
oath to uphold the Constitution when they entered the voting
booth—I believe these legislators should be free to do so,
unimpeded by Proposition 22. Indeed, I believe that article XIV,
section 4, entitles them to have the last word on that issue, since
the definition of “independent contractor” in Proposition 22
“would, in effect, be an amendment of the Constitution, if
accepted as authoritative.” (PG&E, supra, 180 Cal. at p. 500.)
For that reason alone, we must resolve the conflict of
concurrent powers issue presented in this case in favor of the
Legislature. And in doing so, we need not handle this conflict as
a zero-sum showdown in which one of the two must nullify the
24
other. All we need do to resolve this case is recognize that we are
in a “zone of twilight” (Youngstown Sheet & Tube, supra, 343 U.S.
at p. 637 (conc. opn. of Jackson, J.) where both powers are
operative, while resolving the conflict on this particular record
based on a tie-breaking principle drawn from article XIV,
section 4. In practical terms, what that means is simply this:
When the electors choose to legislate on the topic of workers’
compensation by ballot, they must do so in a manner that is
consistent with any prior exercise of article XIV, section 4, power
by the Legislature.
Proposition 22 fails that test. The Legislature used its
article XIV, section 4 power in passing Assembly Bill 5, which
was enacted in 2019 to clarify that app-based drivers and
couriers are included within the existing workers’ compensation
and occupational health and safety systems. Because the
Legislature had acted previously on this topic pursuant to article
XIV, section 4, when Proposition 22 came to the ballot, and acted
decisively, I believe the voter electors were required to respect
what the Legislature had done and lacked power to countermand
it. To borrow Justice Jackson’s famous phrasing, Proposition 22
having been adopted in the face of the “expressed will” of the
Legislature, the voters’ power to legislate by initiative on the
same topic was at its “lowest ebb” in those circumstances
(Youngstown Sheet & Tube, supra, 343 U.S. at p. 637), and must
be deemed subordinate.
Courts have recognized a number of implied limitations on
the initiative powers. Although article XVIII, section 3, of the
25
Constitution expressly declares that it can be amended by
initiative, the initiative cannot be used to revise the Constitution.
For example, the power of statutory initiative cannot be used to
order the Legislature to pass a resolution because article II,
section 8(a) speaks only of the “adoption or rejection of
‘statutes.’ ” (American Federation of Labor v. Eu (1984) 36 Cal.3d
687, 708.) Nor can the power of statutory initiative be used to
regulate the Legislature’s internal operations, because that
would conflict with the express grant of power in article IV,
section 7(a) authorizing the Legislature to run its own affairs.
(People’s Advocate, Inc. v. Superior Court (1986) 181 Cal.App.3d
316, 327.) To this list, I believe we must add that the power of
statutory initiative cannot be used to subvert an express
constitutional power delegated to the Legislature, and to the
Legislature alone. (Ibid. [“Only by means of an initiative
constitutional amendment may the people modify or impinge
upon the freedom of the Legislature to exercise its
constitutionally granted powers.”].)
D. Applying “The Relevant Constitutional Provision”
(Article XIV, Section 4) to the Terms of the Statute
Involved (Proposition 22)
Although Justice Jackson’s Youngstown Sheet & Tube
concurrent powers framework provides important analytical
guidance for resolving the conflict of legislative powers problem
before us—and answers the question presented here simply and
easily when applied in light of cases recognizing certain implied
limits on the initiative power—the specific mode of analysis must
26
come from McPherson, since in that case our Supreme Court laid
the groundwork for us in a case that is unquestionably the closest
analog to our own. Footnote 9 of the McPherson opinion states
that, should a conflict of the kind it contemplates arise in some
future case, the analysis will be governed by the “application of
the relevant constitutional provision or provisions to the terms of
the specific legislation at issue.” (McPherson, supra, 38 Cal.4th
at p. 1044, fn. 9.)
As I noted above, the independent contractor definition in
Business & Professions Code section 7451 is the “specific
legislation” at issue here. This provision overturns the
Legislature’s judgment in Assembly Bill 5 that app-based drivers
must be treated as employees until putative employers prove
otherwise in case-by-case adjudication within the workers’
compensation system. Because class-wide exclusion of app-based
drivers from the workers’ compensation system is the main
objective of Proposition 22—in direct conflict with Assembly
Bill 5—the analysis under footnote 9 of the McPherson opinion
turns on the language of article XIV, section 4.
Settled principles of interpretation govern the inquiry.
Because it is elementary that we avoid interpreting our
Constitution in a manner that renders any of its provisions
nugatory, inoperative or meaningless (Lungren v. Deukmejian
(1988) 45 Cal.3d 727, 735), we must strive to give article XIV,
section 4, meaning. To discern the intent of the people acting in
their sovereign capacity in 1918—and not project our modern
understanding in retrospect—we must apply “ ‘ “the same
27
principles that govern statutory construction.” [Citation.] Where
a law is adopted by the voters, “their intent governs.” [Citation.]
In determining that intent, “we turn first to the language of the
statute, giving the words their ordinary meaning.” ’ ” (People v.
Raybon (2021) 11 Cal.5th 1056, 1065.)
If it is necessary to go beyond plain meaning and consider
context, historical context is relevant, as our Supreme Court
recognized in addressing the parameters of a similar
constitutional provision, article XII, section 5, in McPherson.
(See McPherson, supra, 38 Cal.4th at pp. 1039–1040.) And as
with statutory construction, the language in question “ ‘must also
be construed in the context of the statute as a whole and the
overall statutory scheme. [Citation.] We apply a presumption, as
we similarly do with regard to the Legislature, that the voters, in
adopting an initiative, did so being “aware of existing laws at the
time the initiative was enacted.” ’ ” (People v. Raybon, supra,
11 Cal.5th at p. 1065.)
1. Text: Detailed and expansive grant of power “unlimited
by any provision of this Constitution” to “create” and
“enforce” a “complete system of workers’ compensation”
Applying these principles to article XIV, section 4, two
things are immediately apparent. First, what is most striking
textually about article XIV, section 4, is its detail and specificity.
There is a reason for that. When “California joined a rapidly
growing number of states in adopting . . . workmen’s
compensation act[s]” in the early twentieth century (Mathews,
28
supra, 6 Cal.3d at p. 729), 15 our state was one of only a small
group that specified “definite forms of compensation laws” in its
Constitution. (Schneider, The Law of Workmen’s Compensation
(1922) Ch. 2, § 4, p. 10, fn. 5.) 16 The Proposition 22 proponents
largely ignore the specificity of the text, dealing with it
dismissively by suggesting that article XIV, section 4, is archaic
(which is incorrect, as I have noted above), and is in any event
redundant to the police power (which is equally incorrect, as I
have also noted above).
The breadth of the constitutional language is also striking.
Article XIV, section 4, grants “plenary” legislative power
“unlimited by any provision of this Constitution.” (Art. XIV, § 4,
italics added.) The Proposition 22 proponents contend a literal
reading of this expansive language would lead to absurd results,
for it would free the Legislature from constitutional constraints
such as the gubernatorial veto. But that is a strawman
15 See Fishback & Kantor, The Adoption of Workers’
Compensation in the United States, 1900–1930 (1998) 41 J. Law
& Econ. 305, 319, table 2 (listing 48 states that adopted workers’
compensation statutes between 1910 and 1918); id. at p. 319
(“ ‘No other kind of labor legislation gained such general
acceptance in so brief a period in this country.’ ”).
16See Dinan, Foreword: Court-Constraining Amendments
and the State Constitutional Tradition (2007) 38 Rutgers L.J.
983, 995–996 (noting that constitutional provisions “guaranteeing
the constitutionality of workers’ compensation programs” were
adopted in New York, Ohio, Arizona, California, Vermont,
Wyoming, Pennsylvania, and Texas); id. at p. 995 (describing
these provisions as “court-preempting constitutional
amendments”).
29
argument. The breadth of the grant is easily explained by the
variety of constitutional objections that had been raised prior to
1918. Due process objections were a concern, but not the only
ones.
Nor is there any need to read this language as giving the
Legislature wholly unchecked power, which was the contention
Justice Henshaw made in Western Indemnity I, as echoed by the
Proposition 22 proponents here with their “absurd results”
argument. Respondents readily concede the Legislature must
pass “appropriate legislation” under article XIV, section 4, before
its views are binding on anyone. Naturally, that requires
adherence to all requisite procedures under article II, section 8,
subject to gubernatorial veto. Without following those
procedures, no bill the Legislature passes is binding on anyone.
It is not law at all, and is certainly not “appropriate legislation.”
Similarly, a statute that exceeds the Legislature’s authority
under article XIV, section 4, would not be “appropriate
legislation.” 17
Second, functionally, the constitutional text charges the
Legislature with the responsibility not only to “create” a
“complete system of workers’ compensation”—in effect, codifying
in the Constitution itself the statutory system then in existence—
The Proposition 22 proponents make no claim that
17
Assembly Bill 5 exceeded the Legislature’s authority under
article XIV, section 4. Nor could they, since Assembly Bill 5
simply codified the Supreme Court’s holding in Dynamex
Operations West, Inc. v. Superior Court (2018) 4 Cal.5th 903
(Dynamex).
30
but also to “enforce” that system “by appropriate legislation” in
the years ahead, while making clear that the pre-1918 “social
public policy” reflected in the Legislature’s “complete system of
workers’ compensation” has constitutionally binding effect on
“all departments of State government.” (Art. XIV, § 4, italics
added.) There is nothing time-limited about any of this. Article
XIV, section 4, gave the Legislature a constitutionally designated
role in 1918, and that role continues today. Nowhere is there any
hint or suggestion that voter electors may commandeer it by
ballot statute.
Here again, the Proposition 22 proponents respond with
overstatement. In their view, any suggestion the Legislature has
a singular role to play under article XIV, section 4, must mean an
exclusive role—preemptive of action by any other governmental
actors. As framed in McPherson, that was indeed the issue
addressed, there in the context of article XII, section 5. But not
here. Respondents make no attempt to read article XIV,
section 4, so broadly that it would nullify the power to adopt
initiative statutes or place the Legislature’s judgments beyond
judicial review. Nor do I. What we are called upon to decide is
whether, absent a constitutional amendment, the text of article
XIV, section 4, requires “all departments of State government” to
give deference to the Legislature’s specifically delegated power to
“create” and “enforce” a “complete system of workers’
compensation.” This is why I believe it is crucial to appreciate
that we are dealing with two concurrent powers here, and that
the clash between them on this record must be resolved by
31
examining whether we can discern in the text of article XIV,
section 4, any preference for either of these two forms of statutory
lawmaking when they are in conflict.
I think the answer is yes, we can discern such a preference,
and it is for lawmaking by the Legislature. The grant of
lawmaking power in article XIV, section 4, is not self-executing.
While voter electors passing an initiative statute have no duty to
do anything—their lawmaking power is purely discretionary—
article XIV, section 4, charges the Legislature with an ongoing
responsibility to carry out and abide by the terms of a
constitutional mandate. “[A]rticle XIV, section 4 . . . defin[es] the
necessary provisions for a complete workers’ compensation
system, and le[ft] it up to the Legislature to enact laws to give
effect to each provision.” (Bautista v. State of California (2011)
201 Cal.App.4th 716, 729; see ibid. [“only the Legislature has
constitutional authority to create and enact the workers’
compensation system”].) This special role is not discretionary.
Under article XIV, section 4, “[t]he Legislature must act to fulfill
its constitutional mandate to create the workers’ compensation
system, and the judicially enforceable rights are the laws it
enacts.” (Bautista, at p. 729, italics added.) 18 So long as the
18 My colleagues observe that there is “no logical conflict
between article XIV, section 4 needing implementing legislation
and the voters retaining their initiative power in the same field”
and that “both can coexist.” (Maj. opn., ante, at p. 14, fn. 7.)
I have not suggested to the contrary. They seem to miss the
point of my focus on the need for implementing legislation: The
Legislature has a constitutional duty to act under article XIV,
32
Legislature acts within the boundaries of its authority under
article XIV, section 4, its specially designated constitutional role
demands deference.
Just as with Youngstown Sheet & Tube, there is a federal
model to draw upon here. This model reflects a great deal of
accumulated judicial wisdom applying what are perhaps the most
well-known implementing clauses in American constitutional
law—Congress’s “power to enforce, by appropriate legislation, the
provisions of ” the Fourteenth and Fifteenth Amendments. (U.S.
Const., XIV Amend., § 5; XV Amend., § 2.) The high court has
consistently held that “Congress’ judgment regarding exercise of
its power to enforce the Fourteenth and Fifteenth Amendments
warrants substantial deference.” (Shelby County v. Holder (2013)
570 U.S. 529, 566 (dis. opn. of Ginsburg, J.).) It is true, of course,
that Congress’ powers are limited by enumerated grants, while
the police power is a general power to govern by legislative
prescription and needs no affirmative grant. But that makes no
difference. The lesson to be drawn from these venerable grants of
Congressional authority is that a constitutionally conferred
prospective enforcement power carries with it an ongoing
responsibility to apply and determine the meaning of
constitutional language when faced with changing conditions.
Legislative judgment in discharging that constitutional
responsibility merits substantial deference.
section 4, while voters going to the polls do not. A conflict arises
only if ballot electors seek to obstruct the Legislature from
discharging its duty.
33
To protect the enforcement prerogatives given to the
Legislature in article XIV, section 4, substantial deference is due
here as well—and not just from courts, but also from ballot
statute electors. Unless we treat ballot statute electors as
somehow outside our basic plan of government, they are bound to
give the Legislature deference, just as courts are. Giving the
expansive constitutional text its plain meaning, I believe we
should read article XIV, section 4, as a command that when the
Legislature enacts “appropriate legislation” designed to
implement the constitutionally mandated “complete system of
workers’ compensation,” its judgment trumps that of unelected
voter lawmakers in the face of a conflict. So understood, article
XIV, section 4, does not oust voter electors of power to legislate on
the topic of workers’ compensation; it simply requires them to
respect prior determinations of the Legislature as to what the
Constitution requires.
2. Historical context: Augmentation of the original 1911
amendment, former article XX, section 21, and
significance of Western Indemnity I to the issue of
pro tanto repeal
The historical context surrounding the adoption of article
XIV, section 4, supports this reading of the text. The original
1911 amendment was relatively brief. It stated, simply: “The
legislature may by appropriate legislation create and enforce a
liability on the part of all employers to compensate their
employees for any injury incurred by the said employees in the
course of their employment irrespective of the fault of either
34
party. The legislature may provide for the settlement of any
disputes arising under the legislation contemplated by this
section, by arbitration, or by an industrial or accident board, by
the courts, or by either any or all of these agencies, anything in
this constitution to the contrary notwithstanding.” (Former
art. XX, sec. 21, added by Prop. 10, Special Elec. (Oct. 10, 1911);
Ballot Pamp. text of Prop. 10 at [as of Mar. 13, 2023].)
By the time the Legislature proposed an amendment to
former article XX, section 21, in 1918, the 1917 Act and the
Boynton Act together established the statutory workers’
compensation scheme. These statutes marked the first time an
act of the Legislature occupied any of the “zone of twilight” where
two concurrent powers may operate, as Justice Jackson put it in
his concurrence in Youngstown Sheet & Tube, supra, 343 U.S. at
page 637. Had the voters adopted a workers’ compensation
statute of their own as a replacement for the Roseville Act, or
made some amendment to the 1917 Act or the Boynton Act by
initiative statute prior to November 1918—before the then
existing statutory scheme was given constitutional sanction—
they would have had a free hand to legislate within the limits
prescribed by article XIV, section 4, since their will would have
been memorialized in the constitutional language. But that is
not what happened. Instead, a century later we are faced with
an effort to undermine the constitutionally sanctioned intent of
the Legislature.
35
In Western Indemnity I, the Supreme Court held that the
original 1911 version of former article XX, section 21 was enough
to provide a constitutional foundation for the Boynton Act. But
two Justices disagreed, as I have noted above. We may fairly and
reasonably infer that, when asked to do so by the Legislature in
1918, voters responded by placing a detailed description of the
Boynton Act, as improved and refined in the 1917 Act, into the
Constitution itself, thereby adopting as their own the
Legislature’s pre-1918 interpretation of what it means to have a
“complete system of workers’ compensation.” (Mathews, supra,
6 Cal.3d at p. 733 [“The proposed amendment duplicated in large
measure section 1 of the 1917 act”].) As the Mathews court
explained, article XIV, section 4, was “ ‘a necessary amplification
and definition of the constitutional authority vested in the
legislature by the amendment to the Constitution adopted
October 10, 1911, to enable the enactment of a complete plan of
workmen’s compensation, which amendment failed to express
sanction for the requisite scope of the enactment to make a
complete and workable plan.’ ” (Mathews, supra, 6 Cal.3d at
p. 733, fn. 11.)
This 1918 grant of augmented power—the police power
plus, if you will—could not have been reserved by the people in
1911 because it did not exist yet. It placed the Legislature at the
apex of a complex new administrative system that cut across all
three branches of government, requiring regulation of workplace
safety and workplace accident insurance, adjudication of claims,
and ongoing assessment of the adequacy of medical treatment
36
and compensation. The broad array of responsibilities detailed in
article XIV, section 4, required the exercise of quasi-legislative
powers that only the Legislature could undertake. 19 Ballot
statute electors were—and still today are—constitutionally
unable to fill the role. (American Federation of Labor v. Eu,
supra, 36 Cal.3d at p. 708 [article II, section 8(a), legislative
“powers are limited . . . to the adoption or rejection of ‘statutes.”].)
We may glean from this historical context that voters
supporting the 1918 amplification of former article XX,
section 21, intended that it be given priority over other provisions
in the Constitution in the event of conflict. In addition to the
singular role given to the Legislature, as facilitated by the
binding effect clause and the language “unlimited by any
provision of this Constitution”—two key features of the operative
language that were added in 1918—there is another reason to
draw this inference. These voters must be deemed to have been
aware of the Supreme Court’s holding in Western Indemnity I
that former article XX, section 21, as originally adopted, “worked
a repeal, pro tanto, of any conflicting provision which may have
been in force theretofore.” (Western Indemnity I, supra, 170 Cal.
at p. 695.) It makes no sense that voters would have intended
the 1918 version of former article XX, section 21, to have any
19 Wilson v. Hidden Valley Mun. Water Dist. (1967)
256 Cal.App.2d 271, 279 (“The Legislature and administrators
exercising quasi-legislative powers commonly resort to the
hearing procedure to uncover, at least in part, the facts necessary
to arrive at a sound and fair legislative decision.”).
37
lesser priority than the original version upheld in Western
Indemnity I.
Not surprisingly, courts have faithfully recognized the
position of relative superiority within the Constitution that the
plain language of article XIV, section 4 requires. In an opinion
that supplies the most recent evidence of article XIV, section 4’s
continuing vitality today, a First District, Division One panel
held only a few years ago that article XIV, section 4, “supersedes
the state Constitution’s due process clause with respect to
legislation passed under the Legislature’s plenary powers over
the workers’ compensation system.” (Stevens v. Workers’ Comp.
Appeals Bd. (2015) 241 Cal.App.4th 1074, 1093 (Stevens).)
Stevens also held that article XIV, section 4, “trumps the
separation of powers clause [art. III, § 3] under the state
Constitution’s plain terms,” since article III, section 3, expressly
contemplates exceptions stated within the Constitution itself.
(Stevens, at p. 1092.) Notably, to support these holdings, the
Stevens court cited implied pro tanto repeal cases that trace back
to Western Indemnity I. (Stevens, at p. 1093.) 20
McPherson provides a useful point of contrast. Article XII,
section 5, the plenary power clause at issue there—which
20 Hustedt, supra, 30 Cal.3d at page 343 (“It is well
established that the adoption of [Section 4] ‘effected a repeal pro
tanto’ of any state constitutional provisions which conflicted with
that amendment”); Greener v. Workers’ Comp. Appeals Bd. (1993)
6 Cal.4th 1028, 1037, (article VI of the state Constitution
governing courts’ jurisdiction inapplicable to extent Legislature
has exercised its powers under Section 4).
38
includes no language equivalent to the binding effect clause or
the “unlimited by any provision of this Constitution” language in
article XIV, section 4—was another ballot constitutional
amendment passed in the October 1911 special election.
(McPherson, supra, 38 Cal.4th at p. 1042.) The McPherson court
observed there was no reason to believe that, in October 1911,
voters would have intended to limit their own power under a
simultaneously adopted amendment giving them broad new
authority to adopt initiative statutes. (Ibid.) By contrast, there
is reason to believe that, in the event of a conflict between
article II, section 8(a), and the later adopted article XIV,
section 4, in 1918, the latter would prevail. That reason is this—
between 1911 and 1918, the pro tanto repeal holding in Western
Indemnity I was handed down and, as noted above, knowledge of
it is chargeable to voters. 21
21 The parties argue the issue of implied pro tanto repeal at
length. In my view, the plain language of article XIV, section 4,
is abundantly clear on this issue, and there is no need to resort to
extrinsic interpretative aids. But even if the text were
ambiguous, the position taken by the Proposition 22 proponents
to the contrary—embraced by my colleagues in their construction
of article XIV, section 4—rests on “clear statement” cases decided
long after 1918. (E.g., California Cannabis Coalition v. City of
Upland (2017) 3 Cal.5th 924; Kennedy Wholesale, Inc. v. State
Bd. of Equalization (1991) 53 Cal.3d 245, 252.) Respondents
correctly point out that these cases concern the issue whether
procedural limitations not stated in article II, section 8(a), may
be imported into it, which is not the issue here. But the more
basic problem with the line of argument the Proposition 22
proponents pursue on the issue of pro tanto repeal is that we
must seek to discern contemporaneous voter intent in 1918, not
39
3. Structural context: The binding effect clause in
article XIV, section 4, is unique
Finally, we must consider the language of article XIV,
section 4, within the context of the Constitution as a whole.
Upon an overall examination of our state charter, article XIV,
section 4, stands in sharp contrast to all other, similar provisions
in the document. The Constitution gives the Legislature plenary
power to do things in five places, 22 none of them providing
ongoing enforcement authority, and nowhere else have the people
said in the charter document that other actors in state
government are bound by a “social public policy” enacted by the
Legislature under a specifically delegated implementation power.
These features in the operative text are unique.
The language “binding upon all departments of the State
government” is particularly notable. The plain meaning of “all
departments of State government” encompasses not only the
judicial, legislative, and executive branches of government
(Brydonjack v. State Bar (1929) 208 Cal. 439, 442), but county
and local government as well (City of Sacramento v. Industrial
Acc. Com. (1925) 74 Cal.App. 386 (City of Sacramento)). As noted
above, unless we treat voter legislators as somehow exogenous to
voter intent based on tools and standards of interpretation that
were unknown at the time. In November 1918, Western
Indemnity I established the state of the law on the issue.
22 In addition to article XIV, section 4, see article XII,
section 5 (Public Utilities); article XVI, sections 11 and 17 (Public
Finance); article XIII, section 8.5 (Taxation); article XI, section 5
(Local Government).
40
our plan of government, they are encompassed within the phrase
“all departments of the State Government” when they adopt
ballot statutes. This leaves voter electors ample room to make
discretionary policy choices by ballot statute in the field of
workers’ compensation. While voter electors cannot stand in the
shoes of the Legislature under article XIV, section 4, they can
always serve as “shadow” lawmakers, exercising the police power
to enact any workers’ compensation legislation they wish—so
long as they respect prior “appropriate legislation” enacted by the
Legislature in this special arena.
Several of the amici supporting the Proposition 22
proponents urge us to harmonize article XIV, section 4, and
article II, section 8(a), powers by construing these two sources of
legislative power in a manner that recognizes the Legislature and
voter electors may share authority to act on the topic of workers’
compensation. While I agree with that perspective in principle,
Proposition 22 resists harmonization—unless we declare a naked
preference for article II, section 8(a), which is ultimately what
these amici invite us to do, invoking platitudinous statements
about the importance of direct democracy, and ignoring the
usurpation of the Legislature’s delegated article XIV, section 4
power that this mode of “harmonization” implies. To resolve the
conflict before us correctly, we must recognize that this case is
one of a kind. By making its independent contractor definition
unamendable, Proposition 22 effectively appoints voter electors
the sole and exclusive expositors of what the constitutional term
“employment” means, thereby displacing the Legislature entirely
41
on that topic. This feature of Proposition 22 simply cannot be
reconciled with article XIV, section 4.
All bets are off when it comes to ballot constitutional
amendments, since voter electors exercise sovereign power in
that context and may always substitute their judgment by
adopting paramount law. Indeed, in the final analysis, that is
what this case is about. Until and unless voter electors escalate
things to the level of a proposed constitutional amendment, the
Constitution expressly gives our elected Legislature a unique
role—I believe the preeminent role—when statutes are enacted
pursuant to article XIV, section 4. 23 If, as of 1918, the California
judiciary was bound by the Legislature’s pre-1918 interpretation
of what constitutes a “complete system of workers’
compensation”—in the sense courts were expected to give
deference to the Legislature’s original judgment about the “basic
features” of that system, just as the Supreme Court did in
23 A commonly used metaphor in cases involving challenges
to initiative measures is that the initiative power in article II,
section 8(a) is “in essence a legislative battering ram which may
be used to tear through the exasperating tangle of the traditional
legislative procedure and strike directly toward the desired end.”
(Amador Valley Joint Union High Sch. Dist. v. State Bd. of
Equalization, supra, 22 Cal.3d at p. 228.) In statutory initiative
cases, however, we must take care to consider whether, in some
circumstances—depending on the constitutional provisions we
are dealing with—“the exasperating tangle of the traditional
legislative procedure” (ibid.) is better described as a deliberative
process for resolving competing interests by compromise and
accommodation that the people have affirmatively expressed a
preference for using.
42
Western Indemnity I—then it seems to me voter electors passing
initiative statutes were similarly bound, and are still bound
today.
Disagreeing with me on this point, the majority opinion
points out that the binding effect clause “appears to have been
intended only to mean that the workers’ compensation system
applies to the state and local governments as employers.” (Maj.
opn., ante, at p. 27.) For this idea, the majority cites dicta from
Bautista v. State of California, supra, 201 Cal.App.4th 716, and
City of Sacramento, supra, 74 Cal.App. 386. Neither of these
cases holds, nor has any case ever held, that local governments
are the only “departments of the State government” covered by
the binding effect clause in article XIV, section 4. Nor do these
cases give any reason for why the binding effect clause should be
read as an implicit partial definition of “employer,” a term that
appeared nowhere in the constitutional language and was
already fully defined in the statutory scheme the voters ratified
in 1918. (1917 Act, Stats. 1917, ch. 586 § 7.) 24
Of the two cases, only City of Sacramento—decided less
than a decade after article XIV, section 4 was adopted—merits
24 A notable difference between the language of former
article XX, section 21, as originally adopted in 1911, and the
language of former article XX, section 21, as amended in 1918,
was that the 1911 version used the terms “employer” and
“employee” (former art. XX, § 21, added by Prop. 10, Special Elec.
(Oct. 10, 1911)), while the 1918 amended version omitted them
(former art. XX, § 21, amended by Prop. 23, Gen. Elec. (Nov. 5,
1918)), since by 1918 there were detailed definitions of those
terms in the 1917 Act.
43
full discussion. That case is illuminating, though not for the
reason my colleagues cite it. There, a city employer tried to
invalidate death benefits awarded to the widow of one of its
employees who was killed in the course of his employment.
(City of Sacramento, supra, 74 Cal.App. at p. 387.) The city
employer tried to argue the widow’s claim was “a matter purely
within the jurisdiction and control of the city of which such
persons may be officers or employees” (id. at p. 388), which today
we would call the “home rule” doctrine, our state equivalent of
federalism in the national government.
The contention was that the “compensation” of municipal
employees falls within the exclusive domain of local government
under former article XI, section 8½ of the Constitution. (City of
Sacramento, supra, 74 Cal.App. at pp. 387–388.) This was not an
argument directed to whether the statutory reach of workers’
compensation extends to municipalities as “employers” within the
meaning of the 1917 Act, the operative statute at the time.
Under the plain language of the 1917 Act, “[t]he term ‘employer’
. . . shall be construed to mean . . . [t]he state, and each county,
city and county, city, school district and all public corporations
therein.” (1917 Act, Stats. 1917, ch. 586, § 7.) The statutory
definition of “employer” was never mentioned, which confirms
that there was no genuine issue at the time about whether cities
were included in that term. Rather, the city employer advanced a
jurisdictional argument that the Legislature had no
constitutional power to address the “compensation” of local
employees.
44
The Court of Appeal rejected the city employer’s attempt to
circumvent the 1917 Act by constitutional interpretation, holding
that the word “compensation” in former article XX, section 21, as
amended in 1918 (and now included in article XIV, section 4) was
not used “in the sense of meaning wages” (City of Sacramento,
supra, 74 Cal.App. at p. 392), but rather “in the sense of making
amends for losses sustained, or the paying of an indemnity or an
equivalent, so far as it is possible to do so in money value, to
those dependents who have suffered such losses” (id. at p. 395).
Because the Legislature was given plenary power to enforce
liability for workers’ compensation awards against “all persons,”
the court held, the term “persons” included municipalities. (Ibid.)
Unpersuaded by the municipal employer’s attempt to offer a
reading of former article XX, section 21, that implicitly exempted
it from the “complete system of workers’ compensation” the
Legislature created, the court pointed out that a municipality is a
“department[] of the state government” bound by the “ ‘social
public policy’ ” codified in former article XX, section 21. (City of
Sacramento, at p. 395.)
In my view, the lesson to be drawn from City of Sacramento
is this. What came before the court there—just as it does in this
case—was an attempt to place an entire class of workers outside
the workers’ compensation system based on a narrow
construction of the Legislature’s constitutionally delegated power
to create a “complete system of workers’ compensation” and
enforce compensation awards against covered employers. The
City of Sacramento court was alert to the evasion and relied in
45
part on the “social public policy” embodied in former article XX,
section 21 to reject it. I regret that we have not been similarly
alert to the constitutional evasion Proposition 22 represents. As
“shadow” legislators, the Proposition 22 voter electors are
similarly bound by the “social public policy” codified in article
XIV, section 4, particularly after the Legislature expressed its
view of how that “policy” applies in passing Assembly Bill 5.
III. Proposition 22 Violates Article III, Section 3
A. A Dispute Resolution System Within the Judicial
Branch Is a “Basic Feature” of the Pre-1918
“Complete System of Workers’ Compensation”
Among the “basic features” of the pre-1918 workers’
compensation system set up by the Legislature were these three:
(1) the definition of “employee” included “[e]very person in the
service of an employer . . . under any appointment or contract of
hire or apprenticeship, express or implied, oral or written,” 25
(2) a “ ‘contract of hire’ mean[t] a contract for personal services,
as is indicated by the fact that the basis of compensation provided
by the act is the amount of wages earned,” 26 and (3) disputes over
the employment status of wage-earning workers—including on
the issue of whether they were employees or independent
contractors—were to be decided by the Industrial Accident
1917 Act, Statutes 1917, chapter 586, section 8(a);
25
Western Indemnity Co. v. Pillsbury (1916) 172 Cal. 807, 810
(Western Indemnity II); Press Pub. Co. v. Industrial Acc. Com.
(1922) 190 Cal. 114, 119.
Press Pub. Co. v. Industrial Acc. Com., supra, 190 Cal. at
26
page 119; Western Indemnity II, supra, 172 Cal. at page 810.
46
Commission (IAC), subject to judicial review in the Court of
Appeal. 27
This last element—administrative resolution of workers’
compensation claims, with review channeled directly into the
Court of Appeal—was added in 1913, when the statutory workers’
compensation scheme evolved from a purely voluntary system in
the Roseville Act into a compulsory system in the Boynton Act.
Designed to ensure expeditious resolution of claims by a decision
maker imbued with judicial power, this feature of the Boynton
Act has long been recognized by our Supreme Court as a “basic
feature” of the pre-1918 workers’ compensation system. (Hustedt,
supra, 30 Cal.3d at p. 343.)
Article XIV, section 4, includes definitional language
describing this dispute resolution process in detail. The first
paragraph of article XIV, section 4, begins: The Legislature is
empowered to make “full provision for vesting power, authority
and jurisdiction in an administrative body with all the requisite
governmental functions to determine any dispute or matter
arising under such legislation, to the end that the administration
of such legislation shall accomplish substantial justice in all cases
expeditiously, inexpensively, and without incumbrance of any
character.” (Art. XIV, § 4, italics added.) The second paragraph
271917 Act, Statutes 1917, chapter 586, sections 17–19,
65–67; PG&E, supra, 180 Cal. at page 499; see Murray v.
Industrial Acc. Com. (1932) 216 Cal. 340, 344 (employer has
burden of proof on whether worker is independent contractor);
Drillon v. Industrial Acc. Com. (1941) 17 Cal.2d 346, 350 (same)
(Drillon).
47
of article XIV, section 4, fleshes this out in more detail, and
specifically provides that any administrative decisions “shall be
subject to review by the appellate courts of this State.” (Ibid.,
italics added.)
In the pre-1918 statutory scheme, resolution of workers’
compensation disputes was given to the IAC, and in some
instances to appointed “referees” whose decisions it reviewed
(Boynton Act, Stats. 1913, ch. 176, § 76), 28 but the above-quoted
constitutional language confirms that this statutorily defined
mode of dispute resolution was understood, ultimately, to be
lodged in the judicial branch of government. The Supreme Court
so held in 1916. Upholding the constitutionality of the Boynton
Act in Western Metal Supply Co. v. Pillsbury (1916) 172 Cal. 407,
the court rejected a claim that the use of an administrative
agency for resolution of workers’ compensation claims
“transcends constitutional limitations in attempting to vest in the
[IAC] the power—asserted to be judicial in its nature—to assess
compensation and award it to such dependents.” (Id. at p. 410.)
28 Many years later, the Legislature enacted provisions
permitting arbitration of workers’ compensation disputes in some
instances. (Lab. Code, § 3201.5.) That arbitral scheme, however,
remains within the same administrative system for workers’
compensation that has existed since enactment of the Boynton
Act, with judicial review available in the Courts of Appeal, and
has been held to be consistent with article XIV, section 4. (Costa
v. Workers’ Comp. Appeals Bd. (1998) 65 Cal.App.4th 1177,
1184–1186.) Our system of judicial arbitration (Code Civ. Proc.,
§ 1141.10 et seq; see Rivera v. Shivers (2020) 54 Cal.App.5th 82,
90), another arbitral scheme established by statute within the
judicial branch of government, provides a rough analogy.
48
“[I]n exercising [its] . . . powers,” the Western Metal Supply court
held, the IAC “is performing precisely the same functions that are
performed by any court in passing upon questions brought before
it.” (Id. at p. 412; see Yosemite Lumber, supra, 187 Cal. at p. 779
[“The power to determine whether or not the liability referred to
in the first paragraph of the above section and imposed by this
law exists against any person is judicial power.”].)
B. Proposition 22 Defeats or Impairs a “Core” or
“Essential” Governmental Function
It is not the change in the substantive law governing the
issue of employee versus independent contractor status that
concerns me here—I have addressed that issue above in
discussing the conflict between concurrent article II, section 8(a),
and article XIV, section 4, powers (see parts II.C.–II.D. above)—
but, rather, I am focused on the dismantling of the
constitutionally ratified workers’ compensation dispute resolution
scheme that free-agent workers analogous to app-based drivers
today were entitled to invoke in 1918. This aspect of
Proposition 22 raises an independently fatal constitutional
deficiency.
Because the pre-1918 workers’ compensation system
established by the Legislature called for the resolution of claims
by an administrative agency exercising judicial power,
Proposition 22 took away an adjudicative function that was
constitutionally committed to the judicial branch in article XIV,
section 4. It therefore violates article III, section 3, which states
that, within our “triune” scheme of government (Lorraine v.
49
McComb (1934) 220 Cal. 753, 756), “[p]ersons charged with the
exercise of one power may not exercise either of the others except
as permitted by this Constitution” (art. III, § 3). Electors
exercising the police power to adopt statutes as “shadow”
legislators, as I have noted above, must qualify as “[p]ersons
charged with the exercise of one power” within the meaning of
article III, section 3. (Ibid.)
If the Proposition 22 proponents are correct that the
holding in McPherson should be extended and applied in this
case, they run headlong into a separation of powers problem
under article III, section 3. That is because, even assuming voter
electors may exercise a “similar power” (McPherson, supra,
38 Cal.4th at p. 1033) to that of the Legislature in a manner that
permits them to legislate freely in the field of workers’
compensation, without limitation—indeed, in derogation of the
Legislature’s article XIV, section 4, power—there remains a
critical difference. Voter electors exercising the police power to
pass ballot statutes are subject to separation of powers
constraints, while the Legislature in the exercise of its power
under article XIV, section 4, is not. (Stevens, supra,
241 Cal.App.4th at p. 1092.) Thus, in addition to the separation
of powers violation identified by the majority opinion, another
article III, section 3, problem comes to the forefront if the
Proposition 22 proponents’ reading of McPherson is correct: By
destroying a mode of dispute resolution that app-based drivers
are entitled to utilize within the judicial branch of government,
Proposition 22 unconstitutionally usurps judicial power.
50
To uphold this initiative would “defeat or materially
impair” the exercise of a judicial function (Hustedt, supra,
30 Cal.3d at p. 338) as well as its doctrinal cousin, the rule
against impairment of “core” (Carmel Valley Fire Protection Dist.
v. State of California (2001) 25 Cal.4th 287, 297) or “essential”
(Butt v. State of California (1992) 4 Cal.4th 668, 700, fn. 26)
governmental functions. As the Supreme Court explained many
years ago in Brydonjack v. State Bar, supra, 208 Cal. at page 442:
“The sum total of this matter is that the Legislature may put
reasonable restrictions upon constitutional functions of the courts
provided they do not defeat or materially impair the exercise of
those functions.” Fair questions are often raised in cases
involving this kind of constitutional issue as to whether the
challenged infringement is of a power that may truly be
considered “core” or “essential,” but in this case it seems to me
those questions are answered by the simple fact that, as courts
have long construed it, we are dealing with a power that is
delineated in the constitutional text.
The issue of whether a given worker under a contract for
hire has the status of an employee or an independent contractor
is a judicial question. The Supreme Court so held in Drillon,
supra, 17 Cal.2d 346. At issue in Drillon was the employment
status of one Claude Hooper, a thoroughbred racing jockey who,
while free to work for any horse owner, raced a particular horse
subject to guidance and instruction from the horse’s owner. (Id.
at pp. 348–350.) The IAC concluded that Mr. Hooper was an
51
employee of the owner (id. at pp. 348–349), and the Supreme
Court affirmed (id. at pp. 355–356).
Faced with an argument from the employer that
Mr. Hooper’s status was controlled by the rules of the horse-
racing board promulgated pursuant to statute (Drillon, supra,
17 Cal.2d at pp. 352–354), the court held that “the issue of
whether or not a person is an employee or an independent
contractor is a judicial question and not a legislative or executive
one. Legislative and administrative regulations relating to the
affairs of persons furnishing services to the persons to whom
furnished cannot control the judicial branch of the government in
its determination of that question.” (Id. at p. 355.) The same
analysis applies here.
The Attorney General argues that independent contractors
were always “excluded” from workers’ compensation coverage,
and that in Proposition 22 the voters simply changed the
substantive law determining whether app-based drivers have
that status. There was no such “exclusion”; one was not needed.
Because independent contractor status is, and always was, the
opposite of employee status, and because article XIV, section 4,
today, as under its previous incarnation in former article XX,
section 21, covers only workers in an “employment” relationship,
independent contractors have always been outside the reach of
our statutory workers’ compensation system. (PG&E, supra,
180 Cal. at p. 500; Carstens v. Pillsbury (1916) 172 Cal. 572, 580.)
That is nothing new, and it is not a matter of statutory
“exclusion.”
52
The question here is not whether independent contractors
per se are outside the system. It is whether app-based drivers, a
category of wage workers that did not exist prior to 1918, may be
expelled from the present-day workers’ compensation system by
labelling them independent contractors, thereby depriving them
of any ability to have their employment status determined within
the system. Wage workers in the position of today’s app-based
drivers (who kept driving during the pandemic at great personal
risk to themselves and their families) have always been
presumptively within the workers’ compensation system, which is
why respondents cite Drillon. Neither the Attorney General nor
the interveners discusses Drillon or offers any meaningful
response to the point respondents make on the strength of it.
The critical point Drillon illustrates is that, prior to 1918,
independent contractor status was nothing more than an
employer’s defense, subject to proof on a case-by-case basis in
workers’ compensation proceedings before an administrative
tribunal that exercised judicial power, subject to review in the
Courts of Appeal. The 1917 Act defined the term “independent
contractor” in the same way the Supreme Court applied the
concept twenty-five years later in Drillon. That statute provided,
“The term ‘independent contractor’ shall be taken to mean, for
purposes of this act: Any person who renders service, other than
manual labor, for a specified recompense for a specified result,
under the control of his principal as to the result of his work only
and not as to the means by which such result is accomplished.”
53
(1917 Act, Stats. 1917, ch. 586, § 8(b).) And the people gave
constitutional sanction to this definition in 1918.
In passing Assembly Bill 5, the Legislature codified the
ABC test recognized in Dynamex, supra, 4 Cal.5th 903, as the
general rule to test for employee versus independent contractor
status under the “suffer or permit to work standard” (id. at
pp. 965–967), “subject to a series of statutory exemptions” for
various industries (People v. Uber Technologies, Inc. (2020)
56 Cal.App.5th 266, 277, fn. 5). Under Assembly Bill 5, all
putative employers must continue to litigate any issue of
employee versus independent contractor status before the
Workers’ Compensation Appeals Board, and ultimately in the
Courts of Appeal if necessary. Assembly Bill 5 simply confirmed
that, as a general matter, the ABC test for independent
contractor status governs, except that the more complex, multi-
factor Borello test governs in certain industries. Both tests are
consistent with the 1917 Act’s “independent contractor” definition
and the “suffer or permit” standard courts use to apply that
definition. 29
29Dynamex, supra, 4 Cal.5th at pages 961–962 (“[T]he
suffer or permit to work standard, by expansively defining who is
an employer, is intended to preclude a business from evading the
prohibitions or responsibilities embodied in the relevant wage
orders directly or indirectly—through indifference, negligence,
intentional subterfuge, or misclassification. It is well
established, under all of the varied standards that have been
utilized for distinguishing employees and independent
contractors, that a business cannot unilaterally determine a
worker’s status simply by assigning the worker the label
54
Thus, in Assembly Bill 5 the Legislature did not remove
any of the “exempted” industries from the workers’ compensation
system entirely, which is the much more radical step that
Proposition 22 takes. In the one situation prior to Proposition 22
where the Legislature created an industry-specific definition of
the term “independent contractor” (see Bus. & Prof. Code,
§ 10032, subd. (b) [certain real estate licensees]), there is an
exception for workers’ compensation (ibid. [“For purposes of
workers’ compensation the characterization of the relationship
shall be governed by section 3200, and following, of the Labor
Code”]).
That makes Proposition 22 the first attempt in the history
of California workers’ compensation to drop a class of wage
workers in one industry entirely from the workers’ compensation
system. A telltale sign of the constitutional questions raised by
this bold and unprecedented move is that, under the new
statutory definition of “independent contractor,” the eligibility of
app-share drivers for workers’ compensation benefits is
determined strictly by the employer’s unilateral designation in a
form contract—contrary to the “suffer or permit” standard—
‘independent contractor’ or by requiring the worker, as a
condition of hiring, to enter into a contract that designates the
worker an independent contractor.”); id. at p. 944 (“[A]t the time
the suffer or permit language was initially adopted as part of a
wage order in 1916, such language ‘was already in use
throughout the country in statutes regulating and prohibiting
child labor (and occasionally that of women), having been
recommended for that purpose in several model child labor laws
published between 1904 and 1912.’ ”).
55
which effectively bars these drivers from seeking to have their
“employment” status adjudicated by a decision maker within the
judicial branch. The route is indirect, but the result is a violation
of article III, section 3.
IV. The Argument That, in Proposition 22, Voter
Electors Made a Garden-variety Policy Choice,
Is Incorrect
Pointing to the language in article XIV, section 4,
empowering the Legislature to establish a “complete system of
workers’ compensation” for “any or all . . . workers,” the
Proposition 22 proponents contend that it was perfectly
permissible to withdraw a segment of workers from the original
statutory system by initiative statute, and to provide workplace
accident benefits for that segment of workers in a new and
different way. 30 There is nothing of significance to see here, we
30 The actual constitutional language quoted by the
Proposition 22 proponents on this textual point is that the
Legislature shall have the power to create and enforce a system
of workers’ compensation “liability on the part of any or all
persons”—in context, clearly meaning employers—“to compensate
any or all of their workers”—in context, clearly meaning
employees of said employers (hence the phrase “their workers”)—
“for injury or disability, and their dependents for death incurred
or sustained by the said workers in the course of their
employment, irrespective of the fault of any party.” (Art. XIV,
§ 4, italics added.) It is a stretch to construe the phrase “any or
all . . . workers” to refer broadly to statewide coverage, rather
than, as the context more naturally suggests, to the coverage
status of workers who may have been hired by a given employer,
an issue that seems amenable to determination only on an
employer-by-employer basis. But because the Proposition 22
proponents’ “any or all . . . workers” argument is flawed for other,
56
are told, because the voter electors simply exercised a policy
choice concerning whom to cover in the statutory workers’
compensation system differently than the Legislature did.
But this was no ordinary policy choice. Proposition 22
overturned a constitutionally ratified “social public policy” choice
the Legislature originally made in the Boynton Act and the 1917
Act, and then reaffirmed in Assembly Bill 5 in the course of
discharging its ongoing duty to implement the “complete system
of workers’ compensation” prescribed by article XIV, section 4. It
does not matter whether voter electors now have their own view,
different from the Legislature’s view, of the appropriate statutory
reach of that system. What is dispositive in the face of conflict on
this issue is that, as conditions changed over time, the
Legislature was specifically tasked with making the call based on
its view of what the Constitution required, and its call must be
respected.
The “any or all . . . workers” language in article XIV,
section 4, simply confirms that, by 1918, the Legislature had not
covered all potentially eligible wage workers. The coverage of the
statutory scheme improved considerably between 1911 and 1918,
but it was not fully comprehensive. While the objective of moving
from a voluntary system to a compulsory system was to extend
the reach of the statutory scheme, the 1917 Act still fell short of
universal coverage. The 1917 Act did not cover “casual” workers;
more fundamental reasons, as explained below, I am willing to
indulge arguendo the broader construction of these words that
they put forth.
57
it did not cover “domestic” workers; and it did not cover
“agricultural” workers. 31 “Casual” workers were not considered
wage workers under contracts for hire. They were either
volunteers or nominally paid temporary workers, and their status
as non-employees continues to be reflected in the Labor Code to
this day. 32
Why the Legislature created express carveouts for
“domestic” workers and “agricultural” workers is another story—
and not a happy one, since these two groups consisted largely of
workers of color—but what matters here is that the language
“any or all . . . workers” was broad enough to do two things. It
not only described the less-than-comprehensive coverage the
31 1917 Act, Statutes 1917, chapter 586, section 8(a).
32 See Labor Code, section 3352, subdivisions (a)(4), (a)(5),
(a)(7), (a)(9), (a)(10) and (a)(11); Gund v. County of Trinity (2020)
10 Cal.5th 503, 510 (“volunteers are typically not eligible” for
workers’ compensation benefits). The Proposition 22 proponents
cite Graczyk v. Workers’ Comp. Appeals Bd., supra,
184 Cal.App.3d 997, for the proposition that, because the
workers’ compensation system is statutory, a ballot statute may
be used to “withdraw” previously covered workers from the
system. They misread Graczyk. At issue there was a statutory
provision (former Lab. Code, § 3352, subd. (k)) added in 1981 to
clarify that unpaid student athletes were not employees eligible
for workers’ compensation. (Graczyk, at pp. 1001–1005.) Non-
wage workers were not covered in 1917, and in 1981 the
Legislature simply confirmed that this class of workers remained
outside the system. Graczyk would be on point if Proposition 22
simply clarified that casual carpool commuter drivers (the ones
that disappeared in the COVID-19 pandemic, while paid app-
based drivers kept working) are not covered by workers’
compensation.
58
Legislature had enacted by 1918, but at the same time confirmed
the Legislature’s authority—if it wished to exercise it—to allow
further expansion in the future. The Legislature exercised that
authority in 1959 for “agricultural” workers 33 and again in 1975
for “domestic” workers. 34 Voter electors could have done the
same thing by initiative statute. The holding in McPherson
confirms that. The eventual inclusion of those disfavored groups
was a mark of progress, but it did not throw open the original
core of the “complete system of workers’ compensation” for later
statutory revision. The holdings in Mathews and Hustedt confirm
that.
The Proposition 22 proponents cite various cases holding
that the workers’ compensation system is statutory and that
Statutes 1959, chapter 505, page 2466, section 1
33
(repealing statutory language carried over from the original
exclusion in the 1917 Act of “any employee engaged in . . . in
farm, dairy, agricultural, viticultural or horticultural labor, [or]
in stock or poultry” work, then codified at former Labor Code
section 3352 (Stats. 1937, ch. 90, § 3352(c), p. 267)); see S. G.
Borello & Sons, Inc. v. Dept. of Industrial Relations (1989)
48 Cal.3d 341 (after removal of the exclusion, addressing
circumstances in which agricultural workers may be considered
employees).
Statutes 1975, chapter 1263, page 3315, section 5.5
34
(repealing statutory language carried over from the original
exclusion in the 1917 Act of “any employee engaged in household
domestic service,” then codified at former Labor Code
section 3352 (Stats. 1937, ch. 90, § 3352(b), p. 267)); see In-Home
Supportive Services v. Workers’ Comp. Appeals Bd. (1984)
152 Cal.App.3d 720, 735 (after removal of the exclusion,
addressing circumstances in which domestic workers may be
considered employees).
59
courts have consistently rejected constitutional challenges when
the Legislature has made changes to the system. (See, e.g.,
Stevens, supra, 241 Cal.App.4th at pp. 1094–1096 [statute
eliminating workers’ compensation board’s authority to
determine medical necessity of treatment and vesting such
authority in an independent medical review organization];
Wal-Mart Stores, Inc. v. Workers’ Comp. Appeals Bd. (2003)
112 Cal.App.4th 1435 [statute limiting workers’ compensation
liability for benefits payable to treat psychiatric injury to workers
employed for more than six months, unless the injury was caused
by a sudden and extraordinary employment condition, held to be
consistent with article XIV, section 4].)
These cases resolve nothing here. Courts have long
recognized that the Legislature must be given wide berth to
adjust and refine the workers’ compensation system to meet the
needs of the day, a notion that is fully consistent with my reading
of article XIV, section 4. (See Stevens, supra, 241 Cal.App.4th at
p. 1096 [“it is not our place under the state Constitution to
‘second-guess the wisdom of the Legislature’ in making these
determinations”]; Facundo-Guerrero v. Workers’ Comp. Appeals
Bd. (2008) 163 Cal.App.4th 640, 651 [same].) No one disputes
that statutory changes may be made to the workers’
compensation system at any time if the Legislature determines
changes are necessary to ensure its effectiveness. And by
initiative statute, the voters may join in that ongoing project—
since they, too, may legislate on the same topic—as long as they
do so in a manner that is consistent with prior judgments of the
60
Legislature made by “appropriate legislation.” What is key is
that no case in this line either holds that “basic features” of the
pre-1918 workers’ compensation system may be eliminated by
statute, or involves a direct conflict between an initiative statute
and a statute passed by the Legislature.
It is ironic that the Proposition 22 proponents rely on the
phrase “any or all . . . workers” in support of their argument that
the scope of statutory workers’ compensation may be freely
expanded or contracted, accordion-like, without limit by statutory
initiative. They now seek to justify for app-based drivers the
same kind of second-class citizenship treatment that agricultural
and domestic workers were given in the original policy debate
over the reach of workers’ compensation coverage. The
interveners are quite explicit about this. They point to the
pre-1918 exclusion of agricultural and domestic workers as proof
that app-based drivers may be statutorily excluded today. About
that, all I will say is that the United States Supreme Court may
be willing to read the federal Constitution in a manner that
doubles down on the mistreatment of fellow citizens who were not
considered “full and equal” when the Fourteenth Amendment
was adopted (Dobbs v. Jackson Women’s Health Org. (2022)
___U.S.___ [142 S.Ct. 2228, 2329] (dis. opn. of Kagan, J.)), but in
reading our state Constitution, I am not. These exclusions are a
historical embarrassment, not a license to create new social
hierarchies by statute.
Which brings me back to footnote 9 in McPherson. Moved
to do so by the same progressive reform movement that brought
61
us the powers of direct democracy, the Legislature made a
fundamental choice to cover as many wage workers under
contracts for hire as it could between 1911 and 1917. Many
decades later, in Assembly Bill 5, the 2019 Legislature expressed
its view that the inclusion of app-based drivers within the
workers’ compensation system is required. As I see things, the
Legislature’s determination—which represents its interpretation
of what the term “employment” means for purposes of the
“complete system of workers’ compensation” codified in article
XIV, section 4—must prevail over that of the Proposition 22 voter
electors. Voter electors retain ultimate power to override the
Legislature, but in this context they must do so by constitutional
amendment. (See Legislature v. Deukmejian (1983) 34 Cal.3d
658, 674 [“[I]t was at no time intended that . . . permissive
legislation by direct vote should override the other safeguards of
the constitution. If an amendment of the constitution were
intended, the provision requires steps to be taken that will
apprise the voters thereof so that they may intelligently judge of
the fitness of such measure as a constituent part of the organic
law.”].)
V. Conclusion
“[I]f any portion, section, subdivision, paragraph, clause,
sentence, phrase, word, or application of ” Business and
Professions Code section 7451—the “independent contractor”
definition adopted in Proposition 22—“is for any reason held to be
invalid by a decision of any court of competent jurisdiction, that
decision shall apply to the entirety of the remaining provisions of
62
this chapter, and no provision of this chapter shall be deemed
valid or given force of law.” (Bus. & Prof. Code, § 7467, subd. (b).)
I believe the “independent contractor” definition in Proposition 22
is constitutionally infirm and that, as a result, the entire
initiative by its own terms must fall.
It is undoubtedly true that “[t]he amendment of the
California Constitution in 1911 to provide for the initiative and
referendum signifies one of the outstanding achievements of the
progressive movement of the early 1900’s.” (Associated Home
Builders etc., Inc. v. City of Livermore (1976) 18 Cal.3d 582, 591.)
But article XIV, section 4, was another outstanding achievement
of progressive era reform, and it must be treated with equal
dignity.
In article XIV, section 4, the people gave constitutional
sanction to an elaborate, pre-1918 statutory workers’
compensation scheme. To implement the scheme, they conferred
plenary power on the Legislature, and the Legislature alone.
This expansive and singular delegation of authority, a species of
power that may be found nowhere else in our charter document,
is “unlimited by any provision of [the] . . . Constitution,” and
imposes on the Legislature an ongoing duty of enforcement. (Art.
XIV, § 4.) The pre-1918 statutory scheme itself, together with
“appropriate legislation” enacted from time to time pursuant to
the Legislature’s enforcement power, reflects a “social public
policy” that is “binding upon all departments of the state
government.” (Ibid.)
63
The central dilemma posed by this case is that
Proposition 22 flies in the face of the Legislature’s declared
“social public policy” in the field of workers’ compensation, as
most recently reflected in its enactment of Assembly Bill 5. An
integral part of the workers’ compensation system the
Legislature has been implementing for more than a century
pursuant to its article XIV, section 4, power is that, with two
exceptions that were long ago eliminated, all wage workers are
entitled to have their employment status determined by decision
makers exercising judicial power.
By jettisoning app-based drivers from this constitutionally
mandated system, the independent contractor definition in
Proposition 22 not only violates the plain terms of article XIV,
section 4—because the system that remains is not “complete,” as
the Legislature construed this constitutional requirement in the
pre-1918 system—but conflicts directly with the Legislature’s
recent exercise of its article XIV, section 4, power in Assembly
Bill 5. For both of these reasons, Proposition 22 must be
invalidated as violative of article XIV, section 4. And because
Proposition 22 destroys the ability of app-based drivers to have
their employment status determined within the judicial branch of
government, it must also be invalidated as violative of article III,
section 3.
I would affirm in all respects.
STREETER, J.
64
Trial Court: Superior Court of California, County of Alameda
Trial Judge: Hon. Frank Roesch
Counsel: Rob Bonta, Attorney General, Thomas S. Patterson,
Senior Assistant Attorney General, Mark
Beckington, Supervising Deputy Attorney
General and Jose A. Zelidon-Zepeda, Deputy
Attorney General, for Defendants and
Appellants.
O’Melveny & Myers, Jeffery L. Fisher; Nielsen
Merksamer Parrinello Gross & Leoni,
Arthur G. Scotland, Sean P. Welch, Kurt R.
Oneto and David J. Lazarus, for Interveners
and Appellants.
DLA Piper, Stanley J. Panikowski and Justin R.
Sarno for Former Attorney General of
California Daniel E. Lungren as Amicus Curiae
on behalf of Defendants and Appellants.
Eimer Stahl, Robert E. Dunn and Collin J. Vierra
for Citizens in Charge as Amicus Curiae on
behalf of Defendants and Appellants.
David A. Carrillo, California Constitution Center,
University of California, Berkeley; Benbrook
Law Group, Stephen M. Duvernay for
California Constitution Center as Amicus
Curiae on behalf of Defendants and Appellants.
Davis Wright Tremaine, Rochelle L. Wilcox and
Alexa A. Graumlich for California Chamber of
Commerce as Amicus Curiae on behalf of
Defendants and Appellants and Interveners
and Appellants.
Castellanos v. State of California – A163655
1
Holtzman Vogel Baran Torchinsky Josefiak, Alex
Vogel, Edward M. Wenger and Andrew Pardue
for California Policy Center as Amicus Curiae
on behalf of the Defendants and Appellants and
Interveners and Appellants.
Willenken, Kenneth M. Trujillo-Jamison for
California Asian Pacific Chamber of Commerce,
California Hispanic Chambers of Commerce,
Los Angeles Metropolitan Churches, National
Action Network Los Angeles, National Action
Network Sacramento Chapter Inc., National
Asian American Coalition, and National
Diversity Coalition (“Communities-of-Color
Organizations”) as Amici Curiae on behalf of
Defendants and Appellants and Interveners
and Appellants.
Baker Botts, Michael W. Ward for Arnold
Schwarzenegger as Amicus Curiae on behalf of
Defendants and Appellants and Interveners
and Appellants.
Howard Jarvis Taxpayers Foundation, Jonathan M.
Coupal, Timothy A. Bittle and Laura E.
Dougherty for Howard Jarvis Taxpayers
Association as Amicus Curiae on behalf of
Defendants and Appellants and Interveners
and Appellants.
Jenner & Block, Laurie J. Edelstein and Adam G.
Unikowsky for Chamber of Commerce of the
United States of America as Amicus Curiae on
behalf of Defendants and Appellants and
Interveners and Appellants.
Castellanos v. State of California – A163655
2
King & Spalding, Albert Giang, Jeffery Hammer
and Ramon A. Miyar for Marketplace Industry
Association, Inc. as Amicus Curiae on behalf of
Defendants and Appellants and Interveners
and Appellants.
Buchalter, Steven G. Churchwell and Berit Elam
for Daniel Schnur, T. Anthony Quinn and
Robert M. Stern as Amici Curiae on behalf of
Defendants and Appellants and Interveners
and Appellants.
weintraub tobin chediak coleman grodin,
Brendan J. Begley for Independent Drivers
Alliance of California, Kelly Rickert, Ali
Mazhin and Stephanie Whitfield as Amici
Curiae on behalf of Defendants and Appellants
and Interveners and Appellants.
Olson Remcho, Robin B. Johansen, Richard R. Rios,
Deborah B. Caplan, Benjamin N. Gevercer;
Altshuler Berzon, Stephen P Berzon, Scott A.
Kronland, Stacey M. Leyton, Juhyung Harold
Lee for Plaintiffs and Respondents Hector
Castellanos, Joseph Delgado, Saori Okawa and
Michael Robinson.
Service Employees International Union, Nicole G.
Berner and Steven K. Ury for Plaintiff and
Respondent Service Employees International
Union.
Richard L. Hasen; Public Counsel, Mark D.
Rosenbaum and Kathryn Eidmann for
California Election Law Professors as Amicus
Curiae on behalf of Plaintiffs and Respondents.
Castellanos v. State of California – A163655
3
Hina B. Shah, Women’s Employment Rights Clinic,
Golden Gate University School of Law
Benjamin Beach, PowerSwitch Action;
Nayantara Mehta and Brian Chen, National
Employment Law Project, for Gig Workers
Rising, Mobile Workers Alliance, Rideshare
Drivers United-California, We Drive Progress,
A Better Balance, ACCE Institute, Action
Center on Race & the Economy, Asian
Americans Advancing Justice – Los Angeles,
Bet Tzedek, California Employment Lawyers
Association, California Immigrant Policy
Center, Centro Legal de la Raza, Chinese
Progressive Association, Economic Policy
Institute, Jobs With Justice Education Fund
and Jobs With Justice San Francisco, Lawyers
Committee for Civil Rights of the San Francisco
Bay Area, Legal Aid at Work, Los Angeles
Black Worker Center, Maintenance
Cooperation Trust Fund, National Black
Worker Center, National Council for
Occupational Safety and Health, National
Domestic Workers Alliance, National
Employment Law Project, Pilipino Workers
Center, PowerSwitch Action, Public Rights
Project, Santa Clara County Wage Theft
Coalition, Women’s Employment Rights Clinic
of Golden Gate University School of Law and
Worksafe, as Amici Curiae on behalf of
Plaintiffs and Respondents.
Bush Gottlieb, Julie Gutman Dickinson, Hector
De Haro, and Luke Taylor for International
Brotherhood of Teamsters Local 848 and The
Los Angeles County Federation of Labor, AFL-
CIO, as Amici Curiae on behalf of Plaintiffs and
Respondents.
Castellanos v. State of California – A163655
4
City of San Francisco, David Chiu, City Attorney,
Sara J. Eisenberg, Chief of Complex and
Affirmative Litigation, and Molly J. Alarcon
Deputy City Attorney; City of Oakland,
Barbara J. Parker, City Attorney, Maria Bee,
Chief Assistant City Attorney, Zoe Savitsky,
Supervising Deputy City Attorney and
Katherine Read, Fellowship Attorney; City of
Los Angeles, Michael N. Feuer, City Attorney,
Kathleen Kenealy, Chief Assistant City
Attorney, and Michael J. Bostrom, Senior
Assistant City Attorney for the Cities of San
Francisco, Oakland and Los Angeles as Amici
Curiae on behalf of Plaintiffs and Respondents.
Eric M. Overholt and Andrew W. Lockard for
California Applicants’ Attorneys Association as
Amicus Curiae on behalf of Plaintiffs and
Respondents.
Catherine L. Fisk, University of California,
Berkeley; Veena Dubal and Joseph Grodin,
Emeritus Professor of Law, University of
California, Hastings College of Law, for Labor
Law Professors Sameer Ashar, Veena Dubal,
Catherine Fisk, Charlotte Garden, Joseph
Grodin, William B. Gould IV, Stephen Lee,
Sanjukta Paul, Leticia Saucedo, Reuel Schiller,
Katherine Stone and Noah D. Zatz as Amici
Curiae on behalf of Plaintiffs and Respondents.
Castellanos v. State of California – A163655
5