Understanding The California Ruling That Said Prop 22, The Gig Worker Ballot Initiative, Was Unconstitutional
from the laborious-constitutional-interpretation dept
Opponents of Uber et al. have been cheering the recent California court decision declaring Proposition 22 unconstitutional. Proposition 22 was a ballot measure passed to override significant parts of the legislature’s AB 5 bill, which affected all sorts of untraditional employment arrangements, including those of “gig workers.” Some people unhappy with the policy effects of Proposition 22 then sued to challenge its validity under the California Constitution. And, at least initially, have won.
Whether it actually is a victory for labor is debatable but also somewhat besides the point. The relative merits of any of these things (Uber, Prop. 22, AB 5) is not what’s at issue. Instead, the question is whether the decision correctly interprets the California Constitution.
The California Constitution is, let’s face it, kind of weird. Many state constitutions mirror the US Constitution with the way they are articulated. Not so the California Constitution, which reads much more like a laundry list of specific policies. As a result, it is more changeable than other constitutions, although given all the specific policies that can get baked into it, perhaps not always changeable enough.
Ultimately the court found two aspects of Prop. 22 (but only two aspects, despite the challengers’ arguments) to be unconstitutional given the current incarnation of the California Constitution: the language in Section 7451 about Workman’s Compensation, and the language in Section 7465(c)(4) about amending the law put on the books by the proposition. Each had a different constitutional problem.
The problem with the latter related to those potential future amendments. In general, the legislative effects of propositions are not easily changeable, which is something we’ve taken issue with before. If they were easily changeable then the legislature could simply pass a new law by a normal majority. But with ballot measures, the changes would require another ballot measure, which someone would need to be able to spend millions of dollars to underwrite. It’s this unchangeability that has allowed wealthy parties to essentially blackmail the legislature into passing laws they like (see, e.g., the CCPA), because if they were able to fund the passage of these laws directly via a ballot measure, then there would be little the legislature could ever do to change them later, no matter how much they might need to be changed. Sometimes ballot measures include eased requirements for the legislature to make changes, but by default they do not. As the decision itself explains:
The California Constitution provides that the people of the state may enact laws through the initiative process. (Cal. Const. art. I, ? 8.) When the people pass an initiative statute, the Legislature’s power to amend that statute is limited by the California Constitution: ?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.? (Cal. Const. art. I, ? 10, subd. (c).) Because the voters have the power to limit or allow amendment to their initiative statutes, they also have the power, a fortiori, to attach conditions to permissible amendments. (See Howard Jarvis Taxpayers Assn. v. Newsom (2019) 39 Cal. App.5th 158, 167.) [p.5]
In other words, as the decision itself drives home, the California Constitution, by design, creates a path for essentially permanent legislating via direct democracy. Whether that’s a good thing or not is a separate question. But that’s what the Constitution allows, so that’s what the proponents of Prop. 22 did. The question before the court is if there might have been some exception to that general rule that this particular proposition fell into. Here the court found one, after digging around a bit.
First, it was fine that the proposition made it hard to be changed, because it was always going to be hard.
Everything in Section 7465 is in the nature of an exception to the default amendment rule in Article I, Section 10, Subdivision (c). If Section 7465 had not been included, the Legislature could amend Proposition 22 by a simple majority vote according to each house’s rules, followed by a popular referendum. With Section 7465 enacted, the Legislature can still amend Proposition 22 by a simple majority vote according to each house’s rules, followed by a popular referendum. (See Cal. Const. art. II, ? 10., subd. 4 (2). All Section 7465 provides is another way to amend the initiative statute, albeit one that is difficult to the point of near impossibility. [p. 6-7]
But the court decided it didn’t like how Prop. 22 made it hard.
Proposition 22 also included an unusual provision allowing the Legislature to amend its provisions using an unusual procedure. The legislature may amend Proposition 22 ?by a statute passed in each house of the Legislature by rollcall vote entered into the journal, seven-eighths of the membership concurring, provided that the statute is consistent with, and furthers the purpose of, this chapter.? (Bus. & Prof. Code, ? 7465, subd. (2),) ?Any statute that amends Section 7451 does not further the purposes of this chapter.? (Id., subd. (c)(2).) [p. 5]
Therein lies the rub, because what did it mean to be “for purposes of this chapter”? The proposition had some language on that, which the court cited:
Proposition 22 also provides two additional specific definitions of what constitutes an amendment: ?[a] 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? (id, subd. (c)(3)) and a ?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? (id., subd. (c)(4)). [p. 5]
And that’s the language the court took issue with. Not with (c)(3) ? that language passed muster ? but with the language of (c)(4), because it appeared to impact collective bargaining relationships.
Subdivision (c)(4) is not so simple. There is no other language in Proposition 22 that 20 directly relates to labor representation or collective bargaining. The Proposition proponents argue that independent contractor status is incompatible with collective bargaining: that ?[o]ne of the fundamental issues Prop 22 addresses is the right of app-based drivers to work as independent contractors?a status that precludes them from collective bargaining under a century of state and federal law.? (Proponents? Mem. P&A Opp. Pet. at p24.) They further argue that ?any subsequent attempt by the Legislature to reimpose on app-based drivers traditional employment relationships like collective bargaining rights would ?undo? this choice.? (ibid) But the most maximal state law covered only by Subdivision (c)(4) would create a guild through which independent contractors would bargain collectively their contract terms and working conditions. This may alter their bargaining power vis-a-vis the network companies they contract with, but the Court cannot find that it would diminish their ?independence? or transmute them into employees. The Court therefore finds that Subdivision (c)(4) unconstitutionally purports to limit the Legislature’s ability to pass future legislation that does not constitute an ?amendment? under Article I, Section 10, Subdivision (c). [p. 9-10]
To the court, the problem was that propositions were limited to being only about a single “subject”:
Initiative statutes must be limited to a single ?subject? (Cal. Const. art. I,? 8(d) [?An initiative measure embracing more than one subject may not be submitted to the electors or have any effect”) Courts interpret the term ?subject? liberally to uphold initiative statutes ?which disclose a reasonable and common-sense relationship among their various components in furtherance of a common purpose.? (Brosnahan v. Brown (1982) 32 Cal 3d 236,253.) The general test is whether the parts of a statute are ?reasonably germane to a common theme, purpose, or subject.? (Brown v. Superior Court (2016) 63 Cal 4th 335, 350.) [p. 10]
And to the court, the collective bargaining had nothing to do with the subject of the proposition.
The common ?theme, purpose, or subject? of Proposition 22, then, is protecting the opportunity for Californians to drive their cars on an independent contract basis, to provide those drivers with certain minimum welfare standards, and to set minimum consumer protection and safety standards to protect the public. [?] No other part of Proposition 22 deals with collective bargaining rights other than Section 7465, subdivision (4), and it does so only obliquely and indirectly, as a side effect of a contested construction of certain antitrust laws as barring independent contractors from bargaining collectively. This is related to Proposition 22’s subject but it is utterly unrelated to its stated common purpose. A prohibition on legislation authorizing collective bargaining by app-based drivers does not promote the right to work as an independent contractor, nor does it protect work flexibility, nor does it provide minimum workplace safety and pay standards for those: workers. It appears only to protect the economic interests of the network companies in having a divided, ununionized workforce, which is not a stated goal of the legislation. [p. 11]
Meanwhile, the court also found a constitutional problem with Section 7451, which impacted eligibility for Workman’s Compensation. The issue here is that one of the policies baked into the Constitution is that the Legislature gets to decide who is eligible for Workman’s Compensation. The court calls this power “plenary,” meaning that it’s built into the Legislature, and, per the court, can only be affected by a constitutional amendment. Now, a constitutional amendment could be passed by ballot initiative, but Proposition 22 was an initiative statute, not an initiative amendment, and so, per the court, it did not get to meddle with that Legislature authority.
Proposition 22 is constitutionally problematic for another reason that defies such easy resolution. Petitioners and amici law professors also make the more subtle argument that the Independent Energy Producers case is distinguishable because the statutory initiative in that case increased the power to the Public Utilities Commission, whereas Proposition 22 limits a power vested in the state legislature by the Constitution. (See Independent Energy Producers Assn., (supra, 38 Cal.4th at p.1044 fn.9.) Article XIV, Section 4 also provides that the Legislature shall have the power to create worker’s compensation laws ?unlimited by any provision of this Constitution.? (Cal. Const. art. XIV, ? 4.) However, the Constitution also provides that the Legislature may not act to amend or repeal an initiative statute without a subsequent vote of the 5 people. These two provisions are in conflict. If the Legislature?s authority is limited by an initiative statute, its authority is not ?plenary? or ?unlimited by any provision of [the] Constitution? (Cal. Const. art. XIV, ? 4); rather, it would be limited by Article I, Section 10, subdivision (c). The Supreme Court has held that, as an interpretive guide, the initiative power should be zealously protected and ?any reasonable doubts? should be resolved ?in favor of the exercise of this precious right? (Kennedy Wholesale, Inc. v. State Bd. of Equalization (1991) 53 Cal.3d 245,250) But here, the plain language of Article XIV, Section 4 indicates that it is ?unlimited by any provision of? the California Constitution. (Cal. Const. art. XIV, ? 4) When Section 4 was ratified in 1918, the statutory initiative power already existed in the Constitution. The grant of plenary power to the Legislature conflicts with a limitation on its power to amend an initiative statute under Article If Section 10. The grant of power is not ?plenary? if the Legislature’s power to include app-based drivers in the worker’s compensation program is limited by initiative statute. It is not ?unlimited by any provision of this Constitution? if it is limited by an initiative statute. The plain meaning of Article XIV, Section 4’s plenary-and-unlimited clause governs over the more general limitation on amendment in Article II Section 10. In short, if the People wish to use their initiative power to restrict or qualify a ?plenary? and ?unlimited? power granted to the Legislature, they must first do so by initiative constitutional amendment, not by initiative statute. [p. 3-4]
In theory neither of these apparent defects, with subsection (c)(4) of 7465 or Section 7451, would be fatal to the entire proposition, as both included severability provisions, meaning that if either were found to be invalid then they could essentially be excised from the proposition but the rest could stay in force. But, in a mere two lines, the court decides that Section 7451 was not severable, and so therefore the entire proposition is nullified. [p. 12].
So that’s where we are, with the whole statute thrown out. But whether that will remain the result after appeal, or should remain the result, is far from clear.
First, the decision hardly defends why Section 7451 is not severable. While its analysis on why either it or Section 7465(c)(4) is unconstitutional may end up vindicated later, it seems quite possible that a later court would instead opt to excise the problematic language without invalidating the whole statute (or at least require the question of whether it could be to be reconsidered).
As for the unconstitutionality of each provision, it is really more a matter of constitutional interpretation than a matter of labor policy, although as it happens the interpretation is not completely divorced from labor policy since aspects of labor policy are inherently part of the constitutional language, at least vis-à-vis the Workman’s Compensation part affected by Section 7541. If the court is right about the significance of that “plenary” language, then it may also be right that the Prop. 22’s proponents used the wrong type of ballot initiative to speak on that issue with this proposition.
But the “single subject” analysis of the Section 7465(c)(4) changeability provision seems more dubious. The court here made much more of a leap to decide that collective bargaining arrangements were beyond the “purpose” of the proposition, and thus not something that the changeability provision could impact. Here the court appeared to be projecting a more subjective understanding about the interplay between gig workers and collective bargaining arrangements, and in a way that, while perhaps plausible, does not seem to be well-substantiated or inevitable. This aspect of the holding seems vulnerable to being overruled on appeal, or at least remanded for further consideration.
In sum, it’s all an uncertain mess of questionable labor policy and quixotic constitutional mechanics for which we are a long way off from any resolution. Proposition 22 will continue to grind through the courts, but even once all the proceedings have been concluded we’ll still be left with thorny questions of both labor policy and constitutional order.
The effective permanence of ballot measures raises huge governability issues. It is hard to thus root for any ballot measure, regardless of topic, to succeed in the face of a constitutional challenge. At the same time, however, as long as those are the rules, those are the rules, and we shouldn’t favor a proposition being treated differently by the courts just because we might prefer the substantive result if it were.
On the other hand, while we tend to favor legislating by legislature in order to avoid those problems of permanence, labor policy is one area where the legislature did not get things right. One other upside to legislating via legislature is that legislation can be more considered and based upon a thorough and thoughtful legislative record. But AB 5 is an example of one law where the legislature failed to adequately consider the impact of its policy and ended up producing a law that had all sorts of harmful consequences to the very worker interests they were ostensibly trying to support.
For better or for worse, the ballot initiative process exists for cases like these where the legislature gets policy wrong and then doesn’t fix it (although, to be fair, it did mitigate a few of the problems with AB 5, but not all). Overturning a legislative decision via direct democracy is exactly the political process the California Constitution envisions and invites with the ballot initiative process. Whether, however, it’s a process the California Constitution should invite may be something worth reconsidering, particularly in its current, overly-permanent form. But not on the backs of a single initiative on a single issue of political contention. Because even if you hate the policy that resulted from it, it’s the result of the system working as designed. To change that result you have to first change the system.