Advisory Measures and the Legislature’s Power to Investigate
In a recent decision, Howard Jarvis Taxpayers Assn. v. Padilla, the California Supreme Court considered the validity of Proposition 49, which would have asked the electorate whether Congress should propose, and the legislature ratify, a federal constitutional amendment overturning the U.S. Supreme Court decision in Citizens United v. Federal Election Comm. (2010). The court held that the legislature’s investigative power permits it to place advisory measures on the ballot for the electorate’s consideration.
For what it’s worth, we concur.
The court’s order that the measure be taken off the ballot seemed dramatic enough, as the general rule is that the court will wait to review constitutional challenges to ballot measures after an election, “in the absence of some clear showing of invalidity.” Brosnahan v. Eu (1982) at 4. Or is it? There are enough exceptions to that general rule to make it less of a rule and more of a guideline. A claim that a measure violates a state constitutional provision (e.g., that it violates the single-subject rule) may be considered before the election. Cal. Const., Article II, section 8, subdivision (d); Senate of the State of Cal. v. Jones (1999) at 1153-54. And pre-election claims that procedural requirements for qualifying a measure (e.g., lacking enough qualified signatures) are equally proper. Costa v. Superior Court (2006) at 1006. Thus, pre-election relief is available whether a measure’s claimed defect is substantive or procedural. That being so, the court’s action in ordering Proposition 49 off the ballot was unusual only in that the court rarely finds it necessary to do so, and not as a novel development in initiative law. It was necessary here because if the measure had been left on the ballot, the harm (if any) could not be remedied: regardless how the voting turned out, the state legislature would already have the results of its poll. Overturning Proposition 49 after a vote would achieve nothing, particularly given the court’s reluctance to issue advisory opinions of its own. Pacific Legal Foundation v. California Coastal Comm. (1982) at 170.
Still, as Professor Michael B. Salerno observed in his Daily Journal article (October 1, 2015) on the order to show cause that removed Proposition 49 from the ballot, on first impression this case seemed to raise important separation of powers issues. Those issues took two forms: the judicial branch could be impairing a core legislative power by removing Proposition 49 from the ballot, and the legislature could be exceeding its power with its attempted advisory measure. But the court rejected those concerns and upheld the legislature’s action. The first issue (impairing legislative power by blocking the measure) turned out to be at most a temporary matter, now resolved with the decision that such measures are proper. At most, then, the legislature’s exercise of its power has been delayed, not denied.
On the second issue (the legislature exceeding its power), there is little concern that the powers of the other branches will be impaired if the legislature asks for a nonbinding expression of popular will. It is difficult to see how an opinion poll (as Proposition 49 essentially is) can impinge on the powers of the executive or the judiciary. There is even less concern about affecting the electorate’s powers. It is true that for many purposes, the electorate’s direct democracy powers exist separately from the legislature’s, and there are legitimate separation of powers issues that can arise between the legislature and the electorate. In fact, the original purpose of creating the initiative was to empower the electorate to evade the legislature entirely and legislate directly—thus the moniker direct democracy.
But in other situations, the electorate and the legislature operate in tandem: in the former indirect initiative, in the current legislatively proposed amendment-and-revision process under Article 18, section 1, and in the convention procedure under Article 18, section 2. And from a certain perspective the legislature and the electorate each exercise a portion of the whole legislative power of the state, because the electorate acts as a legislative entity when using its initiative power. Santos v. Brown (2015) at 409. As Professor Salerno pointed out, the state constitution contemplates a “mix-and-match approach” to legislation. If the state constitution permits such a power-sharing and dual-action scheme to take action on the high level of constitutional change, surely a legislative inquiry into voter approval of what amounts to a nonbinding resolution poses no threat to the balance of power in the state government.
There is little incentive, given the nature of a state constitution, to take a crabbed view of the legislature’s power. On the contrary, the natural course when the legislature and the electorate would exercise their political power in concert—as they do here—is for the judicial branch to demur; as the opinion notes, it would be no small matter for the judiciary to annul such an act. And it is not difficult to justify this decision on either theoretical or structural grounds.
From a doctrinal perspective, just as in the speech context—where the remedy for discomfiting speech is more speech, not a restraint—the solution here is more democracy, not foreclosing a line of communication between the people and their elected representatives. It is no objection to point out that the electorate already communicates its feelings about policy issues and candidates through polls and elections; the existence of some political tools in this context should not be taken to preclude others. Similarly, nothing prevents electors from directly communicating with their representatives, or simply obviating the entire communication issue and enacting their own laws. These obvious means of instructing one’s representatives should neither be exclusive nor be used as a justification for precluding reasonable alternate means.
What would a contrary decision from the court look like? One could rely on transaction costs through post-passage remedies like waiting to vote legislators out of office, recalls, or reversing unpopular measures with the initiative or by other means (e.g., the 18th and 21st amendments to the federal constitution). One might argue that a special election has associated costs. But if the advisory measure is on the regular primary or general election ballot, the incremental cost of one more box to tick and one more ballot pamphlet argument to read is negligible. It is true that permitting advisory measures as a new species of initiative measure permits competing advisory measures on the same issue. This could lead to several more boxes to tick and several more ballot pamphlet arguments to read, in an arms race of advisory measures championed by competing interest groups. For example, the legislature proposes outlawing puppies, and in retaliation Howard Jarvis gathers signatures for an advisory measure on whether puppy ownership by members of the legislature should be mandatory. But the initiative system already permits determining public policy through competing initiative measures. And even if there is a campaign battle over the associated advisory measures, will that cost be greater than the expenditures of an angry electorate seeking to reverse a legislative action? In this scenario the relative harms of asking permission and begging forgiveness seem equivalent.
Structurally, the opinion poll can be viewed as the reverse of the petition right. The federal constitution protects the people’s right “to petition the government for a redress of grievances.” U.S. Const., 1st Amend. That right arose from “the dialogue of petition and response between inhabitants and colonial assemblies” and “was intimately related to the structure of colonial politics.” Stephen A. Higginson, A Short History of the Right to Petition Government for the Redress of Grievances (1986) at 145. When the colonial petition right was included in the 1789 federal constitution, a citizen was not merely entitled to an opportunity to petition. The legislature was required to hear it and respond: “That the Framers meant to imply a corresponding governmental duty of a fair hearing seems clear given the history of petitioning in the colonies and the colonists’ outrage at England’s refusal to listen to their grievances.” Id. at 155. The text of the California constitution makes the relationship even more clear: Article 1, section 3, subdivision (a) provides that the people “have the right to instruct their representatives [and] petition government for redress of grievances . . . .” (emphasis added). It is difficult to imagine that the delegates to the 1878 California convention would have viewed that line of communication as an exclusively one-way street. And if the legislature wants to ask the electorate to instruct it, why is it necessary to read into the people’s right to instruct a limitation that the instructions must be sua sponte, and cannot be in response to a question?
This decision leaves open an interesting question. The decision is based on the legislature’s investigative power, which necessarily means relying on the advisory aspect of advisory measures. What should happen if the legislature places a question on the ballot, such as whether the legislature should take certain action; the electorate votes in the affirmative; and the legislature fails to act? As the opinion noted, a power or right to instruct representatives raises important conceptual questions: “Are representatives independent or agents? Do they represent the constituents of their district or the entire state or country? If a right to instruct were granted, would instructions be binding?” Padilla at 37. The conundrum posed by those questions led to an instruction right being omitted from the federal constitution, but there it is in Article 1 section 3 of the California constitution. The opinion notes that the court has never delineated the bounds of that right. It is easy to foresee, now that a procedure exists for the legislature to ask for instructions, a future case arising that requires examination of those difficult agency questions and the Guarantee Clause if the legislature does not follow the electorate’s instructions. David A. Carrillo and Stephen M. Duvernay, California Constitutional Law: The Guarantee Clause and California’s Republican Form of Government (2014).
Questions about the power of California government start from the premises that all political power is inherent in the people (Article 2, section 1) and that the legislature has plenary power to enact laws Marine Forests Society v. California Coastal Comm. (2005) at 31. It is true, as the opinion noted, that the legislative power at its core is to enact laws, and it is true that the initiative power has been limited to measures that enact laws. But that reductive description of the broad power to legislate is inadequate support for the argument that an advisory measure is unconstitutional because it does not actually enact a statute. Absent a structural limit (like a separation of powers concern, dismissed above), when the electorate and the legislature act in concert, that is powerful medicine; indeed, it is enough to revise the constitution itself (via Article 18). Again, the fundamental principle here is that the California constitution is a limitation of the plenary power of the state government, and the general rule is that the state legislative power is broad enough to encompass any act not forbidden by the state constitution. Small wonder then that when presented with the question “Can the legislature work with the electorate to resolve a policy issue?” the court’s answer is, “Yes, it can.”
Center fellows Megan Somogyi, Stephen M. Duvernay, and Brandon Stracener contributed to this article.
- A trivia challenge for the SCOCA staff attorneys - June 30, 2021
- It’s official: SOS chooses easy recall qualification procedure - June 16, 2021
- Event announcement: California redistricting 2021 - April 20, 2021
 Cf. Captain Barbossa, Pirates of the Caribbean: The Curse of the Black Pearl (2003): “[T]he code is more what you’d call guidelines than actual rules.”