California’s Electorate Runs the Game in Recall Elections

Democracy Law and Human Rights
Written By David A. Carrillo, Joshua Spivak, and Brandon V. Stracener
L-jorge-maya-untZdZHOpDc-unsplash.jpg

Abstract

California voters hold a high degree of power in the state’s recall process. Combined with the state’s other direct democracy tools (the initiative statute, the initiative constitutional amendment, and the referendum) the state electorate has power that approaches that of a state government branch. We measure that power by comparing the California electorate to the other government branches and by comparing California’s recall to that of other states. We provide recall examples and conclude by evaluating the arguments for and against vesting this degree of power in the voters. We conclude that concerns about the recall power itself are unwarranted; yet calls for reforming its procedures are meritorious.

Compared with other California branch actors, the voters are relatively powerful

California is currently experiencing a spate of recalls: against two San Francisco School Board members, a Los Angeles City Councilperson, two San Diego Superior Court judges, the Los Angeles, San Francisco, and Sonoma County District Attorneys, and Governor Gavin Newsom. See Cowan, Local Recalls, by the Numbers, The New York Times July 9, 2021; Spivak, The Year in Recalls, The Recall Elections Blog December 30, 2020. The Governor and Sonoma County District Attorney recalls are heading to the ballot, and proponents might still qualify the recalls against the San Francisco District Attorney and the San Francisco School Board members. One causal factor for the current amount of recall activity in California is the high degree of political power its electorate holds in general, and specifically the state’s comparatively simple recall election qualifying requirements.

The California electorate holds four direct democracy tools that permit the voters to make major changes to the state’s constitution, statutes, and roster of elected officials. (California’s constitution uses electorate for the more colloquial voters; we use both here interchangeably.) Those direct democracy tools are the recall, referendum, initiative statute, and initiative constitutional amendment. Cal. Const. art. II, sections 8(a), 9(a), 13; art. XVIII, section 3. The recall permits California’s voters to remove an elected official in a special election before that official’s term expires. Cal. Const. art. II, section 13. The referendum is the electorate’s power to veto statutes passed by the state legislature. Id.art. II, section 9(a). And the voters can use the initiative to legislate by enacting statutes or constitutional amendments through measures placed on the ballot that may pass by majority vote. Id. art. II, section 8(a); art. XVIII, section 3.

Although two additional tools could be considered direct democracy -- the constitutional revision and the constitutional convention (Cal. Const. art. XVIII, sections 1, 2) -- we omit those from our discussion here because exercising those powers requires the state legislature to participate. The voters can employ the four tools we discuss here with no one’s leave but their own.

Compared with each other, these four direct democracy tools all require approximately the same degree of effort to qualify for a ballot. To qualify a statewide officer recall, a petition must be signed by a number of registered voters equal to 12% of the last vote for that office. Cal. Const. art. II, section 14(b); Cal. Elec. Code section 11221(c). The petition must also include signatures from at least five different counties that equal at least 1% of the last vote for that office in each of those five counties. Initiative statutes and initiative constitutional amendments have a lower signature threshold requirement: 5% and 8% respectively of the votes for all candidates for governor in the last gubernatorial election. Cal. Const. art. II, section 8(b). The referendum has the same signature requirement (5%) as initiative statutes. Id. art. II, section 9(b).

These tools all operate almost entirely free from interference by government officials. Once an initiative measure is passed by the electorate, the legislature cannot alter that measure without the electorate’s consent and there is no executive veto; similarly with the referendum the legislature cannot override the voters by simply re-passing the same statute. Cal. Const. art. II, section 10(a), (c) (permitting the legislature to “amend or repeal a referendum statute” but not “an initiative statute” without voter approval); Perry v. Brown (2011) 52 Cal.4th 1116, 1126. No provision permits the legislature to alter initiative constitutional amendments; similarly with the recall, the legislature and the executive branches lack power to stop a recall election once it qualifies. The methods that other states or countries commonly use to limit the electorate’s lawmaking power (double passage, legislative approval of initiatives, or a cooling-off period) do not exist in the California initiative process. The California constitution imposes just a few limits: article II, section 8(e) requires that initiative measures have statewide effect, and provisions cannot be dependent on meeting a certain vote percentage; article II, section 12 bars initiatives from naming an individual to any office or identifying “any private corporation to perform any function or to have any power or duty[.]”

Among California’s direct democracy tools, only the referendum has a subject matter limit: voters may not approve or reject (even in part) “urgency statutes, statutes calling elections, and statutes providing for tax levies or appropriations for usual current expenses of the State.” Cal. Const. art. II, section 9(a); see Geiger v. Bd. of Sup. of Butte Cty. (1957) 48 Cal.2d 832, 839–40 (explaining the rationale for this limit is to prevent disruption of state operations). But all other statutes enacted by the legislature, in whole or in part, remain subject to voter approval or rejection under the referendum. The combined result of these initiative and referendum principles is that California courts view the substantive lawmaking powers of the state legislature and the voters as “coextensive” -- but the procedural limits that apply to the legislature are presumed not to apply to the electorate. California Cannabis Coalition v. City of Upland(2017) 3 Cal.5th 924, 942. And because the voters can use the initiative and referendum to override the legislature proactively or reactively, in combination these principles that define the electorate’s legislative powers mean that the electorate may hold greater (or at least more final) power than the California legislature. Carrillo & Chou, California Constitutional Law (West Academic Publishing 2021) at 159, 162.

California’s recall has no substantive limits: the voters can remove any state officer, whenever they want, for any reason -- or for no reason at all. “Sufficiency of reason is not reviewable.” Cal. Const. art. II, section 14(a). Instead, proponents of a recall effort merely need to obtain the required number of signatures within a 160-day period. Ibid. An official is removed from office in a recall election if the majority votes “yes” on the recall question. Id. art. II, section 15(c). Local California officials are also subject to a recall procedure. Sheriff removal procedures, SCOCAblog March 29, 2021.

The judicial tools for addressing challenges to California’s direct democracy acts are few, limited, and lightly applied. One limiting principle is the “single-subject” rule requiring that an initiative embrace no “more than one subject.” Cal. Const. art. II, section 8(d). But California courts have interpreted that constitutional provision to require only “a broadly deferential standard -- one that rarely requires striking down an initiative measure, to the extent that some question the single-subject rule’s effectiveness.” Carrillo, Duvernay, Gevercer & Fenzel California Constitutional Law: Direct Democracy (2019) 92 S. Cal. L.Rev. 557, 613. A review of 69 single-subject challenges found that the California Supreme Court used the rule to invalidate an act eight times; 12 of those cases involved initiative measures, and the court relied on the single-subject rule to invalidate all or part of an initiative just twice. Ibid. The deference California courts show to initiative measures is apparent in the judicial metric for single-subject rule compliance: it is an “accommodating and lenient” standard. Briggs v. Brown (2017) 3 Cal.5th 808, 829. Both doctrinally and empirically, the single-subject rule has not proved to be a significant limit on initiative measures.

The other restriction courts apply to California’s robust direct democracy provisions is the difference between an amendment to the state constitution (which the voters can do alone by initiative) and a revision (which requires the voters and the legislature to cooperate). Voters “may amend the Constitution by initiative.” Cal. Const. art. XVIII, section 3. But a “revision of the Constitution” requires the legislature to move first with a two-thirds majority vote to propose a revision for voter approval, or a two-thirds majority vote to ask voters whether to call a constitutional convention. See id. art. XVIII, sections 1, 2, and 4. The same two-thirds threshold for the legislature applies to its proposal of constitutional amendments, a sharp contrast from the simple majority required for the voters to approve an amendment. Compare id. art. XVIII, sections 1 and 2, with id. art. II, section 8(e).

To distinguish between a permissible amendment and an impermissible revision when analyzing an initiative, California courts apply a qualitative-quantitative analysis. An initiative is an impermissible revision if it makes excessive changes in the constitution (a “quantitative” effect), or if the initiative makes substantial changes to the state government even if the number of changes are few (a “qualitative” effect). Legislature v. Eu (1991) 54 Cal.3d 492, 506; see Bruce E. Cain et al., Constitutional Reform in California: Making State Government More Effective and Responsive at 279 (Bruce E. Cain & Roger G. Noll eds., 1995). But because the California constitution “provides no guidance on the distinction between an amendment and a revision,” courts have been forced to define “the limits of what the electorate can accomplish by initiative.” Carrillo, Duvernay & Stracener, California Constitutional Law: Popular Sovereignty (2017) 68 Hastings L.J. 731, 739.

The amendment-revision distinction has been challenging for courts to apply and has achieved at best modest and at worst inconsistent results. “With only two exceptions, the California Supreme Court has refused to find that any voter initiative measure revises, rather than amends, the state constitution.” Carrillo & Chou, California Constitutional Law at 183. Beyond that, courts have reached inconsistent results in applying the qualitative–quantitative analysis. In one case, a court invalidated a measure adding 21,000 words to the constitution and affecting 15 of its 25 articles; in another, the court rejected the claim that changes to eight articles and 37 sections was a quantitative violation. Compare McFadden v. Jordan (1948) 32 Cal.2d 330, 333–47, 349-51, with Amador Valley Joint Union High Sch. Dist. v. State Bd. of Equalization (1978) 22 Cal.3d 208, 224. The qualitative analysis is similarly unreliable: in one case, abolishing the state right of marriage by constitutional amendment was not an impermissible revision; in another, abolishing state substantive rights for criminal defendants that altered the authority of state courts to independently interpret criminal law was impermissible. Compare Strauss v. Horton (2009) 46 Cal.4th 364, 388, abrogated on other grounds by Obergefell v. Hodges (2015) 576 U.S. 644, with Raven v. Deukmejian (1990) 52 Cal.3d 336, 350–56. See also Carrillo, Duvernay & Stracener, 68 Hastings L.J. at 740, 752–67 (discussing these inconsistencies).

The electorate’s initiative power has even won out in conflict with another constitutional provision. See California Cannabis Coal. v. City of Upland (2017) 3 Cal.5th 924, 930 (a conflict between the initiative power and a constitutional limitation on local government taxing power). In California Cannabis the California Supreme Court explained that its analysis in multiple decisions “consistently begins with the presumption that the initiative power is not constrained, then searches for clear evidence suggesting that electors could reasonably be understood to have imposed restrictions upon their constitutional power.” Id. at 947; see also Carrillo & Shanske, California Constitutional Law: Interpreting Restrictions on the Initiative Power (2017) 51 U.C. Davis L.Rev. Online 65, 66-67, 74-75 (discussing that decision’s reliance on the well-established presumption in favor of liberally construing the initiative power).

In sum, California’s referendum, recall, and initiative are powerful direct democracy tools with low threshold requirements, few substantive restrictions, weak procedural rules, and light judicial oversight.

Compared with other state electorates, California’s voters are relatively powerful

To compare the California electorate’s degree of power relative to other states, we focus on the recall to illustrate by practical example -- specifically the pending recall proceeding against California Governor Gavin Newsom. State recall laws vary greatly across the country, and the distinctions can be substantive or procedural:

  • Is there a legal reason needed for the recall?

  • How many signatures need to be gathered?

  • Who can sign?

  • How much time is allowed to collect the signatures?

Eleven states are, like California, political recall states: the recall can be launched for political reasons or almost no reason at all. These states include Arizona, California, Colorado, Idaho, Louisiana, Michigan, Nevada, New Jersey, North Dakota, Oregon, and Wisconsin.

The other eight states with recall procedures require a specific, stated, statutorily delineated reason for the recall. In these malfeasance standard or judicial recall states, a judge or an election official can quash the recall for not meeting the requirements, such as: violations of the “oath of office,” “failure to perform duties prescribed by law,” “Conviction for a Felony,” “Corruption,” “Incompetence in office,” or “Malfeasance in office.” These states include Alaska, Georgia, Kansas, Minnesota, Montana, Rhode Island (which exempts the legislature), and Washington. Illinois, which only allows recalls for the governor, does not list specific reasons, but requires that a bipartisan collection of legislators sign the petition for it to move forward. Nearly all recalls of state officials have occurred in political recall states: California, North Dakota, Wisconsin, Oregon, Idaho, Michigan, Arizona, Colorado. The only state official to ever face a recall vote in a malfeasance standard state was a state senator in Washington in 1981. And since that incident, the Washington Supreme Court has tightened the malfeasance standard laws.

California has arguably least burdensome procedure for recalling its governor. Among the 19 or 20 states that allow for a gubernatorial recall (Virginia’s law is not completely clear if a state-level official can face a recall), California stands out. See Ann Bowman & Richard C. Kearney, State And Local Government (Wadsworth Publishing, 10th ed. 2016) at 98. Proponents need to gather only 12% of the vote cast in the last election and they have a leisurely 160 days to do so. And that date can be extended: based largely on concerns about the coronavirus pandemic’s impact on the political process, in November 2020 a judge granted the Newsom recall effort an additional 120 days to collect signatures.

In all states with a recall, the signatures required and who can sign vary, and here is where California stands out as having the lowest signature requirement compared with other states. The signature laws have various metrics for the total required: the requirement may be based either on a percentage of how many voters turned out at the last election, or on a percentage of how many voters are registered or eligible in the state. Comparing these various signature requirements only permits a rough ranking because state populations vary widely. Assuming identical population percentages, the order from most to least onerous would be: eligible voters, most onerous; second-most onerous, registered voters; and least onerous would be turnout. On that assumption, a state that uses a turnout standard likely will require significantly fewer signatures than a state that uses an eligible or registered voter standard.

Using those assumptions, California has the lowest signature requirement:

At first glance Montana’s 10% requirement seems to be easier to reach than California’s 12% mark. But further examination shows that California is much easier: Montana requires the signatures of 10% of registered voters; California requires 12% of voter turnout in the last election. In practice, this means that the proponents needed 1,495,709 signatures to qualify the Newsom recall. If California had used Montana’s law, then based on voter registration data from February 2021 the proponents would have needed 2,215,430 signatures -- that’s over 700,000 more signatures. And there are other reasons why qualifying a recall in Montana is vastly harder. Montana is a malfeasance standard state, so it is extremely difficult to get the petitions approved absent an actual crime. And the Montana signature collection timeframe is just 90 days compared with 160 days in California. The upshot: qualifying a recall in California is much easier than in Montana. And that makes California the overall easiest state to qualify a recall petition.

California’s lengthy signature gathering period contributes to that conclusion. The nearly six-month-long signature gathering period in California is the fifth longest of any state. (Alaska has no time limit; New Jersey grants 320 days; Washington provides 270; Louisiana gives 180.) Aside from having the easiest-to-meet signature requirement, California’s comparatively generous gathering period advantages proponents with a significant amount of time to qualify a recall. And courts have discretion in extraordinary circumstances to grant proponents more time to gather signatures. Cal. Elec. Code sections 11110, 11328 (recalls are conducted “in substantially the manner provided by law for a regular election for the office”); Burton v. Shelley (Aug. 7, 2003) No. S117834, 2003 WL 21962000 at *1, 3 (Secretary of State can exercise discretion over procedures); Laam v. McLaren (1915) 28 Cal.App. 632, 638 (the recall “power is given [to the electorate] by the constitution and statutes enacted in aid of this power should be liberally construed.”). The extension granted to the Newsom recall proponents due to the coronavirus pandemic shows this flexibility.

Comparing California’s pending gubernatorial recall with the last gubernatorial recall that went to the ballot also supports our conclusion that California has the easiest recall qualifying procedures. The most recent example is the recall against Wisconsin Governor Scott Walker in 2012, where Wisconsin required petitioners to amass 25% of the vote for the governor’s office in the last election -- in only 60 days. Wisconsin did have one (very unusual) lenient factor: the signers only had to be eligible voters, not registered voters. It is unclear why Wisconsin has this unusual standard; it may simply have been a language choice in 1926 when the provision was adopted. Still, the Wisconsin proponents needed to gather approximately double the number of signatures, in approximately half the time, as did the Governor Newsom recall proponents.

Finally, California’s recall procedures do not permit the governor, executive appointees, or other officials to stop a recall effort. If a recall is not scheduled, another official or a judge will quickly step in. In the state’s one other gubernatorial recall, against Gray Davis in 2003, California’s highest court acted quickly to resolve procedural disputes and move the recall along to the ballot, noting that the state constitution requires recalls “to proceed in expedited fashion.” Burton, 2003 WL 21962000 at *3. This is yet another example of California courts protecting electorate power; comparatively, in other states this has not worked as well. Several examples show that other state courts frequently fail to act with similar speed, or with similar solicitude for voter power.

  • A disputed recall effort against the three Las Cruces, New Mexico school board members has been pending in that state’s high court since 2019. Miranda Cyr, Las Cruces school board outlines candidate criteria, OKs investigation following Dallman resignation, Las Cruces Sun News June 8, 2021.

  • The town of Newton Falls in Ohio could not get a majority of the city council to schedule a recall against one of its members. The Ohio Supreme Court ruled that the council’s failure to act required the recall to be struck from the ballot, where it presumably will be scheduled on a later date. Court rules no recall election for Newton Falls Councilwoman, WKBN May 27, 2021.

  • Nevada’s Supreme Court invalidated the recall law against judges -- after delaying a ruling for 18 months -- even though its state constitution clearly includes judges as public officers. The dissent argued that the majority’s reasoning was “contrived” and that “Nothing -- not text, context, history, the ballot materials the voters received or the pronouncements of this court and Nevada’s lead constitutional scholars -- supports that our citizens gave up the right to recall judges[.]” Ramsey v. City of N. Las Vegas (2017) 392 P.3d 614, 623.

  • Similarly, in 2011 recalls in both Alaska and Texas were canceled due to a failure to schedule them.

In contrast to those examples of death by bureaucratic delay, in all three modern high-profile California recall campaigns (Davis in 2003, Persky in 2018, Newsom in 2021) the officials with relevant ministerial duties did everything required to swiftly place the matter before the voters. Nor would anything in California law have permitted them to substantially delay or defeat those issues. And California courts have consistently acted to expedite recall processes. Killing a recall by slow-walking it is highly unlikely to succeed in California.

The overall impact of California recall campaigns has been small

Now that the second-ever recall petition against a California governor has qualified for the ballot, numerous commentators have suggested that it is too easy to qualify a recall for the ballot in California. See, e.g., Brownstein, The Trouble With the Gavin Newsom Recall, The Atlantic May 13, 2021. But despite the comparative ease of qualifying recall petitions in California discussed above, historical data show that in California recalls rarely qualify for the ballot and that successful recalls are even more rare.

Recall petitions against California state officials rarely qualify for the ballot, resulting in few state officials being recalled. Despite 179 recall attempts since 1913, only eleven have qualified for the ballot, and only six were successful. California Secretary of State, Recall History in California (1913 to Present). The upshot: in over 100 years 6.1% of all statewide recall attempts qualified for the ballot; just 3.4% of those attempts succeeded in ousting the official. And for the governor specifically (now that the Governor Newsom recall has qualified), only two recall efforts have qualified among 55 attempted gubernatorial recalls -- that’s a 3.6% success rate. Even Governor Newsom faced five previous recall attempts before the current attempt qualified.

Recalls of judges, including recall attempts, are rare as well. Since the recall’s birth in 1911, there have been only 27 attempts to recall individual California Supreme Court justices, including one attempt in 1966 to remove all seven justices. California Secretary of State, Recall History in California (1913 to Present). Aside from the recall of California Superior Court Judge Aaron Persky in 2018, the last successful recall of any California judge at any level was in 1932. See Joshua Spivak, California’s Recall: Adoption of the “Grand Bounce” for Elected Officials, (2004); see also Wilbank J. Roche, Judicial Discipline in California: A Critical Re-Evaluation, (1976) (discussing 1932 recall of Los Angeles Superior Court judges John L. Fleming, Dailey S. Stafford, and Walter Guerin); Joshua Spivak, California: Santa Clara County Superior Court Judge facing recall threats, June 7, 2016. (We are compelled to remind everyone that the 1986 election in which the voters ousted three California Supreme Court justices was a regular retention election, not a recall.)

Given these low historical qualifying and success rates for recalls, we conclude that although the seemingly low bar to success may encourage many recall attempts, the data show that California voters are reluctant to qualify recall petitions, and even more circumspect about recalling officials. We cannot speculate about why voters are less likely to eject officials than petitions are to qualify; yet that is what the data show.

Evaluating practical examples shows that successful recalls are rare

Two recent examples of successful recall campaigns, resulting in removal of the government official, show how rare successful recalls are and the confluence of factors necessary for a successful recall.

The first example is Gray Davis, the only California governor to be recalled. Indeed, no statewide California official had ever been recalled before the Davis recall in 2003, despite every governor since Ronald Reagan in 1968 being subjected to a recall effort. Several unique factors contributed to the Davis recall. Davis blocked Proposition 187 -- an initiative that had garnered 58.93% of the vote -- by refusing to appeal a federal district court ruling that it was unconstitutional. Davis also signed two new restrictive gun-control laws and tripled the vehicle license fee. Perhaps most importantly, California citizens were upset about the state’s electricity crisis, which tripled consumer energy costs and caused rolling blackouts throughout the state. See Huffington, Gov. Davis and the failure of power, Salon Jan. 27, 2001. That Davis accepted $550,000 from a large power company was probably unhelpful. Id. Recall proponents charged Davis with “[g]ross mismanagement of California Finances by overspending taxpayers’ money,” cutting funds to local governments, “failing to account for the exorbitant cost of the energy fiasco,” and generally failing “to deal with the state’s major problems until they get to the crisis stage.” Voter Information Guide for 2003, Special Election (2003) at 5. The Davis gubernatorial election also saw the lowest-ever voter turnout by percentage in California history, ensuring that the signature requirement was proportionally the smallest ever. Voters Views of Politics in California: Dissatisfaction, Distrust, and Withdrawal, Public Policy Institute of California Research Brief, November 2004. These combined extreme circumstances of the (so far) only successful gubernatorial recall in California, coupled with the many failed efforts, suggest that a unique confluence of extraordinary factors may be necessary.

The recall of California Superior Court Judge Aaron Persky is another example of a recall campaign succeeding due to a convergence of multiple unusual circumstances. The recall of a California judge is a rare event, with no California judge having been recalled “since 1932, when Los Angeles voters removed three Superior Court judges accused of taking kickbacks.” Egelko, Fate of Judge Persky, symbol within #MeToo movement, heads to voters, San Francisco Chronicle May 27, 2018; see also What does California’s experience with recall of judges teach us? SCOCAblog December 27, 2016. But Judge Persky “drew national attention in 2016 when he sentenced a Stanford student to just six months in jail for assaulting an unconscious woman.” Astor, California Voters Remove Judge Aaron Persky, Who Gave a 6-Month Sentence for Sexual Assault, The New York Times June 6, 2018. “[I]n the minds of the public, the Turner case has come to symbolize judicial collaboration in the sexual exploitation that gave rise to the #MeToo movement.” Egelko, Fate of Judge Persky, symbol within #MeToo movement, heads to voters, San Francisco Chronicle May 27, 2018. The Judge Persky recall was an unusual situation where a woefully short sentence (even if the defendant was also required to register as a sex offender) garnered national attention in a wave of renewed focus on the inadequate societal response to sexual assault.

Despite the comparative ease of qualifying a recall in California, that ease does not translate into frequent qualification or high success rates. And the Persky and Davis recalls, with their similarly unusual factual scenarios, suggest that the rarity of successful recalls stems from the unique circumstances necessary for a successful recall.

California’s secretary of state and lieutenant governor have broad discretion

In recall elections for California governors, the California secretary of state and lieutenant governor have broad discretionary power over election procedures. The secretary of state can determine how the recall election will be conducted -- she need only conduct the election “in substantially the same manner provided by law for a regular election of the office.” Cal. Elec. Code section 11328. That “substantially the same” frame gives the secretary of state discretion to set the qualifying requirements for replacement candidates. Brandon Stracener and Bob Wu, Two state officials will shape the recall election, SCOCAblog March 26, 2021. And the California Supreme Court interpreted that provision to give the secretary of state great deference in her decision. Burton, 2003 WL 21962000 at *1. That discretion permits the secretary of state to do what her predecessor did in the 2003 Gray Davis recall: require only 65 signatures from a candidate’s fellow party members. Cal. Elec. Code sections 8062(a)(1), 8103(a)(1), 8106(a)(3). Alternatively, she could have required that replacement candidates qualify under the requirements for independent candidates, which would have required 221,5444 signatures to qualify for the ballot. See Cal. Elec. Code sections 8300, 8400, 8550(f); Cal. Sec. of State, Odd-Numbered Year Report of Registration: February 10, 2021 at 1; Cal. Sec. of State, Report of Registration: August 8, 2003 at 1. The secretary of state also enforces other laws on recall candidates, such as a 2019 California law requiring recall candidates to submit five years of tax returns. Cal. Elec. Code section 8902; Cal. Sec. of State, Secretary of State Announces Qualifications for Potential Recall Candidates. For example, when Governor Newsom missed a statutory filing deadline to have his party designation listed on the recall ballot, the secretary of state refused his request to permit a tardy filing. Cal. Elec. Code section 11320; Sacramento Superior Court case no. 34-2021-80003666 (July 12, 2021 order upholding the secretary of state’s decision).

The secretary of state’s broad discretion interacts with the discretion the lieutenant governor has to set the date of the recall election. The lieutenant governor has a 20-day window within which to set the election. Cal. Const. art. II, section 15; Cal. Elec. Code section 11100. Her choice affects how long potential replacement candidates have to qualify for the recall ballot: potential replacement candidates could have anywhere from 1 to 21 days to qualify for the ballot. Cal. Const. art. II, section 15(a) (requiring an election to be scheduled within 60-80 days after a recall petition is certified); Cal. Elec. Code section 11381(a) (requiring nominating petitions to be filed with the secretary of state no later than the 59th day before the election); Brandon Stracener and Bob Wu, Two state officials will shape the recall election, SCOCAblog March 26, 2021. A large signature requirement and a short time frame to qualify could result in very few candidates making it onto the recall ballot.

California’s secretary of state decided to apply the same easy primary candidate qualification procedure from the 2003 Davis recall. See It’s official: SOS chooses easy recall qualification procedure, SCOCAblog June 16, 2021. This discretionary choice matters (and to a lesser extent so will the lieutenant governor’s discretion) because repeating the 2003 Davis recall qualifying procedure could easily result in well over 100 candidates appearing on the recall ballot. In 2003, the recall ballot featured 135 names; many were not serious candidates. See Lyman, The California Recall: The Candidates; California Voters Wonder: Is Anyone Not Running?, N.Y. Times Aug. 16, 2003; Calif. Recall Ballot Will Have 135 Names, AP News Aug, 14. 2003. On July 1, 2021 the secretary of state certified the recall petition, and the lieutenant governor has set the recall election date on September 14, 2021. Cal. Sec. State, SOS Certifies Gubernatorial Recall Petition; Cal. Lt. Gov., California Gubernatorial Recall Election Proclamation. That election date is 75 days from the certification. Forty-six people filed to run as a replacement candidate by the July 16, 2021 deadline. Rosenhall and Kamal, Who's running in the Newsom recall?, CalMatters July 21, 2021.

We resolve arguments about the recall in its favor

The fact that successful examples of California’s recall are rare suggests that its relative ease of use -- lower signature thresholds than other states and no required reason to start a recall -- has produced a well-balanced system for California. Despite being relatively easy to invoke, few recall efforts have qualified for the ballot. And even fewer recall efforts have succeeded in removing an official from office. Although a regular election held every four years typically is enough for voters to manage their elected officials, voters should have the option to respond more quickly when the situation -- or the official -- requires it. This recall option should be difficult enough to exercise that it does not become a substitute for regular elections, but not so difficult that it is practically unavailable. So far, historical outcomes in California recall campaigns suggest that the recall is properly calibrated to be readily available but not overused.

Despite the rare use of California’s recall, its current framework ensures that officials facing a recall cannot game their way out once voters do qualify a recall for the ballot. An official facing a recall cannot run as a replacement candidate. Cal. Const. art. VI, section 15(d); Cal. Elec. Code section 11381(c). If a vacancy in the relevant office occurs (such as by the official resigning or dying) “after a recall petition is filed against the vacating officer, the recall election shall nevertheless proceed.” Cal. Elec. Code section 11302(a). The only exception to that is if an insufficient number of signatures “were filed as of the date of the vacancy.” Cal. Elec. Code section 11302(b)(3). And California law anticipates devious maneuvers -- it prohibits a person “subject to a recall petition” from being “appointed to fill the vacancy in the office that he or she vacated.” Cal. Elec. Code section 11302(b)(4).

An elected official facing a possible recall has few options to avoid it. The official could perhaps resign before a recall petition gains sufficient signatures, but the official cannot be appointed back into that position. This threat itself is a recall feature (or bug, depending on one’s perspective): a credible recall campaign can achieve its end with minimal expense if the official resigns. That resignation achieves rather than undermines the goal of a recall, especially because the official cannot be appointed back into the position. But once there are enough signatures, the recall will go forward -- vacancy or not. Cal. Elec. Code 11302(a), (b)(2). That could produce a scenario where an official resigns in shame and is further shamed by being recalled in absentia.

Even endorsing a possible successor can undermine an official’s chances of surviving a recall race. Although Gray Davis did not endorse a successor, many observers thought that the emergence of Lieutenant Governor Cruz Bustamante as a serious backup candidate further harmed Davis’s chances of surviving the recall. See, e.g., Bustamante Criticizes Davis’ Aides, Los Angeles Times Aug. 18, 2003. That history may have motivated California Democrats to (thus far) avoid naming a backup in the upcoming Newsom recall. See In California’s recall, Democrats would rather not have a backup plan, NBC News May 24, 2021. History also shows that the party that launches the recall frequently overperforms in the replacement race. In the five legislative recalls with a clear political bent, the party of the recall proponents captured the seat. Joshua Spivak, Recall history shows why the Democrats were probably right to scorn the replacement race for governor, The Recall Elections Blog, August 25, 2021.

Additionally, the discretion held by the secretary of state and the lieutenant governor in determining recall procedures and timing reduces a targeted official’s ability (even a governor) to influence those procedures or timing to the official’s advantage. At most, an official can exert informal political pressure. But such pressure lacks power when applied to a lieutenant governor or secretary of state, who are independently elected officials. Cal. Const. art. V, section 11. A governor cannot fire the independently elected secretary of state or lieutenant governor, but the voters certainly can respond to procedural chicanery by a lieutenant governor or secretary of state -- by recalling them or voting them out at the next regularly scheduled election. The secretary of state refusing to overlook Governor Newsom’s failure to designate his party on time illustrates this dynamic: It is possible that Secretary Weber was less awed by the governor’s political pressure and more concerned for her own political survival.

Some argue that recalls are an unnecessary waste of resources when regular elections occur often enough. For example, the Davis and Newsom recall elections both fall about a year from their regularly scheduled term’s end. Given that the Newsom special election will occur on September 14, 2021, the June 2022 primary means California may spend hundreds of millions of dollars to potentially shave a little over a year off Newsom’s term. Lieutenant Governor’s Proclamation Calling for the California Gubernatorial Recall Election; see Democrats push for speedy Newsom recall as new analysis pegs cost at $215 million, Los Angeles Times June 10, 2021. The United States has survived countless bad officials without a recall for federal officeholders, and voters in about half the states have no trouble waiting until the next regularly scheduled election to oust problematic officials.

Also, the fact that qualifying a recall petition in California is not so difficult raises questions about whether a politically polarized environment will make the recall a tempting option for a minority of voters to relitigate elections. See, e.g., Brownstein, The Trouble With the Gavin Newsom Recall. California’s two-step recall process could result in the replacement for a recalled official having received fewer votes -- a near-majority of 49.9% of voters could vote “No” on the recall, resulting in recall, while the winning replacement candidate garners a far lower percentage of votes (but has the highest number relative to other replacement candidates). See id.; compare Cal. Elec. Code section 11383–84, with Cal. Elec. Code section 11385. This has happened before, as in the 2018 recall of state senator Josh Newman, where Newman was recalled even though he received 15,982 more votes than his replacement. On this view, an organized minority could pervert the democratic process and drain state resources by repeatedly qualifying meritless recalls, serving their disruptive ends even if none succeed at the ballot.

These concerns about political polarization, cost to taxpayers, and potential for factions abusing the process all seem plausible. Yet the historical evidence to date does not suggest that those concerns reflect reality. Despite numerous recall attempts, few qualify for the ballot, and even fewer succeed. Nor is the current polarized partisan political culture historically unique. See Jon Grinspan, The Forgotten Precedent for Our ‘Unprecedented’ Political Insanity, PoliticoApril 24, 2021. Instead, we think that several secondary factors are at work. The hidden costs of recall campaigns are a significant factor in mitigating their apparent danger. For example, the cost to obtain sufficient signatures is a major threshold barrier that excludes all but the most professional operations. Using professional signature-gathering companies -- typically a practical necessity to run a statewide campaign in California -- requires millions of dollars. Carrillo, Duvernay, Gevercer & Fenzel, 92 S. Cal. L.Rev. at 597–98. The Newsom recall campaign, which was largely volunteer-run, is an exception that proves the rule, and it only succeeded by getting an extension. California’s electorate is generally conservative at the ballot in the sense that it rejects far more measures than it approves. Id. at 580–81 (around 34% of all initiatives pass, with a maximum possible success rate for initiatives on any given ballot regardless how many initiatives are qualified). And California voters can experience voting fatigue when faced with repeated questions on the same issue. For example, just two years after the voters rejected Proposition 8 in 2018, the voters rejected the similar Proposition 23 in 2020 by an even larger margin. For recalls against governors particularly, and except for the Newsom recall effort, repeated recall attempts have not resulted in greater success — those attempts instead resemble futile tilts that succeed only by fluke.

Policy recommendations: clarify and reform recall procedures

Needed recall reforms lie in ensuring a smoother procedure rather than attempting to constrain the electorate’s access to the ballot. One future recall development that could benefit California would involve codifying the procedures that currently are discretionary. See Cal. Elec. Code sections 11110 and 11328. As the California Supreme Court has noted: “The current recall provisions contain ambiguities which require the Secretary of State to exercise his discretion. If the Legislature disagrees with the manner in which the Secretary of State has exercised his discretion, it is within the Legislature’s province to specify other procedures.” Burton, 2003 WL 21962000 at *3. Thus, the legislature could codify a recall election procedure. Codifying the procedure for replacement candidate qualification would prevent the qualifying scramble spectacles that delight outside observers and frustrate Californians. See Calif. Recall Ballot Will Have 135 Names, AP News (Aug. 14, 2003).

Ultimately, despite its relative ease of qualifying and approving, the California recall has been a historical rarity. That should mitigate concerns about the state’s recall process being overpowered, and it certainly undercuts claims that it is prone to overuse. Instead, concerns about California’s recall machinery should focus on the under-developed procedures for running the recall election itself, which unnecessarily create uncertainty and can result in many frivolous replacement candidates appearing on the ballot. The first flows from the fact that the qualifying process is unclear; the second from the fact that in both 2003 and 2021 the Secretary of State chose the easier qualifying process. As powerful as it is, the recall process could benefit from more procedural certainty, which in turn would give this significant electoral decision the solemnity it merits.

About the Authors:

rsz_sq-carrillo_headshot_2020.jpeg

David A. Carrillo, J.S.D. (carrillo@law.berkeley.edu) has been a Lecturer in Residence and Executive Director of the California Constitution Center at Berkeley Law since 2012, where he writes about and teaches courses on the California constitution. Before joining the academy, Dr. Carrillo practiced law for 16 years, as a Deputy Attorney General with the California Department of Justice, as a Deputy City Attorney in San Francisco, and as a Deputy District Attorney in Contra Costa County. He has served on several state and local commissions, including the California Law Revision Commission, the Judicial Nominees Evaluation Commission, and the Committee of Bar Examiners.

 
sq-spivak_headshot.jpeg

Joshua Spivak (joshuaspivak@gmail.com) is a Senior Fellow at the Hugh L. Carey Institute for Government Reform at Wagner College and a public-relations executive. He received his law degree from the Columbia University Law School and for two decades has studied the history and use of recall elections. He has been interviewed and quoted as an expert source by numerous media outlets, has written hundreds of op-eds and book reviews on political and historical topics, and writes the Recall Elections Blog. He is the author of Recall Elections: From Alexander Hamilton to Gavin Newsom.

 
sq-stracener_headshot_2021.jpeg

Brandon V. Stracener (bvstrac@gmail.com) is a California attorney who received his JD from the University of California, Berkeley School of Law. He served as a law clerk for U.S. District Judges Kimberly J. Mueller and Lucy H. Koh. As an author of several publications on California constitutional law, Brandon is a frequent resource for media comment. His work has appeared in the San Francisco Chronicle, the Los Angeles & San Francisco Daily Journal, The Recorder, and multiple law reviews. He contributes often to SCOCAblog, covering the California Supreme Court, and he maintains a role as Senior Research Fellow at the California Constitution Center.

David A. Carrillo, Joshua Spivak, and Brandon V. Stracener
Previous

Embracing a More Honest Reckoning with History — A Historian’s Perspective on Education, Battling the Culture Wars in Schools, and Liberation
Next

Leading in Local Government