A Successful Experiment: California’s Local Laboratories of Regulatory Innovation


Justice Brandeis’s 1932 remarks in his New State Ice Co. v. Liebmann dissent touting states’ roles as “laboratories of democracy”1 remain one of the most frequently invoked passages in U.S. law, cited repeatedly in judicial opinions2 and employed by politicians of diverse political views such as Ronald Reagan3 and Ralph Nader.4  Yet there is comparatively little empirical legal scholarship about the degree to which the laboratories metaphor—and the particular process by which Brandeis suggested it might work5—captures jurisdictions’ actual experience of the spread of regulation.6  While there is considerable evidence that state and local policies often converge over time,7 several mechanisms for such convergence have been proposed, only some of which involve the type of state-to-state learning process Brandeis had in mind.  In particular, questions remain about whether evolving regulatory similarities across state lines are more likely to be driven by conscious decisions by later-acting states to adopt the successful policies of pioneer states or, rather, by external forces that affect many states simultaneously.

While much research about Brandeisian experimentation and imitation has involved states, scholars interested in policy innovation have recently broadened their focus to include local governments.8  Local jurisdictions may have particular promise as pioneers of new policies for a variety of reasons, including their greater responsiveness to citizen concerns, the reduced influence of lobbyists (who are likely to focus resources at the state or national level), and the greater receptivity of citizens to local change.9  Innovations that begin at the local level frequently can later influence policy at the state level, as state legislators seek to gain ideas from successful local experiments and “interest groups . . . utilize local examples to build their case for statewide change.”10  Because local governments may innovate even more than state governments, they offer a rich source of data about the extent of government-to-government influence and the mechanisms by which it can occur.

With these considerations in mind, we seek to examine the real-world descriptive power of Brandeis’s metaphor by looking at the experiences of California’s local jurisdictions in regulating electronic smoking devices.  Between 2009 and 2010, when electronic smoking devices first became widespread in the United States, and 2016, when California adopted comprehensive statewide regulation of such devices, California adopted a largely hands-off approach at the statewide level toward electronic smoking device regulation.  Instead, California encouraged regulation and innovation at the local level.11  As discussed in this Article, such local regulation most commonly took the form of localities that sought to include electronic cigarettes (e-cigarettes) in their tobacco retailer ordinances by defining “tobacco products” to include such devices.

California’s efforts to foster local experimentation were largely successful.  From 2011 (when the first local ordinance defining tobacco products to include electronic smoking devices was passed) until the time of writing (early 2018), thirty-eight localities adopted ordinances specifically regulating electronic smoking devices as tobacco products to some degree, creating momentum that may have contributed to the adoption of similar laws statewide.12  In addition, twenty-two jurisdictions during this time passed laws that, while not specifically mentioning electronic smoking devices, included language that was broad enough to cover them.13  Thus, a total of sixty jurisdictions appear to have passed ordinances during this period with electronic smoking devices in mind.

Notably, however, jurisdictions differed in the manner by which they sought to regulate electronic smoking devices.  Initially, most jurisdictions’ regulations covered electronic smoking devices only insofar as they delivered nicotine or tobacco leaf.  Ordinances passed later, however, were more likely to cover all electronic smoking devices without reference to their nicotine-delivery function.  This Article will explore this phenomenon and focus on the factors—in particular, the publication of a model statute by a nonpartisan policy group, ChangeLab Solutions—that may have influenced localities to adopt this method as opposed to others.

The Article proceeds in two parts.  Part I surveys the public health and political science literature examining the process through which policies diffuse among local governments.  Part II then examines California local governments’ experience in regulating electronic smoking devices and the potential influence of a model statute promulgated by a nonpartisan, state-sponsored organization, ChangeLab Solutions.  The Article concludes by discussing California’s ultimate decision to adopt statewide regulation of electronic smoking devices, which may have been driven by a Brandeisian borrowing from the successful policies of local jurisdictions.

I. Public Health and Political Science Literature Discussing Diffusion of Policies

Although the ubiquity of Brandeis’s laboratories metaphor suggests that the idea of experiment-led policy convergences is both easily comprehensible and resonant, it is nonetheless worth lingering more on the specifics of Justice Brandeis’s argument.  Writing at the height of the Great Depression, Brandeis suggested that the present “emergency”14 called for regulatory experimentation that would mimic “the process of trial and error” that had produced great inventions and scientific discoveries.15  The Lochner-era substantive due process constraints the Court had thus far imposed, Brandeis warned, ran the risk of stifling this necessary process.16

Brandeis’s conception of state laboratories thus had several components.  First, states must be sufficiently unfettered by federal constitutional constraints so that they are able to develop novel, innovative policies.  Second, the results of such experiments must be capable of being evaluated at some point as successes or failures.  Third, Brandeis’s formulation suggested that policies with favorable results should eventually come, like new scientific insights, to be universally, or at least widely, accepted.17

Brandeis then described a plausible process by which jurisdictions might learn from one another’s experiences and pass higher-quality legislation as a result.  Scholars, however, have broadened this view, suggesting several routes by which policy diffusion may occur.  First, there is emulation, under which jurisdictions “somewhat blindly follow other [jurisdictions’] lead.”18  Second, there is learning, a process that appears most consistent with the Brandeisian model where jurisdictions thoughtfully adopt policies that have proven successful elsewhere.  Third, there is competition, which “occurs when the policy creates spillover effects, such as with tax rates or environmental protection legislation, that lead to positive or negative feedback cycles.”19  A fourth version of this process is a situation where disproportionately large or influential jurisdictions may be able to “foist” their standards on less powerful ones, a process sometimes called the “California effect.”20

While not mentioned by all scholars in this area, another possibility is that jurisdictions may engage in deliberate synchronization of their policies with those of their neighbors, which may help to increase compliance and facilitate enforcement across jurisdictional lines.21  Laws across jurisdictions could also converge without direct imitation if many jurisdictions are influenced by a common source, such as a model ordinance or news coverage that draws attention to a problem, a process described later in this Article.

Political scientists have studied how policies spread through each of these primary routes.  Edward M. Rogers and Jeffrey C. Peterson found evidence of learning from other jurisdictions when they studied the spread of clean air regulations across New Mexico counties.22  They found that champions of a particular policy frequently visited nearby communities that had adopted their preferred policy.  These visits enabled proponents to build support and learn about successful political strategies.  Their research also revealed that legislators relied on models from other communities to aid in drafting policies (while also generally adapting such models to local conditions).23

Joshua L. Mitchell and La Shonda M. Stewart found evidence of both learning and competition in their study of antismoking legislation in Missouri.  In Missouri, counties were more likely to adopt antismoking policies where their neighbors had previously done so (evidence of learning) and less likely to do so where adoption of such policies would have made them outliers within a region (evidence of the effects of competition).24  Their research additionally suggested the possibility of variation based on county size: Large counties may have been both more likely to learn from neighbors and more fearful of competitive effects.25  James Macinko and Diana Silver also found evidence of learning by states.  They studied the spread of impaired-driving laws and found that the existence of a neighboring state with such laws helped to explain other states’ adoption of similar laws.26

This Article seeks to supplement this literature by focusing on the overall degree of synchronization among municipalities in California that expanded their definition of tobacco products to include electronic smoking devices.  The previous work discussed above has focused on the effect of policy adoption by neighboring cities and states.  While we accept that such influence may sometimes occur, in this Article we try to assess additional avenues through which cities may develop their policies, such as an influential model ordinance.

II. California’s Experience

A. Electronic Smoking Devices in California

Electronic smoking devices heat up liquid nicotine (or other products such as marijuana), permitting users to inhale the vapor.  Electronic smoking devices were first introduced in the United States in 2006–07, and by the end of the same decade, their popularity attracted a steady flow of media attention.  In July 2009, the Food and Drug Administration (FDA) warned users that the safety of the devices was unknown and that they might not comply with current law, but declined to ban them, citing uncertainty about its own jurisdiction.30

Electronic devices are similar to conventional tobacco products in several important respects.  They contain carcinogenic substances that are also found in tobacco and high levels of nicotine, often physically resemble cigarettes or other tobacco products,32 and involve the inhalation of nicotine.33  Because of their low cost,34 the variety of available flavors, and the perception that they are safer and more environmentally sound than traditional tobacco products, e-cigarettes are particularly appealing to young people who, once in the habit of using e-cigarettes, may become more likely to use other tobacco products as well.35

Yet while traditional tobacco products have long been regulated at the state36 and federal levels,37 electronic smoking devices fell into loopholes in existing tobacco laws, and jurisdictions were initially slow to pass new regulations.38  In 2011, California restricted sales of e-cigarettes to minors, but took no other immediate action at the statewide level.39  A subsequent California Board of Equalization report described the “truly shocking . . . total lack of regulation [of e-cigarettes] at the state and federal level.”40  Given the similarities between electronic smoking devices and more traditional tobacco products such as cigarettes, many commentators noted that this lack of regulation had the potential to undermine the efficacy of public health campaigns to reduce tobacco use.41

As attempts to regulate e-cigarettes at the state and federal levels stalled, many localities in California began to take action.  Thus, for several years in California, the decision whether or how to regulate e-cigarettes was a primarily local one.  The experiences of California local jurisdictions are described in the following two Subparts.

Growing public concern about the hazards of unrestricted e-cigarette use ultimately prompted the state of California and the federal government to take action.  On May 4, 2016, California Governor Jerry Brown signed into law a group of bills expanding the state definition of tobacco products to include e-cigarettes (thus subjecting e-cigarettes to tobacco use restrictions)42 and raising the minimum age to purchase all tobacco products from 18 to 21 (with an exception for service members). The law went into effect on June 9, 2016.44  In addition, California voters on November 8, 2016 overwhelmingly approved Proposition 56, which raised the cigarette tax (on both conventional and e-cigarettes) by $2 per pack with equivalent increases for e-cigarettes and other tobacco products effective on April 1, 2017.45  At the same time, the FDA enacted a new rule that became effective on August 6, 2016,46 allowing it to regulate electronic smoking devices along with other tobacco products and restricting their sale to those under 18.  Despite federal regulation, state and local jurisdictions may continue to regulate more stringently.48

The period when California’s e-cigarette regulation was a largely local affair is thus neatly bracketed by the growing popularity of the devices in 2009–201049 and the enactment of sweeping state restrictions in 2016.  (It should be noted, however, that local jurisdictions have continued to pass ordinances regulating electronic smoking devices even following the statewide legislation.)  The following Subparts discuss the rise and changing shape of local regulation.

B. State Encouragement of Local Efforts

Despite the initial lack of comprehensive legislative action, California public health agencies did not entirely ignore the rise of e-cigarettes.  Rather, they sought to step up regulation of electronic smoking devices by encouraging local jurisdictions to regulate electronic smoking devices as tobacco products.50  The most active endeavor came through the California Tobacco Control Program (CTCP), whose overall mission is to reduce tobacco use in California.  The CTCP seeks not only to promote the enactment of tobacco control policies but also to pursue a broader strategy of denormalization—that is, to persuade Californians through advertising campaigns, bans on public smoking, and other strategies that tobacco product use is not socially acceptable.52  Among the CTCP’s other activities, the program works with local governments to pass effective tobacco control policies.53

The CTCP’s strategies are founded in research suggesting that passing tobacco control policies at the local level is particularly effective in reducing the social acceptability of tobacco use.54  Some evidence suggests that citizens are more likely to favor tobacco control policies that are passed at the local level.55  Further, the process of passing the tobacco control ordinances appears to make the local community more aware of the dangers posed by tobacco products.56  As a result of this awareness, citizens are more likely to favor additional laws that restrict tobacco use.57

The CTCP provides local governments with information on effective tobacco control policies through an organization originally called the Technical Assistance Legal Center (TALC).58  TALC helped local government pass effective tobacco control policies by performing legal research for local governments and developing model ordinances that local governments can use as a template for their own tobacco control ordinances.59

Outside the realm of electronic smoking devices, TALC successfully influenced local governments to adopt recommended tobacco control policies.  For example, Leslie Zellers and Ian McLaughlin have noted the large influence of TALC’s first tobacco retail licensing policy,60 a policy that forbids the selling of tobacco products without a local government license.61  As Zellers and McLaughlin note, twenty-three out of twenty-six local tobacco retail ordinances adopted by local governments “contained four of the five key provisions of the TALC model ordinance.”62

Local governments have presumably continued to adopt such model ordinances because they have found them to be effective at reducing tobacco product use.  There is some empirical support for the ordinances’ efficacy.  For example, John A. Francis, Erin M Abramsohn, and Hye-Youn Park found that adoption of tobacco control policies by local governments with TALC’s support and encouragement was associated with a reduction in the use of tobacco products in California and greater public support for smoke-free environments.63

Today, TALC’s name has changed to ChangeLab Solutions, but it retains the same fundamental approach to tobacco control.  ChangeLab provides local governments with legal research65 and develops model tobacco control policies for local governments to use when enacting their own tobacco control policies.66  For example, ChangeLab developed a model tobacco retail licensing policy designed to limit minors’ access to tobacco products.67  More recently, ChangeLab has promoted the passage of laws defining electronic smoking devices as tobacco products, permitting them to be regulated through a given locality’s comprehensive tobacco retail licensing ordinance.68  As discussed in the following Subpart, ChangeLab’s approach to electronic smoking devices appears to have significantly influenced California jurisdictions.

C. The Spread of Electronic Smoking Device Regulation in California

When e-cigarettes became popular, many localities in California expanded their regulations to respond to the growing prevalence of electronic smoking devices.  Initially, some jurisdictions expanded clean indoor air laws to include e-cigarettes, with Eureka in 2010 apparently the first to do so.  Other cities adopted idiosyncratic approaches, such as the small city of American Canyon’s ban on e-cigarette retailers (which does not apply, however, to larger retailers selling e-cigarettes along with other products).70

Perhaps the most common and effective way of addressing the problems posed by electronic smoking devices is to adopt or amend a tobacco retailer ordinance that includes such devices in the definition of “tobacco products” subject to regulation.  The following discussion looks at the changing pattern of efforts to regulate electronic smoking devices as tobacco products in California.

To begin with, we note that there are three broad ways in which localities have sought to regulate e-cigarettes and similar devices as tobacco products.  First, some jurisdictions have adopted a definition of “tobacco products” that includes nontraditional products that do not contain tobacco leaf but deliver “biologically active amounts of nicotine.”  This approach, which we have labeled Category A, was inspired by a 2005 ChangeLab model ordinance.71  The ordinance was written “to broaden the definition in preparation for the emergence of alternative nicotine products and nicotine delivery devices,” such as snus or nicotine lollipops.72  Soon after the publication of the 2005 ordinance, a number of jurisdictions quickly adopted definitions resembling the ChangeLab one—for example, twelve jurisdictions did so in 2006 alone.73

The publication of the 2005 ChangeLab model ordinance and the associated initial passage of local definitions preceded the widespread use of e-cigarettes in the United States and were thus not a direct response to electronic smoking devices.  At the same time, the ChangeLab language was broad enough to encompass all nontraditional products containing “biologically active amounts of nicotine,” which includes the majority of e-cigarettes and similar products.  After the spread of electronic smoking devices, some jurisdictions adopted the 2005 ChangeLab approach with the intent of imposing more stringent regulations on e-cigarettes in particular.  For example, Los Angeles adopted an ordinance in the Category A vein in late 2013.74

For purposes of our analysis, we have assumed that jurisdictions adopting Category A–style ordinances in 2011 or later did so with at least some understanding and intention that such ordinances would apply to e-cigarettes.  This cutoff date tracks the period during which jurisdictions began adopting ordinances that explicitly mention e-cigarettes (the first of which seems to have been El Monte’s ordinance in November 201175) and thus allows us to examine the changing shape of e-cigarette regulation over time.

The second category (Category B) includes ordinances that explicitly apply to electronic devices, but only insofar as they deliver tobacco leaf or nicotine.76  For example, Hawaiian Gardens defines “tobacco product” as “any substance containing tobacco leaf, including, but not limited to, . . . electronic cigarettes.”77

By contrast, the third category of ordinances (Category C) applies unequivocally to all electronic smoking devices, whether or not they are used to deliver tobacco or nicotine.  Generally, these ordinances accomplish this broad reach by adopting a two- or multi-pronged definition of tobacco product.  For example, Yolo County has adopted a several-part definition of “tobacco product,” including “any substance containing tobacco leaf,” any substance containing “biologically active amounts of nicotine,” and “any electronic device that delivers nicotine or other substances.78  As discussed below, an approach that decouples electronic smoking device regulation from nicotine or tobacco content is more consistent with a 2015 ChangeLab model ordinance.79  Within Category C, all but two jurisdictions80 use language that closely tracks the language of the 2015 ChangeLab model ordinance.81

All forms of electronic smoking device regulation help address the public health issues presented by such devices.  Regulations that apply broadly to all electronic smoking devices—rather than simply regulating them implicitly through their nicotine-delivery function—may be slightly preferable from a public health perspective.82  As a comment to the 2015 ChangeLab model ordinance notes, all electronic smoking devices—whether or not used to deliver nicotine—may expose bystanders to hazardous vapor and serve to normalize e-cigarette use; restricting the use of all such products may therefore better serve public health ends.83  A uniform, unambiguous rule may also make enforcement easier.84  Although there might be cause for concern that Category C ordinances sweep too broadly, it is worth noting most ordinances—whichever category they belong to—contain an exemption for products approved by the Food and Drug Administration for aid in smoking cessation,85 which goes toward addressing fears about overbreadth  (The ChangeLab model ordinance also contains such an exemption86).

As discussed in the previous Subpart, initially ChangeLab recommended a more limited definition of “tobacco products” based on nicotine delivery.  Even before the introduction of e-cigarettes, ChangeLab had anticipated the potential introduction of new tobacco products, and as early as 2005 recommended a definition of “tobacco products” encompassing any product “containing biologically active amounts of nicotine . . . manufactured . . . with the expectation that the product . . . will be introduced into the human body.”87  Ultimately, however, ChangeLab Solutions recommended a broader definition not linked to nicotine content.  In May 2015, ChangeLab published the current version of its Model California Ordinance Requiring a Tobacco Retail License.88  The model ordinance, while dealing comprehensively with a variety of issues pertaining to tobacco retailing, included provisions to address the policy goal many local jurisdictions were interested in pursuing: ending the loophole for electronic smoking devices by treating them exactly like other tobacco products for regulatory purposes.  Specifically, the model ordinance did so by including electronic smoking devices specifically and unambiguously in the definition of “tobacco products,” without making their inclusion contingent on nicotine delivery.89  Notably, while ChangeLab attempted to make the model ordinance customizable by offering options and “plug-ins” based on local needs and preferences, the e-cigarette provision was one of a handful of ordinance elements identified by ChangeLab as “essential.”

D. The Spread of Explicit, Broad Regulation of Electronic Smoking Devices

As Table A shows, there appears to be a trend toward adopting the Category C model and language resembling the 2015 ChangeLab model ordinance.  Initially, many jurisdictions relied on the older, nicotine-reliant language of the 2005 model ordinance and did not mention e-cigarettes explicitly.  In 2011, for example, six jurisdictions passed a Category A ordinance (regulating electronic smoking devices only implicitly), and just one passed a Category B ordinance (explicitly applying to electronic devices that contain nicotine).  In 2012, the numbers were also lopsided (seven Category A and one Category B).91  In subsequent years, more jurisdictions opted for explicit regulation of electronic smoking devices.  Yet even when jurisdictions began to regulate e-cigarettes explicitly, most of them initially used the Category B approach of tying regulation to nicotine content.  The first five ordinances to pass all used a Category B model,92 and three jurisdictions also used the Category A model in 2013.93  Jurisdictions began to use the Category C model in mid-2014, but they were outnumbered by Category B ordinances that year (seven Category B versus five Category C) and in 2015 (six to five).  Further, three jurisdictions in 2014 and two in 2015 continued to use the Category A model.  In 2016, Category C ordinances outnumbered Category B six to three (with only one Category A ordinance), and all three 2017 ordinances used a Category C approach.  Moreover, almost all Category C ordinances—both before and after the publication of the ChangeLab model ordinance—used language that closely resembled that of the 2015 model ordinance.94

Thus, there appears to have been an increase over time in local ordinances that adopted the Category C approach, using language that mostly tracked the ChangeLab model ordinance.  To be sure, local jurisdictions may have been influenced to adopt this approach by factors other than ChangeLab, such as media coverage of electronic smoking device risks,95 the examples of neighboring communities,96 and the passage of statewide legislation with language similar to the ChangeLab-recommended approach (as described below).97  Nonetheless, communities’ wide use of language closely tracking the ChangeLab approach is suggestive.  Further support for a belief in ChangeLabs’s influence comes from the fact that it provided “technical assistance” to many communities contemplating changes to their tobacco retailer licensing ordinances.98  Indeed, several public proceedings or reports considering a proposed ordinance mentioned ChangeLab’s influence99 or cited ChangeLab’s documents.

A second pattern to emerge is that California communities converged on three basic regulatory models.  Most regulations were easy to classify using the A/B/C scheme, often using identical or near-identical language to others in their category.101  It is notable, therefore, that California jurisdictions mostly converged around a handful of approaches rather than tackling the issue from scratch.  Further, the 2015 ChangeLab model emerges as the preferred choice over time, and the overwhelming preference of communities that chose the broader regulatory approach.102  Notably, many jurisdictions also appear to have been influenced by the older 2005 ChangeLab language in adopting Category A approaches.103

The current scholarship suggests that local governments care about adopting the most effective policies and are capable of gathering information relevant to the decision.  For example, Edward M. Rogers and Jeffrey C. Peterson found that cities and counties learn about new policies from each other: Policy advocates go to different cities to learn from their experiences.104  California’s experience shows localities converging around a few basic approaches to electronic smoking device regulation.  California’s experience, as shown here, also demonstrates a related phenomenon: the existence of some degree of synchronization through a well-respected, state-funded promotion of a policy option that appears to have broadly influenced local governments.

The quick convergence by California’s local jurisdictions on a few favored models of regulation, followed by the rapid passage of a similar law statewide (as described in the following Subpart), has both positive and negative aspects.  On the one hand, the growing convergence of electronic smoking device regulation in California makes it more difficult to perform comparisons of the tradeoffs and relative efficacy of various approaches.  For example, not everyone accepts the public health case for stringent electronic smoking device regulation.  Some public health researchers have found that e-cigarettes are beneficial in helping smokers to reduce their use of traditional cigarettes, which have more negative health effects.105  While those findings are disputed by others106 (and partially accommodated in state and local regulation through the widespread exceptions for FDA-approved smoking cessation devices), one could argue that the rapid adoption of the ChangeLab approach to regulation dampened the possibility of local experimentation with different levels of regulation and related effectiveness comparisons.

Localities’ interest in following ChangeLab models also may have hampered them from moving more swiftly to more sweeping regulation.  ChangeLab did not develop a model ordinance explicitly regulating electronic smoking devices until 2015, and as late as 2016, some jurisdictions were still adopting ordinances that used the old approach of the 2005 model ordinance.  Waiting for guidance from ChangeLab may have deterred jurisdictions from engaging in their own fact-finding and adopting broader language earlier.

The pattern also shows local and state governments in California adapting to changing circumstances and gradually converging on the favored, more up-to-date approach.  It seems likely that ChangeLab’s influence has been predominantly positive.  Because ChangeLab’s policies are rooted in research about the most efficacious approaches, they arguably represent a shortcut in the Brandeisian process, allowing local jurisdictions access to a readily usable model statute that is itself based on adopting ideas from other jurisdictions and from a variety of public health research.

E. California’s Experience Demonstrates Synchronization at the State Level

California’s decision to follow the lead of local governments may also be evidence of policy diffusion from the local to the state level.  Historically, widespread adoption of policies at the local level in California has often been followed by eventual decisions at the state level to enact the policy statewide.  John A Francis, Erin M. Ambramshon and Hye-Young Park, for example, observe that local governments in California were the first to pass laws that restricted smoking in workplaces, such as a 1977 Berkeley law dividing restaurants into smoking and nonsmoking.107  In 1995, the state government eventually followed, responding both to public concerns and tobacco industry fears that, in the absence of statewide action, local jurisdictions would develop more stringent restrictions.108

In its sweeping 2016 law, California chose to regulate the sale of electronic smoking devices by defining tobacco products to include electronic smoking devices in laws that regulate the sale of these products.  This state law used language similar to the Category C style of regulations, which had been adopted in a number of local jurisdictions.109  Although it is unclear how much influence the specific content of the local statutes had on the ultimate statewide bill, the bill’s author, California Senator Mark Leno, touted the widespread adoption of local bans on e-cigarettes in support of the bill.110  Further, proponents of the bill made arguments similar to those advanced by ChangeLab in support of the more comprehensive language—citing concerns that any use of electronic smoking devices could serve as a gateway to traditional tobacco products and that carcinogenic chemicals were present in even nicotine-free liquids used in such devices.112

State adoption can sometimes be a mixed blessing for local pioneers.  On the one hand, it may indicate a wider acceptance of successful local regulations.  On the other hand, premature state adoption may result in the passage of laws that are narrower and less effective, while also potentially stifling local experimentation.  Francis et al. argue that this may have occurred in California’s smoke-free workplace legislation.113  Had this legislation been adopted earlier, they contend, it would have been more tobacco-industry-friendly by including more exemptions.114  By contrast, they suggest that had the state waited until after 1995, additional local regulation would have created a greater opportunity to move public opinion, and a stronger and more comprehensive smoke-free workplace law could have been passed.115  From this perspective, California’s multi-year delay in passing comprehensive e-cigarette regulation may have been desirable.  By the time California passed its 2016 law, the broader, ChangeLab-endorsed approach to regulating electronic smoking devices had already become popular among local jurisdictions.


While Brandeis’s laboratory metaphor is one the most frequently invoked passages in law, there is less empirical evidence about how and why legal innovations spread among jurisdictions.  Brandeis’s metaphor rings true in the sense that local governments do pioneer policy innovations and may be likely to converge around those policies they deem the most effective.  California’s experience is different from the Brandeisian process, however, in the degree to which it appears to reflect the influence of ChangeLab, a state-supported organization that promulgates model policies.  A second lesson from California is that state governments can and do allow localities to experiment with and refine policies before adopting the policy as their own.  This requires caution because state governments can stymie innovation if they prematurely adopt statewide policies before a period of experimentation has taken place.  Thus, ideally state governments should adopt local policies only after there is some evidence of effectiveness.  To the extent that the state waited to adopt a broader regulatory policy until after similar local laws had gained in popularity, California appears to have followed a prudent approach to regulating electronic smoking devices.  California’s experience suggests that state encouragement of local regulatory efforts can both facilitate the spread of sound policy and provide the groundwork for future state-level regulation.

Table 1: Local California Ordinances Adopting “Tobacco Products” Definition Including Electronic Smoking Devices, 2011 to Present

Jurisdiction Ordinance month/year Category Citation
Eastvale (Riverside County) January 2011 A Eastvale, Cal., Eastvale Municipal Code § 6.68.020 (2017)
Carpintiera (Santa Barbara County) February 2011 A Carpinteria, Cal., Carpinteria Municipal Code § 8.52.030 (2017)
Jurupa Valley (Riverside County) July 2011 A Jurupa Valley, Cal., Jurupa Valley Municipal Code § 5.68.020 (2017)
Huntington Park (Los Angeles County) November 2011 A Huntington Park, Cal., Huntington Park Municipal Code § 4-19.01 (2017)
Malibu (Los Angeles County) November 2011 A Malibu, Cal., Malibu Municipal Code § 5.42.020 (2017)
Santa Barbara County (unincorporated) November 2011 A Santa Barbara County, Cal., The Code of Santa Barbara County, California § 37A-2 (2018)
El Monte (Los Angeles County) November 2011 B El Monte, Cal., El Monte Municipal Code § 8.10.010 (2017)
Oxnard (Ventura County) March 2012 B Oxnard, Cal., The Code of the City of Oxnard, California § 11-370 (2018)
Carmel by the Sea (Monterey County) May 2012 A Carmel-by-the-Sea, Cal., Carmel-by-the-Sea Municipal Code § 5.56.010 (2017)
Marin County (unincorporated) May 2012 A Marin County, Cal., Marin County Code § 5.70.040 (2018)
Monterey County (unincorporated) May 2012 A Monterey County, Cal., Monterey County Code § 7.80.020 (2017)
Mill Valley (Marin County) September 2012 A Mill Valley, Cal., Mill Valley Municipal Code § 7.20.020 (2017)
Santa Cruz (Santa Cruz County) October 2012 A Santa Cruz, Cal., Municipal Code of the City of Santa Cruz—1995 Edition § 6.07.010 (2018)
Temple City (Los Angeles County) November 2012 A Temple City, Cal., Temple City Municipal Code § 5-2D-2 (2017)
Campbell (Santa Clara County) December 2012 A Campbell, Cal., Campbell Municipal Code, Revised § 5.58.190 (2018)
Oroville (Butte County) March 2013 A Oroville, Cal., Oroville Municipal Code § 5.28.010 (2017)
Hawaiian Gardens (Los Angeles) September 2013 B Hawaiian Gardens, Cal., Hawaiian Gardens Municipal Code § 5.96.020 (2018)
Saratoga (Santa Clara County) October 2013 A Saratoga, Cal., Saratoga City Code § 15-80.130 (2018)
Los Angeles (Los Angeles County) December 2013 A L.A., Cal., Official Los Angeles Municipal Code § 46.90 (2017)
Goleta (Santa Barbara County) May 2014 A Goleta, Cal., Goleta Municipal Code § 5.07.010 (2017)
Gardena (Los Angeles County) May 2014 B Gardena, Cal., Gardena Municipal Code § 5.52.010 (2017)
Rancho Cordova (Sacramento County) June 2014 B Rancho Cordova, Cal., Rancho Cordova Municipal Code § 4.07.030 (2018)
Dublin (Alameda County) July 2014 C Dublin, Cal., Dublin Municipal Code § 8.43.030 (2017)
Hayward (Alameda County) July 2014 C Hayward, Cal., Hayward Municipal Code § 10-1.2782 (2018)
Kern County (unincorporated areas) July 2014 B Kern County, Cal., Ordinance Code of Kern County § 8.60.010 (2018)
Westminster (Orange County) August 2014 A Westminster, Cal., Westminster Municipal Code § 5.38.020 (2018)
El Cajon (San Diego County) August 2014 C El Cajon, Cal., El Cajon Municipal Code § 8.33.010 (2018)
Duarte (Los Angeles County) August 2014 B Duarte, Cal., Duarte Municipal Code § 5.09.020 (2018)
Santa Monica (Los Angeles County) October 2014 B Santa Monica, Cal., Santa Monica Municipal Code § 4.45.020 (2018)
Monterey (Monterey County) November 2014 A Monterey, Cal., Monterey City Code § 19-106 (2017)
Healdsburg (Sonoma County) November 2014 B Healdsburg, Cal., Healdsburg City Code § 8.26.010 (2018)
Portola Valley (San Mateo County) November 2014 C Portola Valley, Cal., Portola Valley Municipal Code § 8.26.010 (2016)
San Mateo County (unincorporated areas) November 2014 C San Mateo County, Cal., San Mateo County Ordinance Code § 4.96.030 (2018)
San Francisco (San Francisco County) December 2014 B S.F., Cal., Health Code § 19H.2 (2018)
Salinas (Monterey County) January 2015 A Salinas, Cal., Salinas City Code § 16-119.1 (2017)
San Luis Obispo (San Luis Obispo County) March 2015 B San Luis Obispo, Cal., San Luis Obispo Municipal Code § 8.14.020 (2018)
Crescent City (Del Norte County) April 2015 C Crescent City, Cal., Crescent City Municipal Code § 17.91.010 (2017)
Sonoma (Sonoma County) June 2015 C Sonoma, Cal., Sonoma Municipal Code § 7.25.010 (2017)
Daly City (San Mateo County) August 2015 B Daly City, Cal., Daly City Code § 8.54.015 (2017)
Culver City (Los Angeles County) August 2015 B Culver City, Cal., The Code of the City of Culver City, California § 11.15.010 (2018)
Berkeley (Alameda County) September 2015 B Berkeley, Cal., Berkeley Municipal Code § 9.80.020 (2018)
Rocklin (Placer County) September 2015 B Rocklin, Cal., Rocklin Municipal Code § 17.69.020 (2018)
El Cerrito (Contra Costa County) October 2015 C El Cerrito, Cal., El Cerrito Municipal Code § 6.100.020 (2018)
San Mateo (San Mateo County) October 2015 C San Mateo, Cal., San Mateo Municipal Code § 7.40.020 (2017)
Pleasant Hill (Contra Costa County) November 2015 A Pleasant Hill, Cal., Pleasant Hill Municipal Code § 9.45.030 (2018)
Brisbane (San Mateo County) December 2015 C Brisbane, Cal., Brisbane Municipal Code § 8.45.030 (2017)
Fremont (Alameda County) December 2015 B Fremont, Cal., Fremont Municipal Code § 8.55.030 (2017)
Tehachapi (Kern County) March 2016 B Tehachapi, Cal., Tehachapi Municipal Code § 8.52.020 (2017)
Wasco (Kern County) April 2016 B Wasco, Cal., Wasco Municipal Code § 8.52.010 (2017)
Morgan Hill (Santa Clara County) May 2016 A Morgan Hill, Cal., Morgan Hill Municipal Code § 5.64.020 (2018)
Nevada City (Nevada County) July 2016 B Nevada City, Cal., Nevada City Municipal Code § 5.32.010 (2018)
Riverside County (unincorporated areas) July 2016 C Riverside County, Cal., Ordinance 838.2 (Jul. 26, 2016)
Calimesa (Riverside County) July 2016 C Calimesa, Cal., Calimesa Municipal Code § 5.15.020 (2017)
Yolo County (unincorporated areas) September 2016 C Yolo County, Cal., Yolo County Code § 6-15.02 (2018)
Woodland (Yolo County) September 2016 C Woodland, Cal., The Code of the City of Woodland, California, 1955 § 15-43 (2017)
Davis (Yolo County) September 2016 C Davis, Cal., Davis Municipal Code § 34.06.010 (2018)
Santa Clara County (unincorporated areas) October 2016 C Santa Clara County, Cal., County of Santa Clara Ordinance Code § A18-368 (2018)
Alturas (Modoc County) February 2017 C Alturas, Cal., The Code of the City of Alturas, California § 30-1 (2017)
Rancho Mirage (Riverside County) June 2017 C Rancho Mirage, Cal., Rancho Mirage Municipal Code § 5.24.110 (2018)
Oakland (Alameda County) September 2017 C Oakland, Cal., Ordinance 13452 (Sept.19, 2017)

1.       See 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting).

2.       To use one measure, Westlaw indicates that New State Ice has been cited in more than 4000 cases and articles; surveying a sample of these suggests that nearly all citations are to Justice Brandeis’s famous words.

3.       President Ronald Reagan, 1983 State of the Union Address (Jan. 25, 1983) (transcript available at http://www.pbs.org/wgbh/americanexperience/features/primary-resources/reagan-union-1983 [https://perma.cc/GT2N-PBFQ]).

4.       Ralph Nader, State Legislatures as “Laboratories of Democracy,” [small-caps]Common Dreams[end-small-caps] (May 31, 2004), http://www.commondreams.org/views/2004/05/31/state-legislatures-laboratories-democracy [https://perma.cc/T6KD-HMUS].

5.       Under Brandeis’s conception of state laboratories, discussed at greater length infra Part I, jurisdictions should be sufficiently free to experiment with new policies, which can then be evaluated and widely adopted if successful.

6.       Much of the legal scholarship on the subject has taken a position of skepticism toward the innovation-diffusion model.  See, e.g., Brian Galle & Joseph Leahy, Laboratories of Democracy?  Policy Innovation in Decentralized Governments, 58 [small-caps]Emory L.J.[end-small-caps] 1333, 1339 (2009) (“State and local governments do innovate.  But they are unlikely to innovate in all instances at the optimal social level, or in a way that captures the true benefits of experimentation.”); Susan Rose-Ackerman, Risk Taking and Reelection: Does Federalism Promote Innovation?, 9 [small-caps]J. Legal Stud.[end-small-caps] 593, 594 (1980) (arguing that reelection pressures faced by politicians, among other factors, tend to stifle genuine innovation).  See also Michael A. Livermore, The Perils of Experimentation, 126 [small-caps]Yale L.J.[end-small-caps] 636, 708 (2017) (describing the negative aspects of policy diffusion, including the production of information that “could potentially be used by self-interested actors in ways that undermine, rather than promote, social well-being”).

[7].       See Katherine Florey, Making It Work: Tribal Innovation, State Reaction, and the Future of Tribes as Regulatory Laboratories, 92 [small-caps]Wash. L. Rev.[end-small-caps] 713, 723–24 (2017) (discussing scholarship providing evidence of state convergence).

[8].       See, e.g., Heather K. Gerken, Foreword: Federalism All the Way Down, 124 Harv. L. Rev. 4 (2010); Rick Su, Intrastate Federalism, 19 [small-caps]J. Const. L.[end-small-caps] 191 (2016).

[9].       See Michael R. Eriksen & Rebecca L. Cerak, The Diffusion and Impact of Clean Indoor Air Laws, 29 [small-caps]Ann. Rev. Pub. Health[end-small-caps] 171, 174 (2008).

[10].    Charles R. Shipan & Craig Volden, Bottom-Up Federalism: The Diffusion of Antismoking Policies from U.S. Cities to States, 50 [small-caps]Am. J. Pol. Sci.[end-small-caps] 825, 826 (2006).  The authors further elaborate that “[w]ith respect to policy-oriented learning, state-level policymakers may view localities to be laboratories of democracy in a similar way to how state and national officials view the policies in various states.”  Id.

[11].    See infra Part II.A.

[12].    See infra Table 1.

[13].    See id.

[14].    See New State Ice Co. v. Liebmann, 285 U.S. 262, 306 (1932) (Brandeis, J., dissenting).

[15].    Id. at 310.

[16].    Id. at 310–11.

[17].    Brandeis’s model could, then, fail in several ways: States might lack sufficient freedom to make innovation possible, or decline to engage in innovation for other reasons; state experiments might not be susceptible to ready evaluation; and states might either lack interest in borrowing from sister states or borrow policies for reasons unrelated to their objective record of success (such as competitive pressures or the influence of fads).

[18].    See Frederick J. Boehmke et al., The Seeds of Policy Change: Leveraging Diffusion to Disseminate Policy Innovations, 42 [small-caps]J. Health Pol., Pol’y & L.[end-small-caps] 285, 290 (2017)

[19].    See id.  See also Livermore, supra note 6, at 650 (characterizing mechanisms of diffusion in a similar fashion).

[20].    See Livermore, supra note 6, at 650 & n.48 (citing the example of a populous jurisdiction “setting a product standard in a large market that all manufacturers must meet”).

[21].    See James Macinko & Diana Silver, Diffusion of Impaired Driving Laws Among US States, 105 [small-caps]Am. J. Pub. Health[end-small-caps] 1893 (2015) (finding that the existence of a neighboring state with impaired driving laws helped to explain states’ adoption of similar laws).  But see Boehmke et al., supra note 18, at 287 (finding that states are more influenced by adoption of new policies by states that had historically served as “policy leaders” than by adoption by contiguous states); Jack L. Walker, The Diffusion of Innovations Among the American States, 63 [small-caps]Am. Pol. Sci. Rev.[end-small-caps] 880, 897 (1969) (finding that states are most likely to be influenced by other states when they view the state that has adopted the policy as a “point of legitimate comparison”).

[22].    Everett M. Rogers & Jeffery C. Peterson, Diffusion of Clean Indoor Air Ordinances in the Southwestern United States, 35 [small-caps]Health Educ. & Behav.[end-small-caps] 683 (2008).

[23].    Id. at 694.

[24].    Joshua L. Mitchell & La Shonda M. Stewart, Emulation, Learning, or Competition?  Examining Inter-County Anti-Smoking Laws in the State of Missouri, 30 [small-caps]Pub. Admin. Q.[end-small-caps] 317, 332–33 (2014).

[25].    Id. at 333 (presenting evidence for these effects while cautioning that it should not be over-interpreted).

[26].    See Macinko & Silver, supra note 21, at 1893, 1899.

[27].    [small-caps]Tobacco Control Legal Consortium, Pub. Health Law Ctr., Regulating Electronic Cigarettes & Similar Devices[end-small-caps] (2017), http://www.publichealthlawcenter.org/sites/default/files/resources/tclc-guide-reg-ecigarettes-2016.pdf [https://perma.cc/XBG2-BLCC].

[28].    The U.S. Customs and Border Protection import notice for some of the first devices, dated August 22, 2006.  Letter from Robert B. Swierupski, Director, Nat’l Commodity Specialist Div., to Mark Weiss, Weiss & Moy, P.C., https://rulings.cbp.gov/index.asp?ru=m85579&qu=NY+M85579&vw=detail  [https://perma.cc/WB8K-DEAA].

[29].    See, e.g., David Kesmodel & Danny Yadron, E-Cigarettes Spark New Smoking War, [small-caps]Wall Street. J.[end-small-caps] (Aug. 25, 2010, 12:01 AM), https://www.wsj.com/articles/SB10001424052748704557704575437710870116450; Kara Newman, The E-Cigarette: Nicotine Fix, Without the Toxins, [small-caps]CNN Money[end-small-caps] (Aug. 13, 2007, 6:57 AM), http://money.cnn.com/2007/08/09/magazines/business2/electronic_cigarette.biz2/index.htm; Katie Zezima, Cigarettes Without Smoke, or Regulation, [small-caps]N.Y. Times[end-small-caps] (June 1, 2009), http://www.nytimes.com/2009/06/02/us/02cigarette.html.

[30].    See Gregory D. Curfman et al., Editorial, Tobacco, Health, and the FDA, 361 [small-caps]New Engl. J. Med.[end-small-caps] 402 (2009); Daniel J. DeNoon, FDA: E-Cigarettes Bad, but Not Banned, [small-caps]WebMD[end-small-caps] (July 22, 2009), https://www.webmd.com/smoking-cessation/news/20090722/fda-ecigarettes-bad-not-banned#1 [https://perma.cc/X9UV-62HQ].

[31].    [small-caps]ChangeLab Sols., Regulating Toxic Vapor: A Policy Guide to Electronic Smoking Devices[end-small-caps] (2014), http://www.changelabsolutions.org/sites/default/files/Regulating_Toxic_ Vapor-FINAL-20140630.pdf [https://perma.cc/WS2L-TSV2].

[32].    See id.

[33].    See id.

[34].    See [small-caps]Tobacco Control Legal Consortium[end-small-caps], supra note 27.

[35].    See [small-caps]ChangeLabs Sols.[end-small-caps], supra note 31.

[36].    See, e.g., [small-caps]Cal. Penal Code[end-small-caps] § 308(a) (2014 & Supp. 2018).

[37].    See Amanda Shanor, The New Lochner, 2016 [small-caps]Wis. L. Rev.[end-small-caps] 133, 169 (2016).  At the federal level, this has been particularly true since 2009, when the Food and Drug Administration was given new powers over tobacco products.  See id.

[38].    See [small-caps]ChangeLabs Sols.[end-small-caps], supra note 31.

[39].    See [small-caps]Cal. Health & Safety Code[end-small-caps] § 119405 (2012) (repealed 2018); [small-caps]Fiona Ma, Cal. Bd. of Equalization, E-Cigarettes: Challenges and Opportunities for California[end-small-caps] 2, https://www.boe.ca.gov/ma/pdf/e-CigReport.pdf [https://perma.cc/M6A8-9YDD].

[40].    See [small-caps]Ma[end-small-caps], supra note 39.

[41].    See [small-caps]ChangeLabs Sols.[end-small-caps], supra note 31. (noting that “increasing popularity of electronic smoking devices, combined with loopholes in some existing tobacco control laws, have the potential to renormalize tobacco use”).

[42].    See [small-caps]Cal. Bus. & Prof. Code[end-small-caps] § 22950.5(d)(1) (2017) (defining “tobacco product” as “a product containing, made, or derived from tobacco or nicotine that is intended for human consumption [including] an electronic device that delivers nicotine or other vaporized liquids to the person inhaling from the device, including, but not limited to, an electronic cigarette, cigar, pipe, or hookah . . . ”).  This expanded definition means that electronic cigarette (e-cigarette) use is prohibited in any situation in which use of other tobacco products is prohibited.  See [small-caps]Cal. Bus. & Prof. Code[end-small-caps] § 22950.5(c) (2017).

[43].    Lisa Aliferis, California Raises Age of Tobacco Purchase to 21 and Tightens Vaping Rules, [small-caps]NPR[end-small-caps] (May 5, 2016, 10:58 AM), http://www.npr.org/sections/health-shots/2016/05/05/476872674/california-raises-age-of-tobacco-purchase-to-21-and-tightens-vaping-rules.

[44].    See id.

[45].    See California Proposition 56, Tobacco Tax Increase (2016), [small-caps]Ballotpedia[end-small-caps], https://ballotpedia.org/California_Proposition_56,_Tobacco_Tax_Increase_(2016) [https://perma.cc/T36L-2VL7].

[46].    See FDA’s New Regulations for E-Cigarettes, Cigars, and All Other Tobacco Products, [small-caps]FDA[end-small-caps], https://www.fda.gov/TobaccoProducts/Labeling/RulesRegulationsGuidance/ucm394909.htm [https://perma.cc/GZJ3-ZRTA].

[47].    See The Facts on the FDA’s New Tobacco Rule, [small-caps]FDA[end-small-caps] (Nov. 9, 2017), https://www.fda.gov/ForConsumers/ConsumerUpdates/ucm506676.htm [https://perma.cc/L9UQ-LKVC].

[48].    See [small-caps]ChangeLab Sols[end-small-caps]., supra note 31.

[49].    See, e.g., [small-caps]Tobacco Control Legal Consortium[end-small-caps], supra note 29.

[50].    See [small-caps]ChangeLab Sols.[end-small-caps], supra note 31.

[51].    [small-caps]Cal. Tobacco Control Program, Cal. Dep’t of Health, California Tobacco Control Program Overview[end-small-caps] (2017), https://www.cdph.ca.gov/Programs/CCDPHP/DCDIC/CTCB/CDPH%20Document%20Library/AboutUS/ProgramOverview/AboutUsProgramOverview081216.pdf [https://perma.cc/F9QR-ZP4C].

[52].    See id.

[53].    See id.

[54].    See Eriksen & Cerak, supra note 9, at 174.

[55].    See id.

[56].    See id.

[57].    See id.

[58].    See Leslie Zellers & Ian McLaughlin, State and Local Policy as a Tool to Complement and Supplement the FDA Law, 2 [small-caps]Hastings Sci. & Tech. L.J.[end-small-caps] 117, 123 (2010).

[59].    See id.

[60].    See id.

[61].    [small-caps]ChangeLab Sols., Tobacco Retailer Licensing: An Effective Tool for Public Health[end-small-caps] (2012), http://www.changelabsolutions.org/sites/default/files/TobaccoRetailerLicensing_ factsheet_FINAL_20120907.pdf [https://perma.cc/HHV9-UAZR].

[62].    See Zellers & McLaughlin, supra note 58, at 124.

[63].    The authors note that, “Since the start of CTCP, overall cigarette consumption in California has declined by 60.8%.”  John A Francis et al., Policy Driven Tobacco Control, 19 Tobacco Control i16, i18 (2010).  Technical Assistance Legal Center (TALC) is part of California Tobacco Control Program (CTCP).  See also Zellers & McLaughlin, supra note 58, at 123.

[64].    See About Tobacco Control, [small-caps]ChangeLab Sols.[end-small-caps], http://www.changelabsolutions.org/tobacco-control/about [https://perma.cc/XUB2-2XCV].

[65].    Id.  (describing ChangeLab Solutions as providing communities with “research on specific legal issues”).

[66].    See Model Policies, [small-caps]ChangeLab Sols.[end-small-caps], http://changelabsolutions.org/landing-page/model-policies [https://perma.cc/YW3L-7Z5H].

[67].    Compare [small-caps]ChangeLab Sols., Model California Ordinance Requiring a Tobacco Retailer License[end-small-caps] (Cal. Dep’t of Pub. Health 2015) [hereinafter [small-caps]Model California Ordinance[end-small-caps]] (containing a requirement that tobacco retailers obtain a license, a requirement that tobacco retailer renew their license annually, a provision that requires payment of a fee that is sufficient to cover the “administration and enforcement” of the policy, a requirement that tobacco product retailers comply with all laws governing the sale of tobacco products, and a provision that imposes fines and the possibility of a suspension or revocation of a license for violating the policy) with [small-caps]The Ctr. for Tobacco Policy & Org., Am. Lung Ass’n in Cal., Tobacco Retailer Licensing Is Effective[end-small-caps] (2013), http://center4tobaccopolicy.org/wp-content/uploads/2016/10/Tobacco-Retailer-Licensing-is-Effective-September-2013.pdf [https://perma.cc/UXE8-F5RE] (noting that the policies that require tobacco retailers to obtain a license to sell tobacco and renew it annually have a fee that sufficiently funds the licensing program and its enforcement mechanisms, requires tobacco retailers to comply with all laws governing the sale of tobacco products, and have penalties that include fines as well as a provision for the “suspension and revocation of the license” and measures for “reduce[d] illegal sales of tobacco products to minors”).

[68].    [small-caps]Model California Ordinance[end-small-caps], supra note 67.

[69].    See [small-caps]Elizabeth Cox et al., Tobacco Control in California, 2007–2014: A Resurgent Tobacco Industry While Inflation Erodes the California Tobacco Control Program[end-small-caps] 136 (2014), https://cloudfront.escholarship.org/dist/prd/content/qt4jj1v7tv/qt4jj1v7tv.pdf [https://perma.cc/5NH9-ABYG].

[70].    See Noel Brinkerhoff, American Canyon Bans Tobacco, E-Cigarette Stores, [small-caps]Napa Valley Reg.[end-small-caps] (Jan. 21, 2016), http://napavalleyregister.com/news/local/american-canyon-bans-tobacco-e-cigarette-stores/article_17e52583-0e97-5d2a-8b0e-4b4e090caffc.html [https://perma.cc/AC76-RVS6].

[71].    The 2005 ordinance defines “tobacco products” as including: "(1) any substance containing tobacco leaf, including but not limited to cigarettes, cigars, pipe tobacco, snuff, chewing tobacco, dipping tobacco, bidis or any other preparation of tobacco; and (2) any product or formulation of matter containing biologically active amounts of nicotine that is manufactured, sold, offered for sale, or otherwise distributed with the expectation that the product or matter will be introduced into the human body, but does not include any product specifically approved by the FDA for use in treating nicotine or tobacco product dependence." See [small-caps]Cox et al.[end-small-caps], supra note 69, at 135 (emphasis added).

[72].    See id.

[73].    In order of passage, those ordinances were [small-caps]Norco, Cal., Municipal Code of Norco[end-small-caps] § 9.80.020 (2017) (passed 2006); [small-caps]Hollister, Cal., Hollister Municipal Code[end-small-caps] § 5.38.030 (2017) (passed 2006); [small-caps]Murrieta, Cal., Murrieta Municipal Code[end-small-caps] § 5.24.020 (2017) (passed 2006); [small-caps]Riverside, Cal., Riverside Municipal Code[end-small-caps] § 6.24.020 (2017) (passed 2006); [small-caps]San Jacinto, Cal., San Jacinto Municipal Code[end-small-caps] § 5.56.030 (2017); [small-caps]Temecula, Cal., Temecula Municipal Code[end-small-caps] § 5.24.030 (2017); [small-caps]Lancaster, Cal., Lancaster Municipal Code[end-small-caps] § 5.42.030 (2018) (passed 2006); [small-caps]Maywood, Cal., Maywood Municipal Code[end-small-caps] § 4-10.101 (2010) (effective Aug. 22, 2006); [small-caps]Banning, Cal., The Code of the City of Banning, California, 1965[end-small-caps] § 5.72.030 (2017); [small-caps]Concord, Cal., Code of Ordinances, City of Concord, California[end-small-caps] § 5.100.010 (2017); [small-caps]Santa Ana, Cal., Santa Ana Municipal Code[end-small-caps] § 18-90 (2017) (passed Nov. 20, 2006); and [small-caps]Beaumont, Cal., Beaumont Municipal Code[end-small-caps] § 5.68.030 (2018) (passed December 2006).

[74].    See [small-caps]Cox et al.[end-small-caps], supra note 69, at 143–46.  This ordinance was adopted notwithstanding intense opposition from local vaping activists, e-cigarette retailers, and electronic smoking device trade associations, which understood the ordinance to be an attack on their product.

[75].    [small-caps]El Monte, Cal., El Monte Municipal Code[end-small-caps] § 8.10.010 (2017).

[76].    See id.  A few jurisdictions adopted unique regulations that were more difficult to categorize.  In such cases, we classified them as Category B if their predominant focus was on nicotine and tobacco.  For example, Duarte’s ordinance defines “tobacco products” as “any substance containing tobacco or derived from tobacco and any substance used in electronic cigarette and vaping devices including but not limited to cigarettes, cigars, e-juice, e-liquid, e-nicotine, smoke juice, pipe tobacco, rolling tobacco, hookah tobacco, snuff, chewing tobacco, dipping tobacco, snus, nicotine gel, nicotine lollipops, or any other preparation of tobacco.”  [small-caps]Duarte, Cal., Duarte Municipal Code[end-small-caps] § 5.09.020(j) (2018).

[77].    See [small-caps]Hawaiian Gardens, Cal., Hawaiian Gardens Municipal Code[end-small-caps] § 5.96.020 (2018).  Other ordinances in this category use a definition of “tobacco product” based solely on nicotine but elsewhere clarify that electronic smoking devices are included.  For example, Tehachapi’s ordinance defines “tobacco product” in a nicotine-dependent manner as “any product that contains tobacco, is derived from tobacco, or contains synthetically produced nicotine and is intended for human consumption.”  See [small-caps]Tehachapi, Cal., Tehachapi Municipal Code[end-small-caps] § 8.52.020 (2017).  Yet the ordinance also specifies that “tobacco product” includes “electronic smoking devices(s) and electronic smoking device paraphernalia.”  See id.  Although the statute is somewhat ambiguous, it appears to apply to electronic smoking devices and paraphernalia only insofar as they are associated with nicotine delivery.

[78].    See [small-caps]Yolo County, Cal., Yolo County Code[end-small-caps] § 6-15.02 (2018) (emphasis added).

[79].    ChangeLab released a model ordinance in 2008 that made minor changes to the definition of “tobacco products” but mainly retained the language and approach from the 2005 version.  [small-caps]Model California Ordinance[end-small-caps], supra note 67.  Because these two ordinances are so similar, we did not consider the 2008 version separately.

[80].    These two jurisdictions are Hayward and Santa Clara.  Hayward’s ordinance defines “tobacco products” in terms of nicotine content, but then provides that it applies to electronic smoking devices “with or without nicotine.”  [small-caps]Hayward, Cal., Hayward Municipal Code[end-small-caps] § 10-1.2782(t) (2018).  Santa Clara County’s ordinance (applicable to unincorporated areas), the only other ordinance that differed significantly from the ChangeLab model, differed in tying the definition of “tobacco products” to a federal standard (Subchapter IX of the Federal Food, Drug and Cosmetic Act and federal regulations interpreting it) and by specifying a comprehensive list of included items.  See [small-caps]Santa Clara County, Cal., County of Santa Clara Ordinance Code[end-small-caps] § A18-368(e) (2018).

[81].    The equivalent provision from the ChangeLab Model Ordinance reads as follows: “'Tobacco Product' means: (1) any product containing, made, or derived from tobacco or nicotine that is intended for human consumption, whether smoked, heated, chewed, absorbed, dissolved, inhaled, snorted, sniffed, or ingested by any other means, including, but not limited to cigarettes, cigars, little cigars, chewing tobacco, pipe tobacco, snuff; and (2) any Electronic Smoking Device. (3) Notwithstanding any provision of subsections (1) and (2) to the contrary, “tobacco product” includes any component, part, or accessory of a tobacco product, whether or not sold separately.  “Tobacco product” does not include any product that has been approved by the United States Food and Drug Administration for sale as a tobacco cessation product or for other therapeutic purposes where such product is marketed and sold solely for such an approved purpose. See [small-caps]Model California Ordinance[end-small-caps], supra note 67.  A comment explains that “[t]he definition . . . includes electronic cigarettes that do not contain nicotine.”  Id.  Ordinances were categorized as closely resembling the ChangeLab model ordinance if they used a multipronged approach to defining “tobacco product” that included electronic smoking devices as a separate prong and did not link their regulation to nicotine content.  Some ordinances within this category incorporated the suggested ChangeLab language nearly exactly, while others used slight variants.

[82].    See [small-caps]ChangeLab Sols. Model Ordinance Regulating Electronic Smoking Devices[end-small-caps] 5 (Cal. Dep’t of Pub. Health 2015), http://changelabsolutions.org/publications/e-cig-ord [https://perma.cc/L9RX-SHEW].  Although ChangeLab first published a model California stand-alone ordinance to regulate electronic smoking devices in 2013, it appears that more jurisdictions have, instead of adopting a stand-alone ordinance, chosen to regulate e-cigarettes via their tobacco retailer ordinance.

[83].    See id.  Note, however, that the same ordinance includes optional alternative language for jurisdictions that prefer to “limit [their] regulation of Electronic Smoking Device sales [as opposed to use] to only those products that contain nicotine.”

[84].    See id.

[85].    See id.

[86].    See id.

[87].    See [small-caps]Cox et al.[end-small-caps], supra note 69, at 166–67.

[88].    See id.

[89].    See id.

[90].    See [small-caps]ChangeLab Sols.[end-small-caps], Licensing Ordinance Checklist (Cal. Ctr. for Tobacco Policy & Org., Am. Lung Ass’n 2013), https://www.changelabsolutions.org/sites/default/files/CA.LicensingOrdinance_FINAL_20130724.pdf [https://perma.cc/KF29-UYNZ].

[91].    See infra Table 1.

[92].    See id.  These ordinances were passed by El Monte, Oxnard, Hawaiian Gardens, Gardena, and Rancho Cordova

[93].    See id.

[94].    Some jurisdictions—for example, Dublin and El Cajon—phrased their ordinances in terms strongly resembling the ChangeLab model ordinance even prior to that ordinance’s publication.  This may reflect informal consultation between ChangeLab and some California jurisdictions, or the fact that ChangeLab’s model ordinance is itself based in part on surveying best practices already in effect.

[95].    See e.g., Zezima, supra note 29.

[96].    In some cases, there are clear examples of transmission patterns that do not directly involve ChangeLab.  For example, Davis and Woodland, communities within Yolo County, have city ordinances that resemble the model tobacco retailer code because they adopted the Yolo County ordinance—which tracks the ChangeLab language closely—as part of their city tobacco retailer codes.  See [small-caps]Davis, Cal., Davis Municipal Code[end-small-caps] § 34.06.010 (2018); [small-caps]Woodland, Cal., The Code of the City of Woodland, California, 1955[end-small-caps] § 15-43 (2017); [small-caps]Yolo County, Cal., Yolo County Code[end-small-caps] § 6-15.02 (2018).

[97].    See infra Part II.E.  Local jurisdictions may adopt statewide provisions into local law to facilitate enforcement.

[98].    See [small-caps]Cox et al.[end-small-caps], supra note 69, at 136.  This behind-the-scenes assistance by ChangeLab may also help account for the fact that some jurisdictions used language resembling that of the 2015 model ordinance even before it was officially published.

[99].    See, e.g., [small-caps]The City of El Cerrito, Agenda: Regular City Council Meeting[end-small-caps] (Sept. 15, 2015), https://www.el-cerrito.org/ArchiveCenter/ViewFile/Item/2417 [https://perma.cc/E7C9-HQFF] (noting that the staff directed to develop a tobacco retailer licensing program “worked with ChangeLab Solutions” as well as “research[ing] best practices in other communities”).

[100]See [small-caps]City of El Cajon, Planning Commission Agenda[end-small-caps] (July 1, 2014), https://www.cityofelcajon.us/home/showdocument?id=2640 [https://perma.cc/7GZH-Q9EA] (attaching ChangeLab materials in connection with agenda item “Electronic Vapor Devices, Vapor Inhalation Substances, and Hookah”); [small-caps]City of Hayward, Planning Commission Report[end-small-caps] (Nov. 21, 2013), http://www.cityofhayward-ca.gov/CITY-GOVERNMENT/BOARDS-COMMISSIONS-COMMITTEES/PLANNING-COMMISSION/2013/PCA13PDF/pca112113full.pdf [https://perma.cc/N6TQ-3R2V] (report summarizing recommendations on tobacco retailer licensing, attaching material from ChangeLab Solutions).

[101].  There are only a few exceptions, such as Duarte’s ordinance, mentioned earlier.  See supra note 76.

[102]See infra Table 1.  Regulations closely tracking the 2015 ChangeLab approach accounted for seventeen of the nineteen total Category C regulations.

[103]See [small-caps]Cox et al.[end-small-caps], supra note 69, at 135–36 (discussing influence of 2005 ChangeLab ordinance on e-cigarette regulation).

[104]See Rogers & Peterson, supra note 22.

[105]See, e.g., Riccardo Polosa et al., Effect of an Electronic Nicotine Delivery Device (E-Cigarette) on Smoking Reduction and Cessation: A Prospective 6-Month Pilot Study, 11 [small-caps]BMC Pub. Health[end-small-caps] 786 (2011), https://bmcpublichealth.biomedcentral.com/articles/10.1186/1471-2458-11-786 [https://perma.cc/9YQ3-AQMB].

[106]See, e.g., [small-caps]ChangeLab Sols.[end-small-caps], supra note 31 (discussing World Medical Association findings that the usefulness of electronic smoking devices in smoking cessation is unproven).

[107].  See John A Francis et al., supra note 63, at i17.

[108]See id.  The state law, while exemption-ridden, did permit localities to develop stricter regulations; however, local efforts at tobacco regulation nonetheless slowed following the passage of the statewide law.  See id. at i17-i18.  In fact, legal scholars have argued that states should give their cities more powers because cities can be “laboratories of democracy.”  Paul Diller, Intrastate Preemption, 87 [small-caps]B.U. L. Rev.[end-small-caps] 1113, 1128 (2007).  Charles R. Shipan & Craig Volden have likewise discussed the fact that “state politicians look to localities for policy ideas” and the ways in which policy adoption by local governments may “signal political viability” and “signal effectiveness.”  See Shipan & Volden, supra note 10, at 826–27.

[109]See Gene J. Puskar, Clearing the Air on California’s New Tobacco E-Cigarette Law, [small-caps]Sacramento Bee[end-small-caps], http://www.sacbee.com/news/state/california/article82815702.html.  California defines tobacco products as “any of the following,” including products “containing, made, or derives from tobacco or nicotine,” “electronic device[s] that deliver[] nicotine or other vaporized liquids to the person inhaling from the device, including, but not limited to, an electronic cigarette, cigar, pipe, or hookah,” and “any component, part, or accessory of a tobacco product, whether or not sold separately.”  See [small-caps]Cal. Bus. & Prof. Code[end-small-caps] § 22950.5(d) (2017).  California’s law also includes an exemption, present in most local ordinances as well, for FDA-approved smoking cessation aids.  See id.

[110].  Lydia O’Connor, E-Cigarettes Could Face Major Bans in California, [small-caps]HuffPost[end-small-caps] (Jan. 27, 2015, 10:12 AM), https://www.huffingtonpost.com/2015/01/26/california-e-cigarette-ban-vaping_n_6551336.html [https://perma.cc/V47E-H62A] (noting that Leno’s office cited “122 cities and counties in California [that] have already prohibited the use of e-cigarettes in some indoor and outdoor spaces”).

[111]See Anna Curry, Tobacco Reform in the Special Session: Vaporized?  The Cloudy Status of New Restrictions and Taxes for the Tobacco Industry, 47 [small-caps]U. Pac. L. Rev.[end-small-caps] 476, 484-85 (2016) (suggesting that the California law was “inspired” by such concerns); Angela Musallam, California Senate Advances Bill Increasing Smoking Age to 21, Regulating Vaping, [small-caps]CBS Sacramento[end-small-caps] (Mar. 10, 2016, 9:05 PM), http://sacramento.cbslocal.com/2016/03/10/california-senate-advances-bills-increasing-smoking-age-to-21-regulating-vaping [https://perma.cc/PMW8-H69P] (quoting Senator Leno describing such devices as “big tobacco’s way to addict the new generation”).

[112]See Musallam, supra note 111.

[113]See John A Francis et al., supra note 63, at i19.

[114]See id.

[115]See id.

About the Author

Katherine Florey is Professor of Law at the University of California, Davis, School of Law. Andrew Doan is a Public Health Law Fellow at the University of California, Davis School of Law, and received his J.D. from the University of California Davis, School of Law in 2016. This research was supported in part by a grant from the California Department of Public Health.

By uclalaw