|Home | About | Journals | Submit | Contact Us | Français|
To amplify earlier studies of unintended consequences of public policies, this article illustrates both negative and positive unanticipated consequences of smoke-free workplace policies in California bars for women of low SES.
The article relies on thematic analysis in 2008 of qualitative data gathered between 2001 and 2007 from three mixed-method studies of tobacco use in and around bars where indoor smoking is prohibited.
Unanticipated consequences primarily occurred when bars did comply with the law and smokers went outside the bar to smoke, particularly when smokers stood on the street outside the bar. Key negative consequences for women who smoked outside of bars included threats to their physical safety and their public image. For women living near bars, increased smoking on the street may have increased their exposure to secondhand smoke and disruptive noise. For some women, however, unanticipated negative consequences were identified with noncompliant bars. Smokers were conjectured to congregate in the smaller number of bars where smoking was still allowed, resulting in increased exposure to secondhand smoke for low-SES women working in these bars. A common positive unintended consequence of the tobacco control ordinance was increased social circulation and solidarity, as smokers gathered outside bars to smoke.
Smoke-free workplace laws in bars can have both negative and positive consequences for workers and smokers, and low-income women in particular.
Social scientists throughout history, including Durkheim1 and Weber,2 have analyzed the ways in which laws and policies intended to alter human behavior often produce unanticipated results. Such results are due partly to the complexity of social systems, as well as to cumulative individual reactions to policy goals not shared by all participants in those systems. Laws requiring workplaces, including bars and restaurants, to be free from tobacco smoke are increasingly popular in communities and states throughout the world.3–5 Inequality in the execution of such laws may result in less thorough compliance and protection for less powerful groups, such as women of low SES.6–8 A recent example of an unintended consequence of a smoke-free tavern law is an unexpected rise in drunk-driving incidents that may, in part, result from driving greater distances to reach bars in which drivers can both smoke and drink.9
Although social science research on the unintended consequences of public policies has tended to highlight negative consequences,10–17 sociologist Robert Merton noted that unexpected policy consequences also may be positive.18 Using thematic analysis of qualitative data from three mixed-method studies of tobacco use in and around bars where indoor smoking is prohibited, this article illustrates both negative and positive unanticipated consequences of smoke-free workplace policies in California bars for low-SES women.
The findings presented in this paper are drawn from three mixed-method studies conducted between 2001 and 2007 in California, where smoking inside bars was legally prohibited as of January 1998 by California Assembly Bill 13 (AB 13). The extension of the previously limited smoke-free workplace law to bars and restaurants (with the loophole of small owner-operated establishments) was enshrined in California Labor Code Sec. 6404.5, requiring patrons and workers to exit enclosed spaces in order to smoke tobacco products. Bars constitute an important social space in which to study smoke-free workplace laws, because of the strong association between alcohol and tobacco consumption,19 even for those who consider themselves nonsmokers but occasionally smoke cigarettes while drinking in such settings.20 The purpose of all three studies was to characterize bars in which indoor smoking continued despite the law, and to assess the reasons for this, giving particular attention to the social–environmental features of bars which may facilitate or inhibit compliance with the law. The studies differed primarily in locale and setting—the first was set in the city/county of San Francisco, the second compared a subset of San Francisco bars with a similar set in Los Angeles County, and the third looked at bars in the large Northern California county of Alameda. All three studies focused on stand-alone bars (those not attached to hotels or restaurants), and incorporated qualitative and quantitative data collected through structured observations of each bar in the sample, as well as data collected through semi-structured interviews with bar patrons and staff.8,21–24 The procedures used in the studies were reviewed and approved by the IRB of the Pacific Institute for Research and Evaluation.
Bars were selected for the three study samples from a universe of bars that researchers compiled for each county from California Alcohol Beverage Control data, listings in local print and Internet bar guides, phone books, and newspapers targeting various ethnic groups.8,22,23 As detailed in the following paragraph, after the observations were conducted in the sampled bars, an opportunistic sampling strategy was employed for the interview phase of each study. A random sample of bar staff and patrons might include many people who would not have the breadth of experience in the bars necessary to provide in-depth quality information about the bar; however, a sample too narrowly defined by a specific number of years of experience was not practical due to time constraints and the ability of field interviewers to ascertain such information about bar staff and patrons. On the other hand, the close-knit nature of bar society meant that bartenders and regular patrons knew each other and in many cases had contact information for each other, facilitating recruitment. Thus bar staff were screened prior to interviews to ensure that they were not new bar employees, and bar patrons were screened to ensure that they had attended the bar more than once or twice. On completion of the interviews, respondents were asked to recommend other staff or patrons they knew who met these criteria.
The first study (Study 1) focused on a random sample of 121 bars in San Francisco, including four observations per bar and 47 semi-structured interviews (with ten female and 11 male bartenders, four female and seven male owner/managers, and two female and 13 male patrons). Following up on findings in Study 1 related to differential policy compliance in bars catering to patrons of particular ethnicities, the second study (Study 2) focused on 165 bars in San Francisco and Los Angeles that primarily served Asian, Latino, or Irish patrons, and included three observations per bar as well as 79 interviews (with 18 female and ten male bartenders, seven female and ten male owner/managers, and 11 female and 23 male patrons). The third study (Study 3) collected four observations per bar and 81 bar interviews (with 16 female and 15 male bartenders, nine female and 11 male owner/managers, and 11 female and 23 male patrons) from a random sample of 104 bars in Alameda County, which encompasses the city of Oakland and smaller cities, suburbs, and rural areas.
Although the interview instrument included questions about educational experience, it did not include income levels. The female bartenders in these studies were classified as low-SES based on 2000 Census data for full-time female bartenders in California, including mean income ($17,527) and educational levels (<10% have college degrees).25 Few of the female bartenders interviewed reported going to college.
In all three studies, pairs of research assistants conducted hour-long structured, naturalistic observations in the selected bars. Each bar was visited at least three times on different nights of the week and at different times of the night. Observers were trained in observational techniques as well as in ways to keep their interactions with bar patrons and staff brief and unobtrusive. Immediately following the observations, the field observers filled out survey instruments using handheld computers and recorded field notes which they then used to prepare written narratives about the bars and their observations. Observers described bar, staff, and patron characteristics, particularly focusing on tobacco-related evidence (e.g., ashtrays, smoke, cigarette butts, and advertising) and interactions (including asking others for cigarettes, confrontations between staff and patrons about smoking, and discussions of the smoke-free bar law). To facilitate data management and analysis, these narratives were semi-structured (i.e., the observers were provided an outline for their narratives with separate fields for detailed information about the bar setting, patrons, staff, social interactions, and smoking behaviors as well as an open comment field).
After the observation phase of each study was completed, trained interviewers conducted semi-structured interviews with male and female bar staff, patrons, and owner/managers to provide a more complete understanding of the contexts in which people encounter the smoke-free workplace policy. The interviews took 40–60 minutes to complete. Each respondent received a $40 stipend for participating. For Study 2, focusing on bars with particular ethnic identities, the cultural identities of observers matched those of bar patrons whenever possible, and several interviews were conducted in Spanish, Korean, Cantonese, or Mandarin by interviewers fluent in the relevant languages. All the interviews were transcribed and, where necessary, translated into English.
Although unintended consequences of the smoke-free workplace law for low-SES women was not a central research question for any of the three studies, the open-ended nature of the qualitative data collection components for these studies allowed respondents in interviews and field observers in their narrative notes to discuss and elaborate on circumstances, events, and situations relevant to these topics. For example, the field observers were required to describe: general conditions and atmosphere inside and around the bars; patrons and staff, including gender, approximate ages, and estimated SES; and observed relationships between patrons and staff as well as observed smoking behaviors within and immediately outside the bars. Relevant questions from the interview guides included:
Observation narratives and interview transcripts were coded using ATLAS.ti 5.0, a software package for qualitative data analysis. The texts were initially coded by indexing for content and broadly emergent themes. Further coding was then conducted, concentrating on respondent descriptions of women in and around bars. The findings presented here are derived from interviews with female bartenders and are further reinforced by other bar patrons’ observations and contextualized by observer narratives of bars with female bar staff that observers characterized as having a predominantly low-SES clientele or as situated in low-SES neighborhoods. Coding took place throughout the time of data collection and thereafter, but the final analytical process for this paper was conducted in 2008. Themes that emerged from this content analysis are illustrated below with quotations from interviews and narrative observations. Quotations from interviews conducted in Spanish, Korean, Cantonese, or Mandarin are presented in their English translations.
Complying with the provisions of California’s smoke-free bar law, those who wished to smoke were obliged to leave the bar interiors. Although a few bars were able to provide patios or other controlled outdoor spaces, for most bars the smoking area became the section of the street immediately in front of the bar—generally at the front door. This meant that many bar patrons and staff were spending some part of their time outside the bar, on the street. In some bars, indoor smoking continued despite the law.8,23 Previous research found that low-SES women experienced unintended consequences of the smoke-free workplace policy both in and around bars.8,23
For many bars in this team’s studies, the environment outside the bar, although rarely as pleasant as inside, was an acceptable place to spend the 10–15 minutes or so needed to smoke a cigarette. The community of smokers could even create an agreeably sociable environment. Some respondents saw smoking outside as a source for social networking and solidarity for both men and women. One male bar patron observed:
You have the smoking circle outside, sort of like we had in high school. We had a circle in the middle of the school yard where the smokers could smoke. They band together because they are smokers. It’s just like the old movies in the 40s. You lit a cigarette, shared a cigarette with a beautiful woman, so you’ve got something to talk about. Some female bartenders reinforced this theme, such as one who said, “I kind of prefer to go outside anyway and kinda have a chat and whatever, like whoever’s going out, have a chat with them.”
For many other bars, however, the situation on the street raised safety concerns, particularly for women. Field observers reported that the environment outside of some bars felt particularly risky or unsafe. One observer described a “tense” block populated with “gangsters” who “stayed outside of the bar”:
Patrons are, for the most part, Salvadorian, working-class, Spanish-speaking males. Merengue, salsa, and pop in Spanish played from the jukebox and for a Saturday night, the environment was surprisingly calm. Most clients sat at the bar having conversations while drinking beer or playing pool. The environment inside the bar contrasted with the tense and unsafe feel of the block where the bar is located. Outside a group of Central-American street gangsters hang out taking a great deal of space, pacing up and down, talking to friends inside another business or stopping by in cars. However, this group of mostly late teens and early twenties males stayed outside of the bar.
Observers similarly contrasted the “calm” inside other bars with “unsafe” environments outside, where one might encounter, for example, “a tall white male dressed in black [who] was unable to walk straight because of his level of intoxication and moved slowly, holding a large metal stick, which he swung in a threatening manner as he staggered around,” and bartenders recounted incidents of violence that had happened outside their bars, such as one female bartender who told of seeing a patron “argue with his sister outside. He had a cast on. He slapped her in the head with the cast. So, I already knew that he didn’t have a problem hitting a woman.” Regarding people outside the bar who could be perceived as dangerous, one female bartender remarked that “You see them hanging out at liquor stores, outside of bars. If you’re a single lady, young lady, you don’t want to come through if some people are standing right outside the door.” Maintaining the bar space as a safe haven in a troubled neighborhood could inspire what one observer described as a “defensive stance” among bar patrons that was “part and parcel of being a safe, welcoming bar in [one working class] neighborhood.” Going outside to smoke potentially put women in harm’s way.
Regardless of neighborhood, the space immediately outside the bar could be a dangerous zone for women. As one female bartender noted, “Being a woman in a bar by yourself, you’re an easy mark—an easy target.” This was particularly the case for female bartenders. A primary function of the bartender, besides providing alcohol, is monitoring the atmosphere within bars and upholding bar policies, which may include ejecting or “86ing” patrons for being rude, unruly, or refusing to stop smoking inside.22 This role was particularly difficult for many female bartenders, whose authority was frequently challenged by male patrons.8 Patrons who got out of line or became rowdy were sent outside where smokers were also expected to congregate, and where different social norms may have prevailed. Female bartenders who wished to smoke were forced by the new law to leave the relative sanctuary of the bar and risk encountering patrons they had ejected. Although all bartenders risk running into people they have ejected from their bars on the sidewalk outside, both male and female study participants noted that this was particularly challenging for women.
Additionally, for many low-SES women who lived in the vicinity of bars, the smoke-free law may have had the unintended consequence of creating unsafe, or at least unpleasant, conditions around their homes. Smokers congregating outside bars may have caused “problems with the neighbors” as one female bartender remarked, particularly when, as another female bartender commented, neighbors lived “very, very close behind the bar.” She added that, “If you come in the back door you have to actually walk up their driveway to get into our back door. They complained a lot about noise, so we had to stop letting people go out there. The police were at the bar a lot because of the noise. Their baby was awakened by the noise.”
In addition to safety concerns, women smoking outside of bars could encounter issues with cultural expectations about how women should conduct themselves. One male bartender depicted the situation:
Here’s a hypothetical. An upscale woman, a prominent woman comes into your establishment, nice fur coat on, has her jewelry on, looking good. She wants to sit down and relax and have a drink and you have to say, “You have to go outside and smoke.” Now, here I am, a so-called high-class woman with all my jewelry on and fur and I’ve got to go outside and stand around people and smoke. First of all, she ain’t going to want to do it. If she’s a real lady, she’s not going to want to do that. Second of all, she’s going to be thinking about what if somebody hits her in the head while she’s out there smoking, what can she do, you know, snatch her purse, grab her coat, you know, something like that.
Respondents from various cultural backgrounds suggested that smoking on the street was particularly problematic for women—from a female bartender at a Korean bar who reported that “It’s still something where people will stare at a Korean woman smoking,” to the male owner of an Irish bar who suggested that smoking on the street can be hard “to deal with” for Irish women who are culturally “more traditional”:
In Ireland, you’d never seen somebody walking on the street with a cigarette in their hand. It was considered to be rude to smoke outside. So, people generally only smoke in coffee shops or in bars. In fact, [the smoke-free law] was one of the hardest things that my mother had to deal with, especially because she was more traditional. To be seen [outside] smoking a cigarette was really unladylike for her.
Smoking inside bars was generally perceived to be less problematic for women than smoking outside. As another female bartender said, “You look worse smoking in the street,” adding that “You don’t want to set a bad example for your kids. You look worse in the street than inside.” Smoking patios and “smoking rooms” established by bars attempting to circumvent the law also provided some protection to a woman’s image and reputation. “You don’t have to actually go outside outside, ‘cause there’s a smoking room outside” (San Francisco). Yet very few bars observed in this series of studies were able to maintain such spaces.
Uneven implementation and enforcement of the law have been shown to result in inequities in exposure to the workplace hazards the law was intended to prevent, specifically secondhand smoke (SHS).8 This situation may have resulted in some low-SES women being at risk for increased exposure to environmental tobacco smoke. Bar observers commented on the “competition for bars,” and noted that some smoking bars might be more heavily populated than they would be “if smoking were disallowed.” Others observed that smoking was “one of the draws” of the bars where it was still allowed. One female manager of an Oakland bar recalled that “Berkeley had a no-smoking ban in place before 1998. So everybody in Berkeley would come here to smoke and drink and carry on,” adding that in situations where “some of the other bars are still letting people smoke…people will go where they can smoke.”
Much research has indicated that important consequences of smoke-free workplace laws are increased quitting and quit attempts among smokers, due to the challenges of smoking as well as concomitant increases in the perceived stigma of smoking.26–28 Interview data from these respondents reinforced these findings to a degree. One female bar owner noted, “Of course [smoking restrictions] have an effect. Those who can quit will; that’s what will happen. But the people who like to smoke, they’ll keep smoking. “ One female bartender observed that, “There are a few people like that [quitting], but not a lot.” This was particularly the case for the many low-SES women interviewed who worked in bars where smoking continued despite the law. A female bartender stated that, “If everybody smoked inside there, if I’m smelling it so much, then I’m going to try it, and then I’m going to like it.” Yet even in bars where the smoking law was upheld, the pressure to quit smoking was perhaps mitigated by another consequence of the law which is clearly unintended: the creation of the practice of smoke breaks, where an employee might legitimately be allowed to take a break to go outside to smoke. This resulted, at least in one case, in a female bartender taking up smoking. As she reported, she smoked “only because of this place—it’s an excuse to leave, y’know.”
Unanticipated consequences of California’s smoke-free workplace law in this study primarily occurred when bars did comply with the law and smokers went outside the bar to smoke, particularly when smokers stood on the street outside the bar. Although many of the negative consequences also apply to men, they were more pronounced for women who smoked outside of bars. They included threats to women’s physical safety and their public image of propriety,29,30 and added to the cumulative risk that these low-income women might already face from such factors as unstable housing, single motherhood, and lack of resources. For women living near bars, increased smoking on the street may have increased their exposure to SHS and disruptive noise. For some women, however, unanticipated negative consequences were identified with noncompliant bars. Many bar patrons noted that smokers congregated in a smaller number of bars where owners still tacitly allowed indoor smoking. Through a process of assortative distribution,31 low-SES women working in and frequenting these bars may have been exposed to increased SHS.8
For smokers, a very common unintended positive consequence of the tobacco control ordinance was increased social circulation and solidarity, as they went outside to smoke. A previous paper discussed the implications of solidarity among bar patrons as well as between bar patrons and staff as both a possible hindrance to the implementation of tobacco control policies in bars and as a possible support for the implementation of such policies: bar patrons may use their collective strength to resist management’s effort to comply with the law, but alternately bar staff may use this solidarity to leverage compliance.22 The effects of bar patron solidarity on smoking cessation or uptake are as yet unexplored. Previous studies on social cuing32 and the disinhibiting effects of alcohol33 may mean that in drinking settings where smoke is present, nonsmokers are more prone to occasional smoking, and smokers may “binge” on cigarettes.20,34 No research has been conducted on the effects on smoking from the unintended consequence of smokers clustering outside bars, yet the findings from this study indicate that just as smoky bars may function as “smoke-easies” where smokers congregate and where smoking behaviors may be reinforced,35 these doorway gatherings may function as “smoking settings” wherein smoking may well be supported and efforts to quit suppressed. The role of peers in establishing and reinforcing tobacco use has been well established for specific populations, including adolescents36,37 and ethnic minorities.38,39 Further research may be called for to investigate this effect specifically among the subgroup of smokers in bars.
Additionally and consistent with reports elsewhere,32 respondents reported quitting and attempting to quit in response to the limited smoking opportunities and increasing stigma attached to smoking following enactment of the law. The counterexample, however, of a female bartender who took up smoking to justify taking more breaks indicates the complexity of the issue. The lure of smoking to justify an outdoor break has been described for other low-SES occupations, including nurses40,41 and military workers.42 Although encouraging tobacco cessation may have been a covert intention of the tobacco control advocates who supported the smoke-free workplace law, the stated intent of the law was to protect worker health, cast within the framework of fair employment practices (workers should not have to choose between working in an unhealthy environment and quitting work, analogous to laws regarding sexual harassment). Although our description of quitting smoking as an unintended consequence may therefore be arguable depending on how one views this aspect of the law, clearly no one intended for the law to encourage workers to take up smoking.
Although there is a lengthy tradition in the social sciences of pointing out the unintended repercussions of policy, especially for less powerful social groups, recommendations for tying these outcomes to specific recommendations for policy formulation have been less common. This study’s findings suggest that as relatively weak provisions for enforcement of tobacco-free laws in bars enable some recalcitrant bar owners to permit smoking, low-SES women working in such bars may still be subjected to substantial SHS exposure.8 Therefore, implications for policymakers are clear: to be fair to women, smoke-free workplace laws need to be enacted with widely understood and readily enforceable repercussions for violations (e.g., substantial fines).24,43 The gender-neutral framing of strong smoke-free workplace policy will support level playing fields in which all employees (female and male) are not obliged to breathe tobacco smoke in order to work.
For individual bar owners, the findings of this study suggest that creating safe outdoor enclaves when possible will facilitate compliance with the smoke-free workplace policy while minimizing threats of harm to the physical safety and culturally salient reputation of women who smoke outside bars. Paradoxically, however, because an unexpected consequence of compliance with smoke-free workplace laws is social solidarity and sociability among women of low SES who smoke, further research is called for regarding how such solidarity may in fact inhibit their attempts to quit.
Research for and preparation of this manuscript were supported by NCI Grant #R01-CA100772 and California Tobacco-Related Disease Research Program Grants 10RT-0276 and 12RT-0116. The authors wish to express their appreciation to the interviewed bar employees and patrons who generously shared their insights with us. We also would like to thank Tamar Antin, Scott Martin, Travis Satterlund, and the anonymous reviewers for their thoughtful editorial suggestions.
No financial disclosures were reported by the authors of this paper.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.