Alcohol policy making at the local level: complex processes in multiple contexts.
Alcohol policy can be defined as measures put in place to control supply and/or affect demand for alcoholic beverages in a population (Babor, 2002). Governing agencies at all levels--international, national, provincial, regional, and local--make decisions on alcohol policies (Babor et al., 2010). Alcohol policy making at the local level is of increasing importance (Giesbrecht, 2007; Holmila & Warpenius, 2007) and several effective policy measures are decided on and enforced at the local level (Treno, Marzell, Gruenewald, & Holder, 2014). The efficacy and effectiveness of alcohol policies at the local level have been assessed in many studies (Babor et al., 2010; Treno et al., 2014). In particular, policies that regulate the availability of alcohol in terms of outlet density and trading hours have been shown to be effective in reducing drinking and related harm (Babor et al., 2010; Campbell et al., 2009; Hahn et al., 2010; Popova, Giesbrecht, Bekmuradov, & Patra, 2009). Therefore, it is noteworthy that so far, little is known about policy-making processes at the local level (Maclennan, Kypri, Room, & Langley, 2013). The actors involved in these processes may often have conflicting interests. Public health and safety interests may be in conflict with the interests of the industry (Babor et al., 2010). Improved understanding of conflicting interests and values in these processes is important for improved political debate and decisions.
Our focus is on local alcohol policy making in Norway, where local government has a long tradition in alcohol policy issues (Andersen, 2000; Horverak, 1979). Here, decisions on trading hours and outlet density are made by authorities at the municipality level. Therefore, local authorities have the power to exercise policy measures that have been shown in international research literature to be effective in preventing alcohol-related harm (Babor et al., 2010; Campbell et al., 2009; Popova et al., 2009; Stockwell & Chikritzhs, 2009). Similar political responsibility at the local level is also found in other European countries such as Belgium, France, and Sweden (Osterberg & Karlsson, 2002). In this study, we looked at closing hours for on-premise alcohol sales. This may serve as an example to illustrate the alcohol policy-making process.
In Norway, local decisions on closing hours are limited by the national maximum closing hours, until 3.00 am. Despite this somewhat limited political scope of action, closing hours for on-premise alcohol sales seem to have been subject to more changes, more political debate, and more media attention over the past 10-15 years than any other part of local alcohol policy making in Norway. Electronic searches in print media in Norway in 2012 showed that one in three articles mentioning "alcohol" were about on-premise sales, and half of these were about on-premise closing hours at night. Scheffels and Lund showed that over the period from 1988 to 2007, articles in Norwegian print media about on-premise sales of alcohol and regulation of these increased markedly (Scheffels & Lund, 2009). Moreover, on-premise closing hours in Norwegian cities were frequently subject to changes in both directions between 2000 and 2010 (Rossow & Norstrom, 2012). On this basis, it seems suitable to examine this specific alcohol policy measure with respect to local alcohol policy making.
A striking observation is that the changes in on-premise closing hours in Norwegian cities in the first decade of the 2000s occurred in two "waves." Extensions of closing hours occurred in 2004, whereas restrictions, which were more numerous, mostly occurred 4 years later, in 2008. Previous analyses have demonstrated that these changes had significant consequences in several respects. Extended closing hours led to increased violence at night time in the city centers (Rossow & Norstrom, 2012) and to increased turnover in bars and pubs (Melberg & Schoyen, 2012), whereas reduced closing hours had the opposite effects, that is, reduced violence and reduced turnover (Melberg & Schoyen, 2012; Rossow & Norstrom, 2012). In this article, we address the processes that underlay local decisions on these important changes in on-premise sales hours in Norwegian cities and try to explain how these policy changes have come about. More specifically, we explore which actors, interests, arguments, and values were important in the processes that led to the policy decisions, how they interacted, and how this may have changed from the liberalization wave in 2004 to the restriction wave 4 years later.
We applied Kingdon's policy streams approach to analyze policy-making processes (Kingdon, 1995). In a later review of theoretical approaches to understanding policy making, John emphasized the importance of Kingdon's approach because it assumes continual policy change (John, 1998). This seems particularly suitable as a guide to analyze the processes underlying decisions on on-premise closing hours in Norway, as these are found to change repeatedly (Rossow & Norstrom, 2012). A previous study of alcohol policy making at the local level has also applied this theoretical approach (Maclennan et al., 2013).
In Kingdon's approach, policy formation is regarded as the result of the following three interacting sets of processes (or streams): problems, policies, and politics (John, 1998; Kingdon, 1995). Problems are public matters that require attention, whether they get defined as important or not. Policies are proposals for change based on the knowledge and development of interest among actors in a policy sector. Such proposals are provided by "policy entrepreneurs" who mobilize opinion and institutions. These entrepreneurs can be politicians, bureaucrats, consultants, journalists, and academics. The third process, or stream, in Kingdon's approach comprises the political processes, such as election results and swings in the popular mood. These influence how the media and other opinion formers define problems and evaluate potential solutions (John, 1998; Kingdon, 1995). Interaction between these streams or processes may lead to a policy change. In Kingdon's approach, the opening of a "policy window" is also essential, that is, the policy change is dependent on timeliness and political opportunity (John, 1998; Kingdon, 1995).
Data and Method
The data collection followed a systematic scheme with respect to selection and limitation. We first identified the 30 largest cities in Norway (by population size in 2010), contacted the city administration in each city, and obtained information about whether closing hours for on-premise licenses had changed in the period 2001-2010. In Norway, local government elections are held every 4 years. Two elections were held in the first decade in this century, that is, in 2003 and 2007. Consequently, a 10-year time window increased the likelihood of changes in election results and swings in the popular mood. Moreover, policy entrepreneurs and their agendas and priorities are likely to change over a decade. Problems may be perceived differently by various actors. Over a 10-year period, both policies and what is regarded as an adequate solution to a problem, are more likely to change. Thus, this time window provided more opportunities for changes in all three policy streams in Kingdon's approach. City administration officers in 24 of the 30 largest cities in Norway reported that there had been a change in on-premise closing hours at least once during the period. These 24 cities represent a variety in terms of population size, main industries, and political composition of the city council, implying multiple contexts for local alcohol policy making. We obtained information from these 24 cities about the processes behind the policy changes from three different data sources; electronically available editorial articles from local newspapers in all cities for relevant periods, proposals to and minutes from local government meetings with on-premise sales hours on the agenda, and interviews with local bureaucrats. These vary in importance. The first two are the most important, while the interviews constitute a minor additional data source. They are all described in more detail in the following.
We identified local newspapers for the 24 cities from a website that gives information about all Norwegian newspapers available on the Internet and where they are published (www.norske-aviser.com). We conducted relevance sampling of articles (Krippendorff, 2013) using free text searches using "on-premise trading hour*" ("skjenketid*") in the newspapers' electronic archives. We included only news articles and newspaper editorials. We restricted the publishing date to the calendar year in which the decision was made and the previous calendar year. We included only news articles and editorials of relevance for the policy-making process in the city, so articles pertaining to neighboring municipalities were excluded. Issues involving a single or a few on-premise licenses were excluded, as our focus was on policy change rather than administrative handling of single cases.
In one city, the local newspaper was not available on the Internet, and in two other cities, there were no relevant articles covering the period when the closing hours changed (one city in 2004 and one in 2008). Therefore, relevant newspaper articles were obtained for 21 of the 24 cities. Among these 21 cities, 6 cities had changed closing hours twice during the decade. For four of these, relevant articles that covered both changes were obtained, whereas in two cities relevant articles covering only the most recent change were available (Table 1). The media articles from the 21 cities covered 26 changes in closing hours, that is, 19 restrictions and 7 extensions, and they amounted to a total of 280 articles and editorials. Each newspaper article often contained interviews with several stakeholders, for instance politicians (often from various political parties), representatives of the hospitality industry, and police officers.
Documents from the city councils were first retrieved from each city's Internet home page, which are individually organized. In this way, we obtained documents containing proposals and minutes from the city council meeting where the decision was made in 15 of the 24 cities. In 5 of these cities, these documents also contained the written proposals and in 10 cities the proposals and minutes were separate documents. For the remaining nine cities, we contacted archival personnel by e-mail and telephone, and we obtained the relevant documents from the city council meetings and preceding procedures in seven of these cities. We did not succeed in obtaining these documents from the two remaining cities, despite repeated reminders.
Finally, we conducted brief interviews with key informants in the cities' administration. In each city, we identified and contacted a bureaucrat in the local administration who worked with on-premise sales. We asked these bureaucrats if we could interview them. The interviews were conducted in 19 of the 24 cities in 2011 as part of a related study on possible effects of changes in closing hours on violence (Rossow & Norstrom, 2012). They were undertaken as telephone interviews and included questions about what the changes were, how many bars and pubs were affected, and an open question about reasons for why the policy change occurred. The answers to these questions were used as additional information in our analyses.
Most of our data (the newspaper articles and the city council documents) were obtained from documents that were produced without researcher intervention (Silverman, 2001). We applied a content analysis of the text data through a classification process of coding and identification of patterns (Hsieh & Shannon, 2005). The analyses were governed by our theoretical approach and were conducted in three steps. First, the actual policy changes were described with respect to content and context. These descriptions were mainly based on the documents from the city council meetings and supplemented by the media articles and the interviews with informants in the city council administration. Second, the data were explored and categorized into a stream according to Kingdon's approach (Kingdon, 1995; i.e., a problem, a policy solution to a problem, or a political process), if applicable. We explored patterns in the data with respect to whether and in what ways the three streams could be identified and which actors or stakeholders forwarded these views. We explored whether there was consistency or dominance in the data both across cities and over time and whether there was heterogeneity in these respects. The main data source for these analyses was the media articles, but city council documents and interviews with city council administrators provided some additional information.
Finally, and again with reference to Kingdon's approach, we explored possible interactions between the three streams that could explain the direction of the policy change, that is, whether an extension or a reduction in closing hours was the outcome. We also explored whether these possible interactions could explain the magnitude of the policy change, that is, how many hours at night the closing hours had been extended or reduced. The main data source for these analyses was also the media articles, and city council documents provided some additional information.
The Policy Window and the Formal Political Process
In Kingdon's approach, policy change is dependent on timeliness and political opportunity, that is, when the policy window is open (John, 1998; Kingdon, 1995). In almost all cases of changes in closing hours examined in this study, these changes were undertaken as part of a revision of the city's regulation of on-premise and off-premise alcohol sales, or as part of the city's alcohol policy plan, which is more comprehensive. Generally, these regulations or plans are subject to revision in an early phase of a city council's 4-year period (Baklien & Krogh, 2011), and these revisions are therefore bound to be on the city council's agenda fairly shortly after the election. Consequently, almost all changes occurred in 2004 or in 2008. Thus, in the case of changes in alcohol sales hours at the local level in Norway, the policy window opens regularly and predictably.
The revision of local regulations or plans regarding alcohol sales (and other alcohol policies) usually undergoes a formal political process involving various bodies at the administrative and political levels in the city (Baklien & Krogh, 2011) and this was also observed here. We do not have complete information about all stages in the formal processes preceding each decision on a change in closing hours, but we have observed the following. In most cases, the revision was first drafted by a committee consisting of local politicians and this was often subject to consultative statements from various stakeholders, for instance, the hospitality industry, police, and health and social services. In all cases, the revision was dealt with at the administrative level in the city, and in some cases also by the executive committee of the city council before finally being dealt with by the city council.
When on-premise closing hours was an item on the city council's agenda, the administrative staff had presented a prepared and written proposal on the matter to the city council. In addition, political party representatives in the city council often presented alternative or additional proposals to be voted on. For instance, when the city council in Bergen dealt with the proposal of altered alcohol sales regulations in 2008, six politicians presented a total of 27 alternative or additional proposals to the initial proposal to be voted on, and 10 of these alternative/additional proposals pertained to changes in on-premise closing hours. Furthermore, the minutes from the city council meetings showed that the proposed change in closing hours was generally one issue among a long list of issues on the agenda. In many cases, the proposed change in on-premise (and off-premise) closing hours was only part of a proposed alcohol policy plan that was dealt with by the city council meeting. The policy decisions on on-premise closing hours were therefore made in a demanding context. Many proposals on closing hours were to be assessed, and they were only a small part of a comprehensive meeting agenda.
Description of the Changes
Of the 29 changes in closing hours that we examined, 20 were restrictions and 9 were extensions. The changes occurred mostly 6 to 10 months after the election of a new city council. All extensions occurred in 2004 and most of the restrictions occurred in 2008. Thus, a liberalization wave in 2004 was succeeded by a restriction wave when the policy window reopened 4 years later (Table 1). In five cities, closing hours were first extended and then restricted. Consequently, the restriction wave mostly included cities that were not part of the previous liberalization wave and only a fraction of those that were.
The changes in closing hours were small; in 8 cases the change was 30 min, in 16 cases 1.0 hr, and in 4 cases 1.5 hr (Table 1). In some cases, the city council documents showed that the changes were rather complex. They differed according to the type and location and/or age limit regulations, beverage type, weekday, and season. However, a more common picture was that the closing hours at night applied to all beverages and all on-premise licenses. Table 1 presents the various main exceptions to the overall changes.
Assessment of the "Streams" in Kingdon's approach
In the media articles prior to the extensions in 2004, problems related to on-premise closing hours were very seldom mentioned. The city council documents showed that in most cases the police had either stated that they did not believe that extended hours would cause a problem or they did not give any statements in the preceding procedures. But concerns about the economic prospects and competitive conditions of the hospitality industry were expressed in the media in most of these cities, voiced mainly by representatives of the industry but also by politicians and bureaucrats. These concerns were generally also reflected in the proposals to the city council.
While alcohol-related violence, heavy drinking, and nuisance received little attention in the media in 2003-2004, these problems were most frequently noted as important issues that called for action and solution 4 years later, in 2007-2008. In particular, the problems were noted by the police. They regarded the problems caused by alcohol as an increasing challenge that required a disproportionate amount of police resources at the cost of fighting more serious crime. In January 2008, the chief police officer in one county drew the media's attention to the increase in violence and stated:
The police can no longer increase the economic and human resources that are spent on this (i.e., nuisance and violence). The consequences would then be that other important areas must be put further down the list of priorities. (County chief police officer 1, 2008)
Another chief police officer said to the media that alcohol-related violence was the biggest challenge for the police force. A newspaper illustrated the violence and nuisance burden to the police by giving a comprehensive summary of the police journal reports from a Saturday night, when overly drunk people, who were incapable of taking care of themselves or who fought or created public disorder, needed, and were offered, police assistance. Other articles presented victims of violence, suffering from health problems and other types of problem. The police have had a clear role to inform people about problems of violence and nuisance, particularly in the period before the restriction wave in 2008. For instance, in 2008, a local newspaper editor wrote:
In particular, it is appropriate to praise the police for their clear message. Based on their experience and statistics, they have told us where the violent incidents occur, when they must attend to violence, and which people commit the violence. The short version of this is that these are intoxicated young men who get into trouble with each other, at night time, at the weekend, and around bars and pubs. (Local newspaper editor 1, 2008)
Notably, the portrayal of the problems of violence and nuisance, and concern about this, were first and foremost presented in the media in 2007 and 2008, that is, just prior to the restriction wave. This fits into Kingdon's problems stream. The police were clearly the actor that most frequently and markedly informed people about this problem, and they did so both in the media and in their consultative statements to the city council. We also found that other actors were important in expressing their concern and call for action in the media in 2007-2008. These included many politicians from most political parties, national representatives of an NGO (Norwegian Policy Network on Alcohol and Drugs), and to some extent representatives of the health services, local bureaucrats, and the hospitality industry. In the latter case, while this problem has been acknowledged, it seems also to have been played down by the hospitality industry. For example, in 2008 an industry representative stated in a newspaper interview:
I do not believe that the violence has increased as shown in the statistics, but rather that we have been better at reporting episodes of violence. (Hospitality industry representative, 2008)
Before the restrictions around 2008, other themes of relevance that we noted in the media articles and city council documents were additional arguments for opposing suggested restrictions, such as less attractive and lively city centers, and trade leakage to neighboring cities. These were not stated directly as problems but noted as concerns in media debates about on-premise closing hours in consultative statements to the city council. They were most frequently noted by representatives of the hospitality industry but also by politicians and bureaucrats.
Proposals for a change in closing hours as a solution to or as a way of dealing with a problem belong to the policy stream in Kingdon's approach. Before the extensions were decided in 2004, clear arguments were seldom found in the media articles about how or why an extension would solve a problem. However, the importance of competitiveness for the hospitality industry was noted in the media in some of the cities, both by representatives of the industry and by politicians. Four years later, this picture changed markedly. The media presented numerous and diverse proposals for a policy change, indicating how or why the change could solve a problem. In particular, the police emerged as a strong and clear advocate for reduced closing hours and argued why this would be an adequate and effective solution to prevent violence and nuisance at night time in and around bars. In a newspaper interview, the chief police officer in Arendal stated:
For each hour on-premise closing hours are extended, people in the city center get more drunk. It is not the number of bars and pubs that worry us, but rather the fact that sales hours are so long. Private parties will usually start around 8 p.m. anyway, and the longer the sales hours, the later people will go into the city center. Consequently, people are more drunk when they come to the city center than would have been the case if sales hours were shorter. (City chief police officer, 2007)
The chief police officer in another county also asked the politicians to combat violence at the weekends by shortening closing hours and stated:
It is not a good thing that night life is so extensive until 3.00 a.m., when we see the results. I have said that 1.00 a.m. could be a suitable time. It is well documented that the longer the closing hours, the higher the level of conflict. (County chief police officer 1, 2008)
The police further argued that the expected effects of reduced closing hours on violence and nuisance would free police resources for other important tasks. In the media debates, these police recommendations were very often referred to by the politicians, who favored reduced closing hours. They were often referred to in the city council documents as important in the chief administrative officer's assessment of reduced closing hours. In line with this, our informants in the city council administrations stated that compliance with the suggestion of the police and a wish to reduce violence and nuisance were the main reasons for restricting closing hours.
However, reduced closing hours was also frequently dismissed as an adequate way to deal with violence and nuisance, and mostly so by representatives of the hospitality industry, who actively lobbied against reduced closing hours. In 2008, the general director of the Norwegian Hospitality Association stated in several local newspapers that his organization did not believe that reduced closing hours would result in less violence and nuisance, and in one interview he continued:
No, it will rather lead to more people drinking in uncontrolled conditions on the streets and in private parties. And this will not lead to less nuisance and commotion than controlled drinking in bars and pubs. (General director of the Norwegian Hospitality Association, 2008)
This refusal to accept an association between closing hours and violence/nuisance was also presented by the hospitality industry in their consultative statements to the city council. A similar line of reasoning was observed among some politicians from various parties, but most consistently among the Progress Party representatives. In 2007-2008, according to media articles and city council documents, some of the proposals for extensions that were forwarded were closely related to these arguments, although not adopted. In both periods, bar owners and some taxi drivers said to the media that they believed that extended closing hours would result in people leaving the city center at different times and thereby reduce violence and nuisance. One politician from the Conservative Party stated:
It is better to keep people as long as possible in the bars, rather than sending them home to private parties with cannabis and moonshine. (City council representative from the Conservative Party 1, 2003)
Political processes that were important for a change in closing hours, that is, the politics stream in Kingdon's approach, were of several kinds, but they were mostly seen in relation to the reduction wave around 2008. One of these was a joint action by the chief police officers in the four largest cities in Norway (Oslo, Bergen, Trondheim, and Stavanger). They sent a letter to all mayors and city council politicians in Norway, stating their concern about police resources spent on alcohol-related violence and suggesting that on-premise closing hours should be shortened. This joint action at the national level received much attention in the local newspapers, and it may have triggered and/or reinforced the local police authorities' presentation of the violence and nuisance problem in their own district, which again appeared to be an important factor in the restriction wave in 2008.
Another such process occurred in 2007/early 2008 when several mayors and cooperating bodies for cities within the same county took initiatives and worked to obtain common closing hours for neighboring cities. One of the goals was to avoid or reduce trade leakage. These initiatives were observed in the media in three counties and applied to nine smaller cities where closing hours were reduced. They were also reflected in the city council documents in six cities. From the newspaper articles, it seems that having an agreement or common understanding between the cities in a county about common and shorter closing hours at night led some politicians to change their view from opposing to supporting restriction.
Moreover, the media's attention to on-premise closing hours increased markedly from 2004 to 2008, as can be seen in Table 1. Whether this increase in media coverage reflected a "swing in popular mood," as suggested by Kingdon, or whether it reflected more complex dynamics, cannot be inferred from our data. Yet, the arguments concerning the problems of violence and nuisance presented in the media were often reflected in the politicians' lines of reasoning, and therefore it seems likely that the increased media coverage on this topic enhanced public debate and helped to clarify and give nuance to the arguments in the political debate.
The local political swings that are manifested in the local election results may have contributed to the policy changes, but the pattern is less clear in this respect. In some cases, the media reported that a policy change was part of political negotiations about several political issues or powerful positions. For instance, the Progress Party in Alesund, which came into power after the 2003 election, gave the position as mayor to another political party in exchange for extended on-premise closing hours. Moreover, in some cases, a change in the political composition of the city council resulted in a political majority for a policy change. However, among the seven largest political parties in Norway, only two parties have been fairly consistent and predictable in their views on on-premise closing hours, that is, the Christian Democrats favoring restrictions, and the Progress Party, mainly favoring maximum closing hours. In general, the political parties have allowed their representatives to be free to vote as they like in local alcohol policy matters. This may partly explain why certain political swings did not necessarily predict what the policy on closing hours would be.
In this context, it may be noted that the newspapers also reported on the grassroot movements, mainly in the form of Facebook groups, which were established and gained many members who protested against suggestions to reduce closing hours. For instance, in Tonsberg, several thousand persons had joined the group "On-premise closing hours to 3.00 am throughout the year. For us who want to decide for ourselves how long we may stay out at night." These movements, which invoked liberal ideas about individual liberty, gained media attention in several cities, but there were no indications in the media or in the city council documents that they were important for the outcome of the policy-making processes.
Explaining the Two Waves of Policy Changes
The liberalization wave in 2004 and the restriction wave around 2008 may be explained as the outcomes of interactions between the three "streams" in Kingdon's approach. Such interaction of factors in the processes underlying the political decisions implies complexity. The extensions of closing hours in 2004 occurred when there was political concern about the interests of the local hospitality industry, little emphasis on the problems of violence and nuisance in the public debate and consultative statements to the city councils, and little media attention about the topic (Table 1). Moreover, in 2003-2004, the city council documents showed that the police did not express their concern or objections to extended hours in the political process. The media articles showed that they were a much less visible and clear actor in the public policy debate compared to in the following years. In addition to the concern about local trade interests, liberal political values were also given as reasons for a more liberal policy on closing hours in 2003-2004. In 2004, a representative of the Conservative Party in Molde said:
Politicians in Molde must start showing confidence in their voters and citizens. We must depart from our strict local alcohol policy and trust citizens to regulate their use of alcohol themselves.... We cannot sit in the city hall and act as nannies for ordinary people. (City council representative from the Conservative Party 2, 2004)
Thus, under the liberalization wave, the interests of the industry and the liberal political arguments emphasizing individual rights and responsibilities were countered or challenged to a limited extent, and therefore they were probably more likely to be pushed through. Consequently, it seems likely that the liberalization wave occurred in the absence of strong actors providing clear counterarguments to extended closing hours.
By 2008, this picture had changed markedly. The problems of violence and nuisance were frequently noted in the public debate. The police more often commented on these problems and consistently favored reduced closing hours in the media and in consultative statements. The arguments of the police were more frequently supported by bureaucrats and by politicians from most political parties.
Media articles and city council documents showed that most of the political parties held different views on closing hours, both across cities and over time. But, from 2004 to 2008, the representatives of these parties became more likely to support reduced closing hours. The media articles suggested that this support was very often justified by concern with the problem of violence and the arguments of the police. Thus, it seems that the police have played an important role in paving the ground for the restriction wave around 2008. It seems likely that the increased emphasis in the public debate on the problems of violence and nuisance, and the stronger and consistent message from the police, contributed markedly to the change from the liberalization wave to the restriction wave.
Political processes also seem to have contributed to the restriction wave around 2008. The media conveyed that cooperation between administrative or political bodies that wanted common closing hours within the county countered the trade leakage argument in about half of the cities that reduced closing hours. Moreover, the initiative from the chief police officers in the four largest cities suggesting reduced closing hours may have supported the local police spokespersons in their initiatives and ways of reasoning. While liberal, individualistic values were conveyed by politicians and by Facebook groups in 2007-2008, there were also advocates for opposite values. A local newspaper editor stated in 2008:
The development over the past few years has shown that citizens are tired of people who get drunk, who fight, and commit vandalism. It is as always the few that cause damage at the expense of the majority, but these few cause a lot of it, particularly at the weekends. The debate on closing hours is very much about taking the city back from those who make it unpleasant. (Local newspaper editor 1, 2008)
Consequently, the values that were expressed in the public debate on closing hours were more diverse and conflicting under the restriction wave than under the liberalization wave.
The Magnitude of the Changes
So far we have mainly addressed factors of relevance to understanding the direction of change. The fact that the changes were generally small also deserves attention. On average, the extensions of closing hours were somewhat larger (mean = 1.11 hr) than the restrictions (mean = 0.90 hr). In all cases of extensions, closing hours were set to 3.00 a.m., that is, the maximum limit by national law. It is noteworthy that several city council representatives, mostly from the Progress Party, even asked for extensions beyond these national limits, preferably allowing bars to trade 24/7, although such extensions can only be obtained by a policy change at the national level.
Theoretically, the reductions in closing hours could have been large, but they were mostly 30 min or 1 hr. The reductions were often smaller than those suggested by the police and by the Christian Democrats, the political party that was most in favor of restrictions. The small changes were commented on in several newspaper editorials, particularly the 30-min restrictions, and they were deemed to be "symbolic actions" and without impact. When it seemed likely that the city council in Sarpsborg would decide on a 30-min restriction in closing hours, the editor of the local newspaper wrote:
However, the political helplessness that is demonstrated when the mayors (in Sarpsborg and the neighboring cities) agree on only half-an-hour restriction in closing hours, is the worst. Not only is it a puny result, but the difference from today's situation is so tiny that it will not be a real test of the theory of the police about the positive effects on violence at the weekend. Now there is the risk that conclusions may be drawn on false premises, and today's closing hours may just as well be kept. (Local newspaper editor 2, 2008)
In some cases a small restriction appeared to be the result of political negotiations aiming to obtain a compromise. In other cases a small change was proposed by the city councilor or the city council executive committee, probably indicating what was assumed to be politically feasible.
Although local alcohol policy making may have a significant impact on the health and safety of the population and on turnover in the hospitality industry, little is known about the processes underlying policy making. This study contributes to a meager literature. We found that extensions of closing hours occurred when the interests of the industry were countered by other interests and arguments but only to a limited extent. On the other hand, restrictions mainly occurred when there were strong and frequently conveyed concerns about the problems of violence and nuisance. Restrictions to curb these problems were often and consistently suggested by the police. These concerns and clear recommendations seemed to outweigh the interests of the industry. Another factor that is likely to have contributed to the restriction wave in 2008 is cooperating bodies aiming for common closing hours to avoid trade leakage. Thus, we found similarities in the complex processes underlying the policy changes across heterogeneous local contexts.
It was useful to apply Kingdon's approach in order to understand the processes underlying the direction of the changes. The three "streams" were clearly found in the data and their interactions provided reasonable explanations as to why the changes had occurred in a certain direction. A similar observation was also made in a study of alcohol policy development in New Zealand communities, where Maclennan and coworkers found that Kingdon's streams model applied in the rural and metropolitan communities (Maclennan et al., 2013). Kingdon's overarching approach also helped our understanding of alcohol policy making at the local level as a continuous dynamic process.
While there is an increasing literature on the effectiveness and efficacy of local alcohol policy measures on drinking and related harm, there are few studies of how local alcohol policy comes about and develops (Maclennan et al., 2013). While there is indeed some knowledge about political processes in response to interventions at the local level (Rossow, Storvoll, Baklien, & Pape, 2011; Toomey & Lenk, 2011), this is of little relevance in our context. It seems likely that the processes leading to policy decisions within the framework of a (research driven) intervention differ from those that occur without any such intervention. Among the previous studies that have addressed political processes in the absence of an intervention, there are similarities in the findings that are worth noting. Andersen explored the political processes underlying liberalization of alcohol availability in two Norwegian municipalities and found that political decisions were mainly motivated by concern with the interests of the hospitality industry (Andersen, 1997), a finding well in line with our observations regarding extensions of closing hours. The influence of stakeholders on the policy-making process, as found by Maclennan and coworkers (Maclennan et al., 2013), was also observed in our study. In particular, the police and the hospitality industry were important actors in this respect. McKee and coworkers compared American cities that had adopted policies to restrict high-strength beer sales to cities that only considered doing so and found that the opposition of the alcohol industry was an important factor in explaining the different outcomes (McKee et al., 2011). The important role of the hospitality industry was also notable in our study, yet its importance seemed to be relative or dependent on counteracting factors, such as the arguments conveyed by the police.
The alcohol policy arena can be described as a sphere of action for opposing views, contending groups and competing interests (Babor et al., 2010). Our results suggest that what characterized the processes were not only opposing views, for example, on the likely effects of reduced closing hours but also opposing ideologies among the politicians. Although the arguments for a restrictive policy included society's responsibility for the health and safety of citizens, the arguments for a liberal policy on closing hours included the value of individual freedom for bar patrons. Focus on public health and prevention can easily be regarded as unacceptable paternalism and as an infringement of individual freedom (Sulkunen, Rantala, & Maatta, 2004).
The opposing ideologies may also be interpreted within the framework of a legitimacy perspective, as there are different views on what the relationship between the will of the people and the politicians' attitudes should be (Andersen, 1997). One view is that politicians should reflect the will of the people, which in this context could legitimize a liberal policy on trading hours. Another view is that politicians should shape attitudes and opinions and be able to make unpopular decisions, such as restricting closing hours. Our findings also demonstrated that there were clearly competing interests in the policymaking process, most markedly between the hospitality industry and the police.
We found that the changes in closing hours were generally small. The extensions were all limited to the national maximum closing hours at night (3.00 a.m.), whereas the reductions were mostly the result of negotiations and compromises, and in principal they could have been larger. As for political decisions more generally, Lindblom noted that these are often made on the basis of relatively small adjustments to the existing situation and limited to short-term assessments (Lindblom, 1979). Even if the overall aims seem to be relatively unproblematic, at the same time they are distant, and can be experienced as unrealistic. Then the small measures that are possible to implement become dominant, though these measures may not be the ones that bring us closer to our aims.
Strengths and Limitations
The fairly large number of policy changes, and the number of cities where these changes occurred, provided a broad basis for assessment of clear and consistent patterns in policy-making processes within a specific policy area. Our use of various data sources also contributed in this respect. Our use of printed media articles provided concurrent statements from the policy-making process, which are likely to be more rich, nuanced and reliable than retrospective assessments.
The selection, coding, and analyses of the data were conducted by one author (I.R.) and the selection and coding were not subject to calibration. Another limitation is that we have only examined political processes that led to changes in policy. We do not know whether processes that have not led to any policy change differ from those that we have observed and therefore it is possible that processes which seem fairly similar may have different outcomes in terms of policy change. Finally, there may well be important factors in policy-making processes that are not explicitly communicated and not easily detectable. For instance, the politicians' interest in avoiding unpopular decisions and losing voters' support may well be of importance in the decision-making process but hidden in the public debate.
Previous studies have shown that changes in on-premise closing hours at the local level can affect violence, drunk driving, and disorderly behavior (Duailibi et al., 2007; Kypri, McElduff, & Miller, 2014; Popova et al., 2009; Ragnarsdottir, Kjartansdottir, & Davidsdottir, 2002; Rossow & Norstrom, 2012; Schofield & Denson, 2013a, 2013b). The present study has added to the sparse knowledge about the processes that underlie such changes. Our findings suggest that media attention to problems related to long trading hours and the role of the police in presenting arguments were important in the policymaking process. Therefore, proponents for a more restrictive local alcohol policy should consider strategies to influence these factors.
A main argument for leaving many effective alcohol policy measures to be administered at the local level is that many alcohol-related problems are best understood and dealt with locally. While decisions are made at the local level, they are often meant to fit in with national goals and policies. Policy makers at the national level make decisions about whether and how alcohol policy measures can be administered at the local level. A better understanding of how and why policy decisions at the local level come about is relevant to these decisions.
We have noted that knowledge is scarce regarding the political processes that precede alcohol policy decisions at the local level, and there are several directions for amending this scarcity. A comparison of processes leading to changes versus processes that do not lead to changes would provide further insight into why policy changes occur. Moreover, it is important to address whether the findings in this study apply to other contexts. Finally, the present study has addressed a single policy measure that is a frequent topic in public debates, and therefore in future studies it is important to assess whether our findings apply to other, less debated, alcohol policy measures.
We are most grateful for the assistance of Oystein Skjadaaen, Unni Krogh, and Jorunn Moen in obtaining some of the data for this study. We are also most grateful for the valuable comments and suggestions by our colleagues Kristin Buvik and Oystein Skjadaaen, and two anonymous reviewers for Contemporary Drug Problems, on an earlier version of this paper.
This article was prepared for the Contemporary Drug Problems Conference: "Complexity: Researching Alcohol and Other Drugs in a Multiple World."
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
This study was funded by the Norwegian Institute for Alcohol and Drug Research.
Andersen, J. (1997). Kommunal alkoholpolitikk i endring [Changing alcohol policy at the local level]. Nordic Studies on Alcohol and Drugs, 14, 273-289.
Andersen, J. (2000). Municipalities between the state and the people. In P. Sulkunen, C. Sutton, C. Tigerstedt, & K. Warpenius (Eds.), Broken spirits. Power and ideas in Nordic alcohol control (Vol. 39, pp. 157-184). Helsinki, Finland: NAD.
Babor, T. (2002). Linking science to policy. The role of international collaborative research. Alcohol Research and Health, 26, 66-74.
Babor, T., Caetano, R., Casswell, S., Edwards, G., Giesbrecht, N., Graham, K., ... Rossow, I. (2010). Alcohol: No ordinary commodity. Research and public policy. Second edition. Oxford, England: Oxford University Press.
Baklien, B., & Krogh, U. (2011). Prosessen, planen ogpolitikken. Rusmiddelpolitiske handlingsplaner (The process, the plan, and the policy. Alcohol and drug policy action plans). Oslo, Norway: Norwegian Institute for Alcohol and Drug Research.
Campbell, C. A., Hahn, R. A., Elder, R., Brewer, R., Chattopadhyay, S., Fielding, J.,... Middleton, J. C. (2009). The effectiveness of limiting alcohol outlet density as a means of reducing excessive alcohol consumption and alcohol-related harms. American Journal of Preventive Medicine, 37, 556-569.
Duailibi, S., Ponicki, W., Grube, J., Pinsky, I., Laranjeira, R., & Raw, M. (2007). The effect of restricting opening hours on alcohol-related violence. American Journal of Public Health, 97, 2276-2280. doi:10.2105/ajph.2006. 092684
Giesbrecht, N. (2007). Community-based prevention of alcohol problems: Addressing the challenges of increasing deregulation of alcohol. Substance Use & Misuse, 42, 1813-1834.
Hahn, R. A., Kuzara, J. L., Elder, R., Brewer, R., Chattopadhyay, S., Fielding, J., ... Lawrence, B. (2010). Effectiveness of policies restricting hours of alcohol sales in preventing excessive alcohol consumption and related harms. American Journal of Preventive Medicine, 39, 590-604.
Holmila, M., & Warpenius, K. (2007). A study on effectiveness of local alcohol policy: Challenges and solutions in the PAKKA project. Drugs: Education, Prevention, and Policy, 14, 529-541.
Horverak, O. (1979). Norsk alkoholpolitikk 1960-1975. En analyse av alkoholpolitiske virkemidler og deres virkninger (Norwegian alcohol policy 1960-1975. An analysis of alcohol policy measures and their effects). Oslo, Norway: Norwegian Institute for Alcohol Research.
Hsieh, H.-F., & Shannon, S. E. (2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15, 1277-1288.
John, P. (1998). Analysing public policy. London, England: Pinter.
Kingdon, J. W. (1995). Agendas, alternatives and public policies. Boston, MA: Addison-Wesley Educational Publishers.
Krippendorff, K. (2013). Content analysis. An introduction to its methodology (3rd ed.). Los Angeles, CA: SAGE.
Kypri, K., McElduff, P., & Miller, P. (2014). Restrictions in pub closing times and lockouts in Newcastle, Australia five years on. Drug and Alcohol Review, 33, 323-326. doi: 10.1111/dar.12123
Lindblom, C. E. (1979). Still muddling, not yet through. Public Administration Review, 39, 517-526.
Maclennan, B., Kypri, K., Room, R., & Langley, J. (2013). Local government alcohol policy development: Case studies in three New Zealand communities. Addiction, 108, 885-895.
McKee, P. A., Nelson, T. F., Toomey, T. L., Shimotsu, S. T., Hannan, P. J., & Jones-Webb, R. J. (2011). Adopting local alcohol policies: A case study of community efforts to regulate malt liquor sales. American Journal of Health Promotion, 26, e86-e94. doi:10.4278/ajhp.100615-QUAL-193
Melberg, H. O., & Schoyen, O. (2012). Hvordan pavirker redusert skjenketid omsetningen i utelivsbransjen? (What are the effects of restricted on-premise trading hours on turnover in the nighttime economy?) Oslo, Norway: Norwegian Institute for Alcohol and Drug Research.
Osterberg, E., & Karlsson, T. (Eds.). (2002). Alcohol policies in EU member states and Norway. A collection of country reports. Helsinki, Finland: Stakes.
Popova, S., Giesbrecht, N., Bekmuradov, D., & Patra, J. (2009). Hours and days of sale and density of alcohol outlets: Impacts on Alcohol Consumption and Damage: A Systematic Review. Alcohol and Alcoholism, 44, 500-516. doi: 10.1093/alcalc/agp054
Ragnarsdottir, T., Kjartansdottir, A., & Davidsdottir, S. (2002). Effects of extended alcohol serving-hours in Reykjavik. In R. Room (Ed.), The effects of Nordic alcohol policies. What happens to drinking and harm when alcohol controls change? (pp. 145-154). Helsinki, Finland: Nordic Council for Alcohol and Drug Research.
Rossow, 1., & Norstrom, T. (2012). The impact of small changes in bar closing hours on violence. The Norwegian experience from 18 cities. Addiction, 107, 530-537.
Rossow, 1., Storvoll, E. E., Baklien, B., & Pape, H. (2011). Effect and process evaluation of a Norwegian community prevention project targeting alcohol use and related harm. Contemporary Drug Problems, 38,441-466.
Scheffels, J., & Lund, I. (2009). Pubs and restaurants in the Norwegian media. In J. Torronen, B. Olsson, & I. Lund (Eds.), Painting the town red. Pubs, restaurants and young adults' drinking cultures in the Nordic countries (pp. 153-171). Helsinki, Finland: NAD.
Schofield, T. P., & Denson, T. F. (2013a). Alcohol outlet business hours and violent crime in New York State. Alcohol and Alcoholism, 48, 363-369.
Schofield, T. P., & Denson, T. F. (2013b). Temporal alcohol availability predicts first-time drunk driving, but not repeat offending. PloS one, 8, e71169.
Silverman, D. (2001). Interpreting qualitative data: Methods for analysing talk, text and interaction. London, England: Sage.
Stockwell, T., & Chikritzhs, T. (2009). Do relaxed trading hours for bars and clubs mean more relaxed drinking? A review of international research on the impacts of changes to permitted hours of drinking. Crime Prevention and Community Safety, 11, 153-170.
Sulkunen, P., Rantala, K.., & Maatta, M. (2004). The ethics of not taking a stand: Dilemmas of drug and alcohol prevention in a consumer society--A case study. International Journal of Drug Policy, 15, 427-434.
Toomey, T. L., & Lenk, K. M. (2011). A review of environmental-based community interventions. Alcohol Research & Health, 34, 163-166.
Treno, A. J., Marzell, M., Gruenewald, P. J., & Holder, H. (2014). A review of alcohol and other drug control policy research. Journal of Studies on Alcohol and Drugs, 17, 98.
Ingeborg Rossow is an epidemiologist and senior scientist at the Norwegian Institute for Alcohol and Drug
Research. Her main research interests include alcohol and other drug policies.
Bergljot Baklien is a retired senior researcher but still active in research at the Norwegian Institute for Alcohol and Drug Research. Her main research interests include qualitative analyses of interventions at the local level.
Ingeborg Rossow  and Bergljot Baklien 
 Norwegian Institute for Alcohol and Drug Research, Oslo, Norway
Submitted November 29, 2013. Accepted for publication May 5, 2014.
Ingeborg Rossow, Norwegian Institute for Alcohol and Drug Research, Oslo, Norway. Email: firstname.lastname@example.org
Table I. Overview of the Data: Direction and Hours of Change, Date of Change, City, Comments on Change, and Retrieved Data. Date of Change Change City Comment I 2.00-3.00 1.7.04 Bergen I 2.30-3.00 1.7.04 Bodo I 2.30-3.00 1.7.04 Drammen I 2.00-3.00 1.7.04 Lillehammer I 2.00-3.00 26.2.04 Molde Applies to weekends I 1.30-3.00 1.4.04 Sandnes I 1.30-3.00 1.7.04 Stavanger I 2.00-3.00 1.7.04 Trondheim I 1.00-3.00 1.4.04 Alesund R 3.00-2.00 1.1.09 Arendal Exception: summer 2009 (until 3.00) R 3.00-2.30 1.7.08 Bergen R 2.30-1.30 1.9.08 Fredrikstad Weekends. Exception: busy period R 2.00-1.30 11.9.09 Gjovik R 3.00-2.00 1.7.08 Halden R 1.30-1.00 1.7.08 Haugesund Exception: a few night clubs R 2.30-1.30 1.7.08 Hamar R 3.00-2.00 8.12.08 Horten R 3.00-2.00 1.10.08 Kongsberg R 3.00-2.00 1.8.08 Kristiansand R 3.00-2.00 1.7.08 Kristiansund N R 2.30-2.00 1.10.08 Larvik R 3.00-2.00 1.7.08 Molde Weekends R 3.00-2.30 11.7.09 Moss Weekends R 2.30-2.00 1.7.08 Sandefjord R 3.00-1.30 1.7.08 Sandnes Exception: nightclubs (to 3.00) R 2.30-1.30 1.4.08 Sarpsborg Weekends. Exception: busy period R 3.00-1.30 1.7.08 Stavanger Exception: nightclubs (to 3.00) R 3.00-2.00 1.7.08 Trondheim R 3.00-2.00 1.7.08 Tonsberg Exception: summer season 29 changes Sum 24 cities w/data Available Available Documents From Media Interview With Change City Council Articles Key Informants I 2.00-3.00 3 3 No I 2.30-3.00 0 2 No I 2.30-3.00 1 7 Yes I 2.00-3.00 2 0 Yes I 2.00-3.00 1 10 No I 1.30-3.00 2 1 No I 1.30-3.00 3 1 No I 2.00-3.00 5 5 No I 1.00-3.00 3 5 Yes R 3.00-2.00 1 14 Yes R 3.00-2.30 2 11 Yes R 2.30-1.30 2 25 Yes R 2.00-1.30 1 5 Yes R 3.00-2.00 2 11 No R 1.30-1.00 1 29 Yes R 2.30-1.30 2 0 No R 3.00-2.00 0 11 Yes R 3.00-2.00 2 3 Yes R 3.00-2.00 2 0 Yes R 3.00-2.00 2 6 No R 2.30-2.00 1 14 Yes R 3.00-2.00 1 15 Yes R 3.00-2.30 1 6 Yes R 2.30-2.00 2 24 No R 3.00-1.30 2 10 Yes R 2.30-1.30 2 23 Yes R 3.00-1.30 2 6 Yes R 3.00-2.00 2 17 Yes R 3.00-2.00 2 16 Yes 29 changes 52 documents 280 media 19 interviews w/data articles Note. I = increase in closing hours, R = restriction in closing hours.
|Printer friendly Cite/link Email Feedback|
|Author:||Rossow, Ingeborg; Baklien, Bergljot|
|Publication:||Contemporary Drug Problems|
|Date:||Dec 22, 2014|
|Previous Article:||Storytelling in drug treatment: how professionals make sense of what they consider inauthentic client claims.|
|Next Article:||"I just drink for that tipsy stage": young adults and embodied management of alcohol use.|