Do Open Meetings Affect Deliberation? A Comparative Study of Political Meetings in Two Institutional Settings

Transparency in politics is a core value of democracy, but has to be balanced against politicians’ need for deliberation arenas. Norway and Denmark have relatively similar political systems at the municipal level, but the balance between openness and deliberation is approached in different ways: Denmark emphasises closed political meetings and deliberation whereas Norway emphasises open meetings and democratic control. This comparative study of two Danish and two Norwegian municipalities shows that closed meetings make politicians more able to deliberate, and that mandatory open meetings move deliberation to informal meetings, and thus reduce the potential for consensus-based politics.


Introduction
Transparency in politics is a core value of democracy, and has become synonymous with good governance. Many attempts have been made to bolster transparency at different levels of government, with the expectation that the quality of government will improve (Cucciniello, Porumbescu, & Grimmelikhuijsen 2017;Kosack & Fung 2014). One aspect of this 'transparency wave' is open meetings, which are studied in this article. The aim of this paper is to study the effect open meetings have on deliberation and the quality of policy processes, based on empirical studies of two systems with different approaches to open meetings. The research question is: how do open meetings affect deliberation in local elected committees and executive boards? To answer this question, in-depth interviews were conducted with 40 informants in Norwegian and Danish municipalities. These countries were selected because their principles of transparency and deliberation differ. In Norwegian municipal law, there is an emphasis on democratic control and open meetings (NOU 1990: 13), while in Danish municipal law, the emphasis is on deliberation and closed meetings (Indenrigsministeriet 1977).
In this article, theoretical arguments for and against transparency in policy processes will be explained. An overview of the selected cases and the national contexts in which they appear is given, and the interviews are presented in an empirical analysis. The paper will conclude that mandatory open meetings do not satisfy politicians' need for deliberation, and that informal meetings gain significance at the expense of formal, open meetings. This is important knowledge that needs to be taken into consideration before implementing transparency acts into legal systems.

For and Against Transparency in Meetings
Openness as a virtue was formulated by Bentham (2001Bentham ( [1790: 277), when he stated that 'the more strictly we are watched, the better we behave'. However, the urge for transparency has risen significantly in the last 25 years, to the point where prominent scholars such as Christopher Hood (2006) claim that transparency has a ' quasi-religious significance'. The organisation Transparency International has promoted transparency as a key component for good governance since 1993. The information revolution with the internet in general and Wikileaks in particular has also raised awareness of transparency in public affairs. Transparency can be defined as the degree to which access to government information is available (Piotrowski & Borry 2010). Meetings are understood as arenas where several elected politicians can discuss local affairs. Such meetings can be formal (committee meetings, executive board meetings and council meetings), or they can be informal (preparatory party meetings).
Arguments for transparency can be made in regard to both the input and output sides of politics. The output side refers to measures of the delivery of public service goods, for example transparency regarding test results in schools, hospitals, etc. However, this study is concerned with the input side of politics, where open political meetings should make it easier for the electorate to hold politicians accountable, based on a democratic control perspective. Open meetings are also regarded as a means for the electorate to become more informed about policy choices, and therefore make more informed choices on election day.
Nevertheless, the need for transparency has to be balanced against another important democratic virtue, namely democratic deliberation. Deliberation is an aspect of democratic politics in which representatives investigate the reasons for and against a matter freely, and are not afraid to venture mutually acceptable reasons to justify the decisions they make (Gutmann & Thompson 2001;Steiner, Bächtiger, Spörndli, & Steenbergen 2004;Przeworksi 1998;Elster 1998). Deliberation is associated with debates and processes of high quality, where participants need to be open minded and willing to change their preferences during the debate. If some meeting participants have decided in advance what their preferences are, and also agreed not to change these preferences during the meeting, they are not taking part in the deliberative aspect of the meeting. Some meetings may therefore be highly deliberative, while others may be less deliberative -based on the set-up of the meeting.
The choice between open or closed meetings in political processes is therefore a trade-off between two key democratic considerations: on the one hand, democratic control, which revolves around citizens' ability to hold their representatives accountable for the decisions they make -and on the other hand, democratic effectiveness, which revolves around politicians' working conditions and need for deliberation to enhance their ability to make the best possible decisions on behalf of the local community. Piotrowski and Borry (2010) developed an analytical framework for open meetings and transparency that employs a consistent list of arguments for and against public meetings, found in a 1962 Harvard Law Review article (Harvard 1962). Their arguments in favour of open meetings are, first, open meetings can provide better public knowledge of policy issues. Second, openness will deter misappropriations of public funds and conflicts of interest among representatives. Third, government officials will be more responsive to the public when there is an opportunity for public participation at a meeting. Fourth, representatives can receive information from the public in cases where the public has more information about a specific local issue. Fifth, citizens may gain more understanding of complex and difficult decisions. Sixth, open meetings can result in better news reporting of governmental issues. On the other side, the article also lists five arguments against open meetings. First, public pressure is negative for representatives' deliberation. Second, open meetings will lead to long and timeconsuming speeches where representatives attempt to make an impression on the press and electorate. Third, there is a disadvantage to government employees who may object to a programme or policy and voice these objections in public, but who in the end may be required to administer the objectionable programme or policy. Fourth, elected officials may harden their positions once they state them publicly. Fifth, there can be a tendency for the press to sensationalise stories and emphasise only controversial topics brought up at meetings. Since this extensive list was created, the public sector and welfare policies have expanded significantly, and theories of participatory democracy and deliberative democracy are now on the agenda. Together with the media revolution of the last 20 years, there is a constant need for debate on how the input side of politics relates to the public.
Later studies do reveal some problems with open meetings. In an empirical document study of parliamentary minutes, Steiner et al. (2004) find a more respectful tone and understanding of the positions of others in closed sessions than in those that are public. In an empirical study of public versus private deliberation, Meade and Stavasage (2006) find that members of a central bank board are less likely to oppose an established view when the discussion is public.
The alternative to open meetings can be ' openness in rationale', formulated by Mansbridge (2009). The argument is that in non-corrupt systems, with selfmotivated actors, too much transparency and surveillance will be counterproductive, because it is associated with a system of distrust. In contrast, openness in rationale is a form of narrative or deliberative control where representatives explain the choices they have made and have ongoing contact with the electorate during the decision-making process. Etzioni (2010) also argues that insights from behavioural economists weaken the premise of democratic control, due to information overload. In this view, voters neither have the capacity to handle the large amount of information that transparency ideals presuppose, nor respond to open meetings in the way the theories of transparency suggest. Heald (2006) examines various types of transparency and develops a framework for thinking about the directions of transparency. Most relevant for open meetings is his label of 'transparency inwards', the ability to see inside an organisation from the outside. Transparency in this case can be made in retrospect or in real time, where open meetings are real-time transparency and minutes are transparency in retrospect. Open meetings means that 'there is never any time when the organization can focus exclusively on its productive activities' (Heald 2006: 33). He also makes a distinction between 'nominal' and ' effective' transparency, where the gap between the two is labelled the 'transparency illusion'. The illusion is that even if transparency appears to be increasing, as measured by an index or transparency act, the reality might be different -for example if informal, closed meetings gain significance at the expense of formal, open meetings. This may be problematic for democratic accountability (Copus and Erlingsson 2013).
Despite the research that has been done, not a great deal is known about how politicians actually behave in closed versus open meetings. One reason for this is perhaps methodological, due to the obvious difficulties of investigating what occurs behind closed doors. This study will examine this by studying two different systems, in which different meeting rules apply.

Transparency in Norway and Denmark
In comparative studies, Norwegian and Danish municipalities are classified as relatively similar systems (Denters & Rose 2005;Loughlin, Hendriks, & Lidström 2011). Both systems are based on the municipal executive board model, which is part of the consensus-based model of government (Hendriks 2010). Unlike the parliamentary-based 'winner takes all' system, the system is based on a broad representation of the executive body (the municipal executive board). The municipal executive board model is older than the party system; the idea is that leading politicians meet in the municipal executive board to find common solutions to local community problems. In its ideal form, participants meet on equal terms and discuss issues according to the deliberation principle. The same principle applies to political committees, which also have a consistent representation based on the size of parties in the municipal council. The significant distinction between the Norwegian and Danish systems, which is investigated in this study, is whether or not it is mandatory that meetings are open to the public. In Norway, the Local Government Act regulates that all meetings of political committees must be open to the press and citizens. In Denmark, the opposite rule applies: meetings are closed, and it is prohibited to refer from the meetings. This provides a good opportunity to study the importance of the public principle of deliberation in a comparative study based on a similar system design principle.
Norway's regulation on open meetings stems from the Local Government Act of 1993, which requires all meetings in both committees and the executive board to be open to the press and public (NOU 1990: 13). It is not allowed for politicians on a committee or executive board to meet 'informally' to discuss matters. If they do, the local press can lodge a complaint with the county governor or parliamentary ombudsman. Such complaints are usually supported (Leirset 2019). The Local Government Act outlines strict instructions for when meetings can be closed. Political discussions and deliberation on difficult matters, for example, are not considered reasons to close meetings. The open meetings regulation has also resulted in a critical public discourse about closed meetings, where stakeholders in the press frame the need for deliberation in private as 'undemocratic' (Bodahl-Johansen 2018).
In Denmark, the municipal act is quite opposite on this matter. All the meetings in standing committees and the executive board must be closed, and participants in these meetings cannot disclose what their peers have said in the meetings. This is to strengthen the meetings as deliberative organs where each politician is free to discuss with his or her peers. The disadvantages of open meetings are also emphasised in several arguments in Danish judicial reports on municipal legislation (Indenrigsministeriet 1977: 46). First, open meetings can lead to the 'real form of debate being relegated to another location'. Second, open meetings may cause 'too much politicisation' of the work done by the committees. Third, the administration may be 'reluctant to present material to committees at the earliest possible stage'. Moreover, consideration of small party groups is also emphasised as an independent point in the arguments for closed meetings (ibid).
As we can see, Norway and Denmark have approached open and closed meetings differently in their municipal legislation. To answer how this difference works in practice, semi-structured interviews were conducted with a range of local politicians in two Norwegian and two Danish municipalities. The municipalities selected have certain characteristics: all employ the municipal executive board model, they must not have carried out special administrative reforms, and they must be 'traditional' municipalities that do not differ significantly from the norm of municipalities in their country. It has also been desirable to find equally large municipalities which are relatively similar in population size. They have quite similar profiles in that they are located outside major cities, yet are still large enough for there to be a local press that has an interest in covering municipal political meetings and issues. All the municipalities have at least one local newspaper that covers local politics to the extent that all politicians are familiar with the editor or journalist who covers them. However, the frequency with which the local press covers political meetings has not been measured, so there may be some variance in the relations between the political community and the local press that is not reflected in this research. In order to distinguish them from each other, the informants are given fictitious names, and municipalities are referred to as Norway North, Norway South, Denmark East and Denmark West.
Interviews were conducted with 40 informants, most of whom are local politicians. Emphasis was placed on interviewing politicians from as many parties as possible, who also have experience with and a good overview of the local political landscape. In addition, in all municipalities at least one politician with a shorter political career is included, in order to get a freshman's view. The majority of interviews were conducted as individual semi-structured interviews, and there was in addition one focus group interview in each municipality, conducted with two or three informants. The informants also include a member of the local press from each municipality, which has afforded valuable insight into how local politics are judged by those who cover them. The members of the local press are also informed outsiders. In addition, I have observed at least one political meeting in each municipality. Taken together, the methodological approach has provided a rich resource of information, which has given me a good overview of how local politics work in the case municipalities.
Norwegian and Danish municipalities are both composed of three bodies, as summed up in Table 1: the standing committees, the municipal executive board and the municipal council. However, political parties also play an important role in municipalities in both countries. Political parties have their own meetings at some stage in the political process. These meetings are not constituted in law, and it is up to the local party how to organise them. They are closed for others than party members in both countries, and there are no formalities, such as minutes, attached to them. They are therefore 'informal meetings' in judicial terms. Even so, they are informally institutionalised as a normal part of the political process in both countries. This means that if the openness principle affects the functioning of formal meetings, we could expect that the significance of informal meetings would be different in the two countries, in accordance with the 'transparency illusion' argument described above (Heald 2006). From this, we can derive a hypothesis that informal meetings play a more important role in Norwegian municipalities than they do in Denmark.

The Functioning of Meetings in Norway and Denmark
In this section, I start out by giving an overview of the main findings from the semi-structured interviews in Table 2, and explain the quotes further down. Table 2 is used to structure the empirical analysis.

Standing committees
Even if the standing committees in Norway are formally open to the public, whether or not the meetings are covered by the local press varies. However, it is impossible to know if a journalist is present or not in advance, so politicians must be prepared for the possibility that journalists will be present. Therefore, in Norway most parties have institutionalised preparatory party meetings in advance of committee meetings. Several informants, from both government and opposition parties, describe these preparatory meetings as significant, as politicians tend to form their opinions in the party ahead of the committee meeting. This fosters internal loyalty within the party group, even if new arguments emerge in the committee.
When we have a preparatory party meeting, we'll come to an agreement. Internally. So there's one person who presents this in the [committee] meet-ing. It's rare for us to change opinion during the process of the [committee] meeting (Norwegian politician Arne).
Leading politician Arne describes how the issues are discussed in the preparatory party meeting, and states that the party makes its real decision there. Several informants find it difficult to imagine not having an informal meeting in advance, where issues can be discussed openly and freely. Open discussions in committee meetings are described as difficult, because of loyalty to what has been decided previously in the preparatory party meeting. However, compared to executive board meetings and council meetings, committee meetings are regarded as better workshops. Journalists are more often present at executive board and council meetings, so there can be a more informal tone and open discussion in committee meetings where no journalist is present. In Denmark, standing committees are viewed as the best arena for deliberation by almost all informants (see Table 3, below), because committee members are experts in their sectors and have more in-depth knowledge of the issues than do party members. This leads to a more knowledge-based debate in the committee, as opposed to preparatory party meetings. The informants state that cross-over relations between parties are important. Several informants use ' engine room' as a metaphor for committee meetings. Although party politics do play a role, institutionalised preparatory party meetings prior to committee meetings are less common than in Norway. In addition, informants tone down the party element in committee meetings. They are socialised into an understanding of the committees as an arena for open debate and deliberation. This is particularly due to the absence of the press, as well as the prohibition to refer to what others have said in the meetings, which is explicitly  If a journalist is present in a meeting, that's the biggest game-changer. Everyone becomes more concerned with what to say, and how to say it. It plays out as a form of acting, if you know what I mean.

Executive board
The executive boards are more powerful organs than the committees. In Norway, they are therefore more frequently covered in the press. One informant describes the executive board meetings in Norway as 'speaking to the public'. Because of this, it is important to be well prepared for the meeting, which means that the preparatory party meeting is important. Asked whether the preparatory party group meeting or the formal council meeting is more important in terms of actual debates and conclusions, one informant responded as follows, after some hesitation: It depends on the issue, of course. However, we meet a lot in advance. It's easier to formulate better decisions in the preparatory party meetings, since we can discuss freely. I don't like to say it, but it may well be that the most important meeting is the preparatory party meeting. (Norwegian politician Eline) Eline leads a committee in Norway North, and is quite hesitant when she evaluates the practice of preparatory party meetings in advance of formal meetings. She reflects on the informal and formal procedures in the local political process, and notes that her party, presently in power, is criticised by the opposition for not including them to a greater extent in debates and conclusions. However, the critique doesn't bite so hard, because she considers democracy to be 'majority rule'. What is more, it is difficult to include the minority group, because the Local Government Act prohibits all politicians from participating in the preparatory meeting. Informants from the opposition in Norway North describe the executive board meeting as a 'theatre for the public'. This means that it is almost impossible to change the views of the governing group in the meeting, because its members are loyal to each other. The informants explained that the governing group votes through whatever it has decided in the preparatory party meeting. This also changes their understanding of the opposition's role. They see it as their role to argue against the governing party -and based on this, deliver their own propositions. These are voted down, but they can demonstrate to the public that they have an alternative policy. In Denmark, the executive board has the same rules as the committee: the press cannot attend, and it is forbidden to refer to what others have said. This rule is meant to build loyalty between the members of the executive board, and also across party groups. This seem to have the intended effect. One informant, a Danish mayor, explained that the closed meetings make it possible to share early thoughts and withdraw proposals -even for him -if good arguments come up. If journalists were allowed in these meetings, he is very sure that it would have a dramatic effect on this functioning of the meeting. When asked about the effect the presence of journalists would have on the meetings, he stated that the current practice would become impossible.
The relatively open or restricted access to debate in meetings is of particular interest to the executive boards, since these are rather powerful local organs. One might assume that the Danish press would be critical of the practice of closed meetings in such powerful organs. However, the Danish informants from the local press did not particularly object to it.
In principle I'm in favour of open meetings, as a journalist. But I can understand politicians' need to discuss freely. My experience is that they talk more to me, as a journalist, when I'm present, than to each other. This certainly is the case at council meetings, where I can attend. They want to be quoted in the newspaper, to be visible for their electorate. (Danish local journalist) The Danish journalists explained that agendas for these meetings are public, so they are able to create new reports based on these, and follow up with the leader of the committee to get more information on the matter after the meeting. The minutes from the meeting include the voting behaviour of each participant, so journalists can contact politicians from all sides to obtain the arguments in each direction and present them to the public. Danish law forbids politicians from disclosing what other politicians have said in the meetings, but it is permitted to refer to one's own statements. The journalists are very aware of the minutes as an important tool for reporting on meetings. The minutes enable them to report political differences even if the meeting is closed.

Municipal council meetings
The rules concerning openness in municipal council meetings are the same in both countries. Informants in both Norway and Denmark report that it is quite common for most issues to be decided before they end up in the council. However, in the Norwegian municipalities there is a tradition for ad hoc, short party group meetings during council meetings. In Norway South, this is very common. In Norway North, such party group meetings during the council meeting occur approximately every other meeting. One such council meeting was observed in Norway South, where council meetings have a short break, and the party groups can gather in various places in the city hall. Here they can quickly discuss party policy before the formal meeting resumes. This is a major difference from the Danish system, as described below. The party groups are then prepared to argue and vote in a certain direction. In addition, some party groups in Norway use online chat actively throughout the council meeting. The party group has a Facebook chat forum where party members can take part in live discussions on the debated topics. When a political opponent takes the rostrum, the party can quickly decide if they want to present a counter argument, how the argument should be phrased, and who should argue on behalf of the party. Online chat may also be used to show approval and give feedback to party colleagues.
The Danish municipal council is the only formal organ open to the public, and is characterised as the main arena for political play. This is where politicians have an opportunity to profile themselves and their politics. Informants explained that this meeting is distinctly different from the other meetings. Ad hoc party group meetings during council meetings, however, are extremely rare. Several politicians reported that they had never experienced such meetings. One politician with 12 years of service had experienced it maybe once, in an exceptional case. The council meeting is considered formal, since everyone is on their guard due to the presence of the press.

Preparatory party meetings
In Norwegian municipalities, preparatory party group meetings are common during several stages of the political process. There are usually at least two institutionalised party meetings in addition to ad hoc party meetings during the council meeting. All informants described Norwegian preparatory party meetings as important policy workshops, since they are the only space in which politicians can speak on current local affairs out of the public eye. Here, party members can speak freely, and several politicians emphasised that strong disagreement within the party is tolerated. However, the normal procedure is to argue and vote within the party -and members are expected to follow the party line in most issues.
Debates in Danish party groups are also described as good workshops for politics, but first and foremost a place where one can try to think more ideologically. The informants reported that representatives usually meet in the standing committees with a loose mandate so they can be free to deliberate. In the Danish system, preparatory party group meetings are less frequent than in Norway. The normal frequency is one meeting during the policy process, whereas the Norwegians tend to have at least two. Ad hoc party group meetings during council meetings, which is normal in Norway, do not occur in Denmark, except in exceptional cases.
For me that's unheard of. I've never heard of anyone having a party meeting during a council meeting. (Danish politician Lars) In the section above, I have shown the function and dynamics of various political meetings in Norwegian and Danish municipalities. In the next section I will discuss the findings and analyse them according to theories laid out earlier in this article.

Discussion of Findings
The function of standing committee meetings is slightly differently in the two countries. The Norwegian representatives describe a tighter mandate from their parties even at this early stage of the policy process. The Danish standing committee meetings are closed to the public, and it is prohibited to refer to what others have said. The purpose of this is to create an atmosphere conducive to deliberation. Based on the interviews with our informants, this seems to have had the intended effect, as Danish politicians described these meetings as very deliberative. This finding is in accordance with Steiner et al. (2004), who find a more respectful tone and understanding in closed sessions in their analysis of parliamentary debates.
The same principles apply to executive board meetings: Norwegian meetings are open, Danish meetings are closed. The informants' descriptions of executive board meetings were quite similar to their descriptions of meetings in standing committees, since they are subject to the same rules. However, these meetings are of greater significance than meetings in the standing committees, so the press is more interested in covering them. This makes the Norwegian executive boards more of a 'theatre', according to informants. However, the Danish press cover executive board meetings by using the agenda and minutes, and based on this, interview politicians on issues they find interesting. This means that even if the meetings are closed, there is a form of accountability, though it is not based on democratic control in a principle-agent relation. Rather, it is a trust-based ' deliberative control', in accordance with ' openness in rationale' (Mansbridge 2009).
Meetings in the municipality councils of both countries are open, and descriptions of them are fairly similar in the two countries. Since the meetings are open, the politicians are concerned with how they are portrayed in the press, which affects the meeting. This is the main meeting where politicians can explain their politics to the electorate. The most striking difference is the role of the preparatory party group meetings. In Norway, it is normal to have ad hoc meetings within the party group during the council meeting. This is not the case in Denmark, where such meetings are not held, except for one extreme situation in the last 12 years, in one municipality. Ad hoc group meetings can have both positive and negative sides. On the positive side, such meetings can be a vibrant political milieu where issues are not decided in advance. On the negative side, it may be a sign that the early stage of the policy process is inadequate. It suggests that there is a lack of deliberative arenas where politicians may discuss freely across party borders. It may also give more significance to the parties than intended in the municipal executive board model. If the parties play such an important role, one may argue that this is majoritybased rather than consensus-based politics. It also raises questions as to council members' loyalty: are they first and foremost loyal to the party and other party members, or are they loyal to the council and other council members? It appears that Danish councils have a more consensus-based culture, where elected politicians feel more loyalty to the council itself, and Norwegians feel more loyalty to the party (Aarts, Thomassen, & van Ham 2014).
The last type of meeting is the preparatory party meeting, which in the judicial sense are informal meetings. These meetings are closed, and considered good arenas for deliberation in both countries. During the interviews, informants were asked which meeting they felt was best for constructive and solution-oriented debate. The responses show an overall trend of formal committees being ranked highest in Denmark. In Norway, preparatory party meetings are ranked as most important. The difference is shown in Table 3.
We see two conditions in Table 3. First, the summary shows that preparatory party meetings are perceived as being the most important deliberative bodies in Norway, while political committees are perceived as most important in Denmark. We can also note that closed meeting arenas are considered best. In Norway, this only applies to preparatory party meetings.
Preparatory party meetings in Norway are deemed more important as deliberative arenas, and occur more frequently in the political process as compared to Danish municipalities. This suggests that the parties play a more important role in Norwegian municipalities than they do in Denmark. Danish politicians have several closed arenas, which give more room for deliberation. In summary, it is to be expected that more room for free discussion results in more ideas and more solutions on the table. It is also to be expected that room for free discussion reduces 'political play' and strengthens the consensus-making aspect of politics.
It appears therefore that the effect of open meetings in Norway is quite significant in regard to how local politics works, but not as intended in the Local Government Act. The main reasoning for requiring Norwegian local political meetings to be open is democratic control and accountability, but the empirical effect increased and empowered preparatory party meetings. The 'real' debates are moved to a closed informal meeting. This is an example of what Heald (2006) describes as 'transparency illusion', where the effect of a policy designed to ensure openness may in fact be the opposite of what was intended. A comparison of the legal frameworks of municipal acts shows that the Norwegian requires much more transparency from the local politicians than the Danish. However, the effect can be that new, unregulated arenas for decision-making gain significance. This indicates that judicial transparency leads to illusory openness (Heald 2006). It is also troublesome to move a part of the policy process to informal institutions, because it has negative implications for democratic accountability (Copus and Erlingsson 2013).

Conclusions
This study shows that there is a significant difference between Denmark and Norway as to where deliberation takes place. Danish local politicians report that formal meetings in standing committees and the municipal executive board are the most important and best meetings in relation to political discussion and finding good solutions. In Norway, the majority of local politicians reported that the preparatory group meetings best fulfil these criteria.
It appears that committee meetings in Denmark function better according to the intention of the municipal executive board model than in Norway. The Danish model makes it easier to build a broad political consensus with the involvement of a wider section of the political spectrum. The reason seems to be that the meetings must be closed. The effect of the Norwegian model, with mandatory open meetings, seems to be that discussion is moved to parties' preparatory meetings. Formal meetings do not appear to meet politicians' need to discuss political solutions across party lines.
The aim of this paper is to study the effect open meetings have on deliberation, and the quality of policy processes. First and foremost, this study shows that openness does affect deliberation. Having journalists present during meetings is described as a game changer for how meetings proceed, and politicians reported that it was difficult to discuss freely in open meetings. The consequence of mandatory open meetings is that deliberation is pushed to other arenas, such as preparatory party meetings -which as a result gain more significance as policy workshops. Thus, we can see that preparatory party meetings are more numerous and reported to be more important in Norwegian as opposed to Danish municipalities. This leads to more party politics in Norwegian municipalities than in their Danish counterparts. Politicians build an increased loyalty to the party at the expense of loyalty to representatives from other parties.
This study uses a comparative design to give insight into how open meetings affect deliberation. The most important contribution is to show how politicians, in two quite similar systems, respond to one critical aspect, namely open meetings. Open meetings do affect deliberation, a finding that supports earlier studies, which demonstrates that deliberation functions better behind closed doors. Mandatory open meetings do not satisfy politicians' need for deliberation. Even if this research only includes four municipalities, the findings accord with earlier research on the effect open meetings have on deliberation. It is therefore suggested that the findings are transferable to other contexts than Danish and Norwegian municipalities. This is thus important knowledge for all governments considering the dilemmas inherent in the balance between deliberation and democratic control in their municipal legislations.
There are, of course, still some unanswered questions. Some informants suggested that open meetings may give more power to the administration, due to the lack of arenas in which bureaucrats and politicians can discuss freely. In the Norwegian system, no such arenas exist. One hypothesis is that this increases the divide between politics and administration, and raises the threshold for administration officials to deliver issues for political procedure. Further research is needed to investigate this.

Competing Interests
The author has no competing interests to declare.