BETWEEN TRUST AND SUSPICION
A COMPARATIVE STUDY OF THE RELATIONSHIP BETWEEN POLITICIANS AND POLITICAL JOURNALISTS IN BELGIUM, NORWAY AND SWEDEN
Abstract
This article presents an empirical study of the relationship between politicians and journalists in three European countries. Based on a survey among political journalists and Members of Parliament in Belgium, Norway and Sweden we ask how "intimate" the relationship between these two groups really is, and if the informality of the relationship also influences the image they have of one another. Our study shows that the degree of informality differs significantly between the three countries, where the Swedes have less informal contact. We believe this country difference can be mainly attributed to the higher degree of political professionalisation. Unlike Nimmo (1964) our analysis does not suggest that the more informal the relationship is, the less suspicious journalists or politicians are towards the other group. Rather our results seem to show that trust and suspicion go hand in hand.
PETER VAN AELST TORIL AALBERG
Peter Van Aelst is Associate Professor at the University of Antwerp and Leiden University; e-mail: peter.vanaelst@ua.ac.be.
Toril Aalberg is Professor at the Norwegian University of Science and Technology (NTNU); e-mail: toril.aalberg@svt.ntnu.no.
00 00 I
ro rv.
o
o rN
oo
£
£ 3
a
0
£ *
1 *
(R 0 C
! m
Introduction
The relationship between politicians and journalists is characterised by mutual dependence (Mancini 1993; Neveu and Kuhn 2002). The modern politician needs the news media to get his message across and to reach out to voters and colleagues. The political journalist needs to know what is going on in the world of politics, needs this information fast and prefers to have it first-hand. This marriage de raison is most often portrayed as a dance, a tango even with almost intimate interactions between both partners (Gans 1979).1 Some authors go even a step further and talk about (strange) bedfellows (Rosenstiel 1993) or call the relationship boldly incestuous (Charron 1994).
Unfortunately, these metaphors are seldom made concrete, or at least not in a systematic comparative way. A long tradition of studies in political communication (e.g. Nimmo 1964; Cook 1998; Sellers 2010), media sociology (e.g. Sigal 1973; Ericson et al. 1989) and case studies of journalists (e.g. Crouse 1974; Rosenstiel 1993; Jones 1995) have given us in depth insight in how the bargaining process of news and information takes place, but these studies almost always focus on how (political) journalists deal with politicians in a certain country. Because of the Anglo-American bias of this literature far less is known about this relationship in other countries than the US and the UK (for an exception see Stromback and Nord 2006), and hardly anything is known on how the relationship differs between countries.2 A comparative perspective should offer more insight in the antecedents of the intimate relationship between journalists and politicians, as well as it consequences.
We believe it is important to focus on the intimate nature of the relationship because the interactions between journalists and politicians are hardly guided by formal rules or institutions. This is not to say that there are no "rules of the game" but rather that the behaviour of both journalists and politicians is guided by informal rules and negotiable agreements. Scholars who would only focus on the more formal aspects (e.g. media policy) or public interactions (e.g. press conferences) y miss the overall insight in the power relationship.3
In this article we will focus on the degree of informality of the relationship between politicians and journalists in three European countries (Belgium, Norway and Sweden). The first more descriptive part is guided by two basic questions. How "intimate" is the relationship between politicians and journalists in three European countries? And how can differences between the "similar" countries be explained? The second part of the article focuses on the consequences of this informality on the perceptions of politicians and journalists on the (power) relationship. Does the informality of the relationship also influence the image they have of one another? ^ Or put differently: Does a more informal relation also lead to a more positive and less suspicious perception of the other?
Our analyses are based on a survey among political journalists and Members of Parliament in Belgium, Norway and Sweden. As these three countries are considered as belonging to the Democratic Corporatist model of political media systems (Hallin and Mancini 2004) we expect the differences in the relationship to be limited,
G
C >
however not absent. Mainly because of a higher degree of political professionalisation in Sweden, where spokespeople to a larger degree act as mediators, we expect p^ the relationship to be more formal in the Swedish case than in Norway and Belgium.
The features of the political media system in the three countries under study will be discussed later. To measure the informality of the relationship between politicians LO and journalists we use four different indicators: the frequency of informal contacts ^ such as lunches, whether one has friends among the other group, how often one asks or gives advice to members of the other group, and whether one exchanges personal phone numbers. All four indicators refer to the reported behaviour of both journalists and politicians, going beyond the perceptions, orientations and values that are used in earlier research (e.g. Pfetsch 2001). Before discussing our research design more in detail, we will theoretically elaborate on the nature and importance of the close relationship between journalists and politicians.
Theoretical Perspectives on the Relationship between Politicians and Journalists
Studying the Relationship between Politicians and Journalists. The relationship between media and politics has been studied by different scholars from different theoretical perspectives. In general, we can distinguish between a system level approach and an actor approach (Van Aelst et al. 2008). A system level perspective sees the relationship as determined by structures and system properties, rather than a relation between individuals or groups. Structuralist theories focus on impersonal mechanisms that bias the political process or news production without necessarily requiring intervention by any particular action (e.g. Thompson 1995). For instance the indexing theory that argues a dominance of elite sources in the news, can be considered an overriding principle that structures how journalists select political sources (Bennett et al. 2007). An actor approach on the contrary places the individual actors, journalist and politicians, at the centre stage. Following the work of Dahl (1998) and others this approach focuses on resources, and even more on interaction. The central idea is that both the work of politicians and journalists is influenced by mutual perceptions and interactions.
Without questioning the structural approach we believe in the added value of studying media and politics from an actor approach for mainly three reasons. First, the daily interactions between journalists and politicians have at least potentially a direct impact on both news making and law making. A study on the interactions between both groups can be seen as first step to better understand these effects (Cook 1998, 13). It might for instance explain why a party got their issue higher on the media agenda or was able to promote their version of the facts (e.g. Sellers 2010). Second, the interactions can provide us with additional information on the democratic role of the media. Of course the independence of the media towards politics is a structural feature of a media system rooted in a broader perspective on media and democracy (Ferree et al. 2002), but an actor approach may be used as a sort of "reality check." In many countries, including Western democracies such as contemporary France (Kuhn 2010), the actual political independence of the media is stronger on paper than in daily practice. The third and perhaps most fundamental reason to study the interactions between politicians and journalists is that they shape or at least influence the construction of the political arena itself. This is what Davis (2009) has labelled "the social construction" paradigm. He argues that political journalists have become a natural part of how politics works and that this influences how politicians think and act. Because both groups operate in the same
networks or subsystems journalists influence, often unconsciously, all aspects of political life, ranging from policy debates to the rise and fall of individual political careers. The fact that politicians ever more "pro-actively" adapt to the media and its logic (Stromback 2008) makes it more difficult to measure media influence and increases the value of studying the daily interactions of both groups.
A Typology of Interactions: Between Harmony and Conflict. Because both journalist and politicians have something to gain by interaction their relationship is often depicted as one of interdependence, exchange, and mutual benefits. At the same time the relationship is inherently guided by tension and conflict (Blumler and Gurevitch 1981). Both partners often disagree on what is considered newsworthy and how it should be reported. Politicians not only look for media attention but also like to stay in control, and consider journalists as too "active" or interpretative in portraying their person or message. Journalists on the other hand often feel used by strategic politicians and their spin doctors in their efforts to communicate with the public or colleagues. Trust and distrust, or love and hate seem both natural parts of the relationship. This has always been the case. In his study of US press-government relations more than 45 years ago, Nimmo (1964) distinguished between three kinds of patterns in the relationship, referring to different degrees of harmony. The relationship can range from cooperative, characterised by common goals and low conflict, over compatible with increasing tensions, to competitive guided by mutual suspicion and mistrust. Although this typology is useful it remains difficult how to determine which pattern is most applicable for the relationship in a certain place and time.
One the basis of public statements of leading politicians and journalists, complaining about how they trouble each other's work, one would be inclined to see the relationship as one of competition. However, behind the surface both partners routinely keep on working in good understanding and cooperation (Kumar and Jones 2005). Therefore, we believe it is important to go beyond the statements of both actors and also look at their contacts and interactions. According to Nimmo (1964, 211-213) each pattern was accompanied by a specific process of interaction. In the cooperative environment the interaction between journalists and politicians (and their spokespersons) is continuous and informal, making more formal forms of interactions such as news conferences and prearranged interviews needless. In the compatible pattern the interaction is more formalised along interviews and Q press conferences, and unstructured forms of contact are still possible, but less common. In a competitive environment interaction is less frequent and almost always conflicting as journalist and politicians question the value of the interaction.
^	Benefits and Drawbacks of an Intimate Relationship. It is clear that both
partners will benefit from the cooperative pattern characterised by a high degree
u
S
G
C >
of informality. This being said, a very close relation might also create problems and raise normative questions. We will briefly discuss advantages and disadvantages of the cooperative model. For political journalists it means having easy and fast access to political information that on the basis of the trust in the
relationship can be considered as highly reliable (Donsbach and Patterson 2004). Furthermore, the journalist often receives not only information about policy outcomes and plans, but also on how these were established. The journalist becomes
II
a first-hand observer of political "behind the scenes" struggles and intrigues. This kind of information becomes more important as media devote more attention to personal struggles and even the private life of politicians (Langer 2010).	^
For politicians the cooperative context offers plenty and easy ways of communicating their message and seeking electoral support. Additionally, close contacts with journalists offer also less evident resources such as the rich information about the political process journalists carry with them. Particularly, for politicians that are located further from the internal party decision-making process, as many MPs are, this information is useful (Van Aelst et al. 2010). Another gratification for politicians to interact with journalists is because they can be considered as experts in the political communication process. Many journalists have a long career in political reporting which makes their advice of high value for politicians (Cook 1998; Davis 2007). Davis (2009, 211) showed that British politicians obtained advise from journalist on the basis of friendships or as part of the professional exchange process: trading information for advice instead of publicity.
Besides these clear benefits, an intimate relationship can have some drawbacks or at least lead to normative questions. This is certainly the case for journalists, who are supposed to be politically independent and keep their source at a certain distance in order to perform their role as a public watchdog (Schudson 2003). The extent to which politicians can be hold to account by journalists is seen as being negatively correlated with being too intimate. In his study on journalists covering EU politics, Baisnee (2002, 122) reports that especially journalists who aim to do investigative reporting deliberately keep personal distance from their sources and refuse "to dine with officials and develop friendship." The so called adversary model even expects that journalists should be somewhat hostile towards politicians to avoid being "in their pockets" (Blumler and Gurevitch 1981). For politicians intimate relations seem to raise less normative objections, but can nonetheless hinder the political process. Certainly in context of coalition governments, political agreements are the results of delicate compromises and secret negotiations. A close and informal relation with journalists makes it more difficult to maintain the necessary secrecy. Particularly by leaking information, politicians can improve their personal relationship with the receiver of the "scoop," but at the same time damage the trust among his or her fellow political actors (Jones 2006).
Press Politics Relations in Democratic Corporatist Countries
According to Hallin and Mancini (2004) Belgium, Norway and Sweden belong to the so called democratic corporatist model of media and politics characterised by three "co-existences" (Hallin and Mancini 2004, 144-5). The first co-existence relates to a high degree of political parallelism - that is, the tendency for the media to express and reflect political or social divisions in society - co-existing with a strongly developed mass circulation press. Second, a high level of political parallelism has co-existed with a high level of journalistic professionalisation. Third, a strong tradition of freedom of the press has co-existed with active welfare state policies and interventions in the media sector. Some of these co-existences are still in existence, even though the news media no longer reflect political and social divisions to the same extent as was previously the case (Allern 2007).
During the party press era, politicians were often recruited as journalists and editors and vice versa, and the same people were active simultaneously in both arenas. The depolarisation of the press in these countries coincided with a professionalisation of journalism, and most newspapers ended their formal ties with the parties during the 1970s and the 1980s. However, the informal ties between politicians and journalists that were part of the political parallelism tradition in these democratic corporatist countries were not abolished. Political journalists rather tried to broaden their informal contacts to politicians not belonging to their traditional political family (Van Aelst 2007; 0stbye and Aalberg 2008).
Another similar feature of the three countries under study is the dominant position of the public broadcaster. In all three countries the recent market share of Public Broadcasting channels ranges between 38 and 44 percent (Aalberg et al. 2010). When it comes to the mass circulation press it must be considered very strong in Sweden and Norway, which are among the leading countries in the world, and more moderate in Belgium (Hallin and Mancini 2004, 23).4
With regard to the political system, all three countries are parliamentary democracies with multi-party and proportional electoral systems, where voters choose between party ballots. Thus, all countries are party-centred as opposed to candidate centred, and have a tradition of strong parties and little room for individual MPs to go against the party line (Laver and Schofield 1998). This is true, even though voters in all countries can express their preference for a candidate. Coalition or minority governments are the rule and single-party governments based on a parliamentary majority are the exception. Furthermore, the number of relevant parties and the high degree of fragmentation of the parties in the respective parliaments are very similar (Klingemann 2005, 36-7). To sum up, all three countries have a complex political institutional setting with little room for MPs to play an independent role. As a consequence we might expect that political journalists have little incentives to invest in informal relationships with MPs, but rather focus their attention on Cabinet Ministers and party leaders. On the other hand the parliaments of the countries under study all provide services or institutions in or around the parliament to facilitate the interaction process between journalists and politicians. It concerns both places in the parliament where journalists are able to work and invite politicians for interviews or more informal places as bars and restaurants.
Although this discussion is by no means exhaustive, it shows that it is reasonable to treat Belgium, Norway and Sweden as similar cases. However, this does not mean that variations between these countries do not exist. Following the experimental similar systems design (Wirth and Kolb 2004) we will focus on one important country difference: the degree of political professionalisation. The ^ concept of professionalisation is frequently and freely used in political communication literature, often meaning different things. It is used to refer to both the changing process of campaigning and dealing with the media, as well as to the actual "professionals" who have a specific expertise in this process (Negrine and
CLilleker 2002). In the context of this study we define professionalisation as the
Hofrroa fr> wnirn mo infor*a nf it~\n nrnrocc hatiAfDon nr^litir'al i/~n imali cfc ann A/fT^c ic
0 £
>
degree to which the interaction process between political journalists and MPs is mediated," meaning that a spokesperson or other employee of the MP is respon-
«IXlVVtlULVVi^ 1XIVWJ. UJ.	LJ IUL M U^/V^I.Wa^'ViLJVl I V/JL VLlLVi Vlll^lVJ VV VI 1.11V IVli 1U
sible for communicating with the press. Contrary to the US were most Congress Members rely since decennia on a full time press secretary (Cook 1989, 72), this p^ is not (yet) the norm in most European countries. Our data show that in Belgium
only 17 percent of MPs has a personal assistant to deal with the media, in Norway this is somewhat higher (28 percent), but hardly comparable with the 80 percent C^ of the Swedish MPs that has personal support for interaction with the media (see ^ also Van Aelst et al. 2010). We believe that such a "go-between" would lead to a less informal relationship between journalists and politicians, because journalists will have frequent interactions with the personal spokespersons of MPs rather than with the MPs themselves. At least a British study showed that journalists have to go more than in the past through the press office or spokesperson to get in contact with an MP, and that these employees often have informal meetings with journalists as part of the politicians' media strategy (Barnett and Gaber 2001, 97-99). Of course this does not imply that MPs with a spokesperson don't interact themselves with journalists anymore, but rather that the general interaction culture is more mediated and less personalised.
Research Design
To study the interaction process between politicians and journalists a survey was conducted among Members of Parliament and (political) journalists in Belgium (Flanders),5 Norway and Sweden. Surveys among MPs (e.g. Thomassen and Andeweg 2004; Thomassen and Esaiasson 2006) or journalists (e.g. Weaver 1998; Donsbach and Patterson 2004) are a common research technique, but as far as we know only a limited number of studies have questioned politicians and journalists simultaneously about their actual interactions and mutual perceptions (Larsson 2002; Stromback and Nord 2006; Van Aelst et al. 2008; Davis 2009).
In the three countries the data were gathered in a similar way between 2006 and 2008.6 All Members of Parliament were surveyed using a written questionnaire that could be filled out on paper or online (Sweden relied on paper only). Each questionnaire was slightly adapted to the national context, but the core questions remained identical. Special care was given to a perfect translation of the questions. In all countries several reminders were used to increase the response rate. This resulted in a satisfactory response rate of approximately 50 percent among Norwegian and Swedish MPs, and 85 percent among Belgian MPs. The higher response rate in Belgium is mainly the consequence of the fact that researchers visited the parliament and personally contacted the MPs who had not yet responded. It is important to note that in all countries the response among parties reflected almost perfectly their strength in the parliament(s). The MPs who participated in our survey were not significantly different from the total population of MPs on age and gender.
In contrast with MPs, the group of journalists that cover domestic politics is less easy to define. Therefore we opted to include initially all beat journalists who cover the day-to-day work of government and parliament, as well as journalists who are specialised in a certain policy field, like environmental or economic policy. In a second step we only withhold those that report on political actors on a regular basis (3 or more articles). This selection was made on the basis of the following (filter) question: "In how many of the last ten articles / news items you made, was a party or politician of your country mentioned?" In the three countries, around half to two thirds of the contacted journalists yielded a useful questionnaire. To improve the comparability of journalists in all countries, we excluded those journalists who did not regularly deal with domestic politics and politicians. As a consequence, a number of questionnaires from journalists were deleted from the database used here (see Table 1).
Table 1: Sample Design for MPs and Journalists
	Belgium	Norway	Sweden
Time of survey	February-March 2006	February-April 2007	November 2007-March 2008
Response rate MPs	85 % (N=202)	51 % (N=87)	45 % (N=155)
Response rate journalists	66 % (N=299)	57 % (N=228)	52 % (N=195)
Journalists with 3 or more articles about national political actors	54 % (N=165)	81 % (N=184)	62 % (N=120)
The survey among political journalists and MPs contained a wide variety of questions regarding their perception of and relationship with each other. In this article we will mainly use the questions that relate to informal contacts and the personal nature of the relationship as well as several items that tap into the dimension of suspicion and distrust.
To analyse the informality of the relationship between MPs and journalists we use four items that go beyond the more "public" forms of contact connected to the process of news making such as press conferences or interviews. The first asks how often they have lunch with members of the other group. The second item measures whether or not they ask or give members of the other group advice related to their work, whereas the third reveals whether or not they consider any member of the other group as friends. Finally, our fourth item indicates whether or not they give out their personal mobile phone number to members of the other group. The four items are not combined in one factor or scale because they touch upon different aspects of the broad and somewhat diffuse concept of informality.7 Instead we will analyse and discuss them separately.
A central aspect of this article is the relationship between mutual perceptions and actual behaviour. Does the informality of the relationship also influence the image they have of one another? In order to answer this question we use 6 questions that measure suspicion and distrust. Both journalists and MPs were asked to indicate how much they agreed or disagreed on a 5 point scale to the following £ statements: (1) The mass media have too much power, (2) The motivation that drives most political journalists is the desire to exercise political power themselves, (3) The main thing journalists are after these days is a sensational story that draws a large audience, (4) Politicians often use journalists by leaking information to them, (5) Politicians would do anything to get attention from the media, and finally (6) It's more important for a politician to get coverage in the media than to work hard. Factor analysis confirmed that these 6 items loaded on two separate dimensions. The first factor can be said to measure suspicion towards political journalists and yj loaded between .76 and .81 (using varimax rotation). An index consisting of these three questions were constructed and a reliability test confirmed a satisfactory Cronbach's alpha of 0.7. The three last items loaded on a factor that taps into suspicion towards politicians. Although the correlation between these three items was not as high as the first three questions, we also created an index to measure
0 £
C >
suspicion towards politicians. These three items loaded between .62 and .82 and received a Cronbach's alpha of 0.6. The scale runs from 1 to 5 where high values qq indicate a high level of suspicion.
Results
The results will be discussed in two parts. First, we investigate whether the 00 informality of the relationship differs between countries and between MPs and political journalists within the countries. In a second part, we will focus on the effect of the informality of the relationship on the mutual (mis)trust.
Degrees of Informality between Journalists and Politicians
The overall impression of Table 2 is that the relationship between MPs and journalists is more formal then informal. Although it is very common to exchange personal mobile phone numbers in all these countries, there is only a minority who can be said to have a rather informal relationship with members of the other group. In Norway and Belgium, between 16 to 36 percent of MPs and journalists meet for lunch at least on a monthly basis. Such a group is almost non-existent in Sweden. Moreover, there is only a small minority who often or sometimes ask members of the other group for advice about their work. A larger share, but still a minority, considers members of the other group friends.
The data presented in Table 2 show clear differences between Belgium, Norway and Sweden. If we consider friendships and exchange of personal phone numbers the Norwegians are indeed less formal then the Swedes and the Belgians. But Belgians are most informal when it comes to meeting for lunch and giving each other advice. For all four indicators the differences between Norway and Belgium on the one hand and Sweden on the other are significant, both for journalists and MPs. Only the percentage of Swedish MPs that consider journalists as their friends is similar to the situation in Belgium and Norway. We will come back to these outspoken country differences in our conclusion.
Table 2: Characteristics of Intimate Relationship between MPs and Political Journalists (in percent).
	Belgium		Norway		Sweden	
	MP	Journalist	MP	Journalist	MP	Journalist
How often do they have lunch with members of the other group?						
Daily	0	0	1	1	0	0
Weekly	3	15	15	10	1	1
Seldom	72	65	55	47	36	32
Never	25	20	30	42	62	68
MPs: How often do you ask journalists for advice about your work? Journalists: How often do politicians ask your advice for their job?						
Often	1	3	0	0	1	1
Sometimes	17	22	10	14	5	4
Seldom	33	29	42	30	25	15
Never	49	46	48	57	69	80
Do they consider any member of the other group as friends?						
Yes	34	22	43	26	43	12
Do they give their personal mobile phone number to members of the other group?						
Yes	94	81	99	93	85	53
Lowest N	169	137	74	158	143	114
Besides the country differences Table 2 shows another clear pattern, namely the systematic difference between MPs and journalists related to having friends among the other group. This difference holds across all the tree countries. Whereas a considerable minority (approximately 40 percent) of the MPs do admit that they have friends among members of the other group, fewer journalists seem to consider politicians as friends. The Swedish journalists are the least "friendly" (12 percent) while 23 and 26 percent of Belgian and Norwegian journalists admit that there are politicians that they would consider as friends. This finding is in line with the so called adversary model, that journalists prefer to be somewhat more hostile towards, or at least not friends with politicians, in order to avoid being "in their pockets" (Blumler and Gurevitch 1981). For the three other indicators of the informality the difference between journalists and politicians was not significant or at least not in all countries.8
Next to country variation and the difference between the two main players in the relationship we test whether there is variation within each group for two of our four indicators using a linear regression analysis.9 Table 3 shows that politicians with more parliamentary experience have more frequent lunch meetings with political journalists. This supports the idea that an informal relationship between politicians and journalists is gradually built up over time. A similar pattern is found among the political journalists where journalists who report more on domestic politics have more frequent lunch meetings with politicians than those who do
Table 3: Explaining Informal Behaviour Based on Country and Political Experience (unstandardised OLS regression coefficients with standard errors)
u
S
i
0
£ *
i *
(R G C
fN 00
	Meeting	for lunch	Asking/giving advice	
	MP	Journalist	MP	Journalist
Country (Norway =ref. cat.)				
Belgium	-.105	.350***	.335***	.396***
	(.084)	(.067)	(.093)	(.075)
Sweden	-.384***	-.200**	-.117	-.240**
	(.074)	(.066)	(.083)	(.075)
Political experience (Years in parliament)	.018** (.006)		-.008 (.006)	
Journalistic focus on domestic		.086***		-.003
politics (# of political articles)		(.012)		(.014)
Frequency of contact	.252***	.160***	.204***	.217***
(never to daily)	(.035)	(.029)	(.038)	(.032)
Constant	.793	.467	.819	.667
	(.148)	(.120)	(.166)	(.134)
Adj. R2	.260	.319	.090	.169
N	377	517	419	562
Note: The two dependent variables are individual items. Meeting for lunch: Never=1, Seldom or monthly =2, Weekly =3, A few times a week=4, Almost every day=5. Asking advice: Never=1, Seldom=2, Sometimes=3, Often=4. Independent variables are categorised with the following values: Belgium=1, else =0. Sweden=1, else =0. Frequency of contact: Never=1, Once a month or less =2, A few times a month =3, A few times a week=4, Almost every day=5. Years in parliament: numeric ranging from 0 to 30 years. Number of political articles: numeric 3 to 10 (In how many of your last 10 articles was a domestic politician mentioned). The data have been weighted so that MPs and journalists from all countries have the same number of respondents: N=150 for MPs, N=200 for journalists. ***P>.00, ** P>.01, * P>.05
so more occasionally. Probably those journalists that report not only on policy issues, but also on a daily basis about the political game itself benefit most from fO their frequent informal meetings with politicians. However, they are not asked more for their expert advice than their colleagues that focus less on the "politique policienne." Not surprisingly more experienced politicians don't ask for more (or less) advice to journalists.
Table 3 also confirms our earlier findings about country differences, with the Swedes being significantly less informal in their interactions. Also the differences between Norwegian and Belgian actors are often significant, with especially the Belgian political journalists being more informal than their Norwegian colleagues.
Finally we test whether there is a strong relationship between how frequently they meet members of the other group and the relationships degree of informality. It is not surprising that there is indeed a positive relationship between how often they have personal contact with members of the other group and the different forms of informal behaviour. The more often MPs and journalists have personal contact the more likely it is that they have lunch and that they give or receive advice. There seem to be no systematic pattern which indicates that personal contact has a stronger impact on informal behaviour among MPs compared to the behaviour of political journalists. There are however, a few distinct differences between the various countries when it comes to the strength of these relationships. Basically frequent contact matters more for informal behaviour in Belgium then it does in Sweden, but also in Norway.
Relationship between Informality and Suspicion
Let us turn now to the relationship between informal behaviour and perceptions of suspicion and distrust. The framework of Nimmo (1964) suggests that the more informal the relationship is the less suspicious they would be towards the other group. In order to investigate the impact of an informal relationship on perceptions of suspicion and distrust we ran two multivariate regression analyses. The first with "suspicion of journalists index" as dependent variable, and the second with the index measuring suspicion towards politicians. The results are presented in Table 4.
There are not particularly large differences between the countries, with the exceptions of Belgians being considerably more suspicious or critical towards politicians' behaviour (see also Van Aelst et al. 2008). Generally both the Swedes and the Belgians seem to be somewhat more critical towards both groups compared to the Norwegians (who are treated as the reference category). Not surprisingly we also find that politicians are much more suspicious of media and journalists than what the journalists themselves are. Similarly, journalists are more suspicious and critical towards the politicians' behaviour compared to what politicians themselves are. Although there is a negative relationship between meeting for lunch and the degree of suspicion (the more often you meet for lunch the less suspicious you are), the general picture is rather that more informal does not lead to more trust. First of all there is no significant effect of having friends among members of the other group or exchanging personal mobile phone numbers. Those who have friends and give out their mobile phone number are just as suspicious as those who do not have friends among members of the other group or those who do not
Table 4: Explaining Suspicion towards Journalists and Politicians Based on Country and Relationship Characteristics (unstandardised OLS regression coefficients with standard errors)
	Suspicious/critical of journalists	Suspicious/critical of politicians
Country (Norway =ref. cat.)		
Belgium	.206** (.060)	.555*** (0.60)
Sweden	.063 (.064)	.092 (.064)
MP or Journalist	-1.334*** (.052)	.204*** (.050)
Relationship characteristics		
Meet for lunch	-.190*** (.038)	-.059 (.038)
Ask/give advice	.112** (.036)	.108** (.036)
Friends	-.025 (.045)	-.016 (.045)
Mobile phone	-.017 (.073)	-.085 (073)
Constant	5.176 (.193)	3.043 (.192)
Adj. R2	.432	.133
N	874	873
u
S
5
0 £
Note: The two dependent variables are two indexes each constructed from three items. Indexl (suspicious of journalists) is based on how much respondents agree or disagree on a 5 point scale to the following statements: (1) The mass media have too much power, (2) The motivation that drives most political journalists is the desire to exercise political power themselves, (3) The main thing journalists are after these days is a sensational story that draws a large audience. Index2 (suspicious of politicians) is based on how much respondents agree or disagree on a 5 point scale to the following statements: (4) Politicians often use journalists by leaking information to them, (5) Politicians would do anything to get attention from the media, and finally (6) It's more important for a politician to get coverage in the media than to work hard. The two indexes measure the average across the 3 items and allows for one missing value. For information on reliability measures see section on research design. Independent variables are categorised with the following values: Belgium=1, else =0. Sweden=1, else =0. MP =1, journalists =2. Personal contact: Never=1, Once a month=2, A few times a month=3, A few times a week=4, Almost every day=5. Meet for Lunch: Never=1, Seldom=2, Monthly=3, Weekly=4, A few times a week=5, Almost every day=6. Ask advice: Never=1, Seldom=2, Sometimes=3, Often=4. Have friends: No friends=1, Yes, one or two friends=2, Yes, three or more friends=3. Give out mobile phone: No=1, Yes=2. The data have been weighted so that MPs and journalists from all countries have the same number of respondents: N=150 for MPs, N=200 for journalists. ***P>.00, ** P>.01, * P>.05
share their mobile phone number. Furthermore, there is actually a significant and positive relationship between how often they ask or give advice to each other and how suspicious they are. In other words, the more often they ask or give advice the more suspicious they are. Finally, meeting for lunch makes MPs less suspicious about the motives of journalists but this effect doesn't seem to work the other way around. Interestingly journalists that have lunch more often with MPs are not significantly less suspicious about how politicians work. We will further elaborate on this finding in our conclusion.
Conclusion
fl
Over time many authors have studied the intriguing relationship between politico cians and journalists. They did so from different angels and perspectives, all trying
to clarify the potential impact of this intimate dance on law making and/or news making. This study has contributed to this research tradition by using a different, LO more comparative approach that resulted in mainly two relevant insights.
A first interesting finding is related to the significant country differences. Although we compared the informality of the relationship between three democratic corporatist countries, with a similar political media system, the results showed significant variation between the three cases. Especially, the Swedish politicians and journalists opt less for a close informal way of interaction and keep more distance compared to their Belgian colleagues, with the Norwegians holding a middle position. It seems that in Sweden the main actors in the political communication process are more cautious of establishing personal relationships, or put differently, keep the distinction between professional and personal interaction at a higher level. We believe this difference is mainly caused by the more professionalised relationship between journalists and MPs in Sweden, with more spokespeople mediating the relationship. We are aware of the fact that these are not the only plausible explanations for this difference,10 but hope we have made a first attempt to incorporate more specific "political communication" variables in comparative political communication studies.
A second, somewhat counterintuitive, finding of this study is related to the consequences of the informality of the interaction for the (lack of) trust in the relationship. Following Nimmo (1964) we expected that the more informal the relationship is, the less suspicion there would be towards the other group. This has proven to be hardly the case. Only one of the four indicators (lunch meetings) contributed to a more trustworthy view of the other side, and this was only so for the perceptions of politicians. In most cases there was no significant effect, or even a reverse effect (asking/giving advice). These results seem to show that love and hate can go hand in hand. Although politicians and journalists frequently interact informal and in some cases even become friends, this has little or no correlation with the deep-rooted suspicion that characterises the relationship. Politicians are forced to work together and are aware of the mutual benefits. Often they enjoy the personal attention from the "other side" but at the same time there are strong doubts whether one can really trust the other. Put in terms of Nimmo's model: A cooperative way of interaction can go together with a rather distrustful, competitive perception of the relationship (see also Kumar and Jones, 2005).
The more normative question is whether this finding is troublesome. We believe it is not. Politicians and journalists have different, partly conflicting, roles to play in society. Journalists need to be "close" to politicians to understand what is going on and to inform the public. At the same time they should control policy makers and critically examine the political process. It would be more worrying if the informal interactions made journalists "easy-going" and less critical. The same can be said for politicians who need journalist to reach out to the public, but would be naive if they thought journalists are the perfect channel to get their message across. Politicians should still try to reach out to voters directly without journalistic mediation. This being said, there is probably a limit to the degree of mistrust that can be considered healthy before mutual benefits become mutual drawbacks. But until that level is reached politicians and journalists seem to cope well with their rather "schizophrenic" relationship between trust and suspicion.
u
s 5
0
£ *
1 *
(ß 0 c
00
Without doubt further research is needed to fully understand the antecedents of the relationship between politicians and journalists as central actors in our democracy. We see particularly two lines of research worth pursuing. First, more comparative research is needed, including countries from other political media systems to better understand the value and interaction of structural features on the one hand and more specific features related to the political communication culture in a certain country on the other. Second, we need to know more about the consequences of this relationship for both news making and law making. This would mean combining survey and in-depth-interview data with a content analysis of news coverage and parliamentary debates and initiatives.
Notes:
1.	Gans (1979) was the first to use this metaphor, but since then it has been used repeatedly as an introduction to studies in political agenda-setting and journalist - source relations (see among others Bartels 1996; Strömback and Nord 2006; Reich 2006).
2.	The most enriching part of the Sigal's classical work on the "special relationship" between US reporters and officials is when the author (briefly) shows how it differs from the British press-government relation (1973, 131-133).
3.	For a similar line of reasoning about (comparative) politics in general see Helmke and Levitsky 2004.
4.	In 2007, newspaper sales per 1,000 adult citizens were about 601 in Norway, 466 in Sweden, and 173 in Belgium (World Press Trends, 2008).
5.	The Belgian survey was only conducted in the Dutch speaking part of the country (Flanders), containing 60 percent of the population.
6.	The surveys in the three countries were coordinated by: by Michiel Nuytemans, Stefaan Walgrave, and Peter Van Aelst (University of Antwerp) in Belgium; Toril Aalberg and Ann Iren Jamtry (Norwegian University of Science and Technology) in Norway; and Jesper Strömbäck and Adam Shehata (Mid Sweden University) in Sweden.
7.	Althought the four variables are all positively correlated to each other, these correlations are not strong enough to create a reliable index of informality. Factor analysis confirm that they all load on one dimension (between .56 and .73) but they receive a rather low Cronbach's alpha of 0.55. The strongest relationship is found between having friends and asking or giving advice. The weakest relationship is between the variable measuring the exchange of personal mobile phone numbers and the other variables. Excluding the variable with the weakest correlation does not improve Cronbach's alpha sufficiently.
8.	The frequency of having lunch was only significant different (Chi-square) between MPs and journalists in Belgium, exchanging phone numbers was significant different in Belgium and Sweden, and asking/giving advice was in non of the three countries significant.
9.	While our dependent variables are not prefectly linear, we also ran ordinal regression analysis showing very similar results. Therefore we feel conficent to present the OLS regression analysis, which is much easier to interpret.
10.	We checked for several alternative explanations such as the overall culture of interaction in the different countries or the level of professionalisation of the journalistic profession. However, none of these factors provided a solid explanation.
References:
Aalberg Toril, Peter Van Aelst, and James Curran. 2010. Media Systems and the Political Information Environment. Supply and Demand of News and Current Affairs across Countries and Time. International Journal of Press/Politics 15, 3, 255-271. Allern, Sigurd. 2007. From Party Press to Independent Observers? An Analysis of Election Campaign
Coverage Prior to the General Elections of 1981 and 2005 in Two Norwegian Newspapers. Nordicom Review 28, 63-79.
Baisnée, Olivier 2002. Can Political Journalism Exist at the EU Level? In R. Kuhn and E. Neveu (eds.),	00
Political Journalism: New Challenges, New Practices, 108-127. London: Routledge.
Barnett, Steven and Ivor Gaber. 2001. Westminster Tales: The 21st Century Crisis in Political Journalism. London: Continuum.
Bartels, Larry M. 1996. Politicians and the Press: Who Leads, who Follows? Paper presented at the Annual Meeting of the American Political Science Association, APSA in San Francisco.
Bennett, W. Lance, Regina G. Lawrence, and Steven Livingston. 2007. When the Press Fails. Political Power and the News Media from Iraq to Katrina. Chicago: University of Chicago Press.
Blumler, Jay G. and Michael Gurevitch. 1981. Politicians and the Press: An Essay on Role
Relationships. In D. D. Nimmo and K. R. Sanders (eds.), Handbook of Political Communication, 467493. Beverly Hills, CA: Sage.
Charron, Jean. 1994. La production de l'actualité. Quebec: Boréal.
Cook, Timothy E. 1989. Making Laws and Making News. Media Strategies in the US House of Representatives. Washington DC: The Brookings Institution.
Cook, Timothy E. 1998. Governing with the News: The News Media as a Political Institution. Chicago: University of Chicago Press.
Crouse, Timothy. 1974. The Boys on the Bus. New York: Ballantine Books.
Dahl, Robert A. 1998. On Democracy. New Haven: Yale University Press.
Davis, Aeron. 2007. Investigating Journalist Influences on Political Issue Agendas at Westminster. Political Communication 24, 2, 181-199.
Davis, Aeron. 2009. Journalist-Source Relations, Media Reflexivity and the Politics of Politics Journalism Studies 10, 2, 204-219.
Donsbach, Wolfgang and Thomas Patterson. 2004. Political News Journalists: Partisanship,
Professionalism, and Political Roles in Five Countries. In F. Esser and B. Pfetsch (eds.), Comparing Political Communication. Theories, Cases and Challenges, 251-270. Cambridge: Cambridge University Press.
Ericson, Richard, Patricia Baranek, and Janet Chan. 1989. Negotiating Control: A Study of News Sources. Milton Keynes: Open University Press.
Ferree, M. Marx, William A. Gamson, Jürgen Gerhards, and Dieter Rucht. 2002. Four Models of the Public Sphere in Modern Democracies. Theory and Society 31, 3, 289-324.
Gans, J. Herbert. 1979. Deciding What's News: A Study of CBS Evening News, NBC Nightly News. New York: Pantheon.
Hallin, C. Daniel and Paolo Mancini. 2004. Comparing Media Systems. Three Models of Media and Politics. Cambridge: Cambridge University Press.
Helmke, Gretchen and Steven Levitsky. 2004. Informal Institutions and Comparative Politics: A Research Agenda. Perspectives on Politics 2, 4, 725-740.
Jones, Nicholas. 1995. Soundbites and Spin Doctors. How Politicians Manipulate the Media and Vice Versa. London: Cassell.
Jones, Nicholas. 2006. Trading Information: Leaks, Lies and Tip-offs. London: Politico's.
Klingemann, H. Dieter. 2005. Political Parties and Party Systems. In J. Thomassen (ed.), The European Voter: A Comparative Study of Modern Democracies. New York: Oxford University Press.
Kuhn, Raymond. 2010. "Les medias, c'est moi." President Sarkozy and News Media Management. Fr Polit 8, 4, 355-376.
Kumar, M. Joynt and Alex Jones. 2005. Goverment and the Press: Issues and Trends. In G. Overholser and K. H. Jamieson (eds.), The Press, 226-247. Oxford: Oxford University Press.
Langer, I. Ana. 2010. The Politicization of Private Persona: Exceptional Leaders or the New Rule? The Case of the United Kingdom and the Blair Effect. International Journal of Press Politics 15, 1, 60-76.
Larsson, Larsake. 2002. Journalists and Politicians: A Relationship Requiring Manoeuvring Space. Journalism Studies 3, 1, 21-33.
Laver, Michael and Norman Schofield. 1998. Multiparty Government: The Politics of Coalition in Europe. Ann Arbor: University of Michigan Press.
Mancini, Paolo. 1993. Between Trust and Suspicion: How Political Journalists Solve the Dilemma.
European Journal of Communication 8, 1, 33-51. Negrine, Ralph and Darren G. Lilleker. 2002. The Professionalisation of Political Communication.
European Journal of Communication 17, 3, 305-323. Neveu, Erik and Raymond Kuhn. 2002. Political Journalism: Mapping the Terrain. In R. Kuhn and E.
Neveu (eds.), Political Journalism. New Challenges, New Practices, 1-21. London: Routledge. Nimmo, Dan. 1964. Newsgathering in Washington. A Study in Political Communication. New York: Atherton Press.
0stbye, Helge and Toril Aalberg. 2008. Media and Politics in Norway. In J. Stromback, M. 0rsten and T. Aalberg (eds.), Communicating Politics. Political Communication in the Nordic Countries, 83-102. Gothenburg: Nordicom.
Pfetsch, Barbara. 2001. Political Communication Culture in the United States and Germany. Press/ Politics 6, 1, 46-67.
Reich, Zvi. 2006. The Process Model of News Initiative: Sources Lead, Reporters Thereafter.
Journalism Studies 7, 4, 497-514. Rosenstiel, Tom. 1993. Strange Bedfellows, How Television and the Presidential Candidates changed
American Politics 1992. New York: Hyperion. Schudson, Michael. 2003. The Sociology of News. New York W. W. Norton & Company Inc. Sellers, Patrick. 2010. Cycles of Spin. Strategic Communication in the US Congress. New York:
Cambridge University Press. Sigal, Leon V. 1973. Reporters and Officials: The Organization and Politics of Newsmaking. Lexington,
MA: D.C. Heath and Company. Stromback, Jesper. 2008. Four Phases of Mediatization: An Analysis of the Mediatization of Politics.
International Journal of Press Politics 13, 3, 228-246. Stromback, Jesper and Lars W. Nord. 2006. Do Politicians lead the Tango? European Journal of
Communication 21, 2, 147-164. Thomassen, Jacques and Rudy Andeweg. 2004. Beyond Collective Representation: Individual
Members of Parliament and Interest Representation in the Netherlands. The Journal of Legislative Studies 10, 4, 1-23.
Thomassen, Jacques and Peter Esaiasson. 2006. Role Orientations of Members of Parliament. Acta Politica 41, 3, 217-231.
Thompson, B. John. 1995. The Media and Modernity. A Social Theory of the Media. Cambridge: Polity Press.
Van Aelst, Peter, Kees Brants, Philip Van Praag, Claes De Vreese, Michiel Nuytemans, and Arjen Van Dalen. 2008. The Fourth Estate as Superpower? An Empirical Study on Perceptions of Media Power in Belgium and the Netherlands. Journalism Studies 9, 4, 494-511. Van Aelst, Peter. 2007. Toeschouwer, speler of scheidsrechter? Een studie naar de rol van de media in de
verkiezingscampagne van 2003. Brugge: Vanden Broele. Van Aelst, Peter, Adam Shehata, and Arjen Van Dalen. 2010. Members of Parliament, Equal
Competitors for Media Attention? An Analysis of Personal Contacts between MPs and Political Journalists in Five European Countries. Political Communication 27, 3, 310-325. ^	Weaver, David, ed. 1998. The Global Journalists: News People around the World. Cresskill, NJ: Hampton
Press.
CWirth, Werner and Steffen Kolb. 2004. Design and Methods of Comparative Political Communication Research. In F. Esser and B. Pfetsch (eds.), Comparing Political Communication. Theories, Cases and Challenges, 87-114. Cambridge: Cambridge University Press.
u
S
£
00 00