Centralizing Government Communication? Evidence from Finland and Sweden

How governments manage their communication is one crucial indicator of the balance of power within the cabinet and inside the executive branch as a whole. Existing research offers few insights into the process by which governments come to choose one form of communication over another and about the factors driving centralization. This article addresses this gap through a comparison of two countries, Finland and Sweden, examining not only the organizational forms of government communication but also the causal mechanisms at work. Combining theoretical lessons from studies in political communication, political science, and public administration, it develops a centralization argument, focusing on the centripetal factors facilitating coordination and control. Drawing on over 40 interviews with journalists and political or media advisors in the two countries and on government documents, the article offers clear evidence of a trend toward centralization, particularly in Sweden. This trend should be understood as part of a broader process whereby prime ministers and their offices establish stronger control of the entire executive branch

Governments are investing more resources in media relations and in communication with the public, both in terms of staff and strategic leadership.The past few decades have seen a gradual transformation from more horizontal to more hierarchical forms of coordination and management of communication, involving the executive center.The concept of government communication has become an umbrella term for denoting these complex patterns of communication by national governments, involving a variety of actors and structures.Here, government communication is defined broadly as an organization or practices for communication purposes exercised by and through political executive authority structures (cf.Canel and Sanders 2013, 2014, 2016;Hiebert 1981;Pasquier 2012;Sanders 2020).Hence, this article focuses on government communication as central executive government communication, encompassing the executive as the cabinet, the prime minister's office (PMO), the line ministries, and all units for communication at this central executive level.
While communication is central to governance, existing research offers few insights into the process by which political executives come to choose one form of communication over another and about potential explanations for centralizing government communication upward to the PMO.Following Ecker-Ehrhardt (2018, 522), we define centralization of government communication as "the development of central administrative capacities tasked with regularly communicating with non-governmental audiences."The research question of this article is what factors explain the centralization of government communication in Finland and Sweden?It moves beyond the state of the art in three prominent ways.First, previous research has primarily addressed larger Anglo-American systems, and hence this article addresses this gap through a systematic and comparative assessment of the two Nordic countries.Second, the main focus is on the causal mechanisms explaining the degree of centralization, topics that, until now, have been largely unexplored by scholars.Through highlighting the pivotal role of party-political media advisors, this article also contributes to the literature on relationships between civil servants and political advisors (e.g., Shaw and Eichbaum 2018).Third contribution of the article is theoretical: it argues that government communication studies have so far not paid sufficient attention to lessons from the disciplines of political science and public administration.Centralization in communication cannot be separated from broader developments in party politics and administration structures.In particular, it highlights the complementary roles of changing media structures, the logic of coalition governments, and horizontal coordination inside the executive branch that contribute to more centralized forms of communication.Examining these factors together, they provide governments and especially prime ministers (PMs) strong incentives to centralize communication in order for the government to speak with one voice and to control the information flows.Hence, this article speaks to research on executive power and empowerment, and highlights the role of more top-down government communication as a factor driving that process (e.g., Poguntke and Webb 2005).
It is also essential to explain why scholars should be interested in examining the organization of government communication and particularly whether it is centralized or not.The form and evolution of government communication are are crucial indicators of the balance of power within the cabinet and inside the executive branch as a whole, a dimension neglected by political scientists.And, as explained above, existing research produced by political communications scholars, while paying attention to the professionalization of government communication, has not really examined factors driving centralization.While one cannot exactly measure the weight of the variables incorporated into the theoretical framework, it is hoped that, through including factors normally not considered by scholars, this article paves the way for further comparative studies on explaining the form of government communication.Centralization also has normative implications, a question returned to in the concluding discussion.
The next section contains the theoretical framework and explains the definition and measurement of centralization used in this article, while the case selection and data are justified thereafter.The results are then reported in the following section.They show that there is a clear trend of centralization in government communication.However, this trend is definitely more evident in Sweden than in Finland.Both countries exhibit stronger control from the PMO, but this development needs to be understood as part of the broader empowerment of PMs and their offices.The findings also emphasize the role of party-political advisors and their often problematic division of labor with civil servants responsible for official government communication.The concluding section summarizes the main results, discusses their implications, and closes with a recommendation for more comparative research, especially in non-Western cases.

Theoretical Framework: Incentives for Centralizing Government Communication
Government communication is a central feature of politics today.To this end, structures are established and staff appointed to communicate what the government does.Yet, given this basic demand for communication, why do political executives choose one form of organization rather than others?And, more directly in terms of the argument of this article, which factors facilitate the centralization of government communication?Drawing on insights from three different strands of literature-political communication, political science, and public administration-it is argued in this section that governments throughout the world have powerful incentives for transferring communication activities to the PMO, the nerve center of modern cabinets.The three main factors explaining centralization are changing media structures, the logic of coalition governments, and horizontal coordination inside the executive branch.
The overwhelming majority of research on government communication has been produced by political communications scholars.While this strand of research has not really focused on centralization as such, it has nonetheless offered evidence in favor of centralization of communication.The volume edited by Sanders and Canel (2013), Government Communication, is the most comprehensive comparison.Summarizing the 15 country studies, Sanders and Canel (2013, 290-1) report, with particular attention to organizational structure: "As communication channels and objectives have become more complex, including-for example-the development of social media and citizen engagement goals, so governments' organizational structure has become more specialized in a number of countries… The units tend to be centrally located within government with communicators assigned specific tasks." In almost all cases, the creation of centralized units showed some development of strategic capacity (Sanders and Canel 2013, 299).Another common theme was the rising number of those employed in communication tasks by governments, including in central executive government communication (Sanders and Canel 2013, 303-4).As for nonpartisanship, where communication is understood as being directed to serve the public rather than the political party in power, Sanders and Canel (2013, 302) report that "non-partisanship of government communication receives the most comprehensive underpinning in Sweden, Germany and the United Kingdom."One indicator here concerns political appointments and the role of government spokespersons, with significant variation between the countries.Other research also indicates more centralized communication structures and strategic thinking (see e.g., Marland, Lewis, and Flanagan 2017;Peters 2016).In brief, the politicization of government communication requires intense control and this, along with advances in information and communications technology and the more decentralized media environment, drives centralization.
With regard to Sweden, based on the country chapter, Sanders and Canel (2013, 306) argue that "it has not developed a high strategic capacity: the changes reported by Falasca and Nord have produced a government communication structure that is flat, decentralized and rather fragmented."In light of the findings reported in this article this was an underestimation, both of the state centeredness of Swedish political culture and of the centralization and coordination of the government communication structure.However, Falasca and Nord (2013, 40-2) suggest that government communication is becoming "more politicized" while concluding that, "government communication in Sweden has become professionalized to a considerable extent due to the expansion of the communication organization and the number of communication professionals as well as to the new emphasis on communication practices and strategies."There is no similar previous empirical article of Finnish government communication.
Research has emphasized professionalization, including the institutionalization of staff functions, as an indicator of centralization of governmental communication (Papathanassopoulos et al. 2007;Sanders and Canel 2013).Professionalization denotes the (social) process whereby an occupation-such as in communication-transforms itself into a true profession and by which the qualified are demarcated from the unqualified.The precise meaning and measurement of the term remain contested, and it also describes the development of a separate group of professionals with their own values and standards (Freidson 2001).Communication professionals can be expected to establish their own professional knowledge and status, and they are not elected politicians but employed in the government offices to perform communication duties.
The literature on government news management is relevant here (see e.g., Arceneaux, Borden, and Golan 2020).To quote Pfetsch (1998, 82): "Every government maintains a more or less political public relations apparatus that varies in the degree to which professional political marketing methods are used to further government policies."Seymour-Ure (2000, 2003) also observes an ongoing process of centralization partly driven by the institutionalization of staff functions, notably the office of press secretary to premiers.This development impacts relations between political staff and civil servants and may disturb such long-established principles as civil service nonpartisanship and ministerial responsibility.And, most notably, it may produce increasing centralization of government media management.Here is a crucial connection with the more diversified media landscape, which arguably facilitates centralization within the executive.Intended or not, responding to media requires the executive to control information in more complex processes of steering or governance (see e.g., Dahlström, Peters, and Pierre 2011;Jacobsson, Pierre, and Sundström 2015).Indeed, from Sweden there are references to so-called "policy professionals" (Garsten, Rothstein, and Svallfors 2015;Svallfors 2017).This broad category, to which press secretaries or media advisors belong, consists of political employees who often have a background in party politics and PR/communication; they are not politicians and not elected to any office.It is also shown that press secretaries are shaping the public statements of ministers and are working both against and with the media and may punish or reward journalists for their work.
To reiterate, governments and political actors have been forced to develop strategies of news management that mainly are entrusted to spin doctors and media professionals.In the process, sources of political information have become more and more professionalized (Mancini 1999, 240).Research on professionalization of political communication usually studies elections and not between elections government communication (Canel and Sanders 2014, 103;Holtz-Bacha 2016).However, it should be pointed out that such communication activities and strategies are not easily separated and resources of the executive may be used to win elections, especially as politics resembles a "permanent campaign" (Blumenthal 1980).Overall, beyond mapping the professionalization of government communication and linking such developments to changes in media systems, existing literature has been rather weak in identifying causal mechanisms that would explain centralization of government communication.
Moreover, as Canel and Sanders (2014, 100) note, previous research-mainly monopolized by political communication scholars-has "explored the exercise of government communication in the context of and in response to highly mediated environments," thus downplaying the role of variables not usually discussed by political communication scholars.
Moving to insights from political science, the research by political communication scholars outlined above obviously recognizes that government communication, in essence, is about politics: how to either attract or avoid media coverage is often crucial in terms of cabinet survival and the careers of the PM and other political actors.However, beyond the rise of populism that body of work has paid very little if any attention to significant changes in party systems and the composition of governments.Party systems have in recent decades become more fragmented, with new cleavages and issues resulting in the formation of new parties.Particularly, noteworthy has been the emergence of various populist or radical right parties across Europe, with several such parties also entering governments.At the same time, traditional mainstream parties, not least center-left Social Democratic Parties, have lost votes.As a result, coalition governments in the twenty-first century are on average more likely to be ideologically heterogeneous, often including new parties that have no previous experience of leading the country (see e.g., Wolinetz and Zaslove 2018).
This presents a potentially serious dilemma for government communication.The PM and PMO strive for streamlined and cohesive communication, while individual parties or ministers may have their own needs for signaling their views to their electorates, especially in coalition cabinets.Naturally, also line ministries and their communication departments have good reasons to oppose centralization to the PMO.Approaching the equation from the perspective of the PM and her office, the PM thus faces the challenge of ensuring that governments speak with one voice.1This provides powerful incentives for central control and coordination, with the PMO increasingly responsible for government communication.Here, one needs to remember that political science scholars have noted a clear tendency toward empowerment of the PMs and their offices, or what Poguntke and Webb (2005) have labeled the "presidentialization" of parliamentary regimes.Several factors have contributed to this development, from the broader personalization of politics (Karvonen 2010;Langer 2011), including extensive media coverage, to increasing international contacts of the PM (Johansson and Tallberg 2010).Hence, there is a need to pay attention to the interplay between government communication and the broader empowerment of the PM, an angle so far neglected by political scientists.
Public administration scholars have in turn documented a clear increase in horizontal coordination by governments across the world.Mainly, to counteract "ministerial government" or departmentalization of decision making, where each minister controls issues under her jurisdiction, governments have introduced mechanisms that improve coordination between ministries.Such instruments include various intersector policy programs, strategy documents, budgetary frameworks, or midterm review sessions.The horizontal coordination processes are, in most cases, led by the PMO.And even without such horizontal programs, governments have invested resources in the steering capacity of the PM.This provides further incentives for centralizing communication to the PMO (see e.g., Dahlström, Peters, and Pierre 2011;Kolltveit 2015;Peters 2016).
Bringing these various threads together, there are three interrelated factorschanging media structures, logic of coalition governance, and increasing horizontal coordination-that facilitate centralization of government communication to the PM and her office.Research on political communication has indeed already offered evidence of such centralization and professionalization, including the use of press secretaries and other media staff, linking it primarily to the changing media structures and the need to control information flows.Political science scholarship in turn shows how party systems have become more fragmented and unpredictable, with cabinets thus more often multiparty coalitions and including new or ideologically more radical parties that have little or no previous experience of holding office.And public administration research has produced strong evidence of increased steering capacity of the PMs and their offices, not least through various types of horizontal coordination instruments.Examined together, these factors all point in the direction of centralized control of communication in order for the government to speak with one voice and to control information flows.
But before moving to the empirical section, it is essential to explain how centralization or decentralization are measured.In his study of the communication structures of international organizations, Ecker-Ehrhardt (2018) emphasizes management tasks (primarily coordination and strategic planning of communication) and the establishment of specific organizational units or departments for carrying out communication.2 Adjusting his approach, this article uses two primary indicators.The first is the distribution of resources and division of labor between the PMO and the line ministries: whether staff is de facto working for the PMO or the individual ministries,3 and how independent are line ministries vis-à-vis the PMO when carrying out communication duties.
Second, it focuses on professionalization and the extent to which the government and specifically the PMO utilizes (party-political) media advisors and other designated communication experts and structures.The use of party-political media advisors is interpreted as a sign of professionalization and strategic planning or behavior, as it indicates that the PMO and the cabinet as a whole is investing resources in news management and media relations.
Examining the two indicators together, the more there is coordination and control by the PMO and the less line ministries carry out (important) communication, the more centralized the approach-and vice versa.A delegated communication structure where the line ministries have more important responsibilities can also qualify as a de facto centralized system, provided that there is active coordination from the PMO.It should also be stressed that it can be difficult to locate exactly where power lies or how much influence particular actors have, especially regarding developments over time.To maximize the reliability of our findings and to get behind the scenes and to "go beyond plain numbers," special attention was thus paid to the selection of the interviewees that, between them, possess considerable experience of how government communication has developed in Finland and Sweden.

Case Selection and Data
Case selection is driven by both similarities and differences between the countries.On the one hand, as two neighboring Nordic countries, Finland and Sweden obviously share many commonalities in terms of political cultures, state administration, party systems (including the recent rise of populist or radical right parties), and media structures.On the other hand, the two countries also differ in important respects, not least regarding government composition.Sweden was for a long time mainly ruled by Social Democratic single-party cabinets but has since the 1990s moved toward rule by coalition governments which have been the standard type of government in Finland.Significantly, in Finland these ideologically heterogeneous cabinets have for the most part brought together parties from both the right and left, whereas Swedish coalition governments have followed the bloc model, consisting of either left or right parties.Finland also had, until the turn of the millennium, a powerful presidency.As far as media systems or models of the media-state interaction are concerned, Finland and Sweden belong to the democratic corporatist model (Hallin and Mancini 2004).Yet, while similar in some respects, the media systems also differ in their legacy relationships between politics and media.
Empirically, the analysis draws on interviews and documentary evidence.The empirical approach has descriptive or exploratory (how government communication is structured in the two countries) and explanatory (which factors explain the degree of observed centralization) objectives.Focus is on developments since the 1990s, mainly because most of the interviewed persons had relevant work experience from that decade onward.Evidence is summarized from a unique series of over 40 interviews with acting journalists, press secretaries, media or political advisors,4 and top-level civil servants.Overall, the interviewees possess extensive experience of government communication: most of the journalists have covered the government and parliament since the 1990s or even longer, while the interviewed government officials, and particularly the political advisors, have on average held their positions for a shorter time.Almost all interviewees held their respective positions at the time of the interviews, meaning that they could reflect on how government communication had changed.The reason for also interviewing journalists is that they are the bestplaced sources for evaluating the importance of various personnel on the government side as they follow closely what the government does and how it communicates about its activities.The Finnish case draws on a total of 21 in-depth interviews with political journalists from all major media (n = 12), civil servants at the PMO who either work primarily on communications matters or supervise such efforts as senior-level bureaucrats (n = 5), and political advisors who have worked for the recent PM (n = 4).In Sweden, interviews were conducted with journalists (n = 10) and government press secretaries (n = 11).The semistructured interviews were carried out in January to May 2016 in Finland and between late 2014 and early 2017 in Sweden, although most of the Swedish interviews also took place in spring 2016.The interviews were based on an identical set of questions that were then supplemented with additional questions depending on the specific expertise of the interviewees.Each interview took one hour on average, the interviewees were guaranteed strict anonymity, and the recorded interviews were transcribed afterward.
Both country cases also draw on a close reading of all relevant documents produced by the Finnish and Swedish governments since the 1990s.These documents were utilized to understand the formal structures of government communication and the communication strategies of the Finnish and Swedish governments.The list of references includes only those official documents most relevant to our analysis.In this article the term "government communication" refers to all communication activities and media contacts of the PMO and the executive, social media included.5

Finland: A Stronger Role for Political Advisors
The Finnish case establishes a relatively clear trend toward centralization of government communications to the PMO, but certainly weaker than in Sweden.Moreover, it underlines the central role of party-political advisors at the PMO, while the civil servant media staff there seek to maintain a neutral position.
As a result of constitutional changes enacted since the late 1980s, the government has emerged from the shadow of the president as the main executive.The PM is the political leader of the country, and how the government and the PM handle their communications is hence much more important than under the old constitution.Moreover, it is necessary to remember that Finland is typically governed by ideologically heterogeneous multiparty coalitions that often bring together parties from the left and right (Karvonen 2014;Karvonen, Paloheimo, and Raunio 2016).This should also create pressure toward more centralized coordination as otherwise potentially contradictory messages from the line ministries or the junior cabinet parties might jeopardize government decision making.
To begin with, one can distinguish between political communication and civil servant communication.The former refers to the immediate political staff of the PM and ministers; that is, the special advisors, which includes political assistants or political advisors, and the latter to the official communication by the civil servant communications staff of the PMO, called the Government Communications Department (GCD), and other ministries' communications departments.The political advisors normally come and go with each minister, whereas the civil servants are bureaucrats who often spend their entire work careers in the same ministry.Since 1970, all ministers have had their own special political advisors, distinct from civil servants in the ministries, and since 2005 ministers may also have their own state secretaries.The number of such party-political advisors of ministers has increased substantially, with most ministers having a state secretary and on average two to five special advisors.Of the approximate ten advisors working for the current PM Sanna Marin, one deals specifically with media relations while another focuses more on internal government communication.Otherwise, it is still rare for a minister to have an assistant that only deals with press matters.Instead, the assistants have broad duties, including acting as policy advisors.Yet, our analysis below clearly indicates that the advisors have become more important players in handling relations with media.
Journalists see the role of political advisors as much more important than the role of civil servants.In terms of the PM, his chief advisor is also the main channel through which he can be reached when direct access is not possible.In contrast, the GCD is seen as having an almost ceremonial role when it handles official government communications.Moreover, one advisor's account of their role is very telling of the position in terms of power: "We, the political advisors, function as a filter between the PM and the GCD."However, according to both civil servants and political advisors, the main ministers (typically the leaders of cabinet parties) have the last say on communication strategies.As a rule of thumb, the more salient the issue, the more top-down coordination there is.The GCD can participate in planning and might make recommendations, but the politicians and their advisors decide on the political substance, and to some extent, on the timing.
Civil servants and political advisors provided a very coherent account of the division of labor between the PMO and the ministries.The GCD is responsible for the PM's communications and governmental communications as a whole, while the communication departments of the line ministries handle their own ministry's/minister's communications.The GCD is responsible for all the main government press conferences that are held in the PMO's conference room.In essence, the more important matters are handled by the PMO/GCD, whereas communications regarding more day-to-day policyspecific issues are the dealt with by the ministries.The journalists do not regard the other ministries or ministers as that important to follow.Some journalists described a hierarchy of importance: after the PMO comes the ministry of finance-and after those, all the rest.This also manifests itself in a tendency of the journalists to focus on the political advisors even more in the case of the other ministries.Thus, as a whole, the civil servant communication staffs of the line ministries are the least important actors in the whole communications scheme.
However, when specifically asked about centralization, the verdict seems somewhat divided.Some agreed that centralization has occurred, particularly through the strategic management doctrines discussed below, whereas others regarded the system as quite decentralized: the ministries do produce a lot of communications of their own, and numerically speaking, they also have most of the government media staff.As one interviewed journalist put it: "It is becoming more and more… like, the PMO's communications are the most important, and at the [other] ministries, they are a bit subordinate to that.They have re-organized communications, you know."Moreover, a civil servant argued that "centralization has not occurred so far.Only when there are horizontal and important matters, they should be labelled as strategic and shared.Thus, they can be centralized."Referring to the Sipilä cabinet appointed after the 2015 elections, one interviewee pointed out that The ministers said from the get-go that they want to assemble communications resources together, even to the extent of appointing political communication people-that is, 'model Sweden'… We have had a lot of talk about if we should have fewer civil servant media staff, and more in the ministers' staffs… you could say, communication professionals.Indeed, in 2016 the working group appointed by the PMO to assess the communications of the government and individual ministries also essentially called for more centralization of communications: Based on its findings, the working group calls for greater coordination between the government and the ministries in how communications activities are carried out… If the recommendation is followed, all the Government communications will be integrated with the primary responsibility assumed by the Prime Minister's Office.(Valtioneuvoston Kanslia 2016) One of the interviewed political advisors also saw increasing pressures toward centralization: There has been an attempt to condense it, but, in my opinion there is much work to do in this regard.Sometimes there was an idea that maybe there could be only one Communications Department, or pool, in the PMO, from which communications staff would be attributed to the ministries.To sort of… disband the ministries' own CDs.Well, this was not done in the state administration reform [valtionhallinnon uudistus] for some reason, but… gradually it will change.At the moment the ministries have good resources to do things.
Overall, an impression emerges that some kind of balance should be maintained between central control and delegation as not everything can or should be done by the PMO: "the ship is just too big for that." When interpreting the findings, one should thus exercise caution.After all, there is no longitudinal interview data, nor is there any previous research on the Finnish case.However, the trend toward more centralized government communication is nonetheless relatively clear.Any causal mechanisms should also be approached with care, but three mutually enforcing explanations emerge.First, inside the government, the role of the PMO has overall become considerably stronger in recent decades.The PM has become the political leader of the country, including in EU affairs and in foreign and security policy, which is codirected between the president and government.As a result, the staff and overall resources of the PMO have grown significantly, and centralization of communication should be viewed as a significant dimension of that development.Second, as mentioned above, a typical Finnish cabinet is a surplus coalition bringing together parties from the left and right.Hence, centralizing communication more to the PMO appears logical if the goal is to ensure that the government speaks with one voice.Indeed, the governments appointed since the 2011 parliamentary elections have been ideologically very heterogeneous, thus strengthening the need for centralized control.Third, the interviewed civil servants saw that the centralization to the PMO was facilitated by strategic management thinking in recent governments.The cabinets appointed since the turn of the millennium have invested resources in improving coordination and strategic planning inside the entire executive branch, for example through various intersector policy programs and government strategy documents (Kekkonen and Raunio 2011).Obviously, this provides a further incentive for centralizing government communications to the PMO. 6n sum, the case of Finland offers support for the centralization argument.The PMO has stronger resources for communication and the PM and individual ministries make increasing use of party-political (media) advisors.The more salient the issue, the more control there is from the center.Yet, as shown next, the level of centralization is definitely higher in Sweden, both regarding organizational structures and strategic leadership from the PMO.

Sweden: Professionalized Coordination from the Center
The Swedish case also identifies a trend of centralization in government communication.It documents major changes in the system of government communication and some of the important transitions in political/governmental communication that have taken place.In brief, professionalization and increased resources dedicated to government communication make news management more efficient and has centralizing effects on executive systems, strengthening the executive center.
Government communication can be characterized as strong.A measure of this is the extent to which resources for this purpose have increased, both in terms of funding and staffing.The resources allocated for government communication have grown significantly over the past 50 years.A very concrete expression of this decades-long process of change is the significant expansion in staff, including staff for press and information, at the Government Offices (GO) and the PMO specifically over this period of time.In the 1960s, there was only one press secretary employed, and the first was then employed for the PM in 1963, when the PMO could still be described as very slim with a staff of just a few.Over the following decades, the number of staff increased to 189 at the end of 2018 (down from a peak of 203 in 2012) (Government Offices of Sweden 2018).
Another measure concerns the structure of government communication; that is, its organization at different levels and elements.This means the structure of information or communication management within the government-how different parts of the government are coordinating their communication efforts.The system of government communication in Sweden has undergone a series of major changes over time.These have strengthened the center.Over the past three decades, the PMs have brought a shift to a more centralized structure for press/media coordination.The Social Democratic-Green Government that came to office in 2014 brought about a further strengthening of the government communication machinery with centralization to the PMO.Two novelties were that all press secretaries became employed by the PMO (and not by specific line ministries) and morning meetings on a daily basis were introduced that were led by the chief press officer at the PMO.Communication has been coordinated to maximize the attention for the issues the government wants to promote.The unit for media coordination located at the PMO has been upgraded with a strengthened position for the chief press officer.Altogether, there is one chief press officer with deputies, plus over 30 press secretaries and a staff of press assistants, press coordinators, and press communicators.Press assistants (and press communicators) are nonpolitical but work closely with the press secretaries and the political staff.
Organizationally, all press secretaries are based at SB Press, directly under the PM.This is also where all ministerial press secretaries in the GO are employed.Press secretaries are all political appointees employed under the GO agreement on politically appointed staff, such as state secretaries, press secretaries, and political advisors.Out of almost 4,800 employees in the GO, some 200 are political appointees (Government Offices of Sweden 2018).
There are four press secretaries to the PM.Two of them do regular press secretary work.One works mainly with social media.One works with communication matters relating to the EU and foreign policy, including the incoming international visits.There is also a press assistant, alongside the chief press officer and the deputy chief press officer.Moreover, there is the deputy chief press officer for the Green Party and a press secretary for the (Social Democratic) minister for EU affairs, both of whom are also based at the PMO.Furthermore, epitomizing the strengthening of the center, a new position was established in 2017 for the PM: media strategist, which is at the level of political advisor with responsibility for the government's long-term and strategic communication.
In all, there is more central coordination, planning, and steering.What follows is a clear tendency toward centralization of government communication as indicated in the interviews.In terms of the government communication process, it is evident from interviews that the job of these staffers is to promote or hinder the publication of information and then, use communication channels to the greatest effect.
RK Kommunikation provides strategic and operative support for the head of communication at the GO as well as technical and practical support.Its work is based on the document "Communication Policy for the Government Offices," which regulates the internal and external communication and is the common basis for communication activities (Prime Minister's Office 2012). 7The policy document describes the division of responsibilities for communication activities in the GO and the prerequisite of coordination for "good communication."The "ultimate responsibility" for the coordination of both internal and external communication lies with the PMO.Another sign of centralization, at least potentially, is the policy document regulating how ministers should act in social media; that is, a kind of social media communication policy (Government Offices of Sweden 2016).
Step by step, the government has taken control of at least parts of its media coverage, as epitomized by the news management in relation to the presentation of the state budget.As one senior journalist said, "you are in their hands."According to the experienced journalists who were interviewed, this is a typical example of how increased resources in government administration and stronger efforts to control the political news agenda are visible on a day-to-day basis.This change was also confirmed by the press secretaries in their descriptions of their work.
Both journalists and press secretaries described increasing resources on the political side-more coordination, more press secretaries, and more active work from sources to influence news reporting.Over time, the system has changed fundamentally in terms of accessibility and management.There is more control of information by press secretaries, according to the experienced journalists.A journalist with 20 years of experience in public service explained how press secretaries are being more active in limiting the possibilities to direct contact with politicians: "they are a filter all the time… not only in contacts with the government but also with members of parliament."All the experienced journalists confirmed this picture of increased information management.
Among the press secretaries themselves, one of them suggested that the expansion in their numbers along with other communication staffers reflected "a kind of increased professionalization not to give a messy impression."One press secretary emphasized the increased speed in media coverage as another explanation for more resources on communication and more coordination within the government: "Everything is much faster… when something is written on Twitter it can be a news article."All press secretaries are formally employed by the PMO, and coordination is strong with daily meetings.One of the press secretaries explained: It is a result of the spirit of the times and the demands on a better control of what different parts are doing.In more intense media coverage, we need to have better control of what we are sending out… Just basic things like ministers not having press conferences at the same time, releasing news competing with each other… We have a never-ending 24-hour news cycle; everything at a crazy speed.
Journalists and press secretaries gave the same general picture of an increased level of planned communication in the government.There are many reasons, some political and some connected to media development.Among the political reasons are a greater need for coordination in coalition governments and increased awareness of the role of communication in politics.
Professionalization of government communication is very much about control of information flows from the government and leading political parties.The press secretaries work constantly to promote good news to journalists and newsrooms and to avoid spreading bad news through the control of information and also, in some cases, in negotiations with journalists.
In summary, there are clear signs of professionalization of government communication and functional pressures for the government to be as well coordinated as possible to be able to assert themselves vis-à-vis the media and public-to speak with one voice-and this requires centralized media/news management and resources; a coherent approach, or simply "not to give a messy impression," as one interviewee put it, an attempt to create a coherent media image of the government.More broadly, and cynically perhaps, this is a way of sustaining the image of the state/government as coordinated, coherent, and controlled (Jacobsson, Pierre, and Sundström 2015, 38).In any event, Sweden is a case in point of growing resources into government communication, a paradigmatic example of professionalized and centralized strategic communication, of central control and coordination.In sum, the case of Sweden offers clear support for the centralization argument.

Conclusions
This article has made three main contributions to the literature.First, it documented the evolution and trend of centralization of government communication over recent decades in two so far relatively unexplored cases, showing that such centralization is more pronounced in Sweden but also present in Finland.Hence, these case studies of the Nordic countries offer further proof of executives adopting more top-down communication methods (see e.g., Marland, Lewis, and Flanagan 2017;Sanders and Canel 2013).Second, theoretically it offered an explanation for the observed centralization trends.In particular, it argued that one should combine insights from various academic disciplinespolitical communication, political science, and public administration-to understand the causal mechanisms behind centralization.Changing media structures, multiparty coalition governments, and horizontal coordination inside the executive branch together provide incentives for more top-down forms of communication.All of them point in the direction of governments needing to speak with one voice and to control information flows.Future studies of explaining government communication should thus pay more attention to what political science and public administration research has to offer, whereas particularly political scientists exploring the empowerment of the PM should consider how chief executives can benefit from centralized communication (see e.g., Poguntke and Webb 2005).Until now, this line of inquiry has focused on election campaigns but what happens between elections is arguably even more important.And third, it highlighted the increasing role of party-political advisors, particularly in the PMO, which shows that the governments are investing more resources in news management and media relations.Here, the key finding is perhaps not so much the purposeful recruitment of communication experts, but the way it can create tensions between civil servants responsible for official government communication and these party-political advisors (see also Shaw and Eichbaum 2018).The indicate that this is a highly sensitive topic that deserves closer exploration.
There are good reasons to expect that these centripetal tendencies will grow even stronger in the future.Changes in media systems are difficult to predict, and it is plausible to predict that party systems continue to be characterized by volatility and fragmentation.Governments also face two external challenges, increasing international collaboration and the need to counter fake news, especially regarding security matters, that call for unified messages (see e.g., Sanders 2020).There are also the challenges of social media, which has generated changes in how governments communicate (see e.g., DePaula, Dincelli, and Harrison 2018).
The paradox or seeming contradiction of a more decentralized media environment and at the same time greater centralization of government communication can be explained by the dynamics of the more diverse media landscape.In these more volatile circumstances, controlling information flows becomes even more important and challenging for the executive center.The more there are media actors contacting the government, and the more channels also the politicians and civil servants inside the executive branch can utilize for disseminating information, the higher the need for the PMO to control information flows.It is probably unrealistic to expect any government to be able to fully control such flows, but centralization should make it easier.
Nevertheless, centralization is also likely to be resisted, not least by junior parties in governments and by line ministries and public agencies.Centralization also has normative implications.The observed patterns certainly facilitate unified government communication.This obviously serves the interests of the PM and it can also be defended on the grounds of accountability: the media and the citizens find it easier to evaluate government performance when the cabinet speaks with one voice.Yet, democracy and public debate might be better served by more decentralized arrangements allowing for diverse arguments instead of centralized political advisor-led communication.Perhaps, more troublesome is the withholding of information or the strategically timed press releases that might benefit the government but clearly do not facilitate societal debate about topical matters.
However, the limitations of our exercise should also be acknowledged.While the interviews provided evidence of the causal mechanisms, it is difficult if not impossible to exactly measure the weight of each of the factors, especially as they are broader developments unfolding over time.As a result, this article hopefully encourages further research on government communication.The explanatory framework utilized in this article could be tested in other countries and through a wider repertoire of methods.Interviews offer valuable insights into behindthe-scenes processes, especially regarding the problematic division of labor inside the PMO between party-political advisors and civil servants, but surveys and network analysis could enable scholars to detect information flows and contact networks more effectively.Moreover, the framework may be refined to incorporate other factors, such as the domestic political context and government composition (majority/minority, coalition/one-party cabinets).The Swedish experience certainly suggests that centralization may be more easily reached in countries less ruled by coalition cabinets.Finally, research on non-Western cases deserves to be encouraged, in Europe and beyond.The overwhelming majority of existing research has dealt with more stable democracies, and it may well be that their lessons do not apply in other parts of the world.

About the Authors
Karl Magnus Johansson is Professor of Political Science at Södertörn University, Stockholm, Sweden.An ongoing area of research for him is central executive institutions in their communication and coordination aspects.He is the co-editor of Close and Distant: Political Executive-Media Relations in Four Countries (2019).He can be reached at karl.magnus.johansson@sh.seTapio Raunio is Professor of Political Science at Tampere University, Finland.His research interests include legislatures and governments, the EU, political parties, and semi-presidentialism.He is the co-author of Semi-Presidential Policy-Making in Europe: Executive Coordination and Political Leadership (2020).He can be reached at tapio.raunio@tuni.fi