Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

Think tanks in the Danish knowledge regime Abstract

This study explores how Danish think tanks within the area of welfare policy relate to what is termed the Danish and European knowledge regimes. In an attempt to strengthen the focus on agency in studies of ideational change a framework for analyzing ideals of different think tanks is suggested. Following Diane Stone the paper argues that think tank ideals can be conceptualized on three dimensions as more or less research-oriented or ideological. On the basis of a case-study including five Danish think tanks, it is concluded that ideological think tanks with more focused on interest representation than research-oriented think tanks have emerged in the Danish knowledge regime. The study also finds that the extent to which Danish think tanks engage in trans-national activity is relatively modest. Keywords: Knowledge regime, welfare, think tank, Denmark, European Union.

Contents 1. Introduction....................................................................................................................................2 2. Modeling welfare............................................................................................................................2 2.1 A Danish knowledge regime ................................................................................................................... 3 2.2 A European knowledge regime ............................................................................................................... 4

3. Think tank ideals and knowledge regimes...................................................................................5 3.1 Definitions and the study of ideals .......................................................................................................... 5 3.2 Think tank myths, ideals and ideology.................................................................................................... 7

4. Case-study of Danish think tanks .................................................................................................9 4.1 Think tank sampling ................................................................................................................................ 9 4.2 Think tank funding and the public interest.............................................................................................. 9 4.3 Producing research and engaging in politics ......................................................................................... 10 4.4 Bridging knowledge and politics........................................................................................................... 11 4.5 Trans-national cooperation.................................................................................................................... 14 4.6 Case-study summary ............................................................................................................................. 16

5. Conclusions ...................................................................................................................................17 6. References .....................................................................................................................................19 NOTE: This paper is an attempt to relate pilot-interviews with experts from the Danish think tanks: ‘Det Nationale Forskningscenter for Velfærd’ (SFI), ‘Arbejderbevægelsens Erhvervsråd’ (AE), ‘Center for Alternativ Samfundsanalyse’ (CASA), CEVEA and CEPOS to theory about knowledge regimes and ideational change. The study is intended to be the first of a series of thematic case studies including dynamics in and across Denmark, Britain, Germany and the EU-level. Feedback on how to turn this paper into an article and broader reflections on how the study of think tanks can be methodologically strengthened are especially welcome.

1

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

1. Introduction Across the planet think tanks that aim at influencing public policy and public opinion have emerged. But so far very little is known about the practice and ideals of think tanks and what impact they have. Many questions regarding think tanks are left wide open. Do think tanks challenge political parties? Are they elitist organizations that lay a distance to the rest of society (Stone 2007: 269) or organizations with a strong virtual presence on the internet that allows a wide segment of the population to participate in public debate? While many studies on ideational processes have focused on the role of institutions rather than agents in the analysis of different knowledge regimes (Campbell and Pedersen 2011: 171), the purpose of this paper is to study the ideals of Danish think tanks. First an account of different conceptions of national welfare and the development of a European knowledge regime is given. On the basis of the conception that think tank are engaged with relating knowledge and politics the ideological affiliation, conception and bridging of knowledge and politics of five Danish think tanks is presented in a case-study based on expert interviews carried out from March to May 2011. Given that the analysis is based primarily on expert interviews, and does not as yet include textual or quantitative analysis of think tanks publications and websites, the scope for generalizations of the findings of the paper are modest. The paper aims at engaging further with the work being done on the role of ideas in politics (Campbell 1998; Gofas & Hay 2010; Béland & Cox 2011).

2. Modeling welfare In the last decades organizations and scholars have developed concepts and typologies that attempt to capture how different states model or should re-model what is commonly understood as welfare policy. Some of the most prominent examples include ‘the worlds of welfare capitalism’ (EspingAndersen 1990), ‘the active society’ (OECD 1988), ‘national competitiveness’ (Porter 1990), ‘national innovation systems’, ‘the knowledge’ or ‘learning’ society (Lundvall og Johnson 1994; Hansen 2010), ‘the competitive state’ (Pedersen 2011) and ‘knowledge regimes’ (Campbell and Pedersen 2011). Although different in content, these models and conceptions share some basic traits. First, institutions at the nation-state level are at the focus of attention in the sense that commonly defined rules, norms and incentives structure the behavior of agents in the different models (whether companies, knowledge institutions or other types of actors). Second, the models

2

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

portray or envision different ways of balancing or integrating the concept of the market and the state in and across polities and policies. In broad terms a move from viewing the market and the state as opposing forces (welfare models) to mutually constitutive forces (active society, national competitiveness, national innovation systems, competitive states) seems to be evident across the different academic conceptions of welfare (Hansen 2010: 32) as well as in political-administrative strategies of welfare reform (e.g. the politics of The Third Way). In the following I present the claim that the Danish polity can be seen as a particular type of knowledge regime.

2.1 A Danish knowledge regime A recent example of the national model perspective is Campbell and Pedersen’s attempt to combine concepts from theories on production and policy-making regimes to create a typology of knowledge regimes. The concept of knowledge regime can be used to capture the ideas whether through data, analysis, reports or policy recommendations that influences policy-making. Following Campbell and Pedersen’s understanding of knowledge regimes I refer to them as ‘the sets of actors, organizations, and institutions that produce and disseminate policy ideas that affect how policymaking and production regimes are organized and operate in the first place’ (Campbell and Pedersen 2011: 167). Comparative political economists distinguish between two types of production regimes. In liberal market economies markets structure economic activity while in coordinated market economies this is primarily done through non-market relationships like formal corporatist bargaining and various forms of state regulation (Campbell & Petersen 2011: 170). When it comes to policy-making regimes a distinction can be drawn between centralized, closed states where policy-making is restricted to a few arenas largely insulated from the dynamics of civil society and decentralized, open states where policy-making authority tends to be shared and delegated to lower levels of government (Katzenstein 1979). The synthesis of production and policy-making regimes results in the following typology of knowledge regimes: Table 1: Typology of knowledge regimes (Pedersen & Campbell 2011: 186).

Liberal market economy oriented

Coordinated market economy

Decentralised, open state

Market regime

knowledge Consensus-oriented regime

knowledge

Centralised, closed state

Politically tempered knowledge Statist-technocratic regime regime

knowledge

3

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

It is commonplace to classify Denmark as a coordinated market economy and a decentralized open economy because of the strong tradition for neo-corporative decision-making and de-centralisation. This categorization of Denmark is the backdrop on which I study think tank ideals. First however I present the claim that an overarching European knowledge regime has emerged.

2.2 A European knowledge regime Although the European Union (EU) is a governance structure comprised of a wide array of national production and policy-making regimes, it is interesting to observe how the EU has gradually attempted to craft an overarching European knowledge regime by balancing state and market dynamics. A major milestone to this end was former European Commission president Jacques Delors’ vision of a ‘European Social Model’ (ESM) built on social protection and justice, which he conceived as common European values (Ross 1995:46). Delors vision manifested itself in gradual increases in the Commissions right of initiative and qualified majority voting in the Council in the areas of social policy through the Single European Act, the Social Protocol of the Treaty of Maastricht and the Social Chapter of the Treaty of Amsterdam (Liebfried 2005: 249ff). Further, the establishment of a Social Dialogue between employer organisations (UNICE, now Business Europe) and trade unions (ETUC) with the Treaty of Maastricht was an attempt at mimicking national bargaining models at the EU-level in the attempt to realize the ESM. The ability to reach agreement on social policy through the adoption of common minimum standards in the Council and agreements between the partners in the Social Dialogue however proved to be modest in the 1990s (Rhodes 2005: 290). Rather than through formal decision-making, one of the key advancements towards establishing a European knowledge regime happened with the adoption and gradual innovation of the Lisbonprocess in 2000 based on the non-binding Open Method of Co-ordination (OMC). The main reason for the success of the OMC in developing a common conception of European welfare was the ability of the strategy to integrate the non-binding, statistical comparisons used in the OMC with the conceptions of welfare models that was developing in the member states i.e. Esping-Andersens Worlds of Welfare Capitalism and the discourses of the active society, knowledge society and national competitiveness mentioned above (Hansen 2010). While modest in terms of output success throughout the 2000s, the Lisbon Process gradually succeeded in integrating the national welfare

4

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

models in the OMC-framework. An important step in this process was the explicit definition of the most successful national models at the EU-level. An excellent example of this is the report crafted by André Sapir presented to the Council under the British presidency in 2005, where Nordic welfare models were heralded in terms of their high level of equity and efficiency as oppose to the Continental, Anglo-Saxon and in particular Mediterranean models (Sapir 2005: 4). This implicit rating of the different models however became subject to much criticism; partly because the ways and processes through which the models could learn from each other remained blurry (Hansen 2010). The Commissions adoption of the concept and common principles of flexicurity beginning with the Employment in Europe report from 2006 (European Commission 2006) can be seen as an attempt to modify the European framework by translating the national models to an overall principle of flexicurity that were gradually accepted by the European institutions and integrated into national reform-programmes and status-reports of the Lisbon Agenda (Hansen 2010: 59). I find the interaction of agents in the tandem development of national welfare models (e.g. EspingAndersen 1990) and the European regime to be theoretically underdeveloped. Within the main strands of neo-institutional theory rather few empirical analysis have focused on how agents behave and change existing institutions (Schmidt 2011). One way of trying to remedy this is to study the role of agents within these knowledge regimes. The following sections on Danish thinks tanks are intended to relate present developments in the ideals, communications, networks and cross-national activities of think tanks to the above conception of the Danish and European knowledge regimes.

3. Think tank ideals and knowledge regimes 3.1 Definitions and the study of ideals Think tanks are of interest to our understanding of knowledge regimes, because they must relate both to political knowledge and ideology. In other words they have to bridge the realm of the scientific and the political. While think tanks can broadly be conceived as independent, non-profit organizations engaged with analyzing policy issues, the many uses of the term think tank has led some scholars to abandon the concept altogether (Campbell and Pedersen 2011, 170). One of the reasons that think tank definition 5

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

vary is that typologies tend to depend on the institutional setting in which they are analysed. Abelson for example distinguishes between ‘advocacy think tanks’, ‘government contractors’ and ‘universities without students’ in his US study (Abelson 2006: 44ff), while Ullrich distinguishes think tanks ‘generating ideas’, ‘policy oriented analysis’ and ‘furthering debate’ in the EU (Ullrich 2004: 54). Stone distinguishes between scholarly ‘ink tanks’ and more activist ‘think-and-do tanks’ (Stone 2007: 262). Instead of studying think tanks in national polity-setting, as is common in the existing literature on think tanks (Ullrich 2004, Stone & Denham 2004; Abelson 2006), the intention in this study is to discover how national think tanks – with Denmark as the first illustration – relate to their national and the European knowledge regimes respectively and in turn may contribute to consolidate or chance them. How then, should think tanks ideals be studied empirically? In the following I develop a two-step approach. First I distinguish between different types of ideas that think tanks engage with. Second I treat conventional myths about think tanks as a reference point for understanding their ideals. John Campbell usefully distinguishes four types of ideas that are typologised in terms of cognitive or normative reasoning and whether they constitute explicit arguments or underlying assumptions in policy debates. Campbell understands background ideas as basic understandings of politics (democracy, rule of law, welfare state) while foreground ideas are concerned primarily with current political debates. While normative framing involves values and attitudes, cognitive framing is characterized by references to science (Campbell 1998: 378). Table 2: Four types of ideas (Campbell 1998: 378).

Foreground ideas

Background ideas

Cognitive

Programs

Paradigms

Normative

Frames

Public sentiments

In this piece I focus on the two types of foreground ideas that can be defined as follows: •

Frames are the symbols and concepts that actors can use to frame solutions to policy problems in normatively acceptable terms through transposition and bricolage (Campbell 1998: 394).

6

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

•

Programs are technical and professional ideas that specify cause-and-effect relationships and prescribe a precise cause of policy action. (Campbell 1998: 386).

Campbell’s typology allows agents to both internalise ideas and to influence knowledge systems whether by promoting specific programs and frames or working as bricoleurs. Bricoleurs can be understood as actors that engage with the ‘rearrangement of elements that are already at hand, but it may also entail the blending in of new elements that have diffused from elsewhere’ (Campbell 2005: 56; Carstensen 2011). The methodological framework for identifying the different types of ideas among think tanks possibly by discourse analysis will need to be strengthened to properly apply Campbells concepts. This promises a way to operationalise how different types of ideas among think tanks influence the national and European knowledge regime.

3.2 Think tank myths, ideals and ideology As mentioned earlier think tanks must by definition relate to political ideology. Ideology in this piece is understood as ‘a wide-ranging structural arrangement that attributes de-contested meanings to a range of mutually defining concepts and competes over the control of political language’ (Freeden 2003: 54f). Diane Stone points out that the understanding of think tank can be based on three idealized myths that think tanks think, i.e. produce research, bridge knowledge and politics and serve the public interest (Stone 2007: 260ff). The myth of serving the public interest is rooted in the conception of think tanks as stimulators of public debate that also educate the citizenry and promote effective governance through policy analysis (Stone 2007: 261). The research myth comes from the expectation that think tanks have something ‘scientific’ about them, due to their educated staff and their interactions with universities. The myth of bridging implies that think tanks organizationally and functionally can be distinguished from other actor like political parties, universities, interest organizations and private companies that allows them to bridge state, society and science. In this study ideological think tanks are considered those who themselves engage in the competition for control over political language. The affiliation with other ideological actors, the claim that ideological orientation and research are compatible and that knowledge and politics can be legitimately bridged by interest representation are all characteristic of ideological think tanks. In contrast to this research-oriented think tanks in this study terms think tanks that consider research

7

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

and ideology incompatible that bridge knowledge and politics primarily through the exchange of information in research networks, and have no institutional links to ideological actors. In the following case-study I use these three myths to characterize the five Danish think tanks by translating them into the following contrasting ideals: 1. Serving the public good. Think tanks are classified depending on whether they are affiliated with ideological actors that compete over the control of political language or state institutions. In practice a distinction is drawn between on the one hand political parties, trade unionsations and market actors and on the other hand relations to public administration. 2. Producing research. Think tanks ideals on this parameter are classified depending on whether they conceive of ideology and research as compatible activities. 3. Bridging research and politics. Think tank ideals on this parameter are classified depending on whether they prefer using interest representation or exchanging information in knowledge networks. Table 3: Two types of ideal-typical think tanks. Think tank is affiliated with ideological actors.

Political ideology and research are seen as:

Think tank gives preference to:

Ideological think tanks

Yes (1)

Compatible (1)

Interest representation (1)

Research-oriented think tank

No (0)

In-compatible (0)

Knowledge networks (0)

In this short paper (at least) two crude methodological simplifications are made. First, think tanks are considered unitary actors, the staff of which is taken to represent the overall views of the think tanks. Second, public administration is considered to be un-ideological because of the balancing role of the state in the Danish knowledge regime. These assumptions are meant to simplify the analysis, but they do challenge the methodological strength of the case-study which will have to be developed further.

8

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

4. Case-study of Danish think tanks Interviews with experts at the five Danish think tanks Arbejdsbevægelsens Erhvervsråd (AE), Det Nationale Center for Velfærd (SFI), CEVEA, Center for Alternativ Samfundsanalyse (CASA) and Center for Politiske Studier (CEPOS) carried out in March-May 2011 work as primary data for the following case-study.

4.1 Think tank sampling The thinks tanks in this study have been selected from Danish think tanks focused on socioeconomy that were founded in different periods, have different sources of financing and engage with different types of actors. AE was established in 1936 by the main Danish Trade Unions.1 The Danish Parliament in 1958 adopted the law establishing SFI as a ‘half way house’ between universities and public administration.2 CASA was established in 1986 as non-profit consultants that unlike most consultant businesses make their work publicly available and used some of their profit to initiate research on their own. The liberal think tank CEPOS was established in 2004 as the first think tank in Denmark purely financed by private funds.3 In 2008 CEVEA followed suit. While primarily financed by the Trade Union, CEVEA explicitly works with an ideational rather than economist perspective on politics.4 In the following sub-sections the framework developed above is applied. First the main sources of funding and affiliations of each think tank is briefly accounted for. Second it is explored what ideals the different think tanks hold concerning the link between knowledge production and ideology – are these seen as compatible or incompatible activities? Third the ways think tanks bridge knowledge and politics is explored. Finally the way each think tank engages in trans-national cooperation is explored in order to estimate how Danish think tanks relate to the European knowledge regime.

4.2 Think tank funding and the public interest It is characteristic for think tanks, that their influence is based, among other things, on independence (Ullrich 2004: 66) or ‘relative autonomy’ (Stone 2004: 2f) from other political agents. Yet it seems reasonable to expect that funding and institutional affiliation affect think tanks ideals about how to serve the public interest. While AE, and CEVEA are financed mainly by the Trade Union and 1

http://www.ae.dk/om-ae/ae-historie. http://www.sfi.dk/Default.aspx?ID=4508. 3 http://www.cepos.dk/om-cepos/hvad-er-en-taenketank/hvornaar-kom-der-taenketanke-i-danmark/. 4 http://www.cevea.dk/om-cevea/idegrundlag. 2

9

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

CEPOS by donations from private businesses both CASA and in particular SFI receive the bulk of their funding from public administration [evidence to be inserted]. On this account AE, CEVEA and CEPOS are considered ideological while SFI and CASA are considered research-oriented on this dimension.

4.3 Producing research and engaging in politics In this subsection it is investigated whether think tanks see a contradiction between producing research and representing an ideological orientation. This tells us something about how they interpret the ideal of producing research. The expert from SFI clearly underlines that SFI as a public research institution is not a think tank if this term is taken to indicate an overall political agenda [SFI 00:50]. SFI is primarily a taker of research commissioned by the state. The expert claims that the type of work that the politicaladministrative system commissions work from SFI is consistent over time in that 90 pct. of SFI’s research is continuous across governments [SFI 09:50]. CASA works as consultants that perform analysis for both private and public customers. The expert does not claim that CASA is a university without students but it is stressed, that CASA aims at producing research of the same quality as one. Every report is published and publicly available [CASA 09:10]. The CASA informant points out that CASA has accepted that we have entered a phase of modernity where everything is in play in politics. But CASA claims to retain a traditionalist approach to research the extent that they find it necessary to meet certain criteria of inquiry in order to have an opinion on a specific topic [CASA 34:30]. AE seeks to combine their ideals with a scientific paradigm [AE 4:10] (my translation): ‘We are not afraid to admit that the themes that we choose to focus on are guided by certain political or socio-economic objectives … After this selection however, our analysis is as credible as were we a government sponsored research institution, a ministry or a university.’ The opportunity to access statistical data, which AE buys at Statistics Denmark is crucial for the ability of AE to engage on a level playing field with government and ministries [AE 1:40] and thereby to gain credibility. This data however leaves some room for interpretation. The informant from AE stresses that their economists reject to stretch interpretations of data beyond what they can

10

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

carry. It acts to the advantage of AE having a brand as being ‘dry and boring’ and thereby also credible [AE 5:26]. CEPOS have enhanced the resources devoted to research since their creation in 2004 [CP 20:20] (my translation): ‘Part of the reason that we conduct our own research and have increasingly strengthened this effort is to also influence the state of the art of the scientific debate. Our ability to publish in academic journals strengthens the credibility of our analysis and policy suggestions. Scientific work is hard simply to reject’. In contrast to the other think tanks, CEVEA does not derive its legitimacy from credible research but rather from the values of the think tank that include freedom, equality and community [CV 0:40]. When asked whether it is a challenge for CEVEA not to be considered neutral the informant responds [CV 02:30] (my translation): ‘No, our legitimacy rests on our ideas, ideals and values. I am a bit provoked by think tanks that claim to conduct neutral and objective research, that produce objectively correct analysis… Keynes said it quite precisely: ‘neutral, practical men that think they operate in a world without ideology are really made into slaves by the ideas that happen to rule’. We find ourselves in a constant and continuous battle of ideas which can only be fought on the basis of ideals, otherwise you will always end up losing.’ To sum up this section, the think tanks have taken different positions towards the relationship between research and ideology. While CEPOS, AE and CASA se research as a methodology that is compatible with a political ideology, SFI and CEVEA see research and ideology as incompatible positions. This finding may suggest that CEPOS, AE and CASA find it easier to integrate framing and programming than SFI and CEVEA. However Campbell’s conception of frames and programs as alternative foreground policy ideas will need to be applied empirically to properly test this hypothesis.

4.4 Bridging knowledge and politics This sub-section is devoted to exploring think tank ideals in bridging knowledge and politics. SFI claims to evaluate itself largely through the quality of its research products ([SFI: 3:54]). The bridging activity of SFI is performed through the close relations to public administration, that

11

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

allows SFI to engage in discussions with politicians and high-level government officials about political programs which SFI has suggested should more systematically be subject to testing before implementation [SFI: 4:20]. Although CASA does have many public customers, they do not have the same institutional access to public administration as SFI. Instead CASA tries to bridge their work with the political agenda by gathering different organisations that represent street-level workers as well as academics to discuss issues like poverty. CASA finds that there is enough analytical evidence to develop public policy through knowledge rather than ideology [CASA 40:40]. The focus on Danish public policy and national actors gives CASA’s bridging activity a national focus. In order to get public exposure, think tanks need to provide information that fulfills the news criteria of the media, most often the written press. One of the most important aspects in the relation with the media concerns the supply of credible information. AE, CEPOS and CEVEA all produce brief analysis involving compiled statistical data. The strategy of CEPOS is aimed mainly at influencing the public debate through the media [CP 14:20] (my translation): ‘We find that the standpoints of politicians are largely determined by their position in the public debate. A good metaphor for this is to view the public debate as a river. If we consider politicians small floating cork props in the river you can approach them in two ways. The first involves lobbying whereby you attempt to move them slightly up or down the stream ... This approach is cost efficient and quite precise. But the long-term effect is also questionable. Politicians are likely to retain the same position or nearly the same position further down the river. What we try to do is to steam up the river and change the direction of the stream.’ CEPOS emphasises the importance of new modes of communication in influencing the media by blogging at different newspapers [CP 26:22]. CEVEA also uses a targeted media approach [CV 29:40] (my translation): ‘Using the media is a very conscious strategy on our side. Because we engaged in a battle of ideas it is important to target Danish opinion makers. In essence we have a small group of 150 to 250 people that we try to target. This includes political journalists, politicians and opinion makers in the Unions and business organisations. We are not in the yellow press or on the street, our activities e.g. seminars are targeted at small groups’.

12

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

Like CEPOS and CEVEA, AE considers the public sphere very important. In terms of targeting the media, independency from the trade unions that pay much of AE’s budget is considered essential [AE 15:16]. Influence is conceived as a process requiring both a conception of supply and demand in the news cycle. According to the AE informant this sometimes implies adapting to the demands of the media. For example a Danish newspaper wanted AE to do an analysis of a political plan. AE was reluctant because this could add to the uncertainty about this policy but agreed to do it anyway, as AE publishes around 40 articles in that newspaper each year [AE 35:00]. AE tries to bridge their knowledge production with the political in a number of different ways. The expert of AE mentions five different target groups for their strategy of influence that besides the media includes key opinion makers, political contacts, government councils and hearings, educational and research institutions and The Danish Confederation of Trade Unions ([AE 8:28], [AE 10:20], [AE 21:50], [AE 22:37], [AE 24:30]). A quantitative perspective on the references of think tanks in Danish nationwide newspapers from indicates that there are substantial differences in media penetration: Table 4: References in Danish nationwide newspapers from 21.04.2009 to 21.04.2011.

Think tank

Center for (CEPOS)

References in Danish nationwide newspapers from 21.04.2009 to 21.04.20115 Politiske

11

85

943

12

39

Det National Center for Velfærd 607 (SFI)

95

3

CEVEA

3

43

10

4

Arbejderbevægelsens Erhvervsråd (AE)

Studier 1.880

Number of Estimated yearly permanent full reference impact time academic pr. employee staff

255

Center for Alternativ 73* Samfundsanalyse (CASA)

*This number may be underestimated due to search difficulties with the term CASA that is used by several organisations.

5

Searched in Infomedia on 21.04.2011 in the Danish newspapers: B.T., Berlingske, Børsen, Ekstra Bladet, Information, Jyllands-Posten, Kristeligt Dagblad, Politiken and Weekendavisen. Search words: ‘Cepos’, ‘Arbejderbevægelsens Erhvervsråd’, ‘SFI’, ‘Cevea’, ‘Center for alternative samfundsanalyse’.

13

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

The AE expert underlines that their activity within the knowledge regime is not a one way process. The overall dynamics of the political debate largely determines the platform that AE works from [AE 20:35] (my translation): ‘During the last 1½ years the war of economic plans between political parties in Denmark has pushed business organizations and union-affiliated organizations apart. The midfield of the debate is almost void. We support reform – but presently we are forced into the red bloc’. The interviews do indicate that the ‘vertical’ dimension of engaging with the public is supplemented with an informal ‘horisontal’ dialogue between think tanks, interest organisations and government agencies [AE 7:39] (my translation): ‘This is not only a war in the media …Our chief economists meet with CEPOS, DI, DE and the Ministry of Finance several times a year for lunch-meetings, where they have another type of dialogue.’ To sum up the this section differences between the think tanks indicate that SFI and CASA bridge knowledge primarily with public administration while AE, CEPOS and CEVEA mainly see bridging as a process of interacting with other actors affecting the political agenda both in and around the media.6

4.5 Trans-national cooperation The official status as a state research institution allows SFI to take part in the transnational Campbell cooperation that is designed to systematically produce oversights of research studies that have been conducted in different countries and policy areas. SFI sees the cooperating as a stepping stone towards utilizing research resources internationally and avoiding repeating the same studies over and over again [SFI 19:20]. Both CASA and SFI have participated in a number of EU projects, but both organisations find this quite cumbersome. For both think tanks the implication is that international cooperation has developed most strongly through personal research contacts ([CASA 26:50], [SFI 15:20]). Additionally SFI states that the EU framework programs do not provide much funding for social science research. Their policy therefore is to find partners for EU financed projects. The rate of 6

In this section it would be interesting to also investigate the question of how and to what extent think tanks have developed mechanisms that allow feedback from society (Stone 2007: 268).

14

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

success in the EU programs is simply so little that they consider themselves too small to apply alone [SFI 26:30]. AE is the only think tank in the sample to have developed institutional relationships in the EU. This happens through The Foundation for European Progressive Studies (FEPS) that is affiliated with, but formally independent from, the Party of European Socialists (PES). AE’s participation was initiated by former Prime Minister Poul Nyrup Rasmussen who used to be at AE and now chairs PES [AE 26:00]. AE’s economic model Heimdal is able to perform economic analysis across Europe, and AE does conduct analysis of economic scenarios at the European level for FEPS. FEPS regularly comes to Denmark and AE goes to Brussels [AE 28:10]. Summing up however, AE sees the Danish agenda, and the agenda of the trade unions as very centered on Denmark. This leaves the bulk of the demand for analysis at the national level [AE 29:00]. CEVEA has also attempted to expand cross-national co-operation. By going to Britain, Sweden and Germany to make contact with other think tanks, [CV 23:50] (my translation): ‘One of the reasons that we have a good cooperation with these types of institutions [foreign think tanks, ed.] is that there are individuals in these organisations capable of acting independently of the party organisations that finance them. This challenges co-operation across borders. In France, Germany and Finland think tanks are typically funded by political parties and this makes it hard to cooperate with them. In general the continental think tanks are less dynamic than Scandinavian and Anglo-Saxon think tanks, that are easier to engage with’. CEVEA expects that the national agenda will continue to be the centre of their work. The attention given to the EU may increases slightly, but it will be of marginal significance [CV 31:40]. CEPOS finds that bridging with foreign think tanks and the European knowledge regime is restricted by two reasons. First, the policy oriented work of CEPOS means, that the work of foreign think tanks cannot be used directly in the Danish context. This means that the inspirations from sister think tanks in other countries works on an ad hoc-basis [CP 41:40]. Second, [CP 45:40], ‘There is no common liberal organisation at the European level. The nationally think tanks are characterised by nuances. If you take the Adam Smith Institute and the Institute for Economic Affairs in London they are very sceptical of the European Union. In CEPOS we don't have a common standpoint with regards to the ‘tax stop’ or the EU.’

15

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

To sum up this sub-section all think tanks see transnational relations as secondary to the national policy agenda. This seems mainly to be related to the Danish orientation of their networks and sources of financing, but also to the difficulties in utilising knowledge and ideas from elsewhere in their national context. This finding suggests that neither research based, nor ideological think tanks integrate the Danish and European knowledge regime to any large extent.

4.6 Case-study summary On the basic of their ideals relating to the three parameters the think tanks are given an ‘ideologyscore’. This is an aggregation of their relation to the parameters of ideals that have been analysed above. Table 5: Think tanks ideology-scores. Think tank is affiliated with ideological actors (yes = 1, no = 0).

Political ideology and research are seen as (compatible = 1, in-compatible = 0).

Think tank gives preference to (interest representation = 1, knowledge networks = 0).

Think tank score (higher = ideological, lower = research oriented).

AE

1

1

1

3

CEPOS

1

1

1

3

CEVEA

1

0

1

2

CASA

0

1

0

1

SFI

0

0

0

0

The case-study suggests that AE and CEPOS are genuine ideological think tanks, while SFI comes out as a research-oriented think tanks. CEVEA and CASA reach intermediate ratings. It is premature, to place any great emphasis on the ideological think tank score. Yet a comparison of the ideology score and the impact rate pr. employee on newspapers suggests that ideological think tanks tend to get much more public exposure for their money than research oriented think tanks.

16

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected]. Table 6: Think tanks ideology-scores and newspaper impact. Think tank

Think tank ideology score (higher = ideological ideals, lower = research ideals).

Estimated newspaper impact for think tanks pr. employee pr. year.

Center for Politiske Studier (CEPOS)

3

85

Arbejderbevægelsens (AE)

3

39

CEVEA

2

43

Center for Alternativ Samfundsanalyse (CASA)

1

4

Det National Center for Velfærd (SFI)

0

3

Erhvervsråd

5. Conclusions Think tanks are not agents that simply serve the public interest by supplying solutions to political problems in a linear and rational process. Rather think tanks find themselves in a complex environment in which they interact both with other think tanks, universities, interest organizations, politicians and with the segments of the public that take part in their activities. This preliminary study argues that the emergence of ideological think tanks and the understanding of the different actors and ideals they represent can potentially be used to understand change and inertia in the Danish and European knowledge regimes. By using a crude distinction between ideological and research-oriented think tanks, the study has presented the main affiliations and analysed the ideals that think tanks hold towards producing research and bridging knowledge and politics. As yet, little is know about the actual impact of ideological think tanks on the Danish knowledge regimes. To understand this detailed empirical studies of think tanks and their impact on public policy must be carried out. The case-study has however produced two findings. First, ideological think tanks have emerged that are more focused on interest representation than more research-oriented think tanks. Second, the extent to which all think tanks engage in trans-national activity is relatively modest. What perspectives do these preliminary findings raise? Although the consensual tradition of the Danish knowledge regime is built on mechanisms of negotiation that may be largely unconcerned

17

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

with the emergence of think tanks the tendency of ideological agents to ‘freeze’ the public debate into fixed position of red and blue may challenge this dynamism of the Danish knowledge regime. If the symbolic use of knowledge in politics is taking precedence over instrumental policy-making so that knowledge legitimizes rather than improves policy solutions, this creates a contradiction between the cry for innovation and economic growth built into the OMC and the European knowledge regime and a trend towards an ideological battle of ideas within the Danish knowledge regime.

18

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

6. References Abelson, Donald E.: The business of Ideas: the think tank industry in the USA. In: Stone, D. and Denham, A. (eds.) (2004): Think Tank Traditions, Policy research and the politics of ideas. Manchester University Press. Manchester. Abelson, Donald E. (2002): Do Think Tanks Matter? Assessing the Impact of Public Policy Institutes. Montreal. McGill-Queen’s University Press. Béland, Daniel & Cox, Robert H. (eds.) (2011): Ideas and Politics in Social Science Research. New York: Oxford University Press. Bouget, D.: Convergence in the social welfare systems in Europe: from goal to reality. In: Social Policy & Administration, vol 37, 2003: 674–93. Bruno, Randolph & Rovelli, Riccardo: Labour Market Policies and Outcomes in the enlarged EU. In: JCMS 2010, vol. 8 no. 3, pp. 661-685. Busch, Klaus (2010): World Economic Crisis and the Welfare State Possible solutions to reduce the economic and social imbalances in the world economy, Europe and Germany. Policy Analysis, Friedrich Ebert Stiftung. Campbell, John L. & Ove K. Pedersen: Knowledge regimes and Comparative Political Economy. In: Béland, Daniel & Cox, Robert H. (eds.) (2011): Ideas and Politics in Social Science Research. New York: Oxford University Press. Campbell, John L.: Where do We Stand? Common Mechanisms in Organizations and Social Movements. In: Davis, Gerald F.; McAdam, Doug and Scott, Richard W. (eds.) (2005): Social Movements and Organizational Theory, New York: Cambridge University Press. Campbell, John L.: Institutional analysis and the role of ideas in political economy. In: Theory and Society (1998) vol. 27: pp. 377-409. Carstensen, Martin B.: Paradigm man vs. the bricoleur: an alternative vision of agency in ideational change. In: European Political Science Review 2011, 3 (1): 147-167. Davis, Gerald F.; McAdam, Doug and Scott, Richard W. (eds.) (2005): Social Movements and Organizational Theory, New York: Cambridge University Press. Diamond, Patrick; Liddle, Roger (ed.) (2009): Beyond New Labour. London. DiMaggio, P.J., & Powell, W.W.: The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. In: American Sociological Review 1983, 48(2), 147516 Esping-Andersen, Gøsta (1990): The Three Worlds of Welfare Capitalism. Polity Press. Cambridge. Euractiv (2009): The changing face of European think-tanks.

19

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

http://www.euractiv.com/en/pa/changing-face-european-think-tanks/article-142652 (accessed on 25.02.2011). European Commission (2010): Governance, tools and policy cycle of the strategy. http://ec.europa.eu/eu2020/index_en.htm (accessed on 25.02.2011). European Commission (2006): Employment in Europe Report. http://europa.eu/legislation_summaries/employment_and_social_policy/situation_in_europe/c11336 _en.htm (accessed on 25.02.2011). Freeden, Michael (2003): Ideology: A Very Short Introduction. Oxford University Press. Oxford. Giddens, Diamond & Liddle (eds.) (2006): Global Europe - Social Europe. Polity Press. Greve, B.: Indications of social policy convergence in Europe. In: Social Policy & Administration, 1996, vol. 30: 348–67. Gofas, A. & Hay, C. (eds.) (2010): The Role of Ideas in Political Analysis. A Portrait of Contemporary Debates. Routledge, Oxon. Hall, Peter A. and David Soskice (eds.) (2001): Varieties of Capitalism. New York: Oxford University Press. Hansen, Magnus Paulsen (2010): Lissabon-strategien og ensretningen af social og økonomisk tænkning - En genealogi over velfærdsspørgsmålet i EU. Speciale på Forvaltning, Roskilde Universitet. Hillebrand, Ernst: A Society of Empowered Citizens. Outline of a social democratic project for the twenty-first century. Internatiol Policy analysis, Friedrich Eberhart Stiftung, December 2009. Hooghe, Liesbet; Marks, Gary and Wilson, Carole: Does Left/Right Structure Party Positions on European Integration? In: Steenbergen, Marco and Marks, Gary: Introduction: Models of Political Conflict in the European Union. In: Marks & Steenbergen (eds.). European Integration and Political Conflict. Cambridge, Cambridge University, 2004), pp.1-10. Katzenstein, Peter J. 2006. “Denmark and Small States.” In: National Identity and the Varieties of Capitalism: The Danish Experience, edited by John L. Campbell, John A. Hall, and Ove K. Pedersen. Montreal: McGill-Queen’s University Press. Katzenstein (1985): Small States in World Markets: Industrial Policy in Europe. Ithaca: Cornell University Press. Kvale, Steiner (2002): Interview. En introduktion til det kvalitative forskningsinterview. København. Hans Reitzels Forlag. Leibfried, Stephan (2005): Social Policy – Left to the Judges and the Markets? In: Wallace, Wallace & Pollack: Policy-Making in the European Union – Fifth Edition. Oxford University Press, Oxford. pp. 243-278.

20

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

Leibfried, S. (2000) ‘National welfare states, European integration and globalisation: a perspective for the next century’. In: Social Policy & Administration, vol. 34: 44–63. Lundvall, Bengt-Åke & Björn Johnson (1994): The Learning Economy. In: Journal of Industry Studies, Vol. 1, No. 2, Dec. 1994, 23-42. Marcussen, Martin (2010): Den danske model og globaliseringen. Frederiksberg. Samfundslitteratur. McGann, James G.; Weaver, R. Kent (eds.) (2000): Think tanks and civil societies : catalysts for ideas and action. New Brunswick, NJ. Transaction Publishers. OECD (1988): The Future of Social Protection. Social Policy Studies No. 6, OECD, Paris. Pedersen, Ove K. 2006a. “Denmark’s Negotiated Economy.” In: National Identity and the Varieties of Capitalism: The Danish Experience, edited by John L. Campbell, John A.Hall, and Ove K. Pedersen. Montreal: McGill-Queen’s University Press. ´ Pierson, P. (1994): Dismantling the welfare state? Reagan, Thatcher and the Politics of Retrenchment. Cambridge: Cambridge University Press. Porter, Michael E. (1990): The Competitive Advantage of Nations. The Free Press, New York. Rhodes, Martin (2005) Employment Policy – Between Efficacy and Experimentation. In: Wallace, Wallace & Pollack: Policy-Making in the European Union – Fifth Edition. Oxford University Press, Oxford. 279-304. Rich, Andrew (2011): Ideas, Expertise and Think Tanks. In: Béland, Daniel & Cox, Robert H. (eds.) (2011): Ideas and Politics in Social Science Research. New York: Oxford University Press. Rich, Andrew (2004): Think tanks, public policy and the politics of expertise. Cambridge University Press. Ross, George (1995): Jacques Delors and European Integration. Polity Press, Cambridge. Sapir, André (2005): Globalisation and reform of the European social models. Bruegel Policy Paper. Schmidt, Vivien A.: Reconciling Ideas and Institutions through Discoursive Institutionalism. In: Béland, Daniel & Cox, Robert H. (eds.) (2011): Ideas and Politics in Social Science Research. Schmidt, Vivian A.: Taking Ideas and discourse seriously: explaining change through discursive institutionalism as the fourth ‘new institutionalism’. In: European Public Science Review (2010): 2:1, pp. 1-25. Schmidt, Vivian A.: Discursive institutionalism: the explanatory power of ideas and discourse. In: Annual Review of Political Science (2008), 11. Pp. 303-326.

21

Think tanks in the Danish knowledge regime. DRAFT – please do not cite. Paper for “Post-crisis post-Lisbon Economic and Social Policy: A New Era?” 15.-16.09.11. Jesper Dahl Kelstrup, PhD fellow, Roskilde University. E-mail: [email protected].

Sinn, Hans-Werner, 1995. " A Theory of the Welfare State," In: Scandinavian Journal of Economics, Blackwell Publishing, vol. 97(4), pages 495-526, December Stone, D. and Denham, A. (eds.) (2004): Think Tank Traditions, Policy research and the politics of ideas. Manchester University Press, Manchester Stone, D.: Recycling Bins, Garbage Cans or Think Tanks? Three Myths Regarding Policy Analysis Institutes. In: Public Administration 2007. Vol. 85, No. 2 (259-278). Torfing, Jacob: Fra ‘Welfare’ til ‘Workfare’. Nye udfordringer til velfærdsprofessionerne. In: Gjallerhorn, no. 10. 2009, pp. 6-12. Watt, Andrew: Economic Governance Reform: The Good, the Bad and the Missing Make For an Ugly Mix. In: Social Europe. Oct. 5, 2010. Weaver, R. K.: The Politics of Blame Avoidance. In: Journal of Public Policy, 1986, 6 (4), pp. 371398. Ullrich, Heidi: European Union think tanks: generating ideas, analysis and debate. In: Stone, D. and Denham, A. (eds.) (2004): Think Tank Traditions, Policy research and the politics of ideas. Manchester University Press. Manchester. Interviews [SFI MIN:SEK] interview with SFI 21.03.2011. [AE MIN:SEK] interview with AE 22.03.2011. [CASA MIN:SEK] interview with CASA 31.03.2011. [CV MIN:SEK] interview with CEVEA 04.04.2011. [CP MIN:SEK] interview with CEPOS 06.0.2011.

22