Jump to ContentJump to Main Navigation
Bridging the GapsLinking Research to Public Debates and Policy Making on Migration and Integration$

Martin Ruhs, Kristof Tamas, and Joakim Palme

Print publication date: 2019

Print ISBN-13: 9780198834557

Published to Oxford Scholarship Online: April 2019

DOI: 10.1093/oso/9780198834557.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy).date: 25 May 2019

The Politicization of Evidence-based Policies

The Politicization of Evidence-based Policies

The Case of Swedish Committees

(p.127) 9 The Politicization of Evidence-based Policies
Bridging the Gaps

Kristof Tamas

Oxford University Press

Abstract and Keywords

The Swedish experience of government committees offers interesting examples of the diversity of efforts to make research relevant for policy-making. The Swedish case is also an illustration of how the research–policy dialogue may suffer from the gap between different research and policy ‘cultures’. These need to be bridged through dialogue and exchange in order to avoid a demise in the relationships between researchers and policy-makers. The aim of this chapter is to draw on the Swedish example with government committees to discuss critically the potential opportunities, benefits, and pitfalls when attempting to bridge the gap between research and policy-making. The chapter will also cast some new light on the claim that the research–policy nexus in liberal democracies is characterized by ‘the simultaneous scientification of politics and the politicization of science’.

Keywords:   committee, research–policy dialogue, think tanks, migration, integration, politicization


Purely scientifically based research funding in liberal democracies tends to limit the utility of research for policy-making. As a result, part of government-funded research is steered towards policy-oriented studies to cater for the government’s interest as well as, allegedly, that of the public. Over the past couple of decades there has also been a general growth in the interest in producing more evidence-based policies (Davies et al. 2000: 2). Different models have been developed in this regard in terms of organizational, institutional, and funding structures (cf. Hoppe 2005).

The Swedish experience of government committees offers interesting examples of the diversity of efforts to make research relevant for policy-making. The Swedish case is also an example of how the research–policy dialogue may suffer from the gap between different research and policy ‘cultures’. These need to be bridged through dialogue and exchange in order to avoid the demise of relationships between researchers and policy-makers (Davies et al. 2000: 360). The aim of this chapter is to draw on the Swedish example of government committees to critically discuss the potential opportunities, benefits, and pitfalls when attempting to bridge the gap between research and policy-making. The chapter will also cast some new light on the claim that the research–policy nexus in liberal democracies is characterized by ‘the simultaneous scientification of politics and the politicisation of science’ (Weingart 1999: 151–61).

After a brief conceptual introduction, I will review the Swedish committee system, in general, and the functioning of migration and integration committees (p.128) working in the 1970s through to the 1990s, in particular. I then discuss a specific case of a committee where this form of research led to a major public controversy. I also discuss the pros and cons of another approach, whereby policy-makers pick and choose from the menu of interpretations offered by think tanks funded by non-governmental sources, or by academia. The chapter concludes by considering alternative forms of policy-driven research in more recent committees.

The Research–Policy Nexus

It is clear that policy-makers often refrain from taking into account available evidence, or even go against sound evidence provided by the academic community, as ‘many alternative examples of policy initiatives…seem to either fly in the face of the best available research evidence on effectiveness or, at the very best, are based on flimsy evidence’ (Nutley and Webb 2000: 13). While this may have contributed to more widespread post-factual politics (Villumsen Berling and Bueger 2017: 332–41), the public trust in science varies according to political ideology and interest group, and also according to social class, ethnicity, and gender, which may also lead to political divisions influencing policy-making (Gauchat 2012).

Existing research has proposed a range of different conceptualizations and ‘models’ of the research–policy nexus. Hoppe (2005) makes a broad distinction between ‘advocacy models’ and ‘learning models’:

In advocacy models, science is considered one among multiple political voices that enable political debate, judgment, and decision. In the learning models, all actors are constructed as ‘inquirers’ engaged in a process of social learning through social debate.

(Hoppe 2005: 211)

The advocacy models portray the research–policy nexus as a non-exclusive relationship, where researchers compete with other actors attempting to influence policy-makers (Tellmann 2016: 14). Researchers are thus not entirely neutral and objective as ‘each voice in the political arena is considered to be an advocacy plea in favour or against positions defended by other political actors’ (Hoppe 2005: 210). In the learning models, research, and policy communities are regarded as equal participants in a forum for debate in the quest to find acceptable solutions to identified problems. Studies of ‘enlightenment’, ‘knowledge creep’, ‘knowledge shifts’, or ‘research as ideas’, have shown how scientific research historically has had various, sometimes unintended, impacts on policy-making. For some researchers, such impacts could probably be welcome if their findings are used as an aid in defining problems and policy options. They could, however, also be detrimental in (p.129) cases where research is being over-simplified and reshaped into unscientific arguments, or used selectively merely to legitimize predefined political positions (Hoppe 2005: 203; Daviter 2015; see also Weiss 1977, 1980, 1991: 311).

While researchers often want to avoid secluding themselves in an ‘ivory tower’ and cannot always control how their output will be used or interpreted, they need to maintain their autonomy in relation to policy-makers (Villumsen Berling and Bueger 2017: 115–19). Moreover, there are also risks with Wildavsky’s notion of ‘speaking truth to power’ (Wildavsky 1979), as an exaggerated emphasis on the production of knowledge for policy-making as steering primarily through knowledge and experts could be to the detriment of open and more inclusive political, democratic deliberations. A further risk would be if the evidence produced for the sake of reinforced policy-making were to end up in the form of empty rituals that policy-makers came to ignore (Ahlbäck Öberg 2011: 764).

Assessments of research–policy dialogues have also pinpointed the different ‘cultures’ that need to be bridged through open exchange and improved commissioning of research in order to close the existing gaps. Davies et al. (2000) consider that:

policy makers and practitioners complain that research is, among other things, slow, impenetrable, irrelevant, expensive, and inconvenient or uncomfortable. However, to some extent this reflects failures in the commissioning process and failures to understand the nature of research rather than failures of research. (Davies et al. 2000: 360)

Finally, research may become politicized, especially in highly contested issue areas such as migration and integration. However, politicization does not necessarily mean reduced utilization of knowledge or evidence in policy-making. It may, nevertheless, shift the position of utilization from instrumental to symbolic. Symbolic utilization either substantiates pre-existing policy positions, or is used to legitimize policy positions (Boswell 2009; Scholten and Verbeek 2015; Chapter 2 in this book).

The Swedish Tradition of Government-funded Committees: Functioning and Critiques

There is a 400-year tradition in Sweden of relying on committees (similar to those in Norway and Denmark) in order to collect information and research, analyse data and prepare proposals as an input to processes of legislative change and broader policy-making. These committees represent different forms of bridging the research–policy gaps.1 Therefore, they serve as a good case for analysing the extent to which they may be mainly characterized as (p.130) advocacy models or learning models, and the extent to which they may avoid politicization.

Most of the major legislative and policy reforms in Sweden have been based on the report of a committee of some sort (ESO 1998: 57). Committees of inquiry, commissions, more long-standing delegations or expert groups may be led by professional experts, such as professors or legal experts, or politicians within the theme to be covered.2 This is a useful aid since Swedish ministries in international comparison have been small in terms of staffing and resources (SOU 2016: 338). This injection of knowledge and evidence could, in an optimistic scenario, bring more well-founded decisions (Zetterberg 1990: 307; Amnå 2010: 556).

The proposals put forward by committees are usually circulated to all concerned stakeholders for input, including government ministries and agencies, labour market bodies (employers’ associations and unions), organizations, civil society, and the private sector. Committees, in the sense of Bordieau’s concept of social capital and from a broader research perspective on social trust, may be seen as instruments for political deliberations with the aim of bringing opposing views closer together (Trägårdh 2007: 254, see also 261).

In recent decades, there have been important changes in the way that committees have worked and in the way they have influenced policy-making. For example, there has been a relative reduction in the share of parliamentary committees involving both political parties and organized interest groups in relation to inquiries headed by a special investigator (ESO 1998: 57). In addition, Swedish interest groups have noted deterioration in the opportunities to influence government policy-making through the committees (Lundberg 2015a: 44, 56; 2015b). There has also been a relative reduction in the share of committees that take on the major political challenges, while there is an increasing share of committees dealing with somewhat marginal and technical issues (ESO 1998: 57).

An evaluation of the committee system in 1997 by the Parliament’s Auditors, found that the quality of the committees’ work had deteriorated, partly as a result of reduced time given to the committees in which to prepare their analysis and provide their results. A 2004 follow-up evaluation by the Swedish National Audit Office noted that these deficiencies remain, that the remit with which working committees are tasked does not correspond to the time allowed for delivery, and that several committees have deficiencies in the data they use as background to the analysis and proposals (Riksdagens revisorer 1997/98: 12; Riksrevisionen 2004: 8). The appointed researchers do not manage to contribute with novel research and analysis, and then the committees, become more reliant on expertise from within the ministries (Amnå 2010: 558). Indeed, one study has shown that Swedish governments (p.131) have exercised increasingly tighter control over the committees (Riksdagens revisorer 1997/98: 70, referring to Johansson 1992).

Similarly, the Administrative Policy Commission (Förvaltningspolitiska kommissionen) noted in its report that, since the 1980s, the committees had been increasingly brought closer to the government offices and that they were compiling existing research, rather than collecting their own primary data and analysing it to generate new knowledge. It argued that the government should, instead, make greater use of the committees to undertake independent evaluations and take on board the full value of research (SOU 1997: 98–9). The Control Commission (Styrutredningen) suggested that the committees should be used more strategically to compile knowledge as counter-narratives whenever a reform of government agencies and their work is to be considered (SOU 2007: 260, 264–5).

It has also been argued that cases of mounted political steering—preordained results and the premature preparation of legislative proposals before committees’ final proposals are completed—risk damaging the overall political system (Erlingsson 2016). In contrast, it has been suggested that it is a myth that the government has good control over and the ability to direct Swedish committees. This is illustrated by how directives are often rather vague and ambiguous, and, in addition, are frequently not properly applied by the committees. Nevertheless, the government continues to appoint committees regularly, as there are relatively few alternatives in Sweden compared to other countries as far as think tanks and other independent foundations for research are concerned (ESO 1998: 57).

To conclude this broad review of committees in Swedish policy-making: in its ideal form, the Swedish committee system may function as a depoliticized forum from which the government may harvest science-based analysis and preparatory work for its legislative reforms and policy development. Recent critiques have, however, pointed to considerable challenges with this system. Moreover, what happens when the policy field to be analysed by a committee becomes highly politicized, as in the case of migration and integration?

Government-funded Migration Research in Sweden

The Swedish government, as well as the Swedish media, started to show an interest in migration research in the 1970s. The government wanted close cooperation between government officials and researchers with a view to providing an improved knowledge-base for policy reforms, and to evaluate ongoing policy frameworks. Researchers, in general, still had a limited impact on the policy-making process, however. The Swedish government, nevertheless, took the initiative to commission various sector-based research projects that would otherwise not have been carried out within the regular university (p.132) system. Such commissioned research encouraged a burgeoning group of scholars to explore the field of migration research. This period was characterized by a realization that immigrants who had arrived as temporarily needed labour were likely to remain for the long term, and that more systematic policy-making and planning was now a necessity. The government asked researchers for empirical analyses of past and contemporary migration patterns, trends, and consequences. The resulting research provided input to the preparations for both integration and migration control policies (Hammar 2003: 10; 2004: 19).

In the period 1975–1983, the government appointed an expert group (Eifo—the Expert Group on Immigration Research) within the government offices to work on immigration and integration issues. It initiated, coordinated, and published policy-relevant studies on the emerging integration issues in Sweden (Hammar 2003: 10; 2004: 19). Several of the researchers employed by Eifo were moved to Stockholm University in 1983, where a new centre (Ceifo—the Centre for Research in International Migration and Ethnic Relations) was established. At the same time, Deifo (the Delegation of Immigration Research) was set up to initiate and share information about research in the area of migration and integration. A government decision in 1990 transformed this body, together with the Delegation for Social Research, into a social sciences research council.3

These changes had partly been motivated by the view that research also needed to be established in this field within the more regular university research structures. It was probably also due to the need to make research more independent from the government. For a period, however, much research continued to be initiated by the government. For instance, the first professorship in research into International Migration and Ethnic Relations (IMER) was set up in 1988 on the basis of a political initiative in the Swedish Parliament (Riksdag) (Hammar 2004: 30, 33).

In the 1980s and 1990s, migration started to become much more of a politicized issue in Sweden (Byström and Frohnert 2017). An increase in immigration, especially that of refugees and family members, from a much wider range of source countries and regions gave rise to a range of challenges for integration, such as segregation, unemployment among immigrants, and xenophobia. Some parliamentarians and researchers were quite critical of what they saw as an increasingly restrictive migration and integration policy at the end of the 1980s (Spång 2008). This was perhaps one reason why the different governments around that time became more sceptical of approaching academic researchers and asking for advice:

Maybe policy makers in this new restrictive migration period give less priority to migration research, not because the field has become less important for politicians (p.133) and for all of us, but quite the reverse, because it has become a politicized and politically sensitive issue.

(Hammar 2003: 11)

Tomas Hammar has noted that, sometimes, politicians appoint a committee and commission research merely as a means of prolonging the decision-making period, or as an excuse for inaction. Nevertheless, commissioned research initiated by politicians, he argued, should be open-ended and not designed in a way that a ready-made answer must be sought by the researchers. He believed that researchers need to be allowed to apply their regular methods of inquiry, seek new knowledge, and interpret it with the same high scientific standards that they would normally apply in a university environment (Hammar 1982: 22).

Overall, the experiences from the period with Eifo, Deifo, and Ceifo in Sweden were—according to Hammar, who was very much involved with all three entities—that research conducted under these premises remained independent and of high quality. The political steering was mainly limited to the themes and topics chosen for the studies, which were often outlined in a very broad, sweeping manner (Hammar 1988: 19). However, the relationship between the political leadership in the ministry and the appointed researchers was also criticized after a while. At the time, there was a lack of immigration research at the universities in Sweden; in addition, there were requests from various stakeholders that the policy-relevant research should be conducted at a university rather than within the structure of a ministry (leading to the setting up of Ceifo) (Wigerfelt and Peterson 2010; see also Hammar 1994: 13–18; 2008: 61).

During this period, relations between researchers and policy-makers were, in general, good, thanks to the engagement of numerous researchers in the work of committees. This enabled many researchers to participate in the ongoing ‘problem definition’: ‘Swedish migration researchers have made critical assessments of migration and integration policies and even highly critical positions seem to have been taken into account’ (Jørgensen 2011: 103). In the early part of the period, the nexus was characterized by consensus, whereby researchers informed policy-makers. Subsequently, however, conflicting researcher positions emerged, which also had an impact on the overall research–policy dialogue (Jørgensen 2011: 105).

This observation regarding the conflicting research positions motivates the case study in the following section, which is presented in order to illustrate the more recent characteristics of the research–policy nexus in the field of migration and integration policies in Sweden.

The Committee that Capsized

A low degree of politicization in the migration field seems to have enabled a larger share of learning (a forum for debate) than advocacy (competition), (p.134) according to the models initially outlined in the research–policy nexus.4 What happens, however, when migration becomes much more politicized, as has been the case in recent decades? It has been noted that ‘[i]n politicized settings, research and expertise are much less likely to be used as an authoritative source of policy-making, as this could be interpreted as a threat to political primacy’ (Scholten and Verbeek 2015: 2, referring to Hoppe 2005). We shall now take an in-depth look at a case where the research, the committee, and the ensuing debate became highly politicized, and the use of knowledge became more symbolic.

An unusual form of committee in Sweden is one constructed in such a way that it should entirely focus on knowledge-production. The government has, in these cases, chosen to make the committee researcher-driven. Work is often linked to an academic environment, or based on reports being commissioned from external researchers. Some of these committees have not even been called upon to make concrete proposals for changes in policy or legislation, in order to make them even more independent, only being requested to take into account research and knowledge, and not being required to take responsibility for policy-making decisions (Ahlbäck Öberg 2004).

One example is the Integration Policy Power Commission (Ministry of Culture 2000). It was set up in 2000 as an important inquiry into integration policy, including discrimination, in Sweden. The head of the committee was Anders Westholm, a professor in political science and a well-known and highly respected researcher on power and influence. Two of the researchers who were appointed as experts on multiculturalism and discrimination in the committee—Paulina de los Reyes, an economic historian, and Masoud Kamali, professor at the Centre for Multiethnic Research—chose to resign after having expressed concerns regarding the working methods and priorities of the committee.

The critique also centred around the argument from de los Reyes and Kamali that the committee’s research had taken a faulty path due to Westholm’s alleged lack of expertise in research on discrimination and multiculturalism. According to de los Reyes and Kamali, too few researchers with adequate research expertise and an immigrant background were involved (SvT Nyheter 2004). When resigning in April 2003, they published a debate article in the daily newspaper Dagens Nyheter (Daily News). They claimed that the committee was driven by the norm that ‘Swedishness’ is self-evident and natural, while immigrants are different and thus subordinated to others in mainstream society. They claimed this to be an example of the very institutional, structural discrimination that should be the real mission of the committee’s study. This was also the reason why, they stated, they chose to resign (Dagens Nyheter 2003).

These claims were refuted by Anders Westholm. The committee had been tasked with analysing the distribution of power resources and influence (p.135) among people with a foreign background and the rest of the population, and to develop and test hypotheses around what affects the power and influence of immigrant groups in Swedish society (Dagens Nyheter 2003). The minister responsible, Mona Sahlin, a Social Democrat, agreed with de los Reyes and Kamali in much of their critique, including their observation that there is structural discrimination in Swedish society. She argued that she did not take sides in the academic debate but, rather, that she wanted to highlight not only the conflict itself, but also that there is structural discrimination within the academic world (Dagens Nyheter 2004a; 2006). Westholm felt compelled to resign (Ahlbäck Öberg 2004) and the government announced the appointment of Kamali to a new committee on Power, Integration and Structural Discrimination (Justitiedepartementet 2004). Moreover, the terms of reference for Kamali’s new inquiry stated basically, as self-evident, that there is structural discrimination in Sweden (Justitiedepartementet 2004).

Subsequently, the eight researchers who remained on the Integration Policy Power Commission wrote a debate article in Dagens Nyheter. They defended Westholm (Dagens Nyheter 2004b), and his interpretation of the terms of reference. They were critical of integration minister Mona Sahlin, arguing that, instead of meeting her responsibility and creating an atmosphere where Westholm could go on working, she had acted in a way that forced Westholm to resign, as he had sensed that she did not trust him to complete his mission. Those researchers in the expert group who had not resigned tried to meet with Sahlin to discuss the matter but they were not presented with any opportunity to do so, according to media sources (Dagens Nyheter 2004b).

Finally, according to the eight researchers, the government should instead have given the research councils (forskningsråden)—for instance, the research council for working life and social sciences (forskningsrådet för arbetsliv och socialvetenskap)—the task of distributing research funding for these kinds of special requests. Researchers could then apply for funding in the usual competitive environment. They stated that the politicization of research by minister Sahlin was very worrying and could harm both the legitimacy of research and the committee system itself (Dagens Nyheter 2004b). As a result, there was extensive criticism of the minister, mainly from conservative media and commentators (Söderlund 2005). Around seventy researchers—among them very senior members of high-profile academia such as political science professors Bo Rothstein and Olof Petersson—wrote to minister Sahlin protesting against the politicization of the committee’s working methods, a situation which the actions of the minister had allegedly created (Svenska Dagbladet 2003; Dagens Nyheter 2004b; Rothstein 2010).

One observer—the political scientist Shirin Ahlbäck Öberg—argued boldly that ‘with her conduct, Sahlin dictates the scientific formulation of concepts and approaches and points out the “right” research orientation. You get the (p.136) impression that the government with such a decision marks what kind of response they want on the questions formulated in the terms of reference’ to the commission (quoted in Borg 2006: 12).5 However, at the same time, many of the academics who took part in the debate reflected a scientific, rather than political, gap between a positivist epistemology and ontology (represented by Westholm) and critical, postmodern, and postcolonial approaches (as promoted by Kamali) (see, e.g., Svenska Dagbladet 2003). A later study questioned whether there was a ‘danger of explaining too much’ in the Kamali approach: ‘Is it possible to capture the empirical complexity of this field aided by one diagnosis alone?’ (Brekke and Borchgrevink 2007: 38).

Kamali’s final report to the government was published in 2006 (SOU 2006), but the responsible Social Democratic minister at that time, Jens Orback, received it with great scepticism, arguing that it lacked clear legal proposals and fundamental calculations of costs (SvT Nyheter 2006) and that many of its suggestions already formed part of Swedish integration policy (Mavi 2007). Meanwhile, Kamali referred to his results as having shown that structural discrimination is both a cause and a consequence of inequality in Sweden, and that it is created and made permanent through the complex interaction between structural conditions, institutional arrangements, and individual actions. Kamali saw current integration policy as counter-productive and as having the wrong focus, and as having therefore stalled. He argued that radical change, including affirmative action, was necessary (SOU 2006:79).

A few years later, the researcher Stefan Jonsson reasoned that, as soon as the ‘immigrants’ had gained the right to formulate the problem of integration, ‘the political and media elite’ in Sweden had chosen to ignore the problem. Jonsson began to argue that discrimination was not a major problem but, rather, only marginal (Jonsson 2008). Despite this, Jonsson argued that Kamali’s committee had not worked in vain. Although very little came out of the committee’s proposals in terms of politics, the ideational and scientific results were notable in that they moved from a narrow group of researchers into a broader societal context (Jonsson 2008).

A harsh verdict on the whole case was written by the political scientist Ahlbäck Öberg (2004), who sided with Westholm. She argued that, in hindsight, it could be noted that the idea of investigations into power relations in the form of major research programmes had been based on what transpired to be a naive assumption: that, within the framework of the committee, it would be possible to maintain a barrier between policy and research. If the government wishes more research in a particular area, it would be better, she argued, if they allocated funds to the traditional research councils instead.

This case has illustrated how the research–policy nexus became highly politicized and how knowledge utilization became symbolic—politicians using research-as-ammunition (cf. Chapters 2 and 3 in this book; Weiss (p.137) 1979). Considering this controversy, which I have depicted partly through biased voices, we may draw the conclusion that governments should ensure that the researchers they appoint to their committees are enabled to work independently, so that any doubts about the legitimacy of their work and findings can be eschewed. This would possibly require a broader reform of the committee system to ensure greater transparency in the formulation of impartial terms of reference for committees, as well as in the selection of their members—for example, by consulting parliament and relevant research councils.

Politicized Research Funded by Non-governmental Think Tanks

It has been suggested that the politicization of a policy field reduces the extent of institutionalized research–policy dialogues (e.g. through government-funded committees) to the benefit of ad hoc ‘bridges’, such as think tanks (Scholten and Verbeek 2015: 2–3). It could thus be argued that, as an alternative to government-funded committees, governments should turn to think tanks or independent academic researchers when seeking scientific evidence for their policy reforms. Government-funded research is, in this regard, subject to competition in the public arena from think tanks funded by other, non-governmental actors.

This section will show, however, that such research is also frequently used for political purposes in order to gain legitimacy for a proposed course of action—the legitimizing function of symbolic knowledge as outlined by Chapter 2 in this book. Moreover, we may also discern the second symbolic use of knowledge as referred to in Chapter 2, the function of substantiation—‘lending credibility to particular preferences or claims’. This case can therefore also be instructive in relation to advocacy models—competition among various research positions—concerning the research–policy nexus and, indirectly, to the role of publicly funded research.

I will highlight these points by discussing the recent debate about low-skilled jobs in Sweden as depicted in two reports. The debate raises questions about the selectivity of research results, rather than the quality of research itself. Political parties and the labour market parties (employers’ associations and unions) use specific research to legitimize their arguments. The research results referred to in the debate may be consistent but (depending on the methods used, the discipline of the researchers, and the time-period covered) results may be contradictory. This is quite common in daily political debate but may be problematic from the perspective of public deliberation.

(p.138) The Swedish think tank, the Centre for Business and Policy Studies (SNS), published their regular Economic Policy Council Report on the theme of inclusive labour market policies in 2017. The report noted that wages in the Swedish labour market are highly compressed and that the labour market is characterized by very low wage dispersion, which has been more or less static since early 2000. This had allegedly made it more difficult for immigrants and newly arrived refugees to establish themselves in the labour market. Moreover, there are very few jobs in Sweden that are suitable for low-skilled employees—in fact, Sweden has the smallest share among all EU member states. The major challenges in the labour market are concentrated on how to enable low-qualified persons with a foreign background to enter the labour market given the relatively high entry-level wages. Therefore, the study suggested policy changes and adjustments in the approach of the labour market parties aimed at lowering the barriers to enable the low-skilled to find jobs, including by abandoning all unnecessary formal qualification requirements. According to the report, ‘it is fully possible to differentiate wages more without lowering the wages of experienced or qualified workers within the same agreements’ (SNS 2017: 4).

Creating more low-skilled jobs for immigrants in this way was a move that would not be against the so-called ‘Swedish model’, whereby the labour market parties agree on wage conditions in line with collective agreements. Another report suggesting similar measures was also published in 2017 by the Swedish Labour Policy Council (AER), set up in 2015 as an expert council, funded by the Confederation of Swedish Enterprise but officially working independently from the funder. This report suggested similar ways to target the non-natives with the weakest attachment to the labour market, including the low-skilled. The report included a survey among employers, a significant proportion of whom said that lower costs for salaries would induce them to hire more low-skilled employees with subsidized employment. Consequently, the report argued that the labour market parties should take the initiative to encourage companies to hire newly arrived immigrants for subsidized employment, thus also assuming their social responsibility (AER 2017: 27).

Both proposals became part of the public debate. These are, however, very controversial issues from the perspective of the labour market parties, as well as in the opinion of the traditional left-wing and right-wing blocs in Swedish politics. The issue of low-skilled jobs has been one of the most frequently debated in Sweden in 2016–2018. Previous decades have seen a cemented impasse between the left-wing (leftist, social democratic, and Green) and right-wing (conservative, centre-right, and liberal) political parties in the country. Both sides tend to refer to research that supports their standpoint. Does the lowering of entry level wages risk trapping non-natives in a permanent (p.139) low-waged position? Would there be a risk of reduced wages across the line, including for more well-established workers? Are more low-skilled jobs the key to reducing unemployment and exclusion among newly arrived non-natives? What are the alternatives? Regarding these kinds of issues, it is likely that evidence-based research becomes politicized, especially if provided by various think tanks, but also if it originates at independent academic institutions. This case is an illustration of an advocacy model, as described initially in this chapter. It is also a case of research-as-ammunition in the symbolic use of knowledge by policy-makers (cf. Chapter 2):

The struggle between group interests functions as variety generator and selection environment for scientific arguments that underpin political positions and decisions. Every interest involved will look for the type of scientific expertise that harnesses and legitimizes its pre-formed political stance.

(Hoppe 2005: 210)

Government-funded committees may want to join the debate by presenting balanced research, showing the pros and cons of the various policy alternatives, or they may abstain from participating in the debate so as not to be dragged into the process of politicization. The next section will discuss what is already being done in a few policy-oriented committees in order to avoid the politicization of government-funded research outputs.

Committees that Enable Policy-relevant Research without Direct Political Involvement

A more recent form of committee is characterized by broad and open terms of reference, encouraging the committee to assign external, independent researchers to produce reports over a longer period of time. This could be one way to generate guidance for policy development, as well as evaluations of the policy that has been implemented. There are currently three such committees in Sweden: the Migration Studies Delegation (Delmi) under the Ministry of Justice and the Expert Group for Aid Studies (EBA) under the Ministry for Foreign Affairs (both set up in 2013), and the Expert Group on Public Economics (ESO), under the Ministry of Finance,6 which was up and running in the period 1981–2003, and has again been in place since 2007. The ESO was inspired by the model of the Netherlands Scientific Council for Government Policy (WRR), discussed in Chapter 6 by Monique Kremer.

As I am the current Director of Delmi, these committees will be discussed on a more personal note. All three committees have board members appointed by the government, but their terms of reference are defined in a way that enables the committees to work independently from both the government and the (p.140) researchers they assign to carry out research. Their tasks include evaluating current policy-making, compiling relevant research in their areas of expertise, and commissioning and producing research reports as an input both to policy-makers and to the broader public debate. In theory, there remain some hazards with this model, although it bears the potential to contribute a sought-after evidence base to permit more balanced policy decisions. Arguably, there is still the danger that these committees may sometimes be too close to the government, presenting a risk that they may abstain from criticism of the policies carried out. At the same time, there may sometimes be the risk that they are too distanced from the government to the extent that their research becomes too theoretical and conceptual, and thus of less value for the daily challenges of a government.

The case of the ESO has been analysed with regard to its position—either too close to or too far removed from the government in its non-traditional task of questioning government policies and programmes, and their implementation (Lemne 2010). Tensions between various categories of actors—politicians, civil servants, and experts who were involved in the committee’s work—has been a recurring phenomenon. Its influence on day-to-day government policy-making was somewhat hampered by the fact that it usually worked more slowly with a more in-depth focus than was normally the case among civil servants within the government offices. A somewhat paradoxical outcome of the ESO’s role in the public debate was that, although it functioned under the leadership of the government, its critical reports instead often became a tool in the hands of the opposition. That the ESO was a government-funded committee was hence both a strength and a weakness for its policy relevance.

Nevertheless, I argue that these committees are closer to the idea of the learning model, which I regard as a more ideal kind of research–policy nexus than the advocacy model in the highly politicized field of migration. In this regard, I agree with Hoppe that policy-making could be likened to social experimentation:

the learning model treats the policy process as a sort of research process in two respects: first, a policy or policy program is viewed as a set of hypotheses about the causal links between certain (collective, organizational) acts and a specified (desirable) future state of affairs. Second, policymaking is social experimentation. By close monitoring of the degree of goals achievement and a careful analysis of the causes of deviation, errors can gradually be eliminated.

(Hoppe 2005: 211)

The research–policy nexus is thus, as always, filled with potential pitfalls. The policy-making process is characterized by gradual learning and indirect effects from research. When there is more politicization, research and policy dialogues tend to become more fragmented, while the use of knowledge in (p.141) policy-making tends to become more symbolic (Scholten and Verbeek 2015; Chapter 2 in this book). Therefore, even if researchers take the opportunity to engage in commissioned studies, they need to stay detached in order to maintain their scientific independence and legitimacy in relation to government-funded committees.


The Swedish case includes, at least, three types of research–policy interactions. The first model is the traditional Swedish government-funded committee that can be steered politically in terms of members and research assignment, and where the results are to be reported to the government and dealt with within a formal process. The second model concerns research by think tanks which receive funding from, for example, the private sector or various interest groups, but which may nevertheless be selectively interpreted and politicized by various political interest groups. The third model entails a more independent form of government-funded committee that has no direct political involvement.

In conclusion, it is clear that one lesson learned for today’s policy relevant research in the area of migration and integration is that government committees benefit a great deal from being able to work independently, without direct political steering. This is in contrast to research from think tanks, which is often politicized and therefore does not provide a real alternative to publicly funded research. Research at universities tends to be carried out at a distance from the daily policy challenges, with the risk of being driven by intrinsic issues and being inaccessible to policy-makers as well as the general public. The field of migration should strive towards the learning models and try to minimize the risks associated with the advocacy models. I see Delmi as an attempt to establish such a learning model if it succeeds in balancing its output accordingly.

According to the initially discussed advocacy models, the research–policy nexus often shows that researchers merely play a complementary role to the sources of influence or inspiration of policy-makers. Still, the learning models tell us that research may not only bring indirect effects, but also have unintended effects on policy-making and the research–policy nexus—such as committee research results being used by the opposition rather than only by the government, thus becoming part of broader political deliberations. Here, there is much to learn from experiences of how committees have been set up and how they functioned in the past. While many researchers would like to contribute to policy-making, they have good reason to be aware that their research may become over-simplified and be remoulded into unscientific (p.142) arguments, or used selectively merely to legitimize predetermined political positions. Such risks are heightened when the field of enquiry itself is highly politicized, such as in the fields of migration and integration.

The experience from the Integration Policy Power Commission illustrates the risks of politicians steering research in a political direction. Politics-based evidence may undermine the legitimacy of policy-making. This is why we need more transparency in the committee system when formulating the terms of reference and selecting the researchers tasked with an enquiry. Committees need to remain at a distance from the government. Their reports should be permitted to be either critical or supportive of government policies. As long as such reports are solid, they nevertheless carry the potential to contribute to policy learning, from failures as well as from good (if not best) practices.


Bibliography references:

AER. 2017. The Duality of the Swedish Labour Market—Summary of the Swedish Labour Policy Council report Tudelningarna på arbetsmarknaden. Swedish Labour Policy Council.

Ahlbäck Öberg, S. 2004. Fri forskning i kollision med regeringens ideologiproduktion. www.temaasyl.se/Documents/Artiklar/Shirin%20Ahlbäck%20Öbergs%20artikel.doc

Ahlbäck Öberg, S. 2011. Kunskap och politik—mellan nonchalans och teknokrati, Svensk Juristtidning, 8: 764–771.

Amnå, E. 2010. Speaking truth to power? Statsvetarna och kommittéväsendet. In Kontraster och nyanser: Svensk statsvetenskap i brytningstid, eds M. Jerneck and B. Badersten. Lund: Statsvetenskaplig Tidskrift.

Borg, H. 2006. Som man frågar får man svar: Masoud Kamali, Mona Sahlin och politiseringen av kommittéväsendet. Stockholm: Timbro.

(p.143) Boswell, C. 2009. The Political Uses of Expert Knowledge: Immigration Policy and Social Research. Cambridge: Cambridge University Press.

Brekke, J. P., and Borchgrevink, T. 2007. Talking about Integration: Discourses, Alliances and Theories on Labour Market Integration in Sweden. Institute for Social Research Report 2007:9. Oslo: Institute for Social Research.

Byström, M., and Frohnert, P. 2017. Invandringens historia: från ‘folkhemmet’ till dagens Sverige. Delmi Kunskapsöversikt 2017:5. Stockholm: Delmi.

Dagens Nyheter. 2003. Vi hoppar av Sahlins utredning. DN Debatt, 2003-04-06.

Dagens Nyheter. 2004a. Sahlin har ny syn på integrationen, 31/5-2004.

Dagens Nyheter. 2004b. Mona Sahlin politiserar forskningen, DN Debatt, 2004-01-20.

Dagens Nyheter. 2006. Nytt val kräver nytt taktikspel, Peter Wolodarski, 1/7-2006.

Davies, H. T. O., Nutley, S. M., and Smith, P. C. 2000. Introducing evidence-based policy and practice in public services. In What Works? Evidence-based Policy and Practice in Public Services, eds H. T. O. Davies, S. M. Nutley, and P. C. Smith. Bristol: Policy Press.

Daviter, F. 2015. The political use of knowledge in the policy process. Policy Sciences 48(4): 491–505.

Erlingsson, G. Ó. 2016. Utredningsväsendets förfall skadar det politiska systemet, Dagens Samhälle, 19 February. Stockholm.

ESO (1998) Kommitterna och bofinken, ESO-rapport, Ds 1998:57. Stockholm: ESO.

Gauchat, G. 2012. Politicization of science in the public sphere: A study of public trust in the United States, 1974 to 2010. American Sociological Review 77(2): 167–87.

Hammar, T. 1982. Forskningen och politiken. Invandrare och minoriteter 4.

Hammar, T. 1988. Kunskapens makt eller vem är det som styr?. Invandrare och minoriteter 4–5: 19–23.

Hammar, T. 1994. Om IMER under 30 år. En översikt av svensk forskning om internationell migration och etniska relationer, Socialvetenskapliga forskningsrådet.

Hammar, T. 2003. Migration and refugee studies: Reflections about the field’s political impact and university status. In Migration and Immigrants: Between Policy and Reality, eds J. Doomernik and H. Knippenberg. Amsterdam: Aksant Academic Publishers.

Hammar, T. 2004. Research and politics in Swedish immigration management 1965–1984. In Towards a Multilateral Migration Regime, ed. ICMPD. Vienna: International Centre for Migration Policy Development.

Hammar, T. 2008. Diskrimineringsutredningen (DU) och IMER-forskningen. In Anders Gustavsson, Sonja Olin Lauritzen, and Per-Johan Ödman (eds), Främlingskap och tolkningen vänbok till Charles Westin. Stockholm: Stockholm Universitets.

Hoppe, R. 2005. Rethinking the science–policy nexus: From knowledge utilization and science technology studies to types of boundary arrangements. Poiesis Prax 3: 199–215.

Johansson, J. 1992. Det statliga kommittéväsendetkunskap, kontroll och konsensus. Stockholm: Akademisk avhandling, Stockholms Universitet.

Jonsson, S. 2008. Blindhet över partigränserna. Dagens Nyheter 28 May.

Jørgensen, M. B. 2011. Understanding the research–policy nexus in Denmark and Sweden: The field of migration and integration. British Journal of Politics & International Relations 13: 93–109.

(p.144) Justitiedepartementet. 2004. Kommittédirektiv 2004: 54, Makt, integration och strukturell diskriminering. Riksdagen.

Lemne, M. 2010. För långt från regeringenoch för nära: Expertgruppen ESO: s födelse, levnad och död. Stockholm: Stockholms Universitet.

Lundberg, E. 2015a. Det postkorporativa deltagandet: Intresseorganisationerna i den nationella politiken, Background study to Swedish Government Offices (2016) Låt fler forma framtiden! Demokratiutredningen, SOU: 2016:5, Stockholm: Wolters Kluwers.

Lundberg, E. 2015b. Injured but not yet dead: A bottom-up perspective on the Swedish governmental commissions. International Journal of Public Administration 38(5): 346–54.

Mavi, D. 2007. Integration i återvändsgränd, Dagens Arena, 2 November.

Ministry of Culture. 2000. Fördelning av makt och inflytande ur ett integrationspolitiskt perspektiv, Kommittédirektiv 2000:57. Riksdagen.

Nutley, S., and Webb, J. 2000. Evidence and the policy process. In What Works? Evidence-based Policy and Practice in Public Services, eds H. T. O. Davies, S. M. Nutley, and P. C. Smith. Bristol: Policy Press.

Riksdagens revisorer. 1997/98. Riksdagens revisorers förslag angående kommittéväsendet, Förslag till riksdagen, 1997/98:RR3.

Riksrevisionen. 2004. Förändringar inom kommittéväsendet, Riksdagstryck.

Rothstein, B. 2010. Bristen på integration vilar tungt på Sahlin, Göteborgsposten, 5 October.

Scholten, P., and Verbeek, S. 2015. Politicization and expertise: Changing research–policy dialogues on migrant integration in Europe. Science and Public Policy 42(2): 188–200.

SNS. 2017. Economic Policy Council Report 2017, Policies for an Inclusive Swedish Labor Market, English summary, Stockholm.

Söderlund, T. 2005. Vi har inte råd att misslyckas med rasismen. Stockholms Fria, 12 August.

SOU. 1997. I medborgarnas tjänst: En samlad förvaltningspolitik för staten, Betänkande av Förvaltningspolitiska kommissionen, 1997:57, Finansdepartementet.

SOU. 2006. Integrationens svarta bok: Agenda för jämlikhet och social sammanhållning, SOU 2006:79, Slutbetänkande av utredningen om makt, integration och strukturell diskriminering.

SOU. 2007. Att styra staten: regeringens styrning av sin förvaltning, SOU 2007:260. Stockholm: Fritzes.

SOU. 2016. Låt fler forma framtiden! Demokratiutredningen, SOU: 2016:5. Stockholm: Wolters Kluwers.

Spång, M. 2008. Swedish Immigration Policy and Democratic Legitimacy, Current Themes in IMER Research 8, Malmö University.

Svenska Dagbladet. 2003. Forskare kritiserar Sahlin, based on news from TT, 20 December.

SvT Nyheter. 2006. Orback kritiserar Kamalis förslag, 17 August.

SvT Nyheter. 2004. Utredningen som havererade, by Peter Bagge.

Tellmann, S. M. 2016. Experts in Public Policymaking: Influential, Yet Constrained. Nordic Institute for Studies in Innovation, Research and Education.

(p.145) Trägårdh, L. 2007. Democratic governance and the creation of social capital in Sweden: The discreet charm of governmental commissions. In State and Civil Society in Northern Europe: The Swedish Model Reconsidered, ed. L. Trägårdh. New York and Oxford: Berghahn Books.

Villumsen Berling, T., and Bueger, C. 2017. Expertise in the age of post-factual politics: An outline of reflexive strategies. Geoforum 84: 332–41.

Weingart, P. 1999. Scientific expertise and political accountability: Paradoxes of science in politics. Science and Public Policy 26(3): 151–61.

Weiss, C. H. 1977. Research for policy’s sake: The enlightenment function of social research. Policy Analysis 3(4), Fall: 531–45.

Weiss, C. H. 1979. The many meanings of research utilization. Public Administration Review 39(5): 426–31.

Weiss, C. H. 1980. Knowledge creep and decision accretion. Science Communication 1(3): 381–404.

Weiss, C. H. 1991. Policy research: Data, ideas, or arguments? In Social Sciences and Modern States: National Experiences and Theoretical Crossroads, eds P. Wagner, C. H. Weiss, B. Wittrock, and H. Wollmann. New York: Cambridge University Press.

Wigerfelt, A., and Peterson, T. 2010. IMER: möjligheter och gränser festskrift till Björn Fryklund. Internationell migration och etniska relationer etableras. Malmö: Malmö Institute for Studies of Migration, Diversity and Welfare, Malmö University.

Wildavsky, A. 1979. Speaking Truth to Power: The Art and Craft of Policy Analysis. Boston, MA: Little, Brown & Co.

Zetterberg, K. 1990. Det statliga kommittéväsendet. In Att styra riket. Regeringskansliet 1840–1990. Stockholm: Departementshistoriekommittén, Allmänna.


(2.) Ibid.

(4.) The title of this section is taken from a report by the Swedish television station SvT Nyheter (2004).

(5.) My translation.

(6.) These three committees work in a similar way to Eifo and Deifo, although more independently from the government.