Draft Version, please do not cite

Transcription

1 ECPR General Conference September 2014, Glasgow Do political parties matter in higher education? An analysis of party preferences in higher education policy in four European countries Jens Jungblut (jungblut@iped.uio.no) Department of Education, University of Oslo Draft Version, please do not cite 1

2 1. Introduction Studies on partisan policies in education have been somewhat of a blind-spot in political science. However, recent years have seen a resurgence of interest in this area, which to a limited extent also addressed higher education (HE) policy (Busemeyer & Trampusch, 2011). Parallel, HE policy itself gained more saliency especially in Europe since massification led to a situation where a significant part of the public budget is spent on HE and the sector is expected to deliver solutions for other policy areas (Gornitzka & Maassen, 2011). The limited existing literature on partisan preferences in HE policy mainly uses data and coding of the Manifesto Research Group / Comparative Manifestos Project (CMP) for large-n analysis of party positions (see for example Ansell, 2008, ; Busemeyer, Franzmann, & Garritzmann, 2013). This is problematic as the CMP coding differentiates neither between different levels of education nor between support for educational expansion and improvement. As contrary to secondary education, access to tertiary education is skewed towards the more wealthy part of the population (Lucas, 2001), the difference between levels of education and the desire for educational expansion versus educational improvement creates meaningful dissimilarities. Therefore, it is necessary to use a more refined approach to uncover partisan preferences and the dynamics of party conflicts in HE policy. Furthermore, the existing studies analyze the parties HE policies solely along one re-distributive dimension. However, political parties also show differing preferences in connection to the way to steer public sectors (Ansell & Lindvall, 2013; Gingrich, 2011). Thus, this study provides an in-depth analysis of party positions in HE policy based on a twodimensional conceptual framework that combines a re-distributive dimension with one focusing on control and steering of HE. Methodologically the paper employs a qualitative content analysis to investigate the parts on HE from election manifestos of all relevant parties from four European countries: the UK, Germany (North Rhine-Westphalia), Norway and the Netherlands. The guiding research questions are: In how far do political parties in the four countries differ in their proposed HE policies? How well do the expectations based on the two-dimensional analytical framework describe these partisan dynamics? The next section will present the two-dimensional analytical framework and will be followed by the presentation of the data and methods employed. The fourth section will present the party positions in HE policy in the four countries, followed by a discussion of the main findings. The conclusion summarizes the results, highlights their relevance for the existing literature and presents avenues for further research. 2. Conceptualizing party conflicts in HE policy The expectation of conflicts between parties starts from an assumption that they represent different groups within a population with differing political views that offer opportunities for parties to get electoral support in exchange for corresponding political actions (Peters, 2005). Therefore, it can be expected that parties offer political programs that are appropriate to their electorate as well as the ideological characteristics of the party. This idea that differences in the political views of a nation s 2

3 population shape the political conflicts between the parties has already been well described in the classic work by Lipset and Rokkan on societal cleavages (Lipset & Rokkan, 1967). Most of the existing studies that analyze political party positions on HE policy do so using one analytical dimension, which captures socio-economic conflicts and the question whether HE policy is used as a tool for societal re-distribution (i.e. Ansell, 2008, ; Busemeyer et al., 2013) and are often based on large-n analysis of the CMP data. Both choices introduce analytical problems as discussed below. Whereas the authors analyzing HE policy along the re-distributive dimension are right in their assessment that political parties differ in this respect, a fact also supported by empirical studies analyzing differences in policy outcomes (i.e. Busemeyer, 2007, 2009; Rauh, Kirchner, & Kappe, 2011), they are wrong in assuming that this is the only dimension that structures political competition on HE. Another dimension where political parties show different priorities is linked to the question how and by whom the control over the HE sector is exercised (Jungblut, 2014).This twodimensional approach to the analysis of party position in HE reflects the general trend in the literature on party competition in other policy areas (i.e. Häusermann, Picot, & Geering, 2013; Kriesi, 1998; Kriesi, ). These two dimensions, their potential indicators and related expectations about party positions in HE policy are presented below. 2.1 The re-distributive dimension The re-distributive dimension of party conflicts in HE policy addresses its potential for socioeconomic change. Conflicts here are mainly about the question whether a party uses HE for redistribution or not. As participation in HE offers the possibilities to upgrade one s skills, to limit the threat of unemployment and to acquire more cultural capital, HE has the potential to serve as a political tool for re-distribution (Ansell, ; Boix, 1997). Re-distribution normally manifests in the form of enlarged public spending in the respective sector. Several studies have found that the partisan composition of the government matters for the level of public HE spending (i.e. Busemeyer, 2007; McLendon, Hearn, & Mokher, 2009; Schmidt, 2007). However, contrary to other more direct forms of re-distributive actions, like welfare state programs, it is not possible to simply link parties of the political left with more public spending for HE. Contrary to primary or secondary education, HE is not a public good accessible by all members of a society (Ansell, ). Rather, participation in HE is skewed towards the wealthy part of the population and socio-economic background of students is found to be positively related to their likelihood of attending university (Lucas, 2001; Raftery & Hout, 1993). This means that the re-distributive capacity of HE is dependent on the existing level of participation. If a HE system has low participation rates, then it is mainly the wealthy part of the population that attends HE and therefore enlarged public spending in this sector without significantly increasing participation, would be a tool of reverse re-distribution (Ansell, ; Rauh et al., 2011). In contrast, if a HE system has a large participation rate, parties of the political left favor higher public spending in the sector because of its re-distributive capacities (Ansell, ). At the same time, they should in 3

4 general favor a widening of access to HE, both concerning the number of students and the entrance requirements needed. 1 Another indicator is the level of private HE spending. Introducing or increasing tuition fees is a way to shift some of the costs of HE to its users. In a situation where a left party is facing a HE system with low participation rates, it might use this tool to limit the public costs for expansion of HE (Ansell, ). In contrast, in a situation where access to HE is already expanded, the preference for expansion of private HE spending shifts to parties of the right, as they will use increased private HE spending as a way to deter further access into HE to protect their electorate s labor market advantages (Wolf & Zohlnhöfer, 2009). Another way to increase the possibilities to access HE for less privileged parts of society is increased student support. Therefore, when analyzing a party s HE policy position, the positions on student support also needs to be taken into consideration. To sum it up, the re-distributive dimension of party conflicts in HE describes whether a party favors a HE system that can be characterized as expansive or whether the party prefers a more restrictive system (Jungblut, 2014). Four main indicators should be taken into consideration: 1. Public spending for HE 2. Access to HE, in terms of student numbers and access requirements 3. Private spending for HE 4. Student support One can formulate several expectations on how different political parties position themselves on this dimension (Jungblut, 2014). Social Democratic Parties (SDP) can be expected to support an expansive HE system combined with limiting private HE spending, as this provides strong re-distributive possibilities to its electorate and allows for social upward mobility. Rooted in the idea that individuals should be able to use their skills to improve their socio-economic status by themselves without state interference as well as the notion that the education system should cater to the labour market s needs to ensure the future well-being of today s knowledge economy, Liberal Parties (LP) can be expected to support an expansive HE system. This is mainly to ensure that the economy is well supplied with qualified graduates. However, due to the Liberals focus on the ability of the individual to shape his/her future, it can be expected that they favour some form of individual costs for attending HE. This would have the positive effect that it would ensure that public spending for HE would remain limited. A Christian Democratic Party (CDP) can be expected to be in favour of a more restrictive HE system, on the one hand because of its desire to limit re-distribution and protect the wealthier part of its electorate, but also on the other hand because of its support for vocational professions. Conservative Parties (CP) can be expected to shield their electorate from re-distribution. Therefore, they prefer a restrictive HE system. Such a system would on the one hand limit public spending and on the other hand protect the labour market advantage of the CP s electorate that profits from the skewed-access to HE. Comparable expectations are harder to formulate for other party families. Anti-establishment parties (AEP) at the left and right fringe of the political spectrum are known for having only very limited positions concerning HE policy, making it hard to conceptually pin point them (for details see Ansell, 1 Policies that loosen the formal entry requirements, by for example allowing people with advanced vocational education to enter HE, could partially balance out social selection during secondary education and make HE access less skewed. 4

5 : 137ff). Green parties (GP) are harder to place because their electorate is more and more drawn from high income and highly educated strata, while their election manifestos hold diverse positions having both politically left and right characteristics (Rauh et al., 2011). Thus, it can be expected that GP as well as AEP at the left and right end of the spectrum favor an expansive HE system The control dimension The second dimension is linked to the question who controls how the HE sector (Jungblut, 2014), and it is based on conceptual considerations from recent contributions to the field of party competition that identify conflicts between managers and socio-cultural specialists or libertarian and communitarian values (Bornschier, ; Kriesi, 1998), in essence highlighting differing preferences linked to the distribution of authority, power, autonomy and steering. Ansell and Lindvall (2013) have shown that political conflicts in education focus also on the amount of direct control of the state visà-vis the educational institutions; different political parties prefer different levels of direct control and that this significantly shapes the structure of the educational sector. Since the early 1980s, the relationship between the state and the public sector has become more open to changes (Gingrich, 2011; Olsen, 1988), which include the rise of new public management (Christensen, 2011) and increasing relevance of markets. While there has been strong discursive convergence, there is divergence with regards to implementation, due to institutional legacies as well as decisions by actors (Pollitt, 2001). Therefore, shifts in the composition of governments can cause changes in the objectives of reforms (Pollitt, van Thiel, & Homburg, 2007). Opening up the state - public sector relationship has led to conflicts related to the question, who holds the effective control over a certain public sector, which encouraged political parties to take diverging positions on this issue (Gingrich, 2011). Gornitzka and Maassen (2000) distinguish between four steering modes, which are useful for describing the relationship between HE and the state: (1) the sovereign rationality-bounded mode, where steering lies with the government, (2) the institutional mode, where steering decisions are taken within autonomous universities, (3) the corporate-pluralist mode, where different stakeholders negotiate the steering, and (4) the supermarket mode, which uses market mechanisms to steer the sector. These modes differ with regard to whether a state control model or a state supervising model is used (Vught, 1997). The different forms of political control over HE lead to differing levels of centralisation of power between the state and the HE sector. As political parties have distinct preferences how to steer a public sector and how much autonomy professional communities should enjoy, it can be expected that these also play a role in their positions in HE policy. To capture the different political preferences towards the state HE sector relationship four indicators will be used that resemble the four steering modes: 1. The dominance of the government in controlling HE; 2. The autonomy of HE institutions to steer themselves; 3. Steering through negotiation between stakeholder groups; 4. Steering through market mechanism and competition. 2 In the case of extreme right parties due to the social composition of their electorate which is not participating in HE so far, or in the case of extreme left and Green parties due to their ideological preference for redistribution. 5

6 Therefore, preferring a strong and active state that shapes the life of its citizens, a SDP will favour a more centralised control of HE, and a dominant role of the government, sometimes in combination with the inclusion of stakeholders such as student unions or trade unions. A CDP, being generally in favour of giving competences to local institutions, and putting an emphasis on the qualitative homogeneity of public services will see an important role for the state in assuring an equal level of quality also in HE. Therefore, they will opt for autonomous HE institutions that steer themselves, while ensuring that the government uses its power to assure an even quality. CPs support more de-centralised control, because they prefer smaller state structures and more streamlined public services, often combined with forms of market competition. Furthermore, CPs are not opposed to heterogeneity in the quality of HE given their preference for market competition. Therefore, they should favour autonomous HE institutions which are competing in the marketplace. Just like CPs, LPs support a small public sector, a more streamlined state structure and market competition. They see heterogeneity in the quality of HE not as a problem for the state but rather as an issue which is regulated through market mechanisms. They also support the autonomy of HE institutions to enable them to compete freely with others and find their niche in the (inter-)national HE market. Thus, LPs can be expected to favour market-based steering while granting HE institutions substantive autonomy. It can be expected that AEPs at the left and right fringe of the political spectrum, would also prefer the government to be the dominant actor in HE steering. GPs, due to their more diverse electorate, should favor a steering approach based on the inclusion of different stakeholder groups. These eight indicators and the related expectations provide the conceptual basis for a qualitative content analysis of election manifestos. 3. Methods and data Even though the data for each country is presented in a concise manner in the following section, the analysis is based on a detailed qualitative content analysis (QCA) of political parties election manifestos. The focus is on those parts of the manifestos that address HE policy. The paper uses a qualitative approach for two reasons. First, the analysis is exploratory, given that so far there is only very limited research analysing in detail the differences between political parties. Second, the CMP does not offer a separate and sufficiently elaborated coding for HE policy, 3 necessary for distinguishing between party positions for different educational sectors or more specific policy preferences (Busemeyer et al., 2013, p.10). Given the aim of the study, it is necessary to employ a more detailed approach and analyze in-depth a limited number of cases to uncover the substantive party positions linked to HE. The QCA performed for this paper is based on a coding scheme, generated using the indicators presented earlier. The coding was performed by a single coder using the MAXQDA software package. The data consists of election manifestos of all relevant parliamentary parties in the four countries under study for two consecutive elections that lead to differing governments. Election manifestos are seen as the best possible data source to extract policy positions from as they are public documents, geared towards a greater audience and written in a rather clear fashion to get policy positions transported to the electorate (Budge, 2001). They are the only collective policy statement a party 3 One code is used combining both expansion and improvement, see: 6

7 makes, written by the party leadership and functioning also as a unifying pressure towards both the members and the leaders (Laver & Garry, 2000). The manifestos have been retrieved from the raw data of the CMP project, the database of polidoc.net and directly from the parties. The countries for this study, Norway, the Netherlands, the UK 4 and the German Bundesland of North Rhine-Westphalia (NRW), were chosen for several reasons. First, as HE policy in Germany is within the competences of the Bundesländer and not the federal level, it is necessary to analyse the partisan dynamics on HE on this level. NRW was chosen because it is the largest federal state. Second, the countries differ in terms of several institutional variables that potentially impact the partisan competition on HE, such as varieties of capitalism (Iversen & Stephens, 2008), HE and skill systems (Graf, 2013), types of democracies (majoritarian or consensus) (Lijphart, 1999). At the same time, all are European countries, ensuring a common socio-cultural and historical background and therefore the applicability of the two-dimensional cleavage structure (Busemeyer et al., 2013 p.531). Focusing on two consecutive elections limits the timeframe of the analysis in order to minimize the influence of intervening supra-national or global dynamics, such as the Bologna Process or shifts in the so-called global scripts (Gornitzka & Maassen, 2011). In total the database consists of the following election manifestos: Table 1: Overview of countries, elections and parties manifestos included. Country Election year Parties UK 2005 Conservative, Labour, Liberal Democrats Conservative, Labour, Liberal Democrats NRW 2005 SPD, CDU, FDP, Greens, WASG 5 SPD, CDU, FDP, Greens, Die Linke Norway 2009 AP, FrP, Høyre, KrF, SP, SV, Venstre 2013 AP, FrP, Høyre, KrF, SP, SV, Venstre Netherlands CDA, CU, D66, GroenLinks (GL), PvdA, PVV, SGP, SP, VVD 2012 CDA, CU, D66, GroenLinks (GL), PvdA, PVV, SGP, SP, VVD 4. Partisan HE policy position This section presents the results of the QCA in a condensed tabular form, where a plus sign symbolizes a party s support and a minus symbolizes a party s opposition. The use of brackets indicates a less clear position, which will be explicated in the text. 4 In the UK the focus will be on England, as Scotland and partially also Wales have the authority to formulate their own higher education policy. Nevertheless, the political debate around higher education in England happens during the UK parliamentary elections. 5 In the 2005 elections both the WASG and the PDS ran as separate parties, with the WASG being the more successful one reaching more than 2% of the votes. In both parties had merged into the new party Die Linke. 7

8 4.1 The UK Of the four countries, the UK election manifestos include by far the fewest statements addressing HE policy, in particular concerning the control dimension. Table 2 presents a summary of the party positions. Table 2: Party positions in the 2005 & UK elections. Control Indicators Public spending Private spending Access (student numbers) Access (qualifications) Student support Redistribution Government control Autonomy of HE institutions Stakeholderbased steering Market mechanisms Conservative 2005 Conservative Labour Labour Lib Dem (+) - - (+) + + (-) Lib Dem With regard to the re-distributive dimension the Conservatives have a mixed profile. Concerning private HE spending they move from proposing in 2005 to abolish tuition fees introduced by Labour, to supporting the Browne Report which argued for significantly higher tuition fees (Shattock, 2012, p.166f), thus supporting an increase of private spending (Conservatives, 2005, p.9;, p.17). They don t have a position related to public spending and the only policy concerning access to HE is the suggestion in to provide extra university places. While the support for an increase in tuition in is in line with the conceptual expectations, the limited expansion in student numbers and especially the anti-tuition fee policy of the Conservatives in 2005 is contradicting them. However, compared to the other two parties the Conservatives still show the most restrictive preferences on the re-distributive dimension. Another explanation for the Conservative s proposal to abolish tuition in 2005 is the 2.5 party system, which creates an incentive to propose opposing policy solutions instead of using issue saliency to distance oneself from other parties. Labour proposes to increase access both in relation to the entrance qualifications as well as a growth in student numbers (Labour, 2005, p.41;, p.3-7). Regarding private spending, Labour advocates in 2005 to continue their policy of a maximum tuition of GBP. They also support increased public spending and the expansion of support for students from poorer families. In Labour still proposes to enlarge student support and public funding, but does not address issues linked to private 8

9 spending. Except for their general support for the existing level of tuition fees, Labour s policies are in line with the conceptual expectation that a SDP should favor an expansive HE system. The support for a limited amount of tuition fee can be explained by either path dependency or as a way to finance an expansion in access without putting too much additional pressure on the public purse (Ansell, ). As the tuition fees are combined with the demand for more support especially for students from poorer backgrounds, their restrictive effect on access to HE is limited. The Liberal Democrats show a stable preference in relation to private spending, as they support to abolish tuition fees throughout the UK (Democrats, 2005, p.12;, p.39). At the same time, they want to enlarge student support, and they have no position regarding public funding. Concerning access to HE, they show a mixed position since they support wider access in relation to the entrance qualifications, but also a balance between HE and vocational training (Democrats,, p.39). The policies proposed by the Liberal Democrats are partially in line with the expectations; although supporting an expansive HE system, as a Liberal party they should also support private funding, which they do not. One explanation for this discrepancy is that by generally opposing tuition fees, they are able to acquire a distinct political profile that separates them from the other two parties. In relation to the control dimension, the Conservatives fulfill the conceptual expectations (Conservatives, 2005, p.9;, p.17). Labour is not positioned on this dimension in either of the manifestos and the Liberal Democrats only address it in and favor, as expected, more autonomy of HE institutions (Democrats,, p.29). Such lack of party competition in the UK on the control dimension is not entirely surprising, considering that ideas of New Public Management have been embraced rather early in the UK (Pollitt et al., 2007), and that the British 2.5 party-system favors political conflicts to be aligned along only one dimension rather than multiple dimensions as it is the case in multi-party systems. 4.2 Germany - NRW The election manifestos for the elections in NRW show a substantial amount of policy positions linked to HE. Table 3 offers an overview of the parties positions. Table 3: Party positions in the 2005 & elections in NRW. Redistribution Indicators Public spending Private spending Access (student numbers) Access (qualifications) Student support CDU 2005 CDU SPD SPD FDP FDP Grüne Grüne WASG (+) (+) + + Linke Control Government control Autonomy of (-) + + 9

10 HE institutions Stakeholderbased steering Market mechanisms (+) + + Overall the party positions on HE policy in NRW show quite stable profiles. In relation to the redistributive dimension, one can identify two blocks that align along the question whether a party favors tuition fees or not. The first block consists of the CDU and FDP, who formed a coalition between 2005 and, promoting increased private spending by introducing tuition fees up to 500 Euros per semester (CDU, 2005, p.21;, p.8; FDP, 2005, p.17ff;, p.14ff), as well as expansion of student support. The Liberals additionally focus on providing more merit-based stipends, financed partially by private benefactors. While the FDP s proposal to expand student support, especially using merit-based measures, fits the expectations for Liberal parties, the CDU s position in favor of more student support contradicts the expectation towards CDPs, as they should be in favor of a more restrictive system. One explanation may be that because the proposal to introduce tuition fees is ground-breaking (there were no general tuition fees so far), aligning it with an increase in student support limits opposition. As the increase in student support is mainly loan-based, additional public costs and the expansive effect on the HE system can be considered as limited. Concerning public spending and access, both parties show slightly different profiles. The Liberals, as expected, support an expansive HE system by promoting more public spending and, especially in, wider access to HE. The CDU proposes in 2005 to initiate a review of the federal regulation on access to universities with the intention to limit the number of newly enrolled students, while at the same time not positioning themselves on public spending (CDU, 2005, p.21). This is also in line with the more restrictive HE system preferred by CDPs. In the CDU proposes to expand public spending, increase student numbers, and widen access by making it easier for people with vocational qualifications to enter HE (CDU,, p.8f). This seemingly contradicts their preference for a more restrictive HE system; however, because the increase in student numbers is linked to an expected increase in the number of secondary school graduates, it mainly serves the purpose to keep the relative level of participation equal. Overall, the Liberals as well as the Christian Democrats confirm the conceptual expectations. The second block is formed by the Social Democrats and Greens (in a coalition since ), and to a certain extend the Left party. These parties, as presented in Table 3, behave in line with the conceptual expectations (GRÜNEN, 2005, p.92ff;, p.80ff; SPD, 2005, p.6;, p.19ff; WASG, 2005). With regard to the control dimension the party differences are also very pronounced. The FDP confirms the expectations by supporting market mechanisms combined with autonomy for universities (FDP, 2005, p.17ff;, p.14ff). The CDU is expected to support a combination of government control and autonomy of universities as a way to steer the sector. In their manifestos they mainly support more autonomy of HE institutions. However, they also stress the need for a diminished role of the state and propose a limited amount of market mechanisms to be introduced (CDU, 2005, p.21). This partial discrepancy can be explained by the fact that the HE system of NRW, like other German Bundesländer, is characterized by a dominant role of the government in steering 10

11 (Vught, 1997). Therefore, even after the implementation of the CDU s reform the government would still have a role in steering HE. The Social Democrats show a mixed profile. While some positions are in line with the expectations - more involvement of stakeholders, in particular internal stakeholders and students and highlighting the role of the government in higher education steering (SPD, 2005, p.6;, p.21) they also propose, much stronger in 2005 than, increased autonomy of HE institutions. Here also the exceptionally strong role of the government in steering HE in Germany is relevant; the proposal of the SPD in 2005 to give universities more room to manoeuver upholds the principle of a strong government influence. Furthermore, in, after CDU and FDP had implemented their autonomy oriented reforms, the SPD opposes most of these reforms in their manifesto. The Greens fulfill the conceptual expectations by supporting the involvement of stakeholders in the steering of HE, combined with the demand for a strong role of the government (GRÜNEN, 2005, p.92ff;, p.80ff). Finally, the Left party argues, as expected, for more state control of the HE sector, also combined with a stronger involvement of stakeholder groups (LINKE,, p.11f). 4.3 Norway In their manifestos to the 2009 and 2013 parliamentary elections in Norway all parties address HE on both conflict dimensions. Even though there are many consensual topics, one can also find party specific profiles (see Table 4). Table 4: Party positions in the 2009 & 2013 elections in Norway. 11

12 Control Indicators Public spending Private spending Access (student numbers) Access (qualifications) Student support Redistribution Government control Autonomy of HE institutions Stakeholderbased steering Market mechanisms AP 2009 AP Høyre Høyre FrP FrP KrF KrF SP SP SV SV Venstre Venstre (+) (+) (+)

13 In general the party competition in HE policy in Norway can be characterized as limited, not due to a lack of positions in the manifestos but rather due to inter-party consensus, in particular, related to the re-distributive dimension. Two very consensual policy issues can be identified: (1) the need to spend more public money on HE and (2) the need to enlarge student support. The introduction of private spending in the form of tuition fees for public universities is a taboo and no party promotes it, while many parties actively oppose the idea, similar to other Nordic countries where fees for national and EU students are also politically impossible, given the Nordic Model of HE with its focus on equal access and public funding (Gornitzka & Maassen, 2011). Access to HE is only a topic for a few parties and there is generally less attention geared towards it. If it is addressed then mainly in connection to widening access by making the transfer from vocational to HE easier. The absence of intense political competition on the re-distributive dimension can be linked to comparatively high participation in HE and the steady income from oil and gas leading to less scarce public funding. Concerning the conceptual expectations along the re-distributive dimension both the Social Democrats and the Socialist Left fulfill the expectation of supporting an expansive HE system (Arbeiderpartiet, 2009, p.24ff; 2013, p.30; Sosialistisk-Venstreparti, 2009, p.17ff; 2013, p.51ff). The liberal Venstre fulfills the expectations only partially. As expected, they support an expansive HE system, however they do not prefer private spending (Venstre, 2009, p.29ff; 2013, p.17ff), primarily because of the mentioned taboo concerning fees. The positions on the re-distributive dimension of both the conservative Høyre and the Christian Democratic KrF are even more contradictory as they do not support a restrictive HE system. However, also here one can contextualize this discrepancy. First, it is questionable whether a really restrictive HE system is per se a viable policy option in a Nordic welfare state. Second, the support of both Høyre and KrF for more public funding can be explained with the exceptionally good funding situation of the Norwegian public sector. Third, the well-funded public sector and the Nordic Model explain why both parties argue for more student support. Noticeably, Høyre plans to increase the cap on personal income with which students are still eligible for student support, to strengthen the rights of students in private HE institutions, promoting increased student numbers in the economically important fields of engineering and technology (Høyre, 2009, p.9ff; 2013, p.15ff). Finally, Høyre is one of only two parties that do not position themselves against private spending in either of its manifestos. Thus, in the context of Norwegian discussions, Høyre can be seen to partially fulfill the expectations towards CPs; they take the most restrictive position in a political discourse generally geared towards expanding HE. Due to the small number of existing European farmer s parties, no conceptual expectations have been formulated in relation to this party family. The Norwegian SP supports on the re-distributive dimension an expansion of the HE system. However, they support an increase in the number of students only in relation to the economically important subjects in the sciences and refrain from positioning themselves explicitly against private spending (Senterpartiet, 2009, p.40ff; 2013, p.59ff). This position is in line with their general profile as a protector of the Norwegian welfare services but also encompasses their earlier orientation to more conservative economic policies (Allern,, p.172). The populist right FrP by and large fulfills the expectations of supporting an expansive HE system (Fremskrittspartiet, 2009, p.62ff; 2013, p.70ff). The parties positions on the control dimension are more diverse and two groups can be identified. The first supports a stronger role of the government in steering, combined with stakeholder involvement. As expected one can find the Social Democrats (Arbeiderpartiet, 2009, p.25ff; 2013, 13

14 p.30) and the Socialist Left in this group, the latter combining this with a greater role of stakeholders (Sosialistisk-Venstreparti, 2009, p.17ff; 2013, p.51ff). Finally, the Farmer s Party SP supports a dominant role of the government in HE steering, combined with a focus on stakeholder involvement, especially concerning representatives of the municipalities (Senterpartiet, 2009, p.40ff; 2013, p.59ff). Even though no conceptual expectations on the position of Farmer s Parties have been formulated, the policies on HE steering proposed by SP fit their general profile of being a left-leaning non-urban party (Allern,, p.172f). KrF s position on the control dimension fits the expectations outlined earlier, given their focus on the coherence of educational offerings throughout the country and common standards for the monitoring of first year students (Kristelig-Folkeparti, 2009, p.71ff; 2013, p.61ff). However, they lack the expected support for more institutional autonomy of universities. The second group favors market mechanisms combined with a focus on the autonomy of HE institutions. This group includes Høyre, Venstre and FrP. Venstre s and Høyre s positions fit the expectations. The profile on the control dimension of the populist right FrP does not fit the expectations outlined earlier in relation to anti-establishment parties at the right fringe of the political spectrum, but rather resembles the profile of a Conservative party, since they promote steering of HE based on institutional autonomy and market mechanisms. This divergence could be explained by the fact that FrP is not a classical case as they have, contrary to other anti-establishment parties, a quite elaborate section on HE policy in both manifestos. Furthermore, this illustrates the normalization of FrP in recent years (a trend that lead them to be part of the government coalition in 2013). Today s FrP is more of a neo-liberal populist party (Allern,, p.211f), and their position on the control dimension also resembles this more liberal shift. 4.4 The Netherlands In the Dutch multiparty system not all parties focus equally on HE policy, but all parties address it in one way or another (Table 5). Table 5: Party positions in the & 2012 elections in the Netherlands. 14

15 Control Indicators Public spending Private spending Access (student numbers) Access (qualifications) Student support Redistribution Government control Autonomy of HE institutions Stakeholderbased steering Market mechanisms CDA CDA 2012 CU CU 2012 D66 D GL GL 2012 PvdA PvdA 2012 PVV PVV 2012 SGP SGP 2012 SP SP 2012 VVD VVD (+) (+) (-) - - (+) + + (-) - + (+) (+) + (+) (+) + (+) (+) (+) (-) (+/-) + (+/-) (+/-) (+/-) (+) (+) (+) (+) + (-) (-) + (+) (+) (+) + (+)

16 In relation to the re-distributive dimension, access to HE is only addressed by a few parties, while all other indicators have been addressed by nearly all parties. With the exception of PVV, who does not take any position, all parties support the idea to expand public spending. In relation to private spending and student support, the parties show diverging preferences. The Christian Democratic Appeal (CDA) by and large fulfills the expectations towards CDPs. They plan to increase private spending (higher tuition fees) and support more public spending, but only in relation to research, innovation and areas of excellence (CDA,, p.28ff; 2012, p.45f). Contrary to the expectations, CDA supports to keep the grant-based student support system stable. Although this can be seen as being in favor of re-distribution, it also reflects the social responsibility inherent in the ideological basis of CDPs. The second CDP, the Christian Union (CU), has a similar profile. They support higher tuition fees, but also more general public spending (ChristenUnie,, p.26; 2012, p.31), as well as grant-based student support. The third CDP, the orthodox Calvinist SGP, has a similar policy profile as CDA and CU, supporting an increase in fees and general public spending (SGP,, p.18f; 2012, p.10), but favoring a diversified support system with additional student loan schemes. All three CDPs, as expected, support a more restrictive HE system, but, in line with the idea of Christian social responsibility, they also maintain the existing grant-based student support system. The manifestos of the populist right PVV offer barely any positions in HE. Concerning re-distribution they propose that all foreign students should pay tuition, leading to a limited increase in private spending (PVV,, p.29&31; 2012, p.43&45). Furthermore, PVV wants to keep the student support system stable. Due to the lack of explicit HE preferences, it is not possible to definitively position PVV on this dimension. The preferences of the liberal VVD are in line with the expectations: more public spending (to enhance the quality of HE), more access, allowing universities to determine fees (thus increasing private spending), shift from grants to loans in student support and abolish additional fees for students who study longer (VVD,, p.13&17; 2012, p.24f). In total, their position is favoring an expansive system in combination with more private funding. The second liberal party D66 supports, as expected, a more expansive HE system that shifts costs towards the users. They advocate an increase in public funding to improve the quality of education (financed by a shift from a grant-based to a loan-based student support) combined with increased grants for students from disadvantaged backgrounds to prevent negative effects on accessibility, expanding access to students younger than 17 (pending necessary qualifications), increase of tuition fees for secondary degrees as well as abolishing fees for students who study longer (D66,, p.37ff; 2012, p.24ff). Although somewhat contradicting, these are only minimal changes to the existing fee regime. Another party that behaves according to the expectations are the Social Democrats (PvdA), as they support an increase in public funding, a shift towards loans combined with scholarships for disadvantaged students, limiting fees for secondary degrees, abolishing fees for longer studying, and an increase in student numbers in technology subjects and the PhD level (PvdA,, p.26f; 2012, p.9f&20). The Greens (GroenLinks) also fulfill the expectations. In they support a limited increase in public spending (focusing on EU spending for HE) and suggesting a student wage for every student to enhance access (GroenLinks,, p.24ff), a position maintained in 2012 through a focus on life-long learning and adult education in HE (GroenLinks, 2012, p.17f). Overall, their idea is 16

17 to limit private spending (reducing fees or abolishing additional ones for longer studies) and increase public spending, shifting student support from grants to loans while increasing the number of grants for disadvantaged students to maintain access. The Socialist SP, as expected, supports an expansive HE system, increasing public spending, keeping the grant system and expanding it for disadvantaged students. In they propose to increase access to HE, keep tuition fees stable (correcting for inflation), while opposing the idea of differentiated fees (SP,, p.25f), while in 2012 they proposes to limit private spending (abolishing additional fees for longer studying, banning higher fees for second degrees) (SP, 2012, p.31ff). In relation to the control dimension the parties show distinct profiles. In general, the three CDPs (CDA, CU and SGP) support, as expected, more autonomy for HE institutions. CU wants to limited the autonomy in relation to the role of humanities and the importance of ethics (ChristenUnie,, p.26; 2012, p.31), which is in line with Christian values. While the CU completely fulfills the conceptual expectations, the CDA and the SGP combine the focus on autonomy with more market mechanisms, giving them a profile that is more in line with the expectations towards liberal or conservative parties. This pattern - parties combining enlarged autonomy with more market mechanisms or a strong role of the government with more stakeholder involvement - is recurrent and will be addressed later on. The two liberal parties VVD and D66 by and large fulfill the conceptual expectations - more autonomy and more market. In D66 limits their support for more autonomy by proposing to set a minimal number of students underneath which study programs are not allowed to be continued and create a new system of Bachelor degrees (group existing programs in more general and efficient clusters). Although both of these proposals limit the universities room to manoeuver, other proposals, i.e. more freedom for HE institutions to decide who to admit enlarges autonomy significantly. PvdA, as expected, focuses mainly on a strong role of the government in HE steering and in 2012 also on stakeholder involvement, especially linked to external stakeholders (regional authorities and businesses). GroenLinks supports government control by regulating stronger the core activities of universities. This focus on the government as the dominant actor is not in line with the conceptual expectations, but can be attributed to the background of GroenLinks being a merger of several antiestablishment parties on the left fringe of the political spectrum, who in turn would be expected to be in favor of a strong government role. At the fringes of the spectrum, SP fulfills the expectations by supporting a strong government but PVV supports more autonomy of HE institutions (proposing more freedom for them in student selection), which is not in line with the conceptual expectations. However, the very limited amount of policies makes it hard to definitively position PVV on the control dimension. 5. Discussion and conclusion The analysis has delivered some interesting results. First, it was shown that parties do offer different policies in HE and that these differences can be structured along two dimensions, one capturing redistributive conflicts and the other conflicts linked to the control of HE. Second, the analysis supports Ansell s finding (Ansell,, p.137ff) that anti-establishment parties at both ends of the political spectrum show less interest in HE (i.e. the German Die Linke or the Dutch SP and PVV). Third, even though most of the parties fulfilled the conceptual expectations, party competition depends not only 17

18 on the party s ideological background but also has a national flavor, since the existing HE system is the point of departure for the parties policy proposals (i.e. the fee taboo in Norway, or the strong role of the government in NRW). This is in line with the argument that the existing HE system creates path dependencies for parties policy proposals and thus limits possibilities to pursue more ideal party positions (Ansell, ). Concerning positions on the control dimension, the parties often combined two of the four steering modes presented in the conceptual framework. Parties that favor a centralized control often combined stakeholder involvement with a strong role of the government, while those that prefer decentralized control combined autonomous universities with the use of market mechanisms. On the one hand, for implementing market mechanisms autonomous HE intuitions are a pre-requisite. On the other hand, the link between stakeholder steering and a strong role of the government reflects the impact of NPM and post-npm reforms (Christensen, 2011). These shifts in the approach to public sector steering made it hard, even for parties that generally favor an interventionist state, to rely solely on classic governmental steering methods. Therefore, involving stakeholder groups, especially those that can be expected to be ideologically closer to the parties positions (i.e. students in the case of SDPs and GPs), combined with significant government influence, offer the possibility to harmonize the parties desire for influence with the specificities of diverse and modern HE systems. Policy harmonization on the European level could generate the expectation of policy convergence. However, the analysis shows that parties offer diverging policies in HE, supporting earlier claims that policy convergence in HE takes place, when governments follow similar policy agendas (Heinze & Knill, 2008). The divergence in the parties preferences towards HE implies that political parties indeed matter in HE policy. In relation to the guiding research questions, the study has shown that political parties do have distinct HE policies and that these differences were by and large in line with the conceptual expectations: The parties propose different policies along both dimensions. On the re-distributive dimension SDPs and GPs support expansive HE systems with limited or no private costs for higher education. LPs also support expansive HE systems; however, they prefer some form of private spending. CDPs showed a preference for more restrictive HE systems; however, they still maintained student support systems and were less restrictive than CPs. In relation to the control dimension, SDPs and GPs preferred more centralized control combining a strong government with stakeholder involvement. LPs and CPs showed a preference for more de-centralized control, supporting autonomous HE institutions and market mechanisms. CDPs showed a mixed profile on this dimension, calling for further analysis. Furthermore, the analysis confirmed that AEPs on both ends of the political spectrum have only limited positions concerning HE. Finally, the results have demonstrated that the structure of the existing HE system influences the party competition by providing path dependencies. This accounts for cross-national variation in HE preferences of parties with similar ideological backgrounds. This has several research implications. Party families were found to show distinct sets of preferences, not only in relation to the re-distributive but also to the added control dimension. Given the limited number of cases, studies including more and different cases using a similar conceptual approach would allow to validate the findings presented above. This can be done for other European countries, given socio-cultural similarities and the common HE area, but it should also be expanded to other 18