Autonomy vs . Control : Quality Assurance and Governmental Policy in Flanders

Higher education in Flanders has seen some major changes in the 1990s. One of the key elements of the new higher education regulations was the quality assessment system. This exemplified best the government's policy of granting all institutions of higher education autonomy, making them responsible for their policies, while still keeping the quality of higher education somewhat under governmental control. In this article, we focus on the tension between the government's aim of improving and controlling the quality of higher education and universities ' concern for their autonomy. We describe the Flemish government's view on issues of quality in higher education and confront these with an account on the basis of case studies of how the quality assurance system was actually implemented in universities. We conclude that the model of the “market state” or the '”evaluative state” is only realised partially in Flanders. The government is still interventionist when it comes to key policy issues.


Introduction
Higher education in Belgium went through some major changes in the 1990s.As a result of the constitutional reform of 1988, the Dutch-speaking and the French-speaking higher education systems were separated.In other words, whereas formerly a national, Belgian education policy was pursued (albeit by two ministers, a French speaking minister and a Dutch speaking minister), now the government of the Dutch-speaking community in Belgium (that is, Flanders) and that of the French-speaking community (that is, Wallonia) each could develop their own higher education policy.The Flemish government, on which we focus in this article, wanted to "do different and better."This led to a new higher-education legislation in 1991 and to a policy based on the principles of deregulation, autonomy, and accountability.One of the key elements of the new higher-education regulations was the quality-assessment system.This exemplified best the Flemish government's policy of granting all institutions of higher education autonomy, making them responsible for their policies, while still keeping the quality of higher education somewhat under governmental control.
But how did the universities interpret the government's ideas about quality assurance, and how did they negotiate between their own wishes and the government's demands?The main focus of this article is the tension between the new policy goals and instruments of the government, aimed at "controlled autonomy" for the universities, and the actual implementation of the quality assurance system in the universities.
We studied governmental higher-education policies and programmes over a period of about 30 years and carried out case studies of universities at the time they were implementing the quality assurance policy (the second half of the 1990s) 1 .With document analysis and interviews with key actors, we were able to establish a detailed picture of different aspects of the interplay between the government and the institutions of higher education (see also Van Heffen, Verhoeven, & De Wit, 1999).In this article, we will first give an overview of past developments and policies in higher education and indicate their relevance for our understanding of the introduction of the quality assurance system in the 1990s.Next, we describe the Flemish government's view on issues of quality in higher education and point at a shift in the model of state control.Third, we take a closer look at the quality assurance system in Flanders as it was articulated in the governmental regulation.We then turn to the quality assurance system as it was actually implemented in the universities.This will allow us to draw conclusions in the fifth section about the tension between the government's aim of improving and controlling the quality of higher education and universities' concern for their autonomy.

Constraints from the Past
The introduction of a quality assurance system in Flemish higher education cannot be fully understood without taking constraints from the past into account.Therefore, we first consider some elements in the history of higher education in Flanders that influenced the situation in the 1990s.
In the 1970s and 1980s, two developments turned political attention away from higher education, the result being what might be termed a period of "non-politics" regarding higher education.First, the unitary Belgian state was gradually transformed into a federal state composed of three Communities and three Regions.Second, the economic crises that followed the oil crises made savings and cutbacks a priority in all policy domains.In the 1980s, the retrenchment policy also reached the higher education sector, so funding was cut back and little room was left for anything other than financial policy matters.
In this period of non-politics, governmental steering of higher education took on different forms for different kinds of higher education institutions.The Belgian higher education system could be described as a "double binary" system.A first divide was one between universities and colleges of higher education.A second divide existed between state institutions (financed and governed by the state) and so-called "free", often denominational institutions (also financed by the state, but largely autonomous in their governance).The four categories of institutions that thus existed (see Table 1), each had a different relationship with the Belgian government.If we restrict ourselves to the universities, the following differences existed.The state universities were steered in a very centralist way.They were directly under the responsibility of the Minister of Education, who was the organising authority of state education.Therefore, these universities were influenced directly by the features of the political system of the time, which was unstable due to linguistic troubles2 and financially restricted because of the increasing public debt.
The "free" universities, that is, universities established on religious or philosophical grounds, were relatively autonomous.According to Verhoeven (1982), this was because university education was always seen as necessarily autonomous and organised by private initiative and because interventions could trigger irreconcilable conflicts.The government was reluctant to regulate in order not to upset the delicate ideological and linguistic equilibrium in the university sector.
When Belgium became a federal state, in Flanders the view took hold that Flanders could "do it differently and better."The Flemish government wanted to treat all institutions on a more equal basis.In the process of constitutional reform, the Communities became virtually autonomous in a number of fields, including education.This provided the Flemish Community with the opportunity to reform thoroughly all sectors of education.Several decrees were issued, with autonomy, responsibility, and scaling-up being the leading ideas.With the Decree on Universities of 1991, the government took an important step towards achieving far-reaching autonomy for all universities.The decrees imposed only formal requirements (length of the course, division in cycles, ability to abridge the course duration, and so on); the content of education (the course programme) could be decided by the institutions themselves.The new legislation made the former state universities autonomous and gave them almost the same responsibilities as the "free" universities.The higher education rules as a whole (including college education) became more integrated.

Governmental Views on the Quality of Higher Education The State Steering Model
The new legislation of the 1990s meant a radical break with the (Belgian) past, at least in the way the policy was conceived.However, it is well known from the literature that higher education tends to be conservative when it comes to major changes (see for example Kerr, 1982, van Vught, 1987, Salter & Tapper, 1994, Maassen & Gornitzka, 1999).Although innovation takes place constantly in a higher education institution, it is difficult to spread these innovations throughout the institution and even more to change the institution as a whole.Moreover, several authors have pointed out the resistance that universities develop against government interference with quality assurance (see for instance El-Khawas, 1998, Bauer & Henkel, 1999, Dill, 2003).
Nevertheless, the chances of a successful implementation of changes through governmental policy also depend on the steering capacity of the government.Many conceptions of changes in this steering capacity have been developed, with amongst the most salient the shift from the "interventionary state" to the "facilitatory state" (Neave, & Van Vught, 1991), the trend towards "steering at a distance " (Goedegebuure et al. 1994), and the rise of the "evaluative state" (Neave, 1998).The "new public management" is considered to be at odds with the collegial governance that is typical of the higher education institution, especially the university (Amaral, Fulton, & Larsen, 2003).
Given the complexity of the Flemish situation, it is useful to use not a dichotomy (as in for instance the facilitatory state vs. the interventionist state), but a grid of four state models, as offered by Olsen (1988).Olsen distinguishes between the sovereign state, the institutional state, the segmented state, and the market state.In the model of the sovereign state, higher education is viewed as an instrument to achieve economic or social goals.This role is then best achieved through strict control by the political authorities.In the model of the institutional state, the government takes a detached stance, allowing institutions of higher education to uphold academic values and traditions independently of short-term political circumstances.In the segmented state, the government is not a unitary actor monopolising power and control but an aggregate of competitive, legitimate centres of authority.Finally, the marketstate model implies a minimum role of the state, with policy being the result of bilateral agreements between governmental actors and social actors.

Quality as the Main Issue
The market state model has been on the rise in the whole of We stern Europe.Since the 1980s, politics in Western Europe has been marked by the rise of neo-conservatism (Brown, et al., 1996) together with the rise of a kind of neo-liberalism.While individual freedom and a free market are promoted, a strong state is still demanded to guarantee the moral and political order.In education, this "ideology" shows in the autonomy and responsibility granted to educational institutions.In this view, flexibility and institutionalised competition combined with quality control should lead to an educational supply that meets the needs of the educational and the labour market.Moreover, the notion of accountability is central.Education has to account for its efficiency and effectiveness.For higher education, this means providing more directly applicable knowledge.The traditional political parties and ideologies, conservatives, Christian-Democrats, as well as the leftist parties, can easily come to a consensus about the necessity of neo-liberal educational accountability and appr ove of essential features of this new way of policy making (compare Husen, Tuijnman, & Halls, 1992).
In Flanders, too, this new policy ideology has become an essential component of education policy.Many policy documents of the 1990s reflected this semi -neutral management ideology to a considerable extent.This does not mean that "old" values like democratisation and accessibility were completely absent, but they seem to have become of secondary importance.With the constitutional reform of 1988, a shift occurred in the normative conception of the government regarding higher education.While Belgian higher education policy could be described as an amalgam of Christian-Democratic, classical Liberal, and Social-Democrat values, the Flemish government's normati ve background developed into a more homogenous neo-liberal value system (or, in Olsen's terms, a market-state model).
Within this market -state rhetoric, the quality of higher education was a central issue for the Flemish government.It assigned priority to a qualitative extension of the educational system in Flanders.If the educational supply must be able to meet the changing needs of our society, so it stated, a real autonomy in policy making for educational institutions has to be made possible.From then on, quality management became a constant factor in Flemish education policy.In other words, from 1989 onwards, the policy of the Flemish government has been focused on applying the concepts of quality, autonomy, and deregulation to the higher education sector.By abandoning detailed central regulation and opting instead for a policy confined to creating the right conditions for maximising the policy space on the local institutional level, the Flemish government wanted to increase the quality of higher education.The content of this "quality" was not clearly stated.But increasing international competition was often mentioned along with the need for universities to be strong enough to meet in this competition.However, for the government this did not mean that the specific role of higher education (offering education at a high level) should be diluted.The Flemish government didn't want the content of education to be dependant on the volatile demands of students and employers.It is not wrong for universities to establish new courses or change the content of existing programmes to meet changing demands, so it argued, but they have to remember that they must always offer high quality education.
The Minister of Education spoke of a "new policy philosophy, a philosophy that rests on a totally new relationship between the government and education, whereby the government sets out the beacons of the policy, provides means for the realisations of this policy (envelope financing), and grants the widest possible freedom to education for this realisation."(Vlaamse Raad, 1993 -94).
As far as the funding was concerned, the Flemish government considered the basic financing of higher education as a task for the government.It strongly believed in the necessity to pay for the normal functioning of higher education with public funds, and in the incidental nature of private financing, so that higher education would be able to fulfil its essential tasks independent of private funding.The major part of the funding of universitie s in Flanders consisted of governmental grants, and this remained to be the case, even up to the present day, although the third-party funding of some institutions is growing.
As a counterpart, the universities needed to be made accountable for the funding they received from the government.The Flemish government aimed at "affordable quality" and demanded institutions to make adequate use of the money they got from the government because the government had to take its budgetary capabilities into account.The optimum use of funds was a constant preoccupation of the Flemish government in the 1990s.Therefore, a new financing system was introduced in the higher education sector.The new system of envelope financing made it possible to control the funding, at least from the government's point of view.

Related Problems
In the context of the quality of higher education, the Flemish government focused on a number of problems other than quality assurance as such.First, there was the problem of low pass rates.Free access to higher education and freedom of choice remained important principles governing the Flemish higher education system.The other side of the coin was the high number of students who failed to complete higher education.Only 47% of the freshmen eventually graduated from a university (Ministerie van de Vlaamse Gemeenschap, 1999, 193).For this reason, the government included an article in the Decree 3 on Universities of 1991 that required that 5% of all academic staff had to provide specific educational guidance to first-year students.Basically, however, the institutions themselves (and secondary education schools) were responsible for optimising their efforts to increase the pass rates of their students.The policy letter of Minister of Education Van derpoorten (2000) also reflected the preoccupation with the maintenance of existing guidance systems for first-year students, course-selection counselling in secondary education, and the easing of the transition from secondary to higher education.
Other problems pointed out by the Flemish government concerned the rationality of the course supply in higher education.The Flemish government wanted to improve the transparency of the university course supply4 .The opacity of the university course supply made it difficult for potential employers to evaluate the particular knowledge and skills of graduates.The relative value of courses and diplomas was not always clear.Next to this, the lack of transparency could cause problems for students trying to choose a field of study.More generally, the lack of transparency resulted in inadequate linkage between higher education courses and the labour market (Van Heffen & Huisman, 1998).
Instead, universities had to strive to become centres of excellence in both education and research.As a consequence, the universities were expected not to maintain qualitatively inferior courses.However, other than the obligation to establish a quality assurance system, which we will discuss below, no direct policy instrument to achieve this goal was formulated.
Here the Flemish government seemed to be trying primarily to "manage by speech ", a technique it often used for higher education.
The Flemish government conferred with the vice-chancellors of the universities and even assigned a s pecial government commissioner to arrive at a more rational and wellbalanced supply of courses.But again, although several official reports were issued, no results were achieved.When a new Minister of Education took office following the elections of 1999, the government commissioner was dismissed and the reports were classified vertically.
A problem implicitly stated in the context of the erosion of knowledge and rapidly changing knowledge was the structure of lifelong learning.This was seen as one of the great challenges of the coming decades (Van Den Bossche, 1995, 8).Here, too, the government seemed in favour of giving as much autonomy to the institutions as possible so they would be able to adapt their supply dynamically to the rising demand (Ministerie van de Vlaamse Gemeenschap, 1996).

The System of Quality Assessment
Granting autonomy and responsibility to all institutions of higher education was the most important instrument for the Flemish government for guaranteeing the quality of higher education in an international context.This was based on the view that autonomous institutions will develop modern, dynamic, and professional management, and therefore will be more able to adapt dynamically to changing demand and European initiatives.
To verify the effects of autonomy on the quality of education, the Flemish government imposed the development of a quality assurance system.The new legislation of 19915 brought the regulations regarding the entire university sector into one coherent framework.This legislation obliged all Flemish universities to develop a system of quality assurance.More specifically, it stated that each university had to provide both internal and external quality assurance, meaning that it must: v monitor the quality of its education and research activities on a continual basis; v set up a system of regular evaluations (at least every 8 years) with other universities in Flanders and abroad (these evaluations have to be made public); v take account of the results of the evaluations in its policy (and report on this in its annual report for the Flemish government).
The role of the government was defined as monitoring and controlling the quality assessment system by examining the quality assurance program set up by the universities, by appointing a commission to do comparative research on the quality of the education activities in a particular course or courses, and by monitoring the way in which universities take account of the evaluations in their policies.
If the quality of a course turned out to be enduringly unsatisfactory, the government could either stop subsidising the university in question for the students in that course or it could withdraw the right of the university to grant an academic degree for that course.
A meta-evaluation by the government took place in 1997 when the Minister of Education appointed an independent commission of experts.This international audit commission had to verify whether the universities had complied with the quality assurance obligations as defined by the Universities Decree of 1991.On the basis of its research and its visits to the universities, the audit commission concluded (Auditcommissie, 1998) that this was indeed the case, notwithstanding the different ways of taking care of quality assuranc e in each university and despite a number of areas that the commission found could be improved (see below for examples).

Quality Assessment in Flemish Universities
We have concentrated hitherto on the quality assurance system in Flanders as it was conceived by the Flemish government.In this section, we turn to the universities themselves.On the basis of case studies of two universities in Flanders, we will describe how a qualityassurance system actually was established.
But first, and more generally, w e must point out the role of the body representing the universities, the Flemish Interuniversity Council.The Flemish Interuniversity Council (Vlaamse Interuniversitaire Raad) was established in 1976 to promote consultation and co-operation between the uni versities and to be an advisory body towards the ministers of Education and Science Policy in university matters.The Flemish Interuniversity Council is composed of the vice-chancellors of all the Flemish universities and a supplementary member from the two large universities.In a number of matters, the Decree on Universities of 12 June 1991 has made the advice of the Flemish Interuniversity Council mandatory.
Even before the Universities Decree of 1991 was passed, the universities conferred within the Flemish Interuniversity Council about how to establish a quality-assurance system.Agreement was reached with the Association of Universities in the Netherlands (VSNU) to implement the Dutch review system in Flanders, in mutual cooperation.The first review was carried out in 1990, and, since 1991, all reviews were based on this cooperation between the Flemish Interuniversity Council and the Dutch VSNU.
The review system consisted of elements of both internal and external quality assurance.A review took a course or group of related courses (both academic and advanced academic courses) as its point of departure and was carried out in all the universities offering that particular course or group of courses.The review process started with a self-evaluation.Each of the relevant study fields at each university had to write a self-study, reporting the results of a critical self-analysis by all groups associated with the course (professors, assistants, and students).On the basis of this self-study, a review committee, consisting of experts who are not employed by any of the faculties concerned, had to determine if a course achieves the objectives set at the outset.For this task, it could use the self-study carried out by the specific faculty, recent evaluation reports, student surveys, courses, text books, visits to the faculties, discussions with interested parties, and so on.The committee then drafted a course report designed to enable each faculty to keep on improving its quality.The faculties had the opportunity to comment on the report.Once all the faculties in a particular field of study had been examined, a general report was produced in which various aspects of the education provided by the institution were placed in a comparative perspective.The final report, comprising a general section and the course reports, was also intended to inform the Flemish Community about the status of education in the specific field of study.

Case Studies of Quality Assurance Systems
How was a quality-assurance system actually established?To what extent did the universities follow the procedure as agreed in the Flemish Interuniversity Council, and did they add specifications to that procedure?For two universities, we will describe some general characteristics, the policies and initiatives of the central governance level of the institution with regard to both internal and external quality assurance, and, where appropriate, developments at the decentral levels of the institutions.The time frame in each case study is what we c alled above the "decade of reform", that is, the 1990s.
Institution A is a former state university that became autonomous as a result of the university decrees of 1991.It is among the largest in Flanders, both in student numbers and in number of faculties (number of courses offered).
University A has developed a decentralised organisation model.The faculties have a large degree of autonomy.The central level's role is to support the faculties and to take guiding initiatives.The quality control structure too is decentrallized, which is reflected in the important role of bodies at the faculty level in the quality assurance structure.
First, in each faculty and for each course or coherent set of courses, there is an education committee.The education committee is responsible for the curriculum of that particular course or courses and also the continuous optimisation of quality.
Since the start of the academic year 1998-1999, each faculty also has one or more education-quality cells.These cells have to support the education committees regarding the quality of education (especially the quality of the content of education and the instructional material).However, the university board does not specify the duties of the quality cells.
In addition to these bodies responsible for quality assurance University A participates (since 1991) in the system of reviews for both internal and external quality assurance.The procedure is as follows.To prepare the reviews, a self-study has to be drafted.At least one year after the final report of the review, a follow-up report has to be presented to the university board (a form of control by the board that indicates the significance attached to quality control, as is also shown in the establishment of the quality cells), addressing the results of an internal critical discussion of the review report, the possible and desirable short-and long-term remedies, and the implementation plan for these remedies.Every two years, the quality assurance system (and particularly the follow-up of the reviews) is discussed by a special meeting of the university board.Overall, the confrontation with the review commission and the review report are seen as useful and informative.
This situation of decentralized autonomy but central interventions led the international audit commission (Auditcommissie, 1998) to the conclusion that "There is a certain degree of imbalance at University A between the competence of steering the policy content and the actual structures that have to implement that policy.The actual distribution of authority between the central and the decentralised bodies is not clear for the audit commission, notwithstanding the detailed documentation and her visit to the university.This concerns not only who does what but also how certain tasks are executed (intervention strategies)." The decentralised model of organisation of Institution A implies that faculties assure quality differently.We confine our analysis to the faculties of Engineering and Economics, because they differ regarding the functioning of the quality cells and the evaluation of the review system.
In the Faculty of Engineering, the quality cell is placed in the decision-making structure between the education committees and the university board.Its task is to guarantee the uniformity of the courses of the faculty, while respecting the autonomy of the education committees in this respect.Unlike the other faculties that established this structure in the academic year of 1998-1999, this structure was not new in the Faculty of Engineering.Comparable committees had already been in operation for several years in this faculty and have served as an example for the current structure in all the other faculties.
The reviews are regarded positively by the Faculty of Engineering.Not so much the actual visit (which is too short to make a thorough evaluation possible) but the drafting of the self-study report is seen as the most important moment in the review process.
The Faculty of Economics established two quality cells, one for economics, and one for applied economics.The quality cells evaluate education and do general preparatory work for the education committees regarding the educational policy of the faculty.They have, for example, drawn up a code of good practice for the guidance of thesis students and attainment targets for the faculty's education programme.
In the Faculty of Economics, important changes in education took place as a result of the reviews.It is clearly stated that, without the obligation to draft a self -study report, one would never have turned to such a general analysis of the educational activity of the faculty involving re-writing the goals, reforming the curricula, introducing more active educational methods, training of teachers, strengthening the relationship with the alumni, and so on.
Finally, governmental influence is also reported as a result of the work of the special commissioner for the optimisation of university education.Some faculties included the proposals of the commissioner in the reform discussions.In the Faculty of Economics, for example, more specialised postgraduate courses were developed.
Institution B is a large non-state university.It is comparable to institution A in both number of faculties and number of students.
The university management views the university as a professional organisation with the academic staff as the core of the organisation.This implies that it refrains from direct actions or interventions on the lower levels.The university management sets out the general policy lines but recognises the faculties as responsible organising entities.
The same pattern emerges with regard to quality assurance.The university management provides a general framework to situate the work of the education committees (see below) and facilitates, supports, and stimulates.Central financial incentives with faculty matching are used as a lever for educational improvement and innovation.
The core of the internal quality assessment system of University B are the education committees.Each faculty establishes an education committee for each course or for a set of related courses.An education committee consists of professors, assistants, students, and possibly alumni.The educational committee serves as the main body for developing curricula and evaluating educational practices.It establishes an educational frame of reference, that indicates the goals and attainment targets and the means to achieve them.It evaluates these programmes, the didactic methods, and the examination system.They are thus the anchor points for the development of the content and organisation of the courses and the development of an educational concept.Nevertheless, the faculty councils are finally responsible for education, the education committees being only advi sory bodies.
The education committees are responsible for the "daily" quality of education practices.In addition to them, the quality of a course is examined systematically each four or five years by an evaluation commission, i.e. a temporary body consist ing of professors, assistants, and students.In effect, the evaluation commission drafts a self-study report, including a strengths and weaknesses analysis, as a preparation for an external review.In general, as in Institution A, the external reviews are viewed positively.They provide useful suggestions for improvement of the curricula, the teaching methods, and the structures and conditions necessary for optimising educational quality.Nevertheless, for Institution B, the external evaluation is less important than the internal quality assurance.It aims at developing a strong and continuous quality cycle (the audit commission, however, remarked that this continuity is not fully achieved).Therefore, in between two external reviews, which take place every eight years, another internal self-evaluation takes place.At the latest one year after the review, a follow-up report has to be drafted by the evaluation commission.This report has to indicate concrete remedies for the remarks of the evaluation commission and the weaknesses that showed up in the self-study.The follow-up report is discussed by those responsible in the faculty and the university management and is then submitted to the university management council.
The review system in University B has three further characteristics.First, an education office provides educational theoretical guidance and practical support of the educational evaluations.Second, the state of affairs of the internal quality assurance is discussed every year by the education council of the university.Third, the different phases of the internal evaluation process and the rules regarding content and procedures are defined in a manual.
At the lower levels of Institution B, internal quality assurance is carried out in the way des igned by the university management.The faculties (we researched the faculties of Engineering, Economics, Law, and Pharmacy) and the education committees within the faculties accept responsibility for quality assurance.They resent strong interference of t he university management and report that there is little such interference.In fact, they even mentioned a further decentralisation tendency in that important meetings and discussions often take place outside and preceding the formal structure.The education committee then serves only as a way to make decisions official.
The description of these two cases allows us to compare the developments within these universities to the Flemish government's views and policies concerning quality assurance in the Flemish higher education system.

Quality Assurance in Flanders: Unequal Governmental Control
The normative conception of the Flemish government on the quality of higher education clearly was based on a neo-liberal, neo -conservative value system.From this, it argued in favour of far-reaching autonomy for the universities but financed primarily by the government and obliged to account for their use of the funds received.Bearing responsibility consisted of, among other things, taking care of quality, in particular establishing structures for carrying out quality assurance systematically.
However, it took ten years to implement this policy in higher education.Autonomy and responsibility were key words in the government's policy right from the moment the Flemish Co mmunity acquired the authority over education (January 1, 1989).The rules laying down a similar quality assurance policy for all institutions of higher education in Flanders were only completed with the Decree on Education X of May 18, 1999.How can we explain this time gap?
In the history of higher education in Flanders, the distinction between state institutions and non-state ("free") institutions has been of particular importance, with the former having less autonomy than the latter.In other words, the Belgian governmental steering mechanisms have always varied.The Flemish government had to take this into account and could only gradually overcome this historical distinction.
In 1991 the differences between state and non-state universities were for a large part undone -leading to a difficult search for a new balance of power within the former state institutions, as we have seen in Institution A.
The consequence of this differential governmental steering policy for quality assurance is the unequal position of the institutions with regard to the establishment of a quality assurance system at the end of our period of study.In the two cases a different system of quality assurance had been set up, at different paces, and with different motives.Both Univers ity A and University B took part in the review system and saw this as an important impulse for internal quality assurance.Since 1991 (the year of the university decree), both had systematised quality assurance, but not in the same way.University A worked with education committees and has also established quality cells.University B had strengthened the quality tasks of its previously established education committees and used a system of temporary evaluation commissions.Both institutions had a decentralised organisation model, and, especially in University A, this led to differences in the way the quality assurance system functioned in the faculties.

The State Model in Flanders: Autonomy Vs. Control
In conclusion, we may consider the trends and policies with regard to quality assurance in higher education in Flanders from the perspective outlined above.Does Flanders follow the trend towards an "evaluative," "facilitatory" state that "steers at a distance"?And how do the institutions of higher education, with their different histories and backgrounds, respond to the new policy?Taking these models of governmental steering as a point of reference, three conclusions can be drawn for the situation in Flanders.
First, the government's normative view and steering model has clearly shifted after the constitutional reform of 1988.Whereas different models applied in Belgium, with elements of the sovereign, the segmented, and the institutional state applying to a different extent for each sector of higher education (Van Heffen, et al., 1999), the Flemish government from 1989 onwards avowed its belief in the market-state model -and this regardless of the political affiliation of the Minister of Education (since 1989, Flanders successively had a Christian-Democrat, a Social-Democrat, and a Liberal6 Minister of Education).Regarding quality assurance, this resulted in a vast number of policy documents stressing the need for institutions of higher education that can and must develop the quality of their education in an international, competitive environment independently of the government.In this, Flanders follows a Western European trend: "[t]he politics of governance in higher education are now embedded in a discourse which assumes the external regulation of academi c activity to be the natural and acceptable state of affairs."(Salter, & Tapper, 2000, 82) Second, the market state model also seems to apply to the higher education policy of the Flemish government in the 1990s.Attempts were made to create a coherent and similar framework of rules for all institutions of higher education, with autonomy and responsibility as the leading ideas.Here the political background of the minister played a more important role -witness, for example, the cancellation of the assignm ent of the commissioner for the optimisation of higher education.Nevertheless, governmental policy consistently imposed a quality assurance system on the universities and evaluated the systems set up by the institutions.This evaluation took the form of, among other things, an international audit commission for university education that compared the quality assurance structures of the different universities and the obligation for all institutions of higher education to report annually on their efforts and achievements concerning the quality of their education.The institutions became responsible for quality assurance, had to account publicly for it, and could, therefore, be more easily scrutinised by their "clients" in function of educational quality.
The evaluation and control by the government point to the related, but contrasting, third conclusion that the market-state model was only realised partially in Flanders.The Flemish government undoubtedly enlarged the scope for autonomous decision making by the institutions of higher education.It withdrew from direct control.But, indirectly, the government still retained considerable influence.It compelled the universities to establish a fully fledged quality assurance system.This was related to its aim of i ncreasing the transparency of the course supply and the obligation for the institutions to report annually about quality measures relat ed to the reviews.Thus, the government brought about important changes , in that t he quality assurance initiatives already present in universities were expanded and systematised.The Flemish case thus points out that a large, government-induced reform can be effective, even when it has to accommodate a complex system with a variation in higher education institutions that are renowned for their resistance against government interference.The changes induced were not the result of a "minimal role" taken by the government, but by clear interventions on key issues.This means that, although the model of the market state has becom e more important in Flanders, it is still more a policy theory or even ideology than a fully realised, policy-making principle and practice.