This chapter takes a closer look at the grants the Flemish authorities allocate directly to higher education institutions in the Flemish Community to support research activities. The Flemish higher education funding system provides significant resources for research directly to universities, including the research component of the operating grant and earmarked grants for fundamental and applied research (the Special Research Funds and the Industrial Research Funds). The Flemish government also allocates a more modest level of funding to university colleges as part of their block grant to support “practice-oriented research” in the programmatic fields in which these institutions are active.
Resourcing Higher Education in the Flemish Community of Belgium
Higher Education
4. Institutional funding for research
Abstract
4.1 Funding for research in higher education in the Flemish Community
The focus on research funding in this chapter, and in the review more generally, is on core funding provided by government to higher education institutions to maintain and develop their in-house research capacity. In line with a widely held view in higher education policy internationally, the analysis assumes that teaching in all higher education institutions should be informed by the latest knowledge in the disciplines taught. Academic and teaching staff in professionally and academically oriented programmes should, as a minimum, be able and willing to engage with new research findings in their field and incorporate lessons from this research into their teaching practice and the broader support they give to students. In research-intensive institutions, such as the five Flemish universities, a research-active academic staff is a precondition for high quality research-based teaching and learning, as well as for high quality basic and applied research that expands and enriches the knowledge base on which societal development depends.
The level and form of public funding for research has an important influence on the capacity of higher education institutions to create research-informed learning environments for students, to train and nurture new generations of researchers and to allow research-active staff to generate new knowledge and expertise through research activity. In the Flemish Community, as in other OECD jurisdictions, external, project-based funding from public research agencies, businesses and non-profit bodies contributes a significant share of total revenue for research in higher education, particularly in the university sector. This chapter considers theses sources of funding to the extent that they impact on the overall resourcing of higher education institutions, but does not assess in detail how competitive research funding systems – such as those operated by research councils – are designed and implemented in Flanders or other OECD jurisdictions. Such an enterprise would require a review in its own right.
Before examining how different aspects of the core research funding system for Flemish higher education institutions are perceived and perform and how they compare to the situation in other OECD jurisdictions, it is helpful to consider the broader research funding landscape within which Flemish universities and university colleges operate. This section provides an overview of the main public research funding mechanisms targeted at Flemish higher education institutions, the different revenue streams that higher education institutions receive for research and the weight of research performed in higher education in the Flemish research system.
Direct grants to institutions are part of a broader landscape of public research funding
Flemish higher education institutions have access to a diversified set of public funding streams to support research. In addition to the foundational support for research provided to universities through the operating grant, which serves primarily to support staff and running costs, the Flemish government provides:
Universities with an earmarked grant for basic research, which they allocate internally: the Special Research Funds (Bijzondere Onderzoeksfondsen or BOF);
Funding for competitive research grants, administered by the Research Foundation – Flanders (FWO), the Flemish research council. This is the main source of funding for doctoral and post-doctoral positions, as well as an important source of funds for basic research (see below);
An allocation of funds to support the development of research capacity in academic programmes transferred from university colleges in 2013, referred to as Supplementary Research Funds (Aanvullende onderzoeksmiddelen);
Funds allocated to the associations of universities and university colleges to support strategic basic research and applied research and to stimulate knowledge transfer and valorisation: the Industrial Research Funds (Industrieel onderzoeksfonds or IOF) and funding for cooperation with the business sector (referred to in Flanders as “interface activities”);
A more modest allocation of funds for “practice-oriented research” (Praktijkgericht wetenschappelijk onderzoek – PWO) in university colleges, allocated as a supplementary component of the operating grant for these institutions (PWO-middelen).
Table 4.1 provides an overview of the target beneficiaries, objectives and allocation methods for these funding mechanisms, along with the total budget allocated to each in 2020. The sections that follow provide a brief review of each the main mechanisms.
Table 4.1. Public funding instruments for research in Flemish higher education
Note: (1) This figure represents 45% of the core operating grant for universities that is allocated using the main formula (excluding supplementary grants added outside the formula) and represents the proportion of funds allocated using research variables. By convention, the Department of Economy, Science and Innovation (EWI) assumes that 25% of the core operating grant is used for research in practice.
(2) The four Strategic Research Centres in Flanders are: imec, the Flemish Institute for Biotechnology (VIB), Flanders Make and the Flemish Institution for Technological Research (VITO).
The Special Research Funds (BOF): earmarked institutional grants for basic research
The Special Research Funds (BOF) are designed to allow universities to maintain and develop their internal capacity to undertake high quality basic research. In 2021, total funding for BOF amounted to EUR 222 million. The funds are provided to universities in the form of three earmarked grants. The largest of these grant components (around 87% of the total allocation in 2021) is provided to allow universities to fund basic research projects and related researcher positions through internal allocations and competitive procedures, over which the universities themselves have a large degree of autonomy. A smaller component (8.6% of total BOF funding in 2021), called the Methusalem programme, is allocated to provide long-term (seven-year) research grants to outstanding senior academics. The final component (4.4% of total BOF funding in 2021) is allocated to pay for tenure track positions for outstanding post-doctoral researchers. Researchers selected for these posts are funded to undertake research for five years, at the end of which, subject to a positive evaluation, they can be appointed as full-time academic staff without the need for a position to become vacant or be opened (Flemish Government, 2020[1]).
The Flemish government’s Department of Economy, Science and Innovation allocates the funds to the five universities using a specific allocation formula that was significantly revised in 2019, following an evaluation of the mechanism in 2018 (Korlaar, Bongers and Groot Beumer, 2018[2]). In the same year, the annual budget for the BOF was increased by more than 25%. This increase resulted from the injection of EUR 37 million of new funding and a EUR 5 million increase already planned as part of the ongoing integration into universities of academic programmes previously located in university colleges. This brought the annual allocation to over EUR 220 million, with the majority of the increases maintained for subsequent years.
The revised formula for distributing the BOF provides 50% of total BOF funds based on universities’ performance in relation to bibliometric and policy-related allocation criteria (parameter groups “B” and “C”) observed in different multi-year reference periods and recalculated annually. The other 50% (parameter “A”) is allocated based on the historical average vales of parameters “B” and “C” for the previous five years. The parameters used in the allocation formula (referred to as the BOF “key”) are summarised in Table 4.2.
Table 4.2. The allocation formula for the Special Research Funds (BOF key)
Source: Flemish Government (2019[3]) Flemish Government Order on the Special Research Funds (BOF-Besluit) https://www.ewi-vlaanderen.be/sites/default/files/bof_-_besluit_van_de_vlaams_regering.pdf (accessed on 4 June 2021).
Unlike its predecessor, the new formula no longer uses the number of previously awarded bachelor’s and master’s degrees as a criterion for allocating the BOF and significantly reduces the weight of PhD awards in the formula. By introducing a substantial fixed element, recalculated only every five years, it aims to create greater stability in funding for universities. Moreover, the new formula reduces the weight attached to traditional bibliometric indicators (publications and citations), but introduces new parameters designed to capture the quality of research outputs in a more nuanced way (citation distribution parameters) and promote international cooperation. From 2024, the new allocation formula will also use a parameter to measure and reward inter-disciplinary research (Flemish Government, 2020[1]). The parameters employed in the BOF key use different reference years, depending on the nature of the indicator. For example, the number of doctorates and the number of publications are calculated as the average for each university for the reference years t-6/t-5 to t-3/t-2 (where “t” is the current budgetary year), while the calculation of the number of citations takes into account research outputs published between 12 and 5 years before the calculation date.
The revised BOF key, like its predecessor, uses the Web of Science (WoS) as the main source of bibliometric data for fields such as engineering, the natural sciences and medicine. To ensure that publications in Dutch are adequately taken into account, the allocation system draws on a dedicated Current Research Information System (CRIS) to capture publications in the social sciences and humanities. The Flemish Academic Bibliographic Database for the Social Sciences and Humanities (VABB-SHW) registers academic publications from the social sciences and humanities authored by researchers affiliated to Flemish universities that appear in journals and publication media approved by an authoritative panel (ECOOM, n.d.[4]). This system is similar to those used in the Nordic countries to record publications in these fields (Zacharewicz et al., 2019[5]).
To ensure that all five Flemish universities have access to funds to develop their internal research capacity, the legislation governing the BOF guarantees minimum shares of total BOF funding for the three smaller universities. The University of Hasselt is guaranteed 4% of total funds, the VUB 10.5% and the University of Antwerp 13% (Flemish Government, 2019[3]). In practice, in 2021, the VUB and the University of Antwerp received more than or exactly their guaranteed minimum shares as a result of the calculation using the standard parameters of the BOF key. The University of Hasselt, which would not have reached 4% of the budget based on the standard calculation, received 4% of the total BOF budget because of the guaranteed minimum.
Alongside the changes to the funding allocation formula, the modifications to the BOF in 2019 also introduced a new component to support cooperative research projects between universities (Interuniversity BOF or “iBOF”). Around one-quarter (EUR 10 million in total) of the additional funding for the BOF provided by the Flemish government in 2019 is pooled between the five universities to fund projects involving at least three researchers from at least two universities. Researchers initially submit outline proposals to the internal research councils in the universities involved in each project proposal, which invite shortlisted candidates to submit full proposals in a second round. In the first call for proposals in 2020, the 30 projects admitted to the second round were evaluated by an international panel of five experts, in a process contracted out to an external provider, with funding allocated to the 15 top-ranked proposals (VLIR, 2020[6]).
The BOF is designed to allow universities to pursue and fund their own research strategies and create additional space for investigator-driven research in the Flemish research system. The Research Foundation – Flanders (FWO) also supports investigator-driven research projects through its competitive calls, meaning there may be a choice of potential funding streams available for a particular research project or initiative. In practice, individual academics and departments are likely to make strategic choices about whether particular projects are more likely to be successful in the external competitions organised by the FWO or the internal competitions and allocation processes organised within universities for the BOF funds. There are clear incentives for universities to encourage academics to apply for external research funding before, or in parallel to, applying internally for BOF funds, so that total institutional revenue is maximised.
The Industrial Research Funds (IOF): earmarked funds for strategic and applied research
The Flemish government provides the Industrial Research Funds (IOF) as grants for strategic basic and applied research in support of knowledge transfer and valorisation (see discussion of terminology below) to the legal entities formed by the associations of universities and university colleges (see Chapter 2). The IOF had an annual budget of EUR 52.4 million in 2021, which is also allocated to the associations as an earmarked grant, which is then distributed internally within the association through formal selection procedures. IOF resources can be used to support research projects and specific staff positions related to knowledge transfer and valorisation, such as business development specialists.
The main objectives of the IOF are to support the development of strategic basic and applied research aligned to the needs of the economy and to bring innovations to the market. The funds are allocated to the associations of universities and university colleges in recognition of the combined role of universities and university colleges in creating links between higher education and other economic sectors, and to strengthen cooperation in research between the institutions in the associations. The IOF is complemented by funding (amounting to EUR 4.5 million in 2021) for “interface activities”, specifically designed to support cooperation between higher education institutions and the business sector, stimulate knowledge transfer and promote the exploitation of research results generated within universities and university colleges across all disciplines. The funds for “interface activities” are provided to central technology and knowledge transfer offices for each association, to support the researchers with the procedures related to issues such as patents, spin-offs, contract research and license agreements.
Table 4.3. The allocation formula for the Industrial Research Funds (IOF key)
Source: Flemish Government (2009[7]) Flemish Government Order on the Industrial Research Funds (IOF-Besluit), https://codex.vlaanderen.be/PrintDocument.ashx?id=1018147 (accessed on 4 June 2021).
As in the legislation governing the BOF, the legislation governing the IOF specifies minimum percentages of the total IOF budget for the three associations that include the three smaller universities. The association of higher education institutions in Limburg (including the University of Hasselt) is guaranteed 4% of the IOF budget, the Brussels association (including the VUB) is guaranteed 9.5% and the association in Antwerp (including the University of Antwerp) is guaranteed 10.4% (Flemish Government, 2009[7]).
Transferred academic programmes receive additional funds to strengthen research capacity
From 2003 onwards, the Flemish government began to provide targeted funding – initially referred to as “academisation” funds – to support research within academic programmes that were at that time delivered in university colleges. These dedicated funds were retained in the 2008 reform of the funding model for higher education institutions. Funding was further increased in 2012 and the allocation renamed “Supplementary Research Funds”, in advance of the transfer of most academic programmes from university colleges to universities. The objective was to allow the transferred programmes and those that remained in university colleges within the Schools of Arts to expand their internal research capacity. The annual budget for the Supplementary Research Funds is established in the Higher Education Code, with funding increasing every year up to 2025. In 2020, the budget for Supplementary Research Funds (for both universities and Schools of Arts) amounted to EUR 86.4 million (Flemish Government, 2020, p. 144[1]).
Funds are allocated based on the average number of credits taken and degrees awarded in the academic programmes transferred to universities and in Schools of Arts. The resources may be used to cover staff, operating and exceptional costs related to increasing research capacity in these programmes. Studies have found that the funds are being used within institutions to support the transferred academic programmes, but progress in building up research capacity varies between faculties and disciplines (de Boer and Jongbloed, 2018[8]; Rekenhof, 2018[9]). Early evaluations of the effects of the Supplementary Research Funds were, however, undertaken before the effects of the funds could realistically be felt, particularly as the initial budget available was comparatively low and is progressively increasing over time (de Boer and Jongbloed, 2018[8]).
In addition to the “academisation” funds allocated by the Ministry of Education and Training, the Department of Economy, Science and Innovation has provided a modest level of grant funding specifically for programmes transferred to universities from university colleges in the integration process. These funds are integrated into the allocation for the Special Research Funds (BOF).
University colleges receive an additional grant for practice-based research
University colleges engage in practice-oriented research (Praktijkgericht wetenschappelijk onderzoek – PWO) as part of their mission to promote the application of new knowledge and innovation in the professional fields in which they operate education programmes. Such research activities, generally undertaken in direct cooperation with employers and professionals, complements the applied, strategic and basic research undertaken in universities.
As noted, the university colleges receive a specific funding allocation for practice-oriented research that is paid by the Agency for Higher Education, Adult Education, Qualifications and Study Grants (AHOVOKS) in addition to the formula-based operating grant. In 2020, the total budget for the practice-oriented research grant, which is allocated to institutions based on their share of enrolled study credits, was EUR 30.2 million (Flemish Government, 2020[1]). In addition, the university colleges receive some financial support for knowledge transfer projects, training for researchers and research equipment from the Flemish Agency for Innovation and Enterprise (Vlaams Agentschap Innoveren en Ondernemen – VLAIO) (VLAIO, 2019[10]). The TETRA (TEchnology TRAnsfer) programme, for example, provides funds to university colleges and universities through open calls for proposal to support technology transfer activities and had an annual budget of almost EUR 9.6 million in 2020 (Flemish Government, 2020[1]). In addition, the VLAIO ran a call for proposals in 2020 with a total budget of EUR 2 million to fund infrastructure investments in university colleges to support research into digitalisation, the circular economy and sustainability and energy and climate (VLAIO, 2020[11]).
Income for research in universities and university colleges
The analysis of the income streams of Flemish universities and those in five other comparable OECD jurisdictions undertaken for this review examined the main sources of university funding for research in the systems covered. Figure 4.1 presents a synthetic breakdown of the proportions of total institutional revenue in the university sectors in the six jurisdictions that come from targeted or earmarked funding for research, as well as core, block grant funding that is allocated for research. In all cases, the proportions shown represent averages for the university sector in the jurisdictions in question, based on consolidated accounting data for the most recent year from which such data were available to the review team. In all jurisdictions, there will be considerable variation between universities in the actual patterns of income.
Figure 4.1 shows the proportions of core funding that are nominally allocated to universities for research within national funding allocation mechanisms. In the Finland, for example, this is 34%. In practice, however, block grant funding for research is not earmarked specifically for research activities and will often be absorbed by general salary and operating costs within universities. In the Flemish Community, 45% of the core operating funds allocated to universities are formally destined for research and this is the proportion used for the calculations underlying Figure 4.1 to ensure comparability with the other jurisdictions shown. However, by convention, the Flemish authorities take into account only 25% of core operating funds for research in the calculation of total public spending on research.
The other income components illustrated in Figure 4.1 represent, as far as possible, funds specifically earmarked for research. In the case of the Flemish Community, the proportion of total university income coming from the BOF and IOF (“targeted institutional research grant”) is highlighted separately, reflecting the fact that these funds must be used on research, unlike research components in block grants. For all jurisdictions, Figure 4.1 distinguishes external research income from the main research councils (the Research Foundation – Flanders (FWO) or the Dutch Research Council (NWO), for example), European Union research funding programmes (Horizon 2020, now Horizon Europe) and other third-party funding for research. The latter category is not recorded at the same level of granularity in the consolidated university accounts for all the jurisdictions covered and it is not possible to determine with certainty whether all the revenue in question is destined for research. Nevertheless, the data provide a reliable general overview of the scale of third-party funding for research.
The proportion of research council funding varies from around 16% in Ireland and 12% in Finland, to around 6% of total university revenue in the Netherlands and the Flemish Community. It is important to keep in mind that these are the average proportions of total revenue – representing the weight of these funding streams – not an indication of the absolute level of funding. The proportion of total university revenue from EU research funding varies between 3% in Scotland to 6% in the Flemish Community, while other external sources of research funding (from the business, non-profit or government sectors) account for 20% of average university income in the Flemish Community, 18% in Finland, 11% in the Netherlands and 4% in Ireland. The figure for Ireland may reflect the inclusion of some types of third-party funding for research in the category of “research council funding” for Ireland that are considered within “other external research funding” in other jurisdictions.
Figure 4.1. Structure of research income in universities in six OECD jurisdictions
Note: Proportions are taken from consolidated accounts for university sector in each system, for most recent year for which this information is currently available. The category “Targeted institutional research grant” refers to the combination of the BOF and IOF in the Flemish Community, the only system where such a mechanism exists. The category “Other external research funding” includes funding for research projects from public and private sources, as classified in national accounting systems and the category “Other funding” includes core funding for teaching, tuition fees and other institutional revenue that is not explicitly for research. Research councils: Ireland: Irish Research Council, Health Research Board and Science Foundation Ireland; Scotland: UK Research and Innovation (UKRI); Flemish Community: FWO; Netherlands: NWO; Denmark: Independent Research Fund Denmark; Finland: Academy of Finland.
Sources: Adapted from: HEA (2019[12]) Financial Trend Analysis – Universities and IoTs 2007-2017, Higher Education Authority; Audit Scotland (2019[13]), Finances of Scottish universities, https://www.audit-scotland.gov.uk/report/finances-of-scottish-universities (accessed on 20 April 2021); Flemish Government (2020[14]) Verslag over de financiële toestand en de evolutie van het personeelsbestand van het hoger onderwijs in 2019 Deel II – Universiteiten (Report on the financial situation and the evolution of staffing in higher education in 2019 – Part II – Universities) https://www.vlaanderen.be/publicaties/verslag-over-de-financiele-toestand-en-de-evolutie-van-het-personeelsbestand-van-het-hoger-onderwijs (accessed on 12 January 2021); DUO (2021[15]) WO Gegevensboek 2015-2019 (Universities Databook 2015-2019), https://www.duo.nl/open_onderwijsdata/databestanden/onderwijs-algemeen/financiele-cijfers/gegevensboeken.jsp (accessed on 24 May 2021); Rathenau Instituut (2020[16]) Ontwikkeling derde geldstroom en beïnvloeding van wetenschappelijk onderzoek (Development of the third funding stream and its influence on scientific research), https://www.rathenau.nl/nl/vitale-kennisecosystemen/ontwikkeling-derde-geldstroom-en-beinvloeding-van-wetenschappelijk (accessed on 6 June 2021); Universities Denmark (2021[17]) Det Statistiske Beredskab - Danske Universiteter (University Statistics – Universities Denmark), https://dkuni.dk/tal-og-fakta/beredskab/ (accessed on 24 May 2021); Vipunen (Vipunen, 2021[18]) University income statement, Vipunen – Finnish Education Statistics, https://vipunen.fi/fi-fi/yliopisto/Sivut/Talous.aspx (accessed on 24 May 2021).
Funding for research in non-university institutions in the five binary systems (there has been no binary distinction in Scotland since 1992) is consistently lower than in universities (see Table 3.2 in Chapter 3). Some governments, such as Ireland and Denmark, provide no grant funding specifically for research to non-university institutions as part of their core public funding. In other systems, such grants represent a small proportion of total institutional revenue, ranging from 2.6% of total revenue in universities of applied science in the Netherlands to 3% in university colleges in the Flemish Community. Of the systems analysed in depth, only Finland provides a substantial institutional allocation for research to universities of applied science, with the research component of the core grant representing over 15% of total institutional income in the sector. In all systems, non-university institutions also generate research income from a variety of external sources, including, as in Flanders, grants from public enterprise and innovation agencies and businesses.
The place of higher education in the Flemish research system
The Flemish Region is one of the most research-intensive regions in Europe. In 2019, total research and development expenditure in the Flemish Region was the equivalent to an estimated 3.35% of regional gross domestic product (GDP), compared to an European Union average of 2.2% in the same year and an increase from 2.08% in 2009 (Debackere et al., 2021[19]). In 2019, the region exceeded the European Union (and Flemish Government) target of spending 3% of GDP on R&D for the first time. Around 70% of total expenditure on R&D in 2019 (EUR 6.6 billion) was spent in the business sector, which is also the sector that has seen the highest rate of growth in R&D spending in the last decade. When spending on research in Flemish higher education institutions located in Brussels is taken into account, around 18% of total R&D expenditure (EUR 1.7 billion) in 2019 occurred in the higher education sector. A further 12% of R&D expenditure was executed in the government sector, including the four Strategic Research Centres (SOCs), and the remaining 0.4% was spent within the non-profit research sector (Debackere et al., 2021[19]).
As shown in Figure 4.2, the share of total research spending (Gross Expenditure on R&D – GERD) in the higher education sector – at 18% - is lower than in many comparable European jurisdictions, although at around the same share as in Germany. This pattern reflects the high level of R&D expenditure in the business sector in Flanders, as well as the existence of large public research institutions outside the higher education sector. Expenditure on R&D in the business sector in Flanders in 2019 was the equivalent of 2.4% of regional GDP, compared to an EU-27 average of 1.46%. Spending on R&D in public research establishments outside the higher education sector represented 0.4% of regional GDP, compared to an EU-27 average of 0.25%. Leaving aside Portugal, which has invested heavily in its university research system in recent years to expand the country’s relatively under-developed research system, Norway and Denmark are notable for the high proportion of total research spending that is undertaken in the higher education sector.
Figure 4.2. Proportion of total R&D performed in the higher education sector
Note: Data for the Flemish Community include spending on R&D in higher education institutions in the Brussels-Capital Region.
Source: OECD (2021[20]) OECD Main Science and Technology Indicators http://oe.cd/msti (accessed on 2 June 2021); Data for the Flemish Community: Debackere et al. (2021[19]) Totale O&O-intensiteit in Vlaanderen 2009-2019 “3% nota” (Total R&D intensity in Flanders 2009-2019 “3% report”), https://www.ewi-vlaanderen.be/sites/default/files/bestanden/3_nota_2021.pdf (accessed on 10 August 2021).
Figure 4.3. Spending on R&D performed in the higher education sector
Note: * Data for Wallonia are for 2017. Data for the Flemish Community on higher education expenditure on research and development (HERD) include spending on R&D in Flemish higher education institutions located in the Brussels-Capital Region, but are referenced to the GDP of the Flemish Region. The EU-27 refers to the 27 member states of the European Union from 2020 onwards.
Source: Eurostat, Main database, https://ec.europa.eu/eurostat/web/main/data/database (accessed on 2 June 2021); Data for the Flemish Community: Debackere et al. (2021[19]) Totale O&O-intensiteit in Vlaanderen 2009-2019 “3% nota” (Total R&D intensity in Flanders 2009-2019 “3% report”), https://www.ewi-vlaanderen.be/sites/default/files/bestanden/3_nota_2021.pdf (accessed on 10 August 2021).
Figure 4.3 shows R&D spending in the higher education sector (Higher education expenditure on research and development – HERD) in OECD jurisdictions as a proportion of GDP and the absolute value of spending on R&D in higher education in euros per capita. This shows that spending per capita on R&D in higher education in the Flemish Community, at around EUR 240 per capita annually, is around 70% higher than the EU-27 average, but only around 80% of the level seen in Austria and Finland and less than half of the level in Denmark and Norway. Viewed as a proportion of GDP per capita, the level of spending on higher education research in the Flemish Community, at 0.6% of GDP in 2019, is around the same levels as in the Netherlands, but lower than in Finland, Austria, Scotland and the three Scandinavian countries. As noted, part of the difference between spending on R&D in the higher education sector in the Flemish Community and in comparable jurisdictions is explained by the significantly larger role in Flanders of public research establishments outside the higher education sector. In the European Union, only Germany spends a higher proportion of GDP than Flanders on research in government research establishments (Eurostat, 2021[21]). Furthermore, the most recent increases in the budget for the BOF and the Research Foundation – Flanders (FWO) are not reflected in the data on higher education expenditure shown in Figure 4.3.
Although levels of spending on R&D in higher education in the Flemish Community were somewhat lower than in other high-spending northern European higher education systems in 2018, academic research undertaken by Flemish researchers (a large proportion of which occurs in Flemish universities) performs highly on international measures of research impact. As shown in Figure 4.4, research outputs from Flanders are cited significantly more frequently (the Mean Observed Citation Rate) than the average of all publications in the same scientific journals or publication media, achieving one of the highest observed citation rates of major European research systems. Against this measure, only Ireland, Denmark and the Netherlands achieve higher impact from their published research among the comparator systems shown.
Figure 4.4. Citation rates in Flanders and other OECD jurisdictions
Note: The Mean Observed Citation Rate (MOCR) is the number of citations in the three years following the publication divided by the number of publications. The Mean Expected Citation Rate (MECR) of a publication is defined as the average number of citations of all publications in the same journal in the same year.
Source: ECOOM (2019[22]) Vlaams Indicatorenboek (Flemish Indicators Book), https://www.vlaamsindicatorenboek.be/ (accessed on 2 June 2021).
4.2 How Flemish institutional funding for research compares and performs
As with the core funding model for higher education institutions, it is helpful to consider criteria that might characterise effective models for the public funding of research in higher education. In the area of research, governments and higher education institutions also look for funding approaches to fulfil a range of objectives. Among the objectives for higher education research funding most frequently cited in policy documents in OECD jurisdictions and in the literature dealing with research funding are:
1. Funding should support a diversified range of research activities, including “blue-sky”, curiosity-driven basic research, strategic research focused on societal challenges (such as climate change, for example) and applied and practice-oriented research focused on solving immediate and practical problems.
2. Funding systems should support – and perhaps promote or incentivise – high quality and impactful research in all domains. The ongoing debates about the design and actual effects of performance-based research funding (PBRF) discussed later in this chapter are directly relevant to this objective.
3. Funding from external (third party) funders of research should provide adequate funding to support legitimate overhead costs associated with running research-intensive organisations, such as higher education institutions. The growth in research projects in the higher education sector that are funded by external public, private and non-profit organisations in recent years has raised increasing concerns about overhead recovery in many OECD jurisdictions (EUA, 2018[23]).
4. Public funding for research in higher education should provide appropriate access to resources for research to all higher education institutions and other research institutions with a relevant role in the research system. In this respect, a frequent concern is how to balance the objectives of supporting world-leading research capacity and research institutions and of promoting high-quality research across a wide range of institutions in a research system, including professionally oriented higher education institutions.
Combining findings from interviews conducted for this review and comparative analysis of models of funding research in higher education in other OECD jurisdictions, the following sections examine how the Flemish funding model performs and compares to other OECD systems in relation to the four objectives above. These sections analyse, in turn:
How the Flemish model handles the balance between basic, strategic and applied research and whether lessons exist from international evidence on what the “right” balance might be;
The way the Flemish system for funding research in higher education seeks to promote the quality and impact of the research funded and undertaken in the sector;
The extent to which legitimate overhead costs in Flemish higher education institutions are covered by external research funding and the effects this has on institutions;
The extent to which the mechanisms in place for funding research in higher education institutions in the Flemish Community provide adequate funding for the different institutions in the system.
Funding basic, strategic and applied research: finding a balance
Research types and basic allocation mechanisms for research funding
Policy discussions about how to create balanced systems of research funding frequently revolve around two related, but ultimately distinct, questions. The first is the appropriate balance between support for different types of research. As set out in Box 4.1, the main distinction traditionally used in classifying research has been between basic research, undertaken to increase the stock of human knowledge, without a specific or direct application in mind, and applied research, undertaken to solve immediate problems.
In practice, an intermediate research type is often introduced in research policy, including in the Flemish Community and in European Union research programmes (Flemish Government, 2009[24]; European Commission, 2021[25]), in the form of “strategic” or “mission-oriented” research. These terms are typically used to refer to basic research undertaken with a specific societal objective in mind (such as tackling poverty, reducing pollution or increasing the energy efficiency of buildings, for example), but where the precise application of the new knowledge to be created is not clear at the outset of the work. It is in this context that a distinction is sometimes made between curiosity-driven research, referring to basic research where the researcher or research team define the research objectives freely, and mission-driven or mission-oriented research, where the research objectives are determined – at least partly – by research strategies defined by universities, research centres, public authorities or research funding agencies.
In Flemish research policy, mission-driven research is sometimes referred to as “strategic basic research” and sometimes as “targeted” (gericht) research. It is contrasted with curiosity-driven, “untargeted” (niet-gericht) research. In the Netherlands, the term “ongebonden” (literally “unbound” or “unfettered”) has been used in recent policy reviews to refer to curiosity-driven research, contrasted with “strategic” research that is explicitly focused on societal objectives (KNAW, 2019[26]). The level of detail in which national research funding agencies prescribe objectives for strategic, mission-oriented or targeted research varies between OECD jurisdictions. Flemish policymakers argue that Flemish government programmes to support targeted research typically allow researchers considerable freedom to specify research objectives.
The second key question for structuring public research funding is the choice of funding allocation mechanisms. Here, the question is the appropriate balance between institutional research funding, where grants are allocated directly to higher education institutions, and external competitive funding, typically awarded by research councils. Discussions on this issue are often linked to those on the balance between different types of research, based on an assumption that competitive research funding is often mission-driven and institutional funding allows more space for curiosity-driven research. While there may be some truth to this in some systems (KNAW, 2019[26]), there is no necessary correlation between the two distinctions. Competitive funding systems can provide grants for curiosity-driven research, while core research grants to higher education institutions may be used to fund institutional or departmental research strategies with clearly defined societal goals.
Box 4.1. Types of research and development: international and Flemish definitions
The OECD’s Frascati Manual (OECD, 2015[27]), the main reference for international data collection on research and development (R&D), defines three types of R&D activity (which can be performed in business, higher education, government or non-profit sectors) based on the objectives of the activity:
1. Basic research (referred to as “fundamental research” in Dutch) is experimental or theoretical work undertaken primarily to acquire new knowledge of the underlying foundations of phenomena and observable facts, without any particular application or use in view.
2. Applied research is original investigation undertaken in order to acquire new knowledge. It is, however, directed primarily towards a specific, practical aim or objective.
3. Experimental development is systematic work, drawing on knowledge gained from research and practical experience and producing additional knowledge, which is directed to producing new products or processes or to improving existing products or processes.
The Frascati Manual also provides a classification of government budget allocations for R&D (GBARD) by socio-economic objective (SEO) that distinguishes between funds that government allocates for specific thematic areas (such as environment or transport) and funds that it allocates for the “general advancement of knowledge”, without specifying a purpose or field. The latter category includes general university funds (GUF), defined as the R&D funding share coming from the general grant that universities receive from government.
In addition to the concepts of “basic” and “applied research” (defined in accordance with the Frascati Manual), the Flemish legislation governing government support for science and innovation (Flemish Government, 2009[24]) distinguishes “strategic basic research” (strategisch basisonderzoek). This is defined as “high-quality generic research that aims to build scientific or technological capacity as a basis for economic and/or societal applications that are not clearly defined at the start of the research and can only be developed through follow-up research”. Strategic basic research may be considered, in the terms of the Frascati Manual, as basic research that is targeted at a specific socio-economic objective, rather than at the general advancement of knowledge.
Source: OECD (2015[27]) Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development https://dx.doi.org/10.1787/9789264239012-en; Flemish Government (2009[24]) Decreet betreffende de organisatie en financiering van het wetenschaps- en innovatiebeleid (Decree concerning the organisation and funding of science and innovation policy), https://codex.vlaanderen.be/Portals/Codex/documenten/1018085.html (accessed on 4 June 2021).
The case for funding basic research is well established. Not only do future scientific advances in applied research depend on foundations laid by previous basic research (Ronai and Griffiths, 2019[28]), but high-quality basic research activity in universities is also crucial for creating conducive environments for training the next generation of researchers and delivering research-based education at undergraduate and post-graduate level (Universities New Zealand – Te Pōkai Tara, 2017[29]). The extended time horizons associated with basic research and the lack of prospects for financial returns in the short to medium-term create forms of “market failure” that mean that public investment is often the only viable option to allow such research to take place. History demonstrates that research driven by the curiosity of individual researchers, studying an aspect of physics, mathematics, psychology or philosophy, for example, purely for the advancement of knowledge, has often been behind major scientific discoveries and advances in thinking that modern societies now take for granted. Equally, however, other breakthroughs in basic research, particularly in fields such as medicine, economics, sociology or law, have been spurred by a desire on the part of researchers to solve (or contribute to solving) a particular societal challenge (Monard et al., 2018[30]).
A recent study by the Royal Netherlands Academy of Arts and Sciences (KNAW) examined the funding allocated by the Dutch Research Council (NWO) for curiosity-driven research (ongebonden onderzoek) in comparison to mission-driven research (strategisch onderzoek), with a view to advising the Dutch Government on the most appropriate policy to adopt (KNAW, 2019[26]). The advisory panel argues strongly that both types of basic research are important for a functioning research system, suggesting a 50:50 ratio in funding for curiosity-driven and mission-driven basic research. It calls for a rebalancing in the funding system in the Netherlands, with an increase in core grant funding (rijksbijdragen) to allow for more curiosity-driven research, pointing to the perceived success of the Excellence Initiative in Germany (BMBF, 2021[31]) in boosting curiosity-driven research, as well as the design of the most recent European Union framework programmes (European Commission, 2021[25]).
The second structural question noted above – the right balance between external competitive research funding and research grants allocated directly to higher education institutions – has also been the focus of research and analysis, although findings are mixed. Most policy-relevant reviews support the principle of open competition as a means to identify and allocate funding to excellent researchers and research. The value of existing competitive research funding systems, such as the National Science Foundation (NSF) or National Institutes of Health (NIH) in the United States, the European Research Council (ERC) in the European Union or the Research Foundation – Flanders (FWO), in supporting excellent curiosity-driven and mission-driven basic research is widely acknowledged (Monard et al., 2018[30]; Wahls, 2018[32]).
However, some analysts – generally academics themselves – have argued that competitive research systems are not as effective in selecting excellent research proposals as often portrayed and that institutional grants for research are a more effective means to promote efficient and effective higher education research systems. On the basis of a quantitative analysis of the research performance of different European higher education systems, Sandström and Van den Besselaar (2018[33]), for example, argue that researchers are motivated to produce high-quality outputs primarily by the challenge of discovery and peer recognition. They point to known disadvantages of competitive funding systems, such as the time used in submitting unsuccessful funding bids, the potential of bias and conservatism in the panels conducting the ex-ante selection of research proposals and the risk that winning prestigious grants is seen as a primary measure of success, rather than the quality of the research actually delivered. Taking into account these factors, they argue that institutional funding for research is the most efficient way of supporting high-quality research.
Other commentators have taken a more nuanced view, noting the value for researchers and research projects of having to prepare research proposals, submitting them to external scrutiny and competing nationally or internationally with the best researchers in a given field. While recently recommending an increase in institutional funding for Dutch universities, the Royal Netherlands Academy of Arts and Sciences does not fundamentally call into question the value of competitive research funding from the Dutch Research Council (NWO) and the European Research Council (KNAW, 2019[26]). A similar line has been taken by the Royal Flemish Academy of Belgium for Science and the Arts (KVAB) (Monard et al., 2018[30]). Moreover, from an empirical perspective, the detailed design of institutional and competitive funding systems, as well as a complex set of contextual factors, including culture and history, are likely to have more impact on research quality and efficiency than the presence of particular ratios among funding streams (OECD, 2018[34]).
The orientation of public funding within the Flemish research system
As noted in Box 4.1, international and national data systems for R&D classify research activities by type (basic and applied) and the sector where they are performed (including higher education), while classifying government spending on R&D by socio-economic objective. Figure 4.5 provides an overview of the level of government allocations for research and development (GBARD) per inhabitant for selected OECD jurisdictions, including Flanders for the year 2019. For each jurisdiction, the figure shows the proportion of spending per inhabitant that is spent on core research grants for higher education institutions (General University Funds – GUF), the proportion that is spent on curiosity-driven (untargeted) research from other sources and the proportion that is spent on mission-driven (targeted) research.
The Frascati Manual (OECD, 2015[27]) assumes that General University Funds are allocated for the “general advancement of knowledge” – in other words for curiosity-driven basic research, without a specific socio-economic objective specified in advance. Other funding for the general advancement of knowledge (shown here as “untargeted” funds from other sources) may be allocated for curiosity-driven basic research by research councils or other research funders. “Targeted” research funding includes funds for mission-oriented basic research and applied research, linked to clearly defined socio-economic objectives, although, as noted, the degree to which public research funders specify detailed research objectives for targeted funding varies between OECD jurisdictions.
Figure 4.5. Orientation of government spending on R&D by socio-economic goals
Note: For the Flemish Government expenditure, the research component of the operating funds for universities makes up the majority of the “General University Funds”, the BOF (highlighted separately) and a majority of FWO funding are classified as non-targeted funding from other sources and the IOF is classified, along with other targeted funding streams, as targeted GBARD. Spending per inhabitant used the population for Flanders in the reference year. A proportion of funds are spent in higher education institutions in the Brussels-Capital Region, outside the territory of the Flemish Region.
Source: OECD (2021[20]) Main Science and Technology Indicators, http://oe.cd/msti (accessed on 2 June 2021); Data for Flanders from Belspo (2021[35]) Government budget appropriations or outlays for R&D, https://www.belspo.be/belspo/stat/b32_en.stm (accessed on 2 June 2021).
Based on the official classification of government R&D spending used by the Belgian authorities for internationally comparable data (Belspo, 2021[35]), Figure 4.5 shows that just over 40% of total Flemish government spending on R&D goes to untargeted, curiosity-driven research. Slightly less than half of this funding for untargeted research is directed through General University Funds (the research component of the operating grant), a further quarter is allocated through the Special Research Funds (BOF) and a majority of the remainder is awarded through competitive procedures by the Research Foundation – Flanders (FWO).
In terms of spending per inhabitant, when the research component of the operating funds and the BOF are combined, the Flemish government provides a similar level of untargeted core funding for research to universities as their Finnish counterparts and more than twice the level of direct grant provided to universities by governments in the United Kingdom and Ireland. In contrast, governments in Sweden, the Netherlands, Denmark, Germany and Norway provide a proportionally higher level of untargeted funding for research to universities – in the case of Denmark, more than twice the funding level per capita.
Two key factors need to be kept in mind when interpreting the international data presented in Figure 4.5. First, the underlying classification of funds as “targeted” (mission-driven) and “untargeted” (researcher-driven) is likely to vary between OECD jurisdictions and may not always reflect reality. Research funding classified as “untargeted” may not always allow full freedom to researchers to define their research objectives. Equally, some funds which are nominally “targeted” by government or research funders may, in fact, allow considerable room for researchers to establish research objectives, within very broad parameters (such as research on climate change or energy, for example).
Second, the constellation of research-performing organisations in domestic research systems influences the level and proportion of public research funding that is directed to higher education institutions. In Flanders, public research establishments play a far more important role than most other OECD research systems, conducting research valued at around 0.4% of regional GDP annually, compared to only 0.15% and 0.12% of GDP for the equivalent sectors in Sweden and the Netherlands respectively. This structural difference is an important factor in explaining the distinctive pattern of Flemish government research spending in comparison to the patterns of government spending in Sweden and the Netherlands – two countries with very similar rates of total government R&D spending to that in Flanders.
The Flemish university sector has called for an increase in untargeted research funding from the Flemish government (VLIR, 2020[36]). It is clear that the government has an ambitious strategic research agenda linked to focus areas (speerpunten) defined at Flemish level, in areas such as digital transformation, health and carbon neutrality (Flemish Government, 2020[37]). This agenda informs the allocation of a proportion of the research funding provided to the four public, thematically oriented Strategic Research Centres (SOCs) and to universities through the Research Foundation – Flanders (FWO). However, there appears to be no solid evidence from the Flemish system or internationally to suggest that current levels of strategic research funding in Flanders are inappropriate or that the proportion of total public research funding for curiosity-driven research is too low. As illustrated by the discussion above, the inherent difficulty of classifying research funding as mission-driven and curiosity-driven complicates matters further and renders debates focused purely on the notional proportion of funding for one or the other type of research sterile and unhelpful.
A 2018 evaluation of public funding for basic research in Flanders (Van der Beken et al., 2018[38]), covering both the BOF and FWO funding, concluded that the mix of funding streams was generally balanced and the different research funding mechanisms are complementary. The BOF was found to allow universities the autonomy to develop their own research strategies, including in blue-sky and inter-disciplinary research, and to support young researchers in developing their research profile and skills, preparing them to compete for external FWO or European funds. At the same time, the evaluation noted that Flanders had – at that time – not yet reached its goal of spending 1% of GDP on public research funding (to contribute one-third of the 3% spending target) and argued that there was scope to direct additional funds to supporting young researchers, to promoting cooperation between universities and for pioneering, inter-disciplinary research.
In the policy position previously cited (Monard et al., 2018[30]), the Royal Flemish Academy of Belgium for Science and the Arts (KVAB) calls for higher core research funding for higher education institutions to bring spending levels in line with high-spending northern European countries (see Figure 4.5). The KVAB argues that additional resources are required to create an environment that supports young researchers to build their research portfolios. It would also allow more researchers to pursue academic careers, particularly in light of comparative high student-to-staff ratios in Flemish higher education (see Chapter 6). Somewhat in contrast, the same policy position suggests that the current funds for the Methusalem component of the BOF, which provides long-term grants for top researchers within individual universities, would be more appropriately situated at Flemish level, as a fully competitive, excellence-oriented funding stream.
Criteria for allocating research funding to higher education institutions
As noted in Chapter 3, OECD governments have increasingly introduced formula-based methods for allocating grant funding to higher education institutions. As part of this process, funding systems have often linked the allocation of the research components of core grants to universities to specific research-related metrics, including PhD graduate numbers, measures of third-party research income and bibliometric indicators of research productivity and impact. A smaller number of OECD countries have linked allocation of institutional funding for research to the results of peer reviews of research units and departments, undertaken in the context of systematic research assessment exercises. Analyses and policy discussions relating to research funding allocation models tend to centre on three main aspects of policy design:
the relative merits of using indicators and peer review for research assessment and funding;
the strengths and weaknesses of different indicators in capturing research quality and;
the positive and perverse effects of performance-based research funding (PBRF) on research productivity and quality in practice.
The use of indicators and peer review in research funding models
In a recent and comprehensive review of performance-based research funding systems in Europe, Zacharewicz et al. (2019[5]) identify three main models that are used to allocate the core institutional research grants to universities:
1. Systems with no performance elements in their models for allocating core institutional research grants to institutions. This includes systems where institutional research funding is allocated based solely on education metrics, such as student enrolment.
2. Systems where institutional core research grants are allocated based on research-related metrics for previous years. Within this group are systems that use bibliometric indicators (publications and impact indicators) in their formulas and those that rely solely on non-bibliometric indicators, such as PhD graduate numbers or research income obtained from third-party funders. Formative peer review of research-performing entities may occur in these systems, but peer review does not play a direct role in funding allocation.
3. Systems that allocate institutional research funding taking into account the results of peer-review-based assessment exercises. Here, a distinction can be made between systems that base allocations on the results of peer review that is informed by research metrics (albeit with differences between fields) and systems that allocate funding based on peer review, without reference to bibliometric data.
Table 4.4 provides an overview of the approaches to allocate core institutional research grants to universities in European countries. This illustrates that formula models using bibliometric indicators are used in ten European jurisdictions, including the Flemish Community. In the Flemish case, this applies to the formulas for both the research component of the core operating grant for universities, which takes into account publications and citations (see Table 3.1 in Chapter 3), and the formulas for the Special Research Funds (BOF) and Industrial Research Funds (IOF), which both include bibliometric indicators, albeit to a limited extent in the case of the IOF (see Table 4.2 and Table 4.3).
Austria, many German federal states and the Netherlands use PhD graduate numbers in their formula, but no bibliometric indicators. The Czech Republic, Italy, Lithuania and Portugal use peer review exercises that draw on bibliometric indicators to different extents in different fields. The United Kingdom, which has a long tradition of performance assessment in higher education research, is the only European country to rely solely on peer review for the allocation of the core research grant to universities. The UK model, first introduced in 1986, inspired the other peer-review systems in Europe and in other parts of the world, such as the system implemented in Brazil by CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Coordination for the Development of Higher Education Personnel).
Table 4.4. Allocation models for core institutional research grants
Source: Adapted from Zacharewicz et al. (2019[5]) Performance-based research funding in the EU Member States – a comparative assessment, http://dx.doi.org/10.1093/scipol/scy041.
Alongside their mapping of research funding systems in Europe, Zacharewicz et al. (2019[5]) identify the most important advantages and drawbacks of the main models of performance-based research funding, based on a wide-ranging literature review (see summary in Box 4.2). In broad terms, bibliometric-based allocation systems have the advantage of their perceived objectivity and low cost, along with the comparatively low administrative burden they place on researchers. At the same time, such approaches, if not applied carefully, can easily disadvantage researchers in the humanities and social sciences, where there is a less dominant tradition of international publishing. As in the Flemish Community, some systems address this challenge through the creation of specific, domestic databases to record publications in the humanities and social sciences (see below). Bibliometric metrics are generally applied at the level of institutions, rather than departments or research groups, owing to data availability. Moreover, as discussed in later sections, the validity as indicators of research quality of the parameters used in bibliometric assessments remains contested within the scientific community and beyond.
In contrast, peer-review exercises allow a more qualitative and differentiated approach that can take into account differences between disciplines and provide assessments at the level of individual research groups. However, peer review is inherently subjective and carries the risk that research is consistently assessed in the light of dominant theories and schools of thought in a way that stifles innovation. Particularly in small research systems with languages which are not widely spoken, the pool of peers available to conduct peer reviews may be too small and too closely inter-connected to permit an objective assessment exercise, although this issue has also been a challenge in some larger systems (OECD, 2018[39]). Crucially, peer review is labour-intensive and expensive, meaning reviews can be conducted less frequently than indicator-based assessments of research output and impact.
Box 4.2. Performance-based research funding systems: strengths and drawbacks
As part of their broad survey of the design of performance-based research funding (PBRF) systems in European countries, Zacharewicz et al. (2019[5]) identify strengths and risks and drawbacks of different models of research performance assessment used for institutional research funding.
Advantages and drawbacks of different PBRF approaches
Source: Adapted from Zacharewicz et al. (2019[5]), Performance-based research funding in EU Member States—a comparative assessment, http://dx.doi.org/10.1093/scipol/scy041.
The choice and effects of indicators in performance-based research funding
In systems that do use bibliometric indicators to measure research productivity and impact, debates are ongoing about the strengths, weaknesses and validity of different metrics. Among the ten European systems using formulas with bibliometric indicators to allocate core research grants to universities that are listed in Table 4.4, seven apply various parameters and weightings to reflect the assumed impact value of the journals where research outputs are published. Authorities have tended to combine indicators of publication volume with journal impact weightings to introduce a proxy for the quality of publications into formulas and to reduce perverse incentives to publish in low impact (low quality) journals to boost output (Zacharewicz et al., 2019[5]). However, the use of journal impact factors has been widely criticised on the grounds that they provide an indication of the general quality of the journal, but not of individual research outputs. Notably, the 2012 Declaration on Research Assessment (DORA), signed by multiple scientists internationally, calls for research funding bodies and research performing institutions to avoid using journal impact factors in the assessment of individual scientists’ research outputs or in hiring, promotion, or funding decisions (DORA, 2012[40]).
Of the same ten systems, only the Flemish Community, Croatia, the French-speaking Community of Belgium, Norway and Sweden use citations as a parameter in their allocation models. Citation data provide a more direct indicator than journal-related weightings of the impact of individual publications within the scientific community. However, some analysts argue that, while citation indicators provide a reasonable measure of the scientific relevance and impact of a research output, they fail to capture other dimensions of research quality, such as the underlying solidity of the work, its originality or its broader societal value (Aksnes, Langfeldt and Wouters, 2019[41]). In their Leiden Manifesto for research metrics, Hicks et al. (2015[42]) acknowledge the value of bibliometric data, but call on them to be used exclusively in combination with qualitative, expert assessment, arguing that “indicators must not substitute for informed judgement”.
Notwithstanding the different conceptual, methodological and practical concerns about the use of peer review and bibliometric indicators in funding allocation models, the authors of recent studies concur that evidence on the effects of performance-based research funding models is mixed and often inconclusive (OECD, 2018[34]; Debackere et al., 2018[43]; Zacharewicz et al., 2019[5]). A first challenge is that it is difficult to observe reliably changes in the behaviour of researchers, research groups and higher education institutions that may result from changes to research-funding models. As Butler (2010[44]) notes in an early review of available evidence: “For every anecdote about a particular response … [to a performance-based research funding system]…, it will nearly always be possible to discover an anecdote that suggests the contrary”.
A second difficulty is that incentives created by the design of the research funding system are only one of the factors that influence researchers and the research system, making it challenging to isolate the specific influence of the research funding model. The level or growth of research funding appears likely to have an important influence on research productivity and impact, for example. The swift increase in the scientific performance of the Danish university system in recent years, for instance, has coincided with sustained increases in public funding. The degree of internationalisation of a research system also appears to have a positive influence on research quality, as international co-publications are more highly cited than single-author or national research outputs. University-level policies and the degree of competition in the system are also likely to influence researcher behaviour and, ultimately, the number and quality of the research outputs they produce (OECD, 2018[34]).
Despite the uncertainties surrounding the precise influence of research funding design on research systems, there is some evidence that fears about the perverse effects of performance-based research funding systems might be exaggerated. In an analysis of the research performance of the higher education system in Australia, Van den Besselaar, Heyman and Sandström (2017[45]) found that the introduction of a new performance-based research funding model in the late 1980s was ultimately associated with an increase in both the productivity and impact of Australian higher education research. This contrasts with earlier findings by Butler (2003[46]) that suggested the new system had led to an increase in research production, but a fall in citation impact of the research produced, suggesting a focus on quantity over quality. Van den Besselaar, Heyman and Sandström argue that this initial drop in citation impact is likely to have reflected a transition phase, as Australian academics sought to publish in higher quality, international journals (Van den Besselaar, Heyman and Sandström, 2017[45]).
New approaches: towards the use of alternative indicators of research impact?
Although the perverse effects of established bibliometric indicators may be more limited than some authors have predicted, like all indicators, they have their limitations. In particular, citation indicators mostly capture the impact of research outputs in the relatively closed – and self-referential – universe of researchers in the same or related disciplines. In the last decade, interest has grown in alternative metrics (“altmetrics”) as a means to capture the wider impact and use of research outputs (Khodiyar, Rowlett and Lawrence, 2014[47]; European Commission, 2018[48]). Alternative metrics measure online activity related to research outputs, which can include preprints and other identifiable documents, as well as finalised publications. The online activities now measured by companies, such as Altmetric or Plum Analytics, include saving and bookmarking references, mentions in online content, including in the press, non-academic literature and blogs, as well as sharing on social media.
Altmetrics are typically seen as complementary to established bibliometric indicators, rather than as a replacement. Many universities now encourage staff to track the impact of their research with altmetrics, as well as traditional bibliometric indicators, particularly as altmetrics can provide faster feedback on the impact a publication has and provide an indication of impact among a wider range of audiences (Imperial College London, 2021[49]). There is some evidence that the performance of research outputs in relation to altmetrics correlates to their performance in relation to traditional citation indicators, lending weight to their potential as complementary indicators of research impact (Glänzel, Chi and Debackere, 2020[50]). However, as synthesised in Box 4.3, a number of questions remain about the validity and proper interpretation of altmetrics, which limits their current potential for use in research assessment and research funding policies. Among the most important of these limitations from a quality assurance and policy perspective is that the online attention cannot be used reliably as a proxy for quality, meaning that use of altmetrics in decisions nominally based on quality assessment would be risky and contestable.
Box 4.3. Strengths and weaknesses of altmetrics
Altmetrics provide an indicator of the online attention received by an article or other research output. In general, their advocates argue that almetrics can complement – but not replace – traditional metrics like citations by capturing interest in research that cannot be measured in a citation count.
The strengths and weaknesses of altmetrics
Source: Based on: Imperial College London (2021[49]), Altmetric, https://www.imperial.ac.uk/research-and-innovation/support-for-staff/scholarly-communication/bibliometrics/altmetric/ (accessed on 3 June 2021).
To improve the range of information available on the impact of the research that they fund, research funding bodies in some OECD jurisdictions have developed dedicated online facilities in which researchers are asked to enter information about the publications, collaborations and intellectual property resulting from their research. A relatively well-established example, originally developed by the UK research funding councils in 2013, is Researchfish in the United Kingdom (Researchfish, n.d.[51]).
In the Flemish Community, the Flanders Research Information Space (FRIS) has been created as a regional platform to gather information on government-funded research activity, researchers, business developers and research outputs directly from publicly funded research organisations (Flemish Government, 2021[52]). The platform provides a single, largely automated reporting facility for researchers in receipt of public research support and aims to simplify the reporting burden in comparison to systems previously used. The non-confidential information collected is made publicly available on the FRIS portal to increase the visibility of Flemish research and stimulate research cooperation, knowledge transfer and valorisation. The information collected also serves to monitor the impact of the publicly funded research and provide transparency for the public. From 2022 onwards, FRIS will be expanded to provide information on patents, research infrastructure and additional data on research publications.
The impact of performance-based research funding in Flemish higher education
The Flemish government linked the allocation of the Special Research Funds (BOF) to publication and citation parameters in 2003 and introduced the number of publications as a parameter for allocation of the research component of the operating grant to universities in 2008. In the period since the introduction of these mechanisms, the output and impact of research from Flemish universities has increased significantly. The number of publications in scientific journals authored or co-authored by researchers based in Flanders almost doubled from 15.7 per 10 000 inhabitants in 2006 to 28.2 per 10 000 inhabitants in 2018 (ECOOM, 2019[22]). The average number of citations per publication also almost doubled in the same period (ECOOM, 2019[22]).
Studies into the effects of the research-funding model in Flanders note this impressive improvement in productivity and impact, but like the international studies discussed above, conclude that it is impossible to dissociate the impact of the performance-based funding models from other contextual factors. In one of the most thorough examinations of the Flemish research funding model and its impacts, Engels and Guns (2018[53]) highlight the sophisticated design of the BOF allocation formula and the significant efforts made to capture publication and impact in the social sciences and humanities through the Flemish Academic Bibliographic Database for the Social Sciences and Humanities (VABB-SHW).
However, they also stress that the direct influence of the bibliometric indicators on the amount of funding individual universities receive is comparatively limited, owing to the guaranteed minimum shares of total BOF funding for the three smaller universities (see above). Moreover, they highlight the growth in public spending on research in Flanders and increase in the number of researchers in the period since the funding models were introduced, as well as a broader trend internationally towards higher rates of research productivity and citation impact. All these factors are likely to have increased research output and impact levels. While the parameters used in the BOF formula may have affected the behaviour of researchers and universities, it is not possible to prove that the increases in research output and impact would not have occurred in the absence of the performance-oriented funding system. Representatives of universities consulted for this review also stressed the strong development of institutional policies to monitor and promote research quality in the last two decades, which is also likely to have influenced the level and impact of Flemish university research.
If it is impossible to prove the direct influence of the research-funding model on research productivity and impact, it is equally hard to prove that the funding system has led directly to perverse effects, such as an excessive pressure on academics to produce research outputs (“publish or perish”) or an unhealthy level of competition between academic staff and institutions. Although there are many reports that pressure on academics has increased in the last decade (Monard et al., 2018[30]), this too may result from broader changes in the environment in which academics work. As research has become increasingly internationalised, the level of scrutiny of research outputs – and probably competition between researchers – has also increased. While, as discussed in Chapter 6, action can be taken to address unhealthy levels of workplace stress in academia, a highly performing research system is, by definition, likely to be competitive and demanding.
Notwithstanding these challenges for assessing the influence of research funding, the recent changes to the design of the BOF key appear positive. They have reduced the weight attached to productivity and citation impact, with 50% of total funds now allocated for five years based on historical funding patterns and new parameters introduced to capture and reward the distribution of citations, interuniversity and international cooperation and, ultimately (from 2024), inter-disciplinary research. The Interuniversity Special Research Funds (“iBOF”) programme explicitly supports cooperation between universities. These changes, which were partly introduced in response to findings from the last evaluation of the funding model (Van der Beken et al., 2018[38]), are widely welcomed by universities and other stakeholders. Moreover, given that Flemish universities already perform well on established international research metrics, it appears coherent to focus attention on potentially neglected aspects of research quality and the promotion of innovative inter-disciplinary research.
The full economic cost of research: paying for overhead
Research projects funded by external public bodies, such as research councils, private businesses and non-profit organisations account for a substantial proportion of the research activity undertaken in higher education institutions in OECD jurisdictions (see Figure 4.1). Particularly as research council budgets have increased and higher education institutions have been encouraged and supported by policy to cooperate more actively with business and diversify their revenue streams, the proportion of total institutional funding coming from third-party sources has generally increased in recent decades. Undertaking third-party funded research requires people, space and equipment and draws on the time and resources of central services in universities and colleges. Against this backdrop, there have been growing concerns in many OECD jurisdictions that levels of third-party funding are insufficient to cover the full economic cost (FEC) of conducting the projects funded and that externally funded research projects are being cross-subsidised by other revenue, theoretically destined for teaching or other activities.
The UK’s Higher Education Policy Institute has calculated that there is a “research deficit” of almost GBP 3.3 billion in non-funded costs across the UK higher education sector, for example (Olive, 2017[54]). This funding gap has traditionally been filled from surpluses generated from non-publicly funded teaching (mostly from fees charged to international students) and other income, including from consultancy and technology transfer. In Australia, a research centre found that one dollar in five spent on research comes from surpluses on teaching, with the majority of the income also coming from fees paid by international students (Norton, 2015[55]).
In systems that are more reliant on public funds, including the Flemish Community, higher education institutions generally have less scope to cross-subsidise research from genuine surpluses on other activities, meaning any shortfall in funding from third-party sources for covering the full economic costs of activities must be paid from the general institutional budget. In the Netherlands, for example, analysis has shown that research funded from external sources consumes a large proportion of the government operating grant to universities for research and puts downward pressure on budgets for teaching (Rathenau Instituut, 2018[56]; VSNU and NWO, 2020[57]). Similar frustrations are widely reported in the Flemish higher education sector (VLIR, 2020[36]).
A key concern in the Flemish context is that the rules governing competitive funding provided by public funding agencies – and notably the Research Foundation – Flanders (FWO) – place limits on the amount of indirect costs (overhead) that can be paid from research project funds. Typically, governments and research funders have sought to limit the amount of overhead costs that they fund, often based on an implicit assumption that overhead is negative and associated with waste and inefficiency. In practice, however, all well-functioning organisations require effective central and support services to manage and administer the organisation and support staff undertaking “core business” activities. Studies in the Netherlands, for example, have found that university overhead costs are not higher than overhead costs in comparable organisations in other economic sectors (Huijben et al., 2011[58]).
Table 4.5. Typical overhead rates applied by research funders in selected OECD jurisdictions
As shown in Table 4.5, public research funding bodies in some OECD jurisdictions do include explicit allowances for indirect costs in their funding rules for research projects as does the European Union’s main research funding programme. In the scope of this project, it has not been possible to undertake a comprehensive view of overhead policies used by research funding agencies in OECD member countries. Moreover, the level and role of core institutional funding for research must be taken into account when interpreting the overhead rates applied in different national contexts. In this respect, approaches vary. Some funding agencies, such as the Dutch Research Council (NWO) pay no or very low overhead contributions, based on a historical convention that the government research grant to universities should cover indirect costs (VSNU and NWO, 2020[57]). Research funding agencies in Ireland pay explicit contributions for overhead in a system where the direct public research grant to universities represents a comparatively small proportion of institutional income (see Figure 4.1). The Independent Research Fund Denmark, meanwhile, applies a comparatively high overhead rate for project funding, in a university system where the direct public research grant to universities is relatively higher in international comparison.
Although far from comprehensive, Table 4.5 shows that the overhead rates applied by the FWO in Flanders (typically 17%) are somewhat lower than those applied in other jurisdictions, including the European Union’s Horizon Europe programme, which applies a rate of 25%.
A challenge for Flemish universities seeking to demonstrate the need for additional funds to cover overhead is the absence of a transparent, system-wide cost accounting model that would allow institutions to demonstrate the direct and indirect costs of different activities in a comparable way. Over the last decade, an increasing number of OECD higher education systems have adopted system-wide protocols for activity-based costing (ABC). Such protocols make it possible to identify the direct costs of different types of activity in operational units – such as departments, faculties or research centres – and to attribute indirect (overhead) costs to these activities with some degree of accuracy.
The United Kingdom was the first European country to introduce a system-wide cost accounting model for universities in 1999 (the Transparent Approach to Costing – TRAC). This resulted in part from pressure from within the university sector itself to demonstrate the costs of delivering externally funded research projects and to make the case for higher funding rates from the national research councils. The introduction of other ABC accounting systems has largely been driven by the financial reporting requirements of competitive research funding programmes, in particular the European Union’s research and development framework programmes (EUA, 2018[23]). Many higher education institutions have developed their own ABC models, with varying levels of sophistication. Outside the United Kingdom, system-wide approaches, with a common set of standards, have been implemented in Ireland, the Nordic countries and the United States. Denmark has been the latest country to introduce a standardised cost accounting model in the form of the Common Chart of Accounts (fælles kontoplan) (Danish Ministry of Education and Research, 2020[59]).
Evidence suggests the impact of new activity-based costing models on internal resource allocation and accounting practices within higher education institutions and on public policy is generally positive. In some cases, it is clear that the introduction of new ABC models has had a profound impact on the way higher education institutions operate. In Sweden for example, the introduction of the SUHF model (for Sveriges universitets- och högskoleförbund – Swedish Association of Higher Education Institutions) led to the majority of institutional income being transferred to departments, which then pay a transparent unit overhead surcharge for each unit of their direct salary and operating costs. In Finland, the use of the ABC model is reported to have increased cost-awareness among staff and made the cost implications of engaging in externally funded projects more transparent (EUA, 2018[23]).
Equitable distribution of research funding in the higher education system
The final aspect of research funding policy for the higher education sector addressed as part of this review has been the extent to which all higher education institutions with the potential to contribute meaningfully to overall research efforts have access to funding to allow them to do so. In this context, two questions arise in the Flemish system: a) the allocation of funds for basic research between the five universities and b) the provision of resources for applied and practice-oriented research to the university colleges.
As noted in Chapter 2 (see Table 2.1), the Flemish university landscape consists of two large, research-intensive institutions (the KU Leuven and Ghent University) and three smaller universities in Antwerp, Brussels and Hasselt, of which Hasselt University is the smallest and most recently established. All five universities have well-developed research profiles and unquestionably generate high-quality research. However, the KU Leuven and Ghent University are ranked within the top 100 institutions in the world in the Times Higher Education and Academic Ranking of World Universities (ARWU) (Times Higher Education, 2021[60]; Shanghai Ranking, 2021[61]) and have demonstrably higher internal research capacity than the other three institutions. In such a university research environment, a potential risk could be that performance-based research funding mechanisms lead to a concentration of research funding in the largest institutions, at the expense of the smaller institutions. Over time, as research capacity in the well-funded institutions is strengthened and capacity in the less well-funded institutions develops more slowly, there is a further risk that the concentration of resources becomes more acute. This is the so-called “Matthew effect”, an expression coined by Robert K. Merton (1968[62]), in reference to the biblical parable of the talents:
“For unto every one that hath shall be given, and he shall have abundance: but from him that hath not shall be taken away even that which he hath” (Matthew, 12:29).
While this problem seems largely to have been avoided in the Flemish Community (see below), several international studies suggest that excessive concentration of research resources in a few institutions has a negative impact on the efficiency of the university research system. Based on a quantitative analysis, Sandström and Van den Besselaar (2018[33]) posit that the comparatively superior research performance (in terms of productivity and impact per capita) of the Netherlands and Sweden compared to the United Kingdom is due to a more even distribution of research resources in the university system. In the United States, Wahls (2018[32]) showed that the research productivity and impact of individual principal investigators peaked when they received a certain critical mass of external research funding (USD 400 000 annually), but declined proportionally with higher levels of funding. He argues that this calls for minimum and maximum levels of funding per researcher.
Although, as discussed above, the Flemish funding system for university research is strongly performance-oriented, the design of the funding allocation models contains specific features to ensure that smaller universities are not unduly disadvantaged and receive resources to build their internal research capacity. Firstly, within the allocation model of the research element of the core operating grant, the base funding component (sokkel) uses a degressive weighting system that provides a proportionally higher level of resources to universities with lower rates of publication and PhD graduate output (see Chapter 3). Secondly, both the Special Research Funds (BOF) and the Industrial Research Funds (IOF) provide guaranteed minimum shares of the total funding to the three smaller universities. This funding model appears to be viewed positively in the Flemish university sector.
The second issue relating to the distribution of research funding in the higher education system concerns the resources for research in university colleges. The Higher Education Code makes clear that practice-oriented research is part of the core mission of university colleges (Flemish Government, 2013[63]). In 2020, the total grant to the 16 university colleges for practice-oriented research was EUR 30.2 million (Flemish Government, 2020[1]), amounting to around 3% of total university college revenue. Despite impressive efforts to develop and promote practice-oriented research in university colleges, the review interviews illustrated that this level of core research funding is widely viewed as inadequate. The current level of the institutional allocation makes it difficult for university colleges to fulfil their potential in creating high-quality research-informed learning environments and contributing to solving real-world problems in cooperation with outside partners.
University colleges in the Flemish Community also receive additional public support for research, through their participation in projects funded by the Industrial Research Funds (IOF) received by the association of which they are members. Moreover, as highlighted earlier, university colleges receive some grant aid from the Flemish Agency for Innovation and Enterprise (Vlaams Agentschap Innoveren en Ondernemen – VLAIO) for research projects, training and infrastructure. Defining the role of non-university higher education institutions in research has been a challenge in other OECD jurisdictions, including Ireland and Denmark. Nevertheless, it is clear that some other OECD countries – such as Finland – allocate substantially more public funds for research to universities of applied science and that these institutions have a valuable role to play in the research system. While recognising the differences between systems and institutional missions in different countries, there is scope for the Flemish Community to build on existing research capacity in university colleges and to learn from the experience of other OECD systems in terms of funding research in non-university institutions.
4.3 Policy issues and recommendations for institutional funding for research
Policy issue 1: Balancing types of research and investment
This review has examined the allocation of funding for research to higher education institutions from a necessarily broad perspective and not examined the focus and distribution of the research projects actually undertaken in Flemish universities and university colleges. Nevertheless, available information on the design of policy instruments and the distribution of research spending by type of research shows that the Flemish research funding system delivers substantial and diversified resources for research to universities. The research element of the core operating grant for universities recognises the need to support research as one of the foundations of the university system. The Special Research Funds (BOF) provide earmarked funding for basic research to permit universities to develop their research profiles, allowing institutions almost complete latitude in the internal allocation and use of these funds. The Industrial Research Funds (IOF) provide significant resources for strategic basic and applied research and promote direct cooperation between universities and university colleges, as well as between the higher education sector and businesses. These internal funds are complemented by the competitive funding system administered by the FWO and extensive contract research activities.
Given the specific role of university colleges in applied and practice-oriented research, the question – specifically raised in the terms of reference for this review – of the balance between funding in higher education for strategic or mission-oriented research and curiosity-driven research only concerns the five universities. As noted in this chapter, the evidence base for recommending a particular ratio of funding between these two types of research is weak. Moreover, national and international data that seek to capture expenditure on the two types of research are potentially misleading because of the inherent ambiguity of the definitions used and the difficulty of classifying government spending by research type in a satisfactory manner. If there is a case to adjust the current balance between strategic and curiosity-driven research in Flemish higher education, that case must be made based on a more thorough-going analysis of research activity in universities and evidence that the current focus of resources is detrimental to particular types of research. Taking into account the limited international evidence, this review concurs with the 2018 evaluation of the Flemish research funding system (Van der Beken et al., 2018[38]) in concluding that there is no compelling reason to alter the current distribution of funds between the different funding streams for research.
The level of funding directed through these funding streams is another matter. It would be disingenuous to argue that research in Flemish universities is poorly funded in comparison to average funding levels in OECD jurisdictions. Nevertheless, as illustrated by Figure 4.3 and Figure 4.5, in absolute terms, Flanders spends less on its university research than some comparable jurisdictions, such as the Nordic countries – research systems that have some of the strongest research performance in the OECD and with which Flanders can legitimately be benchmarked. It is also clear, notwithstanding the limitations of research spending targets, that the Flemish government has not yet reached its own target of spending 1% of GDP on public funding for research and development. In 2019, Flemish government spending on research amounted to around 0.7% of GDP. When the Flemish share of Belgian federal government spending on research is taken into account, this rises to 0.8% and, when funds from European Union framework programmes are included, to 0.87% (Debackere et al., 2021, p. 21[19]). Taking into account the strong performance of the Flemish university research system against objective – if imperfect – measures of research productivity and quality and the good performance of systems with higher funding levels, increased investment in research in the sector could be justified.
Recommendations
As the state of public finances allows, the Flemish Government can legitimately continue to increase public funding for research in higher education, as part of broader efforts to reach the government target of spending the equivalent of 1% of Flemish GDP on research and development from public funds (including federal and European funds).
In line with the considerations highlighted in Policy Issue 4 below, there is a case to increase the share of total public funding for research provided to university colleges to enable them to develop their capacity in practice-oriented research. This might be a priority for allocation of additional funding. In the absence of a strong case for rebalancing the research funding system for universities, other available additional funding could be used to increase the resources distributed through all the main public research funding channels for these institutions.
Policy issue 2: Promoting quality and impact through research funding to universities
The methods used to allocate institutional research grants to universities and associations of universities and university colleges incorporate performance metrics with the aim of promoting the quality and impact of the research funded and undertaken in higher education institutions. The implementation of these performance-based allocation models has coincided with a substantial increase in the productivity and impact of research from Flemish universities, although the precise contribution of the funding models to these trends cannot be proven. In light of the already high level of performance observed in the Flemish university research system, the recent shift to reduce the weight of output indicators in the formula for allocating the Special Research Funds (the BOF key), as well as the introduction of more nuanced indicators to capture cooperation and inter-disciplinary research, is positive. Given the evidence on the effects of indicator-driven funding models, it would be unwise to over-estimate the power of these new indicators to incentivise particular behaviour patterns among researchers and institutions. However, initiatives such as the Interuniversity BOF (“iBOF”) to promote cooperation between universities are already bearing fruit and the design of the funding allocation formula sends a clear and positive signal to the research community about policy priorities.
Both the model for allocating the variable component of the research element of the operating grant and the BOF key seek to reward performance, but the two allocation systems use different parameters and weights. While these differences result from the historical development of the two mechanisms, there is no clear rationale for such a difference. Moreover, the existence of two systems for calculating two separate institutional allocations for research to the same five universities creates additional administrative burden, as two sets of calculations must be made each year.
More generally, both the Agency for Higher Education, Adult Education, Qualifications and Student grants (AHOVOKS) and the Department of Economy, Science and Innovation (EWI) report that the calculation of institutional funding allocations each year represents a considerable effort, but generates relatively small changes to institutional budgets. As discussed in Chapter 3, it is preferable to maintain a relatively close – and regularly reviewed – link between student-related activity and the funding institutions receive, as higher education institutions have only partial control over the enrolment and progression of students. For research, this is less clearly the case. Research strategies and employment of researchers are fully within the control of institutions. Moreover, research is a long-term activity from which outputs and impacts evolve more slowly than in education, providing further justification for a longer time horizon for funding allocations. There could be a case for making research allocations to higher education institutions for longer periods and assessing performance using the established parameters for periods of more than one year.
Recommendations
To streamline the research funding system, the Flemish Government, in cooperation with the universities, should analyse the impact of allocating the research component of the operating grant to universities using the same distribution parameters as for the BOF, allowing only one distribution calculation to be made. This might mean, for example, sharing the 45% of the operating grant to universities assigned for research in the same proportions as the BOF (i.e. based on the BOF key). As well as looking at the direct financial impact of such a change for individual universities, the analysis should consider how measures could be designed to mitigate the effects of the change on the budget received by each institution.
In due course, the Flemish authorities should analyse the detailed effects of allocating the BOF to universities, using the parameters in the current BOF key, for four or five-year periods, aligned with the periodicity of the institutional agreements proposed in Chapter 3. The share of the BOF allocated to each university could remain constant for the same four or five-year period, as could each university’s share of the research component of the operating grant. This would contribute to reducing the burden of administering the funding model. Such a move would also require the development of mechanisms to reduce the financial impact of changes in performance on individual universities when the budget level is recalculated at the end of the four or five-year period. The Industrial Research Funds (IOF) could also be allocated for the same four or five-year periods using the current allocation formula to calculate the budget level.
To ensure accountability and transparency, particularly if there is a move to multi-annual funding, the Flemish authorities should consider requiring universities to establish and publish institutional research strategies, explaining priorities for investment of the public funds allocated to them for research (the research component of the operating funds, the BOF and the IOF). This should be accompanied by a summary of measures taken to ensure the relevance and effectiveness of the use of these funds. These institutional research strategies, which would build on existing institutional strategies, could be integrated into the institutional agreements proposed in Chapter 3, if this mechanism is implemented. Alternatively, they could exist as stand-alone documents.
Institutional strategies could identify priority areas for research investment within existing research portfolios and demonstrate how institutional efforts contribute to societal challenges identified at European, Belgian and Flemish level. However, any centralised prioritisation of research activity should be sufficiently broad to allow individual researchers and research teams to tailor their research efforts to changing needs and circumstances. Moreover, strategies should also make clear how long-term, curiosity-driven research would be promoted and protected.
Goals within institutional strategies need not be focused on traditional productivity or impact metrics: they might, for example, include allocating a higher proportion of funds to curiosity-driven research or increasing the number of cooperative research projects. Light-touch annual reporting on progress towards the goals, building on the annual reporting requirements already required by the Higher Education Code, could form a basis for ongoing dialogue between the five universities and the funding departments and agencies. Such reporting could be integrated into the broader reporting and dialogue linked to a new system of institutional agreements, if this is implemented.
Policy issue 3: Funding legitimate overhead costs and increasing cost transparency
Although it is not possible to quantify, it is clear that Flemish universities are feeling the financial consequences of an increased number of research projects and researcher positions that are funded by external partners, which has increased calls on institutional overheads. When research projects are funded by external partners with insufficient allowance for overhead costs, institutions are forced to make up the shortfall with their existing internal resources. It is important for policymakers to ensure that an ambition to increase research activity is not pursued at the expense of diverting institutional resources away from other core missions, most notably learning and teaching.
Calculations of institutional overhead should be transparent and verifiable by funding bodies, elected representatives and interested citizens. This is currently impossible in Flemish higher education due to the absence of a transparent and consistently applied activity-based cost accounting model. The experience of multiple European higher education systems in implementing such costing models has shown the value of such models for understanding and demonstrating cost structures within higher education institutions.
Recommendations
As resources allow, Flemish authorities should take steps to increase the overhead rates applied for resource-intensive research projects funded through Flemish external competitive public funding instruments (notably the Research Foundation – Flanders) to align more closely with the benchmark level of 25% used in the European Union’s Horizon Europe programme. Lower overhead rates may be appropriate for some types of grant, including fellowships. In revising overhead policies, it would be instructive to consult research-funding authorities in selected comparator systems to learn from their experience. As the BOF and IOF are direct grants to higher education institutions, with a pre-determined maximum budget and over which the institutions themselves have considerable control, there is no obvious case for changing the rules applying to these funds. These measures would provide appropriate recognition of the overhead costs generated by a multiplication of externally funded research projects and introduce consistency between Flemish and European funding streams for universities.
If overhead rates are increased within external competitive public funding instruments, higher education institutions should be required to apply the same rates for research financed by private funders.
To improve transparency in relation to costs and the use of funds within higher education for government and society, the Flemish higher education sector should develop and introduce common standards for activity-based cost accounting – initially for universities and, ultimately, for the whole higher education sector. Such a system should make it possible for higher education institutions to demonstrate the indirect costs associated with different activities in a comparable way and strengthen the case for overhead costs to be funded at a higher level in externally funded, project-based research. Common standards for activity-based costing should exploit, as far as possible, the existing cost accounting models used in universities and be kept as simple as possible in order to minimise administrative burden. However, the standards must allow the generation of reliable, transparent and comparable information. In this respect, there is scope to draw on the experience of the Nordic countries, Ireland and the United Kingdom, which have successfully introduced such systems. It is probable that additional peer learning with these systems would be valuable.
Policy issue 4: Distribution of funding across the higher education sector
As noted, the design of the allocation formulas for research funding to universities appears to be equitable and is not contested within the university sector. The question of additional funding for university research is addressed above. The model for allocating funding for practice-oriented research to university colleges, based on student enrolment is also sound, given the strong link between such research and the professionally oriented education within the university colleges. However, based on the consultations conducted for this review, it appears that the current level of core funding for practice-based research in university colleges is almost certainly too low to allow them to build their capacities in research and to fulfil their potential and their legally defined research mission.
At the same time, while individual university colleges have developed their own strategies for practice-oriented research and are taking steps to develop their internal research capacity, the review team found little evidence of a coherent Flanders-wide strategy for research in university colleges that could help to guide future investments in the sector.
Recommendations
As a basis for further investment in practice-oriented research in universities colleges, aimed at allowing these institutions to make an even greater contribution to Flemish research efforts, the university college sector in the Flemish Community should develop a system-wide strategy for practice-oriented research. Such a strategy could be developed collectively by the university colleges, with support from relevant government services and stakeholder organisations and highlight priorities for capacity development and effective approaches to structuring and implementing practice-oriented research in university colleges.
The Flemish Government should allocate a proportion of the increased public funding for research suggested in Policy Issue 1 to increase the institutional grant for practice-oriented research in university colleges (currently administered by the Ministry of Education and Training). This should be done in such a way as to increase both the level of funding for institutions and the share of the total public funding envelope for research allocated to practice-oriented research.
In return for increased investment, the Flemish authorities can legitimately require university colleges to develop institutional strategies for practice-oriented research – in a similar way to the requirement above for universities to develop research strategies. As with universities, these institutional strategies could be integrated within institutional agreements, if this instrument is implemented, or exist as stand-alone documents.
References
[41] Aksnes, D., L. Langfeldt and P. Wouters (2019), “Citations, Citation Indicators, and Research Quality: An Overview of Basic Concepts and Theories”, SAGE Open, Vol. 9/1, http://dx.doi.org/10.1177/2158244019829575.
[13] Audit Scotland (2019), Finances of Scottish universities, Auditor General for Scotland, Edinburgh, https://www.audit-scotland.gov.uk/report/finances-of-scottish-universities (accessed on 20 April 2021).
[35] Belspo (2021), Government budget appropriations or outlays for R&D, Monitoring and evaluation of research & innovation, https://www.belspo.be/belspo/stat/b32_en.stm (accessed on 2 June 2021).
[31] BMBF (2021), Die Exzellenzstrategie (The Excellence Strategy), Bundesministerium für Bildung und Forschung, https://www.bmbf.de/de/die-exzellenzstrategie-3021.html (accessed on 5 June 2021).
[44] Butler, L. (2010), “Impacts of performance-based research funding systems: A review of the concerns and the evidence”, in OECD (ed.), Performance-based Funding for Public Research in Tertiary Education Institutions: Workshop Proceedings, OECD Publishing, Paris, https://doi.org/10.1787/9789264094611-en.
[46] Butler, L. (2003), “Explaining Australia’s increased share of ISI publications - The effects of a funding formula based on publication counts”, Research Policy, Vol. 32/1, pp. 143-155, http://dx.doi.org/10.1016/S0048-7333(02)00007-0.
[59] Danish Ministry of Education and Research (2020), Fælles Kontoplan: Konterings-og fordelingsvejledningen 2021 [Common Chart of Accounts: Accounting and Distribution Instructions 2021], https://ufm.dk/uddannelse/videregaende-uddannelse/faelles-kontoplan/konterings-og-fordelingsvejledningen/Konteringsogfordelingsvejledningenversion2021excl.kapitel6.pdf (accessed on 19 February 2021).
[8] de Boer, H. and B. Jongbloed (2018), Evaluatie interne allocatiemodellen Vlaamse universiteiten (Evaluation of the internal allocation models in Flemish universities), CHEPS - Center for Higher Education Policy Studies, Universiteit Twente, https://data-onderwijs.vlaanderen.be/documenten/bestand.ashx?nr=11643.
[43] Debackere, K. et al. (2018), Mutual Learning Exercise: Performance-Based Funding of University Research, Publications Office of the European Union, Luxumbourg, http://dx.doi.org/10.2777/644014.
[19] Debackere, K. et al. (2021), Totale O&O-intensiteit in Vlaanderen 2009-2019 “3% nota” (Total R&D intensity in Flanders 2009-2019 “3% report”), ECOOM Expertisecentrum O&O Monitoring, https://www.ewi-vlaanderen.be/sites/default/files/bestanden/3_nota_2021.pdf (accessed on 10 August 2021).
[40] DORA (2012), Declaration on Research Assessment (DORA), https://sfdora.org/about-dora/ (accessed on 3 June 2021).
[15] DUO (2021), WO Gegevensboek 2015-2019 (Universities Databook 2015-2019), Dienst Uitvoering Onderwijs, https://www.duo.nl/open_onderwijsdata/databestanden/onderwijs-algemeen/financiele-cijfers/gegevensboeken.jsp (accessed on 24 May 2021).
[22] ECOOM (2019), Vlaams Indicatorenboek (Flemish Indicators Book), Expertisecentrum Onderzoek en Ontwikkelingsmonitoring van de Vlaamse Gemeenschap (ECOOM), Brussels, https://www.ecoom.be/ (accessed on 2 June 2021).
[4] ECOOM (n.d.), VABB-SHW Database, Flemish Academic Bibliographic Database for the Social Sciences and Humanities (VABB-SHW), https://www.ecoom.be/en/data-collections/vabb-shw (accessed on 4 June 2021).
[53] Engels, T. and R. Guns (2018), “The Flemish Performance-based Research Funding System: A Unique Variant of the Norwegian Model”, Journal of Data and Information Science, Vol. 3/4, pp. 44-59, http://dx.doi.org/10.2478/jdis-2018-0020.
[23] EUA (2018), Accepting University Accounting Practices under Horizon Europe A compendium of national and institutional cases, European University Association, Brussels, https://eua.eu/downloads/publications/compendium%20of%20accounting%20practices_web.pdf (accessed on 31 January 2021).
[25] European Commission (2021), Horizon Europe - the EU Research and Innovation Programme 2021-27, European Commission, Brussels, https://ec.europa.eu/info/sites/default/files/research_and_innovation/funding/presentations/ec_rtd_he-investing-to-shape-our-future.pdf (accessed on 4 June 2021).
[48] European Commission (2018), Mutual Learning Exercise: Open Science — Altmetrics and Rewards, Publications Office of the European Union, Luxembourg, http://dx.doi.org/10.2777/22329.
[21] Eurostat (2021), GERD by sector of performance, https://ec.europa.eu/eurostat/web/main/data/database.
[52] Flemish Government (2021), FRIS Research Portal, https://researchportal.be/en/about-fris (accessed on 12 August 2021).
[1] Flemish Government (2020), Speurgids 2020 Ondernemen & Innoveren (Tracking Guide 2020 - Enterprise and Innovation), Departement Economie, Wetenschap en Innovatie (EWI), https://www.ewi-vlaanderen.be/sites/default/files/speurgids2020.pdf (accessed on 12 January 2021).
[14] Flemish Government (2020), Verslag over de financiële toestand en de evolutie van het personeelsbestand van het hoger onderwijs in 2019. Deel II - Universiteiten (Report on the financial situation and the evolution of staffing in higher education - Part 2 Universities), Regeringscommissariaat Hoger Onderwijs, Brussels, https://www.vlaanderen.be/publicaties/verslag-over-de-financiele-toestand-en-de-evolutie-van-het-personeelsbestand-van-het-hoger-onderwijs (accessed on 12 January 2021).
[37] Flemish Government (2020), Vlaamse Veerkracht: Relanceplan Vlaamse Regering (Flemish resilience: Flemish Government recovery plan), Flemish Government, Brussels, https://publicaties.vlaanderen.be/view-file/39939 (accessed on 27 May 2021).
[3] Flemish Government (2019), Besluit van de Vlaamse Regering tot wijziging van het besluit van de Vlaamse Regering van 21 december 2012 betreffende de financiering van de Bijzondere Onderzoeksfondsen aan de universiteiten in de Vlaamse Gemeenschap (Flemish Government Order on changes to the decision of the Flemish Government of 21 December 2012 concerning the funding of the Special Research Funds for the universities in the Flemish Community), https://www.ewi-vlaanderen.be/sites/default/files/bof_-_besluit_van_de_vlaams_regering.pdf (accessed on 4 June 2021).
[63] Flemish Government (2013), Codex Hoger Onderwijs (Higher Education Code), https://data-onderwijs.vlaanderen.be/edulex/document.aspx?docid=14650#1 (accessed on 28 May 2021).
[7] Flemish Government (2009), Besluit van de Vlaamse Regering betreffende de ondersteuning van de Industriële Onderzoeksfondsen en de interfaceactiviteiten van de associaties in de Vlaamse Gemeenschap (Flemish Government Order concerning support for the Industrial Research Funds and interface activities in the associations of the Flemish Community), https://codex.vlaanderen.be/PrintDocument.ashx?id=1018147 (accessed on 4 June 2021).
[24] Flemish Government (2009), Decreet betreffende de organisatie en financiering van het wetenschaps- en innovatiebeleid (Decree concerning the organisation and funding of science and innovation policy), https://codex.vlaanderen.be/Portals/Codex/documenten/1018085.html (accessed on 2 June 2021).
[50] Glänzel, W., P. Chi and K. Debackere (2020), “Measuring the Impact of Research – from Scholarly Communication to Broader Impact”, in Ball, R. (ed.), Handbook Bibliometrics, De Gruyter Saur, http://dx.doi.org/10.1515/9783110646610-015.
[12] HEA (2019), Financial Trend Analysis - Universities and IoTs 2007-2017, Higher Education Authority, Dublin.
[42] Hicks, D. et al. (2015), “Bibliometrics: The Leiden Manifesto for research metrics”, Nature, Vol. 520/7548, pp. 429-431, http://dx.doi.org/10.1038/520429a.
[58] Huijben, M. et al. (2011), Benchmarkonderzoek overhead universiteiten (Benchmarking study: overhead in universities), Berenschot.
[49] Imperial College London (2021), Altmetric, Imperial College London website, https://www.imperial.ac.uk/research-and-innovation/support-for-staff/scholarly-communication/bibliometrics/altmetric/ (accessed on 3 June 2021).
[47] Khodiyar, V., K. Rowlett and R. Lawrence (2014), “Altmetrics as a means of assessing scholarly output”, Learned Publishing, Vol. 27/September, pp. S25–S32.
[26] KNAW (2019), Evenwicht in het wetenschapssysteem - De verhouding tussen ongebonden en strategisch onderzoek (Balance in the science system - the relationship between researcher-driven and strategic research), Koninklijke Nederlandse Akademie van Wetenschappen, https://www.knaw.nl/nl/actueel/publicaties/evenwicht-in-het-wetenschapssysteem (accessed on 27 May 2021).
[2] Korlaar, L., F. Bongers and T. Groot Beumer (2018), Evaluatie van de Bijzondere Onderzoeksfondsen (BOF) (Evaluation of the Special Research Funds), Dialogic , Utrecht, https://www.ewi-vlaanderen.be/sites/default/files/bestanden/evaluatie_van_de_bijzondere_onderzoeksfondsen_-_managementsamenvatting.pdf (accessed on 12 January 2021).
[62] Merton, R. (1968), “The matthew effect in science”, Science, Vol. 159/3810, pp. 56-62, http://dx.doi.org/10.1126/science.159.3810.56.
[30] Monard, E. et al. (2018), Onderzoeker-gedreven wetenschap: Analyse van de situatie in Vlaanderen (Researcher-driven science: Analysis of the situation in Flanders), Koninklijke Vlaamse Academie van België voor Wetenschappen en Kunsten (KVAB), Brussels, https://www.kvab.be/sites/default/rest/blobs/2016/nw_onderzoekergedreven.pdf (accessed on 5 June 2021).
[55] Norton, A. (2015), The cash nexus : how teaching funds research in Australian universities, Grattan Institute, Carlton, Victoria, https://nla.gov.au/nla.obj-337774472/view (accessed on 5 June 2021).
[20] OECD (2021), Main Science and Technology Indicators, https://www.oecd.org/sti/msti.htm (accessed on 2 June 2021).
[34] OECD (2018), OECD Science, Technology and Innovation Outlook 2018: Adapting to Technological and Societal Disruption, OECD Publishing, Paris, https://dx.doi.org/10.1787/sti_in_outlook-2018-en.
[39] OECD (2018), Rethinking Quality Assurance for Higher Education in Brazil, Reviews of National Policies for Education, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264309050-en.
[27] OECD (2015), Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development, The Measurement of Scientific, Technological and Innovation Activities, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264239012-en.
[54] Olive, V. (2017), How much is too much? Cross-subsidies from teaching to research in British universities, http://www.hepi.ac.uk (accessed on 22 November 2019).
[16] Rathenau Instituut (2020), Ontwikkeling derde geldstroom en beïnvloeding van wetenschappelijk onderzoek (Development of the third funding stream and its influence on scientific research), https://www.rathenau.nl/nl/vitale-kennisecosystemen/ontwikkeling-derde-geldstroom-en-beinvloeding-van-wetenschappelijk (accessed on 21 April 2021).
[56] Rathenau Instituut (2018), Het onderzoek aan universiteiten en umc’s (Research at universities and university medical centres), https://www.rathenau.nl/nl/wetenschap-cijfers/geld/inkomsten-uitgaven-van-universiteiten-en-hogescholen/het-onderzoek-aan (accessed on 6 June 2021).
[9] Rekenhof (2018), Integratie van de academische hogeschoolopleidingen in de universiteiten (Integration of academic university college programmes into universities), Rekenhof, Brussels, https://docs.vlaamsparlement.be/pfile?id=1417483 (accessed on 12 January 2021).
[51] Researchfish (n.d.), Track research and evidence impact with Researchfish, https://researchfish.com/ (accessed on 5 June 2021).
[28] Ronai, I. and P. Griffiths (2019), “The Case for Basic Biological Research”, Trends in Molecular Medicine, Vol. 25/2, http://dx.doi.org/10.1016/j.molmed.2018.12.003.
[33] Sandström, U. and P. Van den Besselaar (2018), “Funding, evaluation, and the performance of national research systems”, Journal of Informetrics, Vol. 12/1, pp. 365-384, http://dx.doi.org/10.1016/j.joi.2018.01.007.
[61] Shanghai Ranking (2021), 2020 Academic Ranking of World Universities, Shanghai Ranking, https://www.shanghairanking.com/ (accessed on 12 August 2021).
[60] Times Higher Education (2021), World University Rankings, https://www.timeshighereducation.com/content/world-university-rankings (accessed on 12 August 2021).
[17] Universities Denmark (2021), Det Statistiske Beredskab - Danske Universiteter (University Statistics - Universities Denmark), https://dkuni.dk/tal-og-fakta/beredskab/ (accessed on 24 May 2021).
[29] Universities New Zealand – Te Pōkai Tara (2017), The links between research and teaching, Universities New Zealand – Te Pōkai Tara, https://www.universitiesnz.ac.nz/latest-news-and-publications/links-between-research-and-teaching (accessed on 5 June 2021).
[45] Van den Besselaar, P., U. Heyman and U. Sandström (2017), “Perverse effects of output-based research funding? Butler’s Australian case revisited”, Journal of Informetrics, Vol. 11/3, pp. 905-918, http://dx.doi.org/10.1016/j.joi.2017.05.016.
[38] Van der Beken, W. et al. (2018), Systeemevaluatie van de financiering van (fundamenteel) onderzoek (System evaluation of funding for (fundamental) research), IDEA Consult, Brussels.
[18] Vipunen (2021), University income statement, Vipunen - Finnish Education Statistics, https://vipunen.fi/fi-fi/yliopisto/Sivut/Talous.aspx (accessed on 24 May 2021).
[11] VLAIO (2020), Subsidie-oproep onderzoeksinfrastructuur hogescholen (Call on research infrastructure for university colleges), https://www.vlaio.be/nl/nieuws/subsidie-oproep-onderzoeksinfrastructuur-hogescholen (accessed on 4 June 2021).
[10] VLAIO (2019), Blikopener wil praktijkgericht onderzoek door hogescholen aanbieden aan kmo’s (“Can Opener” project seeks to provide university college research to SMEs), https://www.vlaio.be/nl/nieuws/blikopener-wil-praktijkgericht-onderzoek-door-hogescholen-aanbieden-aan-kmos (accessed on 4 June 2021).
[6] VLIR (2020), Eerste iBOF-projecten geselecteerd (First iBOF projects selected), Vlaamse Interuniversitaire Raad (VLIR), https://vlir.be/nieuws/eerste-ibof-projecten-geselecteerd/ (accessed on 10 August 2021).
[36] VLIR (2020), Over de impact van de Vlaamse universiteiten en hun rol als motor voor welvaart en welzijn (On the impact of Flemish universities and their role as motors for prosperity and well-being), Vlaamse Interuniversitaire Raad (VLIR), Brussels, https://docs.vlaamsparlement.be/pfile?id=1525519 (accessed on 5 June 2021).
[57] VSNU and NWO (2020), Integraal plan Verlagen druk op het wetenschapssysteem (Integrated plan on reducing pressure on the science system), Vereniging van Universiteiten (VSNU) and the Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO), https://www.rijksoverheid.nl/documenten/rapporten/2020/03/02/bijlage-verlagen-druk-op-het-wetenschapssysteem (accessed on 2 June 2021).
[32] Wahls, W. (2018), “The NIH must reduce disparities in funding to maximize its return on investments from taxpayers”, eLife, Vol. 7, http://dx.doi.org/10.7554/eLife.34965.
[5] Zacharewicz, T. et al. (2019), “Performance-based research funding in EU Member States—a comparative assessment”, Science and Public Policy, Vol. 46/1, pp. 105-115, http://dx.doi.org/10.1093/scipol/scy041.