Where did all the money go? Funding, personnel and expenditure in Swedish universities and colleges 2001–21

Abstract

Swedish universities and colleges have received a substantial funding increase since the turn of the millennium, as part of continued policies of expanding the admission of students to higher education to broader layers of the population and strengthen Swedish public research and development to increase the competitiveness of the Swedish knowledge-based economy. In this article, publicly available statistics are used to trace how this increase in funding has been used by the sector. Comparing figures on income (base grant for research, third-party funding and base grant for education) with statistics on personnel and student enrolment as well as data on actual expenditure, the article draws some conclusions that are used to discuss some common misunderstandings and erroneous beliefs, including claims of a ‘depletion’ of the base grant for research and an uninhibited growth of the number of administrative staff, which are common themes in the Swedish and international debate over higher education.

Keywords:

• Funding
• research
• growth
• administration
• universities
• Sweden

Introduction

The Swedish university system has grown dramatically in the past two decades. Both student and staff numbers have risen by more than 30% and the overall income has grown by almost 70% (adjusted for inflation). Proper analysis of the system-level consequences of this expansion is likely to add important nuance to some controversial debates, including the popular assumptions that the base grant funding for research is in long-term decline in favour of third-party funding in Sweden, that the cadres of administrators at Swedish universities are growing unceasingly at the expense of the teaching and research staff and that temporary employments are more common and permanent (tenured) positions are in decline. Similar claims are also often heard internationally and are usually uttered as part of broader debates over alleged economisation, marketisation, bureaucratisation and managerialist reform of universities, as well as system-wide stratification and resource concentration to élite institutions and pressing and conflicting demands on universities to be engines of growth in the globalised knowledge economy (Engwall & Nybom, 2007; Teelken, 2012; Hallonsten & Silander, 2012; Rider et al., 2013; Ek et al., 2013; Stensaker & Benner, 2013; Holmberg & Hallonsten, 2015; Alvesson & Spicer, 2016; Karlsson & Ryttberg, 2016; Silander & Haake, 2017; Ekman et al., 2018).

Are such claims really true? Did the resource increase in the Swedish university system fund the expansion of the ‘all-administrative university’ (Ginsberg, 2012) or was it all allocated in the form of temporary competitive grants, with strings attached, thus undermining academic integrity and long-term quality of teaching and research? The purpose of this article is to add some (comparably) hard facts to these debates and thus attempt to dispose of some beliefs as myths or misconceptions. This is done with system-wide, publicly available figures on income (base grant for research, third-party funding and base grant for education), statistics on personnel and student enrolment, data on actual expenditure and historical data on salary levels. Conclusions therefore remain on the overall level but nonetheless show some irrefutable patterns. Most of the money went to hiring more permanent teaching and research staff, increasing their salaries and recruiting more students. Administrative staff certainly increased in numbers but not as much as often claimed. Moreover, while third-party funding increased its relative share of university income for research, the governmental base grant for research also grew substantially.

The next section contains a brief historical background to the structure and condition of the Swedish academic system, based on secondary sources. Thereafter, some notes are made on method and the data used in the analysis, followed by a straightforward presentation of the data used. In the second half of the article, the data is analysed and some key conclusions drawn. On the basis of this, in the final section, the results and their explanation are discussed in combination with a broader conceptual contextualisation and some suggestions for future research.

Background

Sweden has a historically strong research system by international comparison, with total annual R&D expenditures above 3% of GDP, which is the highest in the European Union. The system is strongly dominated by the private sector, whose share of the total annual research and development (R&D) expenditures is around 70%, and the universities and colleges, whose share is roughly 25%. This is unusual in international comparison since Sweden never built up any noticeable sector of governmental research institutes to complement universities on the R&D performer side (Hallonsten, 2018). Meanwhile, Swedish higher education has been expanded dramatically in the past century, with more than 40% of each age cohort entering tertiary education today. The current research and higher education system was shaped by the astonishing economic growth in the post-World War II period and the ambitious overall project of building a Swedish universal welfare state and a democratic and egalitarian society, which placed great public faith in the abilities of (higher) education, scientific research and technological development to improve society. Meanwhile, comparisons with other countries and other policy areas yield that Sweden has limited centralisation of authority and rather weak coordination between actors and processes in the area of research policy (Benner & Sörlin, 2007; Engwall & Nybom, 2007; Holmberg & Hallonsten, 2015; Hallonsten, 2020, ch 4).

Before World War II, Sweden hardly had a national research or higher education policy at all. But the post-war expansion of the Swedish welfare state meant significant mobilisation in both academic research and higher education, in order to supply Swedish industry with a qualified labour force, to give the working class access to higher education and to further progress built on science and technology (Elzinga, 1993). Estimations are that the total number of students enrolled in higher education in Sweden increased 30-fold between 1940 and 2012, in a continuous growth with two especially significant waves of expansion in the 1960s and 1990s (Holmberg & Hallonsten, 2015). With time, this expansion necessitated the founding of new universities and colleges, with the 1977 reform bringing the largest number of new organisations to the mix, as previous vocational schools and branch campuses of the universities were converted into regional colleges without university status. This reform also made Swedish higher education a uniform system with central authority (although the old universities mostly kept their privileges (Askling, 1989)) and separated the governmental base grants for education and research. The latter allegedly weakened the links between education and research and, in the decades to come, the system evolved and diversified further (Holmberg & Hallonsten, 2015). The six old and full-breadth universities are today complemented by six specialised (medical, technical, agricultural) colleges with university status, five comparably small and regional universities and eleven regional colleges without university privileges.

Today, therefore, the landscape of higher education in Sweden is essentially a two-tier system of the type common in many European countries, with a smaller number of old universities and an additional group of vocational colleges turned regional colleges (Askling, 1989; Sköldberg, 1991; Kyvik, 2004; Holmberg & Hallonsten, 2015). The privileges of the twelve old universities and specialised schools with university status are seen especially in research, where they absorb 90% of the total available funding (both base grant and third-party funding), with the sixteen colleges and newly created universities sharing the rest. In education, the pattern is less clear, as the twelve first-tier universities enrol 58% of the students (autumn term of 2019).

Until the 1990s, there was very little governmental initiative and strategic prioritisation in the funding for research in Swedish universities. Resource allocation was still done mostly on basis of tradition, with power residing with the university faculties and the chairholder professors, and little or no room for central university management to make any priorities of their own. Third-party funding from the research councils and private foundations was mostly allocated through grants awarded to applicants in open calls within classic subject areas (natural science, medicine, humanities, social science) (Engwall & Nybom, 2007; Holmberg & Hallonsten, 2015).

The 1990s saw several important changes. From 1991 to 1994, five public research foundations were created as a deliberate attempt to introduce ‘strategic’ research funding and break previous path-dependencies in research and make universities more ‘entrepreneurial’ and internationally competitive (Benner & Sörlin, 2007). In 1993, a comprehensive university reform decentralised decision-making and made part of the base grant for education contingent upon throughput of students and strengthened the regional colleges by giving them access to base grant funding for research and the right to appoint professors (Engwall & Nybom, 2007; Holmberg & Hallonsten, 2015). At the end of the decade, the opportunity was opened for colleges to apply to the government for university status and, in 1999, three regional colleges were upgraded to universities by decision of the government, followed by an additional two in 2004 and 2016. In 2001, the research councils were restructured in order to achieve better opportunities for strategic prioritisation and mobilisation in areas of particular importance and, since then, the share of governmental funding channelled through the research councils for specific purposes (including excellence funding, strategic funding and field-specific programmes) has increased dramatically, mostly at the expense of the traditional project grant funding awarded through open calls (Hallonsten, 2020, ch 4).

In international comparison, therefore, Swedish research policy is heterogeneous and weakly coordinated, with policy formulation delegated to several authorities and resources spread over many areas. Occasionally, local initiatives have managed to grow into internationally competitive positions, with some support from specialised funding bodies and benign policymakers and bureaucrats (Benner & Sandström, 2000; Gribbe & Hallonsten, 2017; Hallonsten & Christensson, 2017) but the strength of the system has historically been able to maintain breadth. However, since the beginning of the 2000s, policies have shifted and the government is now more clearly aiming to mobilise strategically in specific areas of strength, so that Sweden can further develop the competitiveness of is knowledge-based economy and meet the growing challenges of globalisation. The result has been clear: first, there is more central steering and strategic initiative on the level of the government and its ministry for higher education and research (Hallonsten, 2020, ch 4), seen in general increases of the base grant for research to the universities and the allocation to the research councils. Second, several billion SEK have been earmarked to specific areas (mostly in medicine, engineering sciences and research for sustainable development), competitive ‘excellence’ funding (Hallonsten & Silander, 2012) and efforts to stimulate more ‘collaboration’ between academia and broader society (Broström et al., 2019). The annual increases of funding to Swedish universities and colleges has been unceasing regardless of political majority in the Swedish parliament.

Although a 2010 reform to the governance of universities and colleges formally took power over the distribution of resources between different scientific areas, faculties and departments out of the hands of the government and gave it to central university management, resources are still, to a great extent, distributed on the basis of tradition and path-dependence (Holmberg & Hallonsten, 2015). The base grant for research, by which is meant the unfettered funding allocated to universities in the annual governmental budget, and doctoral education, amounted to 21.1 billion SEK in total in 2021 and constituted 44.4% of the total funding for research and doctoral education in Swedish universities and colleges that year. Third-party grant funding amounted to 22.3 billion SEK (46.9%). The remaining share, 4.1 billion SEK (or 8.7%), comprised of fees, financial income and income from commissioned research.

The landscape of third-party funders is broad and varied, with several research councils, public and private foundations, non-council governmental agencies with their own research funding activities and foreign and domestic businesses and EU funding. The largest single funder is the Swedish Research Council, with 22.9% of the total third-party funding in universities and colleges, followed by the private Wallenberg foundations (8.8%) and the EU framework programmes (5.9%). All universities and colleges in Sweden are organised as governmental agencies, except three which are run as independent foundations, but all 28 have the absolute most part of their funding from the base grant for research (25.6% of their total income), third-party funding for research (27.0%) and the base grant for education (35.3%).

Method and data

The data used in this article comes from three main sources. First, all data on finances, personnel and students come from the statistics database of the Swedish Higher Education Authority. In the results and analysis sections below, all data reported comes from this source unless otherwise stated. Second, some financial data has also been extracted from the annual reports of the Swedish publicly owned real estate company Akademiska Hus, which has a majority share of the market for campus buildings at Swedish universities and to which universities and colleges pay rents. Third, official salary statistics for academics provided annually by the labour union the Swedish Association for University Teachers and Researchers in their membership magazine, have also been retrieved from the Lund University Library for the years studied. There are gaps in this material, as personnel categories have changed over the years, and the statistics only contain data on (some) teaching and research staff. Complete time series of average salary levels of professors and senior lecturers, the two key academic employment categories, were however possible to obtain for the whole period studied. Average salary levels for doctoral students were possible to obtain from 2004 onwards. In addition, some complementary data have been retrieved from the Eurostat database and the public Swedish statistics agency Statistics Sweden.

The time period studied, 2001–21, was chosen on the basis of the availability of data, especially from the statistics database of the Swedish Higher Education Authority, which contains major gaps for years before 2001. All figures on income and expenditure of the universities were adjusted for inflation and thus converted to 2021 prices. The numbers of registered students are reported for semesters in the database used, which means that the annual numbers reported below were obtained by calculating an average of the enrolment in the spring and fall semesters of each year.

Results

As already noted in the introduction, the total income of Swedish universities increased from 48.9 billion SEK to 82.6 billion SEK, or 68.8%, between 2001 and 2021 (note, as a reminder, that these figures and all similar figures below have been adjusted for inflation and are given in 2021 prices). This can be compared with the overall productivity growth of the Swedish economy, which was 69.2% in the twenty years studied (also inflation-adjusted). The growth of the total income of Swedish universities was rather even over this period, with some hikes here and there, especially in the research funding, but with a steady growth only interrupted by a very minor decrease in the overall funding in 2007 (−0.5%), when the base grant for education fell by 4%. Otherwise, all the three major income sources (base grant for education, base grant for research and third-party research funding) increased in all years, with few and minor exceptions.

The base grant for research increased by 8.82 billion SEK, or 71.7%, in the period studied. Meanwhile, third party funding for research increased by 11.2 billion SEK or 100.2%, which means that the share of base grant funding of the total income for research fell from 46.7% to 44.4%. This was, however, not a uniform development: between 2003 and 2008, the share increased as third-party funding declined somewhat and in the years 2009–10 a major general increase in research funding (see below) gave a 20.6% increase in third-party funding in 2009 and a 12.8% increase in the base grant for research in 2010, which makes the curve deviate from its overall path of increased share of third-party funding for research in the universities.

The base grant for education likewise grew dramatically over the period, by 11.7 billion SEK or 66.9%. The growth was most marked in the first years of the millennium, with a growth of almost 20% in 2002 and almost 5% in 2003. The years thereafter were followed by small increases or small decreases in 2004–08 (neither ever more than 4%), two years of clear growth (8.3% and 6.5%, respectively, in 2009 and 2010) and then more or less only minor oscillations around zero growth in the ten years thereafter. These financial developments are summarised in Figure 1.

Figure 1. Total income of Swedish universities and colleges, 2001–21, divided on major income sources, billion SEK. Note that all numbers have been adjusted for inflation.

Looking at actual expenditure, the three largest items are personnel, rents and other operating costs, which in total make up around 90% of the total expenditure. The total personnel costs increased by 85.4% in 2001–21, whereas the other two major expenditures, namely rents and other operating costs, increased by 40.5% and 34.9%, respectively. The relative shares of these cost categories did not fluctuate much in the period; other operating costs oscillated between 14% and 18% of the total in the whole period, whereas total personnel costs grew from 57.9% to 65.9% of the total and costs for rents decreased its share of the total, from 14.3% in 2001 to 12.3% in 2021.

Not surprisingly, given the growth in income and the growth of personnel costs, the total number of employees (full-time equivalents, FTE) grew by 32% between 2001 and 2021. For teaching and research personnel, the period brought an even steeper growth, especially in the categories professors, senior lecturers, so called ‘career development positions’ (which includes postdocs and associate lecturers) and other personnel with teaching and research duties (mainly researchers and research assistants). Together, these categories of personnel grew by 11,882.6 FTEs or 77%, and especially the career development positions grew significantly (2644.5 FTE or 265.9%). The number of professors and senior lecturers grew by 63.5% and 73.0% respectively, but other categories decreased in numbers, namely lecturers without doctoral degree (−1326.7 FTE or −20.5%) and temporary teachers (paid with fees) (−1012.5 FTE or −65.1%). The various categories of administrative personnel experienced different types of changes. The category administrative staff increased by 5038.1 FTE in the period, or 58.5%, whereas library personnel decreased by 310.3 FTE or 21.5% and technical staff decreased by 902.6 FTE or 11.9%.

In the same period, the salaries of professors and senior lecturers increased by 32.4% and 29.5%, respectively, and in the years 2004–21, the salaries of doctoral students increased by 22.6%. The number of registered students increased significantly between 2001 and 2021, from 283,555.0 to 374,374.5, or 32%, in a development that was not entirely even but varied somewhat, largely following the fluctuations and overall growth of the base grant for education (see above). There was significant growth in 2003–04, some decline in 2005–08, another major growth in 2009–10 and, after that, a net decrease of 8204.5 students in 2010–19 and a subsequent growth with over 30,000 registered students in the final two years of the period. The data presented in this section is summarised in Table 1.

Table 1. Summary of the reported data

	2001	2021	Change
Total income (billion SEK)	48.9	82.6	68.8%
Base grant for research (billion SEK)	12.3	21.1	71.7%
Third party funding for research (billion SEK)	11.1	22.3	100.2%
Base grant for education (billion SEK)	17.5	29.1	66.9%
Total expenditure (billion SEK)	49.2	80.3	63.0%
Personnel costs (billion SEK)	28.5	52.9	85.4%
Rents (billion SEK)	7.0	9.9	40.5%
Other operating costs (billion SEK)	8.7	11.8	34.9%
Total number of employees (FTE)	49,220.0	64,951.2	32.0%
Professors (FTE)	3268.8	5343.0	63.5%
Senior lecturers (FTE)	5715.3	9889.8	73.0%
Doctoral students (FTE)	8166.2	10,528.8	28.9%
Administrative personnel (FTE)	8610.7	13,648.8	58.5%
Average salary professor (SEK)	52,223.0	69,119.0	32.4%
Average salary senior lecturer (SEK)	39,073.0	50,585.0	29.5%
Number of registered students (FTE)	283,555.0	374,374.5	32.0%
Swedish GDP (billion SEK)	32,225.0	54,570.0	69.2%

Note that all numbers have been adjusted for inflation.

Analysis

It was established as a starting point for this article that the Swedish academic sector grew dramatically in volume between 2001 and 2021, due to significant increases of all three major sources of income, namely base grant for research, third-party funding for research and base grant for education. Comparing the data on the vast resource increase in the two decades studied with the other indicators reported above renders important and interesting inferences that can help answer the research question: where did all the money go?

Concerning education, the base grant grew by over 11 billion SEK, or 66.9%, whereas student numbers grew by only 32%. This means, among other things, that the average cost (to the national education budget) per student increased. In 2001, it was 78,201 SEK, whereas in 2021 it was 92,054 SEK (using a simple ratio of total education income divided by the total number of enrolled students), an increase in the average cost per enrolled student of 17.7%. The figures should of course only be used as very rough estimates of a long-term trend, since there are significant fluctuations during the time period studied, and not least also because the reported data do not take into account differences between fields and levels of education (basic and advanced). Nonetheless, there is an evident discrepancy between the growth in the gross income for education and the growth in the number of students enrolled. This discrepancy could, conceivably, be explained by an increase in the quality rather than quantity of higher education overall in Sweden, should a reliable measure of this be available. Which it is not, at least not in the context of the present study.

Concerning research, it shall foremost be noted that the base grant for research and the third-party funding for research have grown at different rates, which has led to a slight tilt in the balance of research income in favour of third-party funding. In 2001, the base grant for research made up 46.7% of the total income for research and in 2021, it was 44.4%. However, this overall relative decline in base grant funding is minuscule in comparison with the overall growth on all accounts, namely a near 70% resource increase across the board. Moreover, since the base grant for research increased even more than the overall income (71.7% compared to 68.8%), this means that the research base grant’s share of the total income to the universities and colleges increased somewhat in the period. Previous studies suggest that the distribution of the increases of both base grant for research and third-party funding for research across the system is strongly unequal and favours the larger and older universities (Stensaker & Benner, 2013; Holmberg & Hallonsten, 2015) but follow-up analyses on this level of detail will have to commence to bring clarity to these issues.

It is interesting to note that the most accentuated growth pattern in the expenditure of the universities and colleges in the period studied is in personnel costs, which grew by 85.4% in 2001–21 and thus increased its share of the total expenditure from 57.9% to 65.9%. Costs for rents also grew (by 40.5%) but decreased its share of the total, from 14.3% to 12.3%, as discussed further below.

The growth in personnel costs has several possible explanations.

First, the overall personnel numbers increased by 32.0% in the period, which is significantly lower than the overall growth in funding (68.8%) and the overall growth in personnel costs (85.4%). This means that counting all personnel categories, the average cost of an FTE university employee was significantly higher in 2021 than 2001, namely 815 kSEK compared to 579 kSEK, a 40.5% increase. For comparison, taken together, rent costs and other operating costs per FTE university employee increased by only 4.1%.

As noted above, different personnel categories increased and decreased quite significantly in number in these years. The FTE of all teaching and research personnel (not including doctoral students) increased by 40.7% and FTE doctoral students by 28.9%. Meanwhile, FTEs of administrative, technical and library personnel together increased by 21.7% and the number of FTE administrative staff separately increased by 58.5%. It is likely that those categories of academic staff that increased in number, namely professors, senior lecturers, career development positions, and ‘other teaching and research personnel’, are more costly than those categories that decreased in number, namely lecturers without doctoral degree and temporary teaching staff, which could account for some of the increased personnel costs overall. Another factor for this development could be the significant increase in the salaries of professors and senior lecturers, 32.4% and 29.5% respectively, as shown in the annual statistics provided by the labour union the Swedish Association for University Teachers and Researchers, in their membership magazine. The increase in number of FTE doctoral students (28.9%) is perhaps of minor importance in explaining the increased personnel costs but their increased salary level (2004–21), by 22.6%, is likely more consequential. In the categories of non-academic staff, no similar conclusions can be drawn, as there is no obvious causal link between on one hand the growth in administrative staff and decrease of library and technical staff, and the seeming growth in cost of personnel; moreover, data on the salary increases of these employee categories is not available.

Three things in the material seem to suggest that the overall resource increase in the years 2001–21 most of all has financed academic positions: first, the growth of personnel costs as share of total expenditure; second, the increase in numbers of professors, senior lecturers and doctoral students; and third, the increased salary levels of all three categories. In connection therewith, it is interesting to note that while the category of administrative staff has grown by 5038.1 FTE or 58.5% in the period studied, the numbers of library staff and technical staff have decreased (by 310.3 and 902.6 FTE, or 21.5% and 11.9%, respectively), which means that the net growth of non-teaching and non-research personnel in the period was only 21.7%.

Of particular interest is also the costs for rents. This expenditure increased by 40.5% in 2001–21, which is significantly less than the increase in personnel costs, which also means that rents as share of total expenditure decreased from 14.3% in 2001 to 12.3% in 2021. A majority of the buildings occupied by universities and colleges in Sweden are owned by the state through its real estate company Akademiska Hus. The annual reports of this company reveals that its market share of university and college buildings in Sweden was 66.5% in 2021 (calculated as the ratio between its total gross rent incomes and the total rent expenditures of the universities and colleges). As a governmentally owned company, Akademiska Hus is obliged to pay annual dividends to the government. Over the years 2001–21, these dividends averaged 1.5 billion SEK per year, which constitutes an average of 2.3% of the annual total turnover of the universities and colleges. Although a minor sum in this context, it is worth noting that this arrangement means that the government every year takes back a few per cent of its funding to the universities and colleges through its real estate company. This share has increased in the years studied, from 0.5% in 2001 to 2.7% in 2021, with one especially exceptional year, namely 2015, when no less than 11.9% of the total turnover of the universities and colleges was channelled back to the governmental budget through a shareholder’s dividend of 8.7 billion SEK (in 2021 prices).

Where the money went

The starting point for the analysis in this article was the simple identification of a significant increase in the total funding for Swedish higher education and academic research in the past two decades. It is in itself remarkable. To reiterate; the base grants for research and education have both increased by around 70% and the total third-party funding has more than doubled. The first major task of the discussion and conclusions of this article will therefore have to be to explain, on the basis of available historical evidence and conceptual tools from the Swedish and international study of academia, why this happened. Similar explanations of the other findings of the article will then follow and hopefully put the findings in proper perspective and secure their relevance for a broader audience.

The major resource increase most likely has a multitude of causes. Leaving the purely political causes aside for a brief moment, it is interesting to note that funding increases of the sort reported in this article have, in a sense, been predicted by scholars studying other university systems, in other time periods, and making general inferences on the nature of university teaching and research as economic activities. Studies have, for example, suggested that university teaching and research have neither the means, nor the incentives, to increase efficiency in the way that the production and service industries do and therefore normally cannot be expected to share in the productivity gains of the rest of the economy (Bowen, 1967; Ehrenberg, 2002). Expansion of universities, allegedly imperative in their organisational logic (Greenberg, 2007), and salary raises, expected by university employees and demanded by their labour unions, need therefore be paid by increased income. Elsewhere, most conspicuously in the United States, tuition fees are therefore growing at a significantly higher rate than consumer prices, whereas in Sweden, as has been shown, the governmental base grants (and third-party funding) have grown dramatically. In Sweden, the overall growth in income of the universities in 2001–21 (68.8%), is on par with the overall GDP growth in the same period (69.2%), which confirms this thesis. The resource allocation system has, in a sense, adjusted to the productivity growth in the broader economy and expanded teaching and research in the universities accordingly.

This has, most of all, been accomplished by political initiative. The review of the research policy developments in the background section above pointed at a general shift in governmental research policy around the turn of the millennium, that upgraded research (and, to some extent, higher education) to a new and strengthened role in the long-term securing of the international competitiveness of the Swedish economy. Due both to globalisation and increased competition from abroad, and the increasing influence of ‘knowledge economy’ discourse and adjoining theories of the role of knowledge and technology for economic competitiveness, the shift is seen rather clearly in the language of the four consecutive Swedish governmental research policy bills of 2000, 2005, 2009 and 2012, which all, with increasing intensity, emphasised the need for strategic mobilisation in research, both across the board and in specific areas of highest priority. Here, the government turned political rhetoric into substantial reinforcements. The excellence funding initiatives from 2009 and on meant general increases of both base grant funding and competitive funding (through the research councils) and seem also to have created a precedent where every government wants to make a mark and launch their own ‘strategic’ funding programmes, be it centres of excellence (Hallonsten & Silander, 2012) or the cultivation of external relations (Broström et al., 2019); regardless of their specific foci, funding has increased as a result of these policies. It has also been argued that the ‘excellence’ funding rhetoric has spread among the actors in the very pluralist and heterogeneous landscape of Swedish research funding bodies and produced isomorphic change (Hallonsten & Hugander, 2014) that has perhaps contributed to the overall growth in funding. On the education side, the political ambition of expansion of higher education in Sweden seems to have remained vivid well into the twenty-first century, which probably explains some of the continued growth of the base grant for education. In addition, the expansion of the system as such, especially the upgrades of two colleges to universities in the time period studied, can probably also account for some of the growth.

It is popular to make claims that the financial situation for Swedish universities has worsened in recent decades, although such claims seldom pass peer review and reach publications and therefore are hard to point at in proper citations. The reasons are, given the above, quite obvious. We have no data about the alleged depletion of resources in the 1980s and 1990s which is said to have led to a ‘sharp reduction in research funding provided directly to the universities by the State’ (Engwall & Nybom, 2007, p. 41). However, on the basis of the data analysed in this article, it can be concluded that in the two most recent decades, any such development has been squarely reversed. It is of course true that ‘competitive funding schemes gradually have gained popularity at the expense of institutional block grants’ in Sweden (Silander & Haake, 2017, p. 2010), although only if adding that the ‘institutional block grants’ (by which is meant the same as base grant for research in this article) have decreased slightly in relative terms but increased dramatically in absolute terms. It is hard to judge whether the block grants are indeed ‘modest’ in Sweden (Håkansta & Jacob, 2016, p. 6), since this depends on the frame of reference. If the comparison is between 2021 and 2001, the base grant funding for research in Swedish universities and colleges is far from modest today.

However, this statement is of course only true on the very general level and taking into account no additional considerations. The alleged depletion of base grant funding for research might be possible to prove with qualitative investigations of the internal governance structures at universities and colleges, unrealistic demands on them, and the suggestions that base grant funding is used for a lot of things that isn’t curiosity-driven or free research but rather education, administration and co-financing of research grants for specifically defined purposes (Marton, 2005; Granberg & Jacobsson, 2006; Håkansta & Jacob, 2016). Likewise, comparisons between universities and colleges seem to suggest a very uneven distribution of funds between them (Holmberg & Hallonsten, 2015) and comparisons between different fields may yield similar results that could warrant the conclusion that in some fields, base grant funding has indeed been shrinking. There is also a point to be made regarding the practical consequences of ‘excellence’ funding to specific areas or with other strings attached, such as building external relations, mobilising in certain cross-disciplinary areas, or supporting promising early-career researchers. Such conditioned funding may increase overhead costs and thus contribute to a relative depletion of general funding for research (or education), although ‘depletion’ is arguably too strong a word to use in this article altogether, given that it demonstrates a long-term, overall growth of funding across the board.

Future research should preferably look into these and other additional angles on the topic, such as internal use of funding at the universities and colleges, salary levels of administrators, changes in services provided by the support functions in universities and colleges due to the decrease in the number of library and technical personnel, possible increases in overhead costs due to earmarked funding and not least qualitative investigations of the often discussed tendency of teachers and researchers at universities spending more time on administration and grant applications and less on actual teaching and research.

Here, it suffices to conclude that while it is clear that the relative share of base grant funding for research has decreased in the past two decades, in favour of a larger share of competitive funding, both have increased dramatically (albeit unequally) in real terms. As already noted several times, the 71.7% increase of the base grant for research is difficult, to say the least, to interpret as a depletion or anything similar. Similarly, the increase of the base grant for education, of 66.9%, should be viewed in appropriately positive terms, perhaps especially also given the slightly more modest increase in the total number of students during this period, which means that the available money per student has grown considerably.

Another popular notion today, in Sweden and abroad, is that academia is being taken over by administrators. Looking only at the net growth in the number of administrative staff in Swedish universities and colleges, which was close to 60% in the time period studied, it might be possible to confirm this suspicion. Classical works in the social sciences have argued that growing administration is an almost unstoppable force of modern organisational life (Weber [1905], 2003; Parkinson 1957; Habermas 1984) and more recent sociological analyses have claimed that current society indeed experiences the ‘era of total bureaucratisation’ that not only has to do with the expansion of the public sector of the economy but also ‘the iron law of liberalism’, namely that any governmental initiative to reduce red tape and promote market forces will unavoidably lead to an increase in the total number of regulations and an increase in the total number of administrative staff (bureaucrats) to maintain them and secure compliance (Graeber, 2015, pp. 9–18). This would perhaps also provide an explanation as to why it seems the numbers of administrative staff at universities and colleges in Sweden have grown uninterruptedly regardless of the ruling policy doctrine (regulation or deregulation) and regardless of the parties in power. Specifically concerning universities, the claims of growing administrative control have been a topic of debate at least since the 1990s (Readings, 1996; Magala, 2009; Ginsberg, 2012; Collini, 2012; Alvesson & Spicer, 2016; Graeber, 2018) and there are many explanations available for why this development has occurred. Readings (1996, p. 8) pointed to the influence of the French-American historian Jacques Barzun in the late 1960s, who called for universities to build up an administrative capacity, populated by non-academic bureaucrats, to exercise proper civil service authority and thus keep the organisations and their operations in check (Barzun, 1969). A parallel and perhaps not unrelated process is connected to globalisation, which took off for real in the 1990s and created what several analysts have identified as an international market of students, researchers and funding (Altbach & McGill Peterson, 2007; Wildavsky, 2010), in whose traces a major international and local effort of evaluation and rankings has been built up (Hazelkorn, 2011; Hallonsten, 2021), sufficiently administratively demanding to warrant the claim that this has created a ‘Frankenstein monster’ of evaluation (Martin, 2011). Needless to say, this development should be visible in the changes in volume of university administrations, along with the rise of managerialism and economisation of higher education and academic research demonstrated in other recent works (Berman, 2012; Münch, 2014). Put differently, the new role of universities as ‘strategic actors’ in today’s society and economy (Krücken & Meier, 2006) seems to have transformed the role of the university administrator, from being a performer of pure support functions to vast strategic and managerial responsibilities. Although this development has been seen as a threat to the collegial governance of universities, it is not necessarily so that its most evident or important manifestation is in (growing) numbers of administrators but perhaps instead the qualitative change in their responsibilities and role (Karlsson & Ryttberg, 2016).

As the data and analysis in this article shows, the picture is in need of some nuancing and deeper study beyond pure numbers and sweeping claims. The growth in administrative staff at Swedish universities and colleges has been significant in the past two decades but in comparison with the vast net growth in FTE academic positions (professors, senior lecturers, doctoral students, career development positions, other teaching and research personnel), the growth is not so overwhelming. Both categories of tenured staff (professors and senior lecturers) have grown at higher rates than administrators and so at least in pure numbers it is possible to refute the claims that universities are taken over by administrators at the expenses of academic staff. Although these claims are often backed by numbers that show that in some universities in the United States and the United Kingdom administrative staff even nowadays outnumber teachers and researchers (Ginsberg, 2012; Alvesson & Spicer, 2016), there has so far been no real evidence, aside from the autoethnographic and anecdotal, to support the claims in the Swedish case. Up to now: as the numbers show, professors and senior lecturers make up 23.5% of the total FTE workforce of Swedish universities and colleges in 2021 (a larger share than 2001, when it was 18.3%) and administrative staff 21.0% (compared to 17.5% in 2001).

Needless to say, the numbers prove nothing about the actual power relationships inside the organisations of universities and colleges but merely report on the relative sizes of personnel categories. As with all statistics, there are most likely errors and hidden features. Issues naturally not covered by the analysis in this article, but nonetheless very suitable for further research, is possible shifts in the overall workload of academics, academic leaders and other staff at universities. It has, for example, been claimed that academics nowadays do more administrative work (Alvesson & Spicer, 2016, p. 35); this assumption should be tested. The role of the administrator has been analysed in depth and with specific attention to the Swedish case (Karlsson & Ryttberg, 2016); further similar studies would be very useful. Another topic that requires qualitative analysis of some depth is the correlation and possible causal relationships between the growth of administrative staff and the parallel decrease in library personnel and technical personnel, shown clearly by the numbers reported in this article, and whether this can be a sign of a takeover of universities by administrators, not at the expense of the professoriate but perhaps at the expense of other important support functions.

A popular point of critique among Swedish academics is the housing arrangements for Swedish universities and colleges. The governmentally owned limited liability company Akademiska Hus owns close to two-thirds of all the buildings they use and their required annual dividends to the government effectively means that a certain share of the public funding to universities and colleges is channelled back to the government each year; an average of 2.3% of their total annual turnover, as noted in a previous section. This is itself problematic and cause for some critique but, in the context of this article, the main issue is the rents’ share of the total expenditure, which has decreased from 14.3% to 12.3% in the years studied. This seems to be contrary to common beliefs and should be put in proper perspective, namely comparison with other expenditure categories.

Rents and other operating costs have increased rather modestly (40.5% and 34.9%, respectively), compared to personnel costs which have increased by no less than 85.4%. As already noted above, it is the academic staff and their salaries that has grown the most: the 32.4% increase in the average salary of professors and the 29.5% increase in the average salary of senior lecturers, are figures adjusted for inflation. This is, in itself, remarkable. One would perhaps expect that, in an expanding system, salaries would have a more modest rise but the data suggests otherwise and here, explanations are multiple and complementary.

It was mentioned in the background section above that weak coordination is a key feature of the Swedish research policy system. The governmental allocation of base grant funding follows largely a political logic, whereas the availability of external grants, also partly political, depends on a much more heterogeneous set of factors, including individual interest and skill in applying for such funding, and the agendas and procedures of both domestic and international, and public and private, funders. Importantly, these processes are almost completely detached from the key mechanism for salary-setting in Sweden, namely the central negotiations between unions and employers (or employer associations) that takes place entirely separately from legislative processes. Labour unions are historically strong in Sweden and while they had once a voluntary restraint on salary increases due to collective solidarity, they later evolved into more autonomous and self-serving organisations catering to the special interests of their members (Steinmo, 2010; Svensson, 2016). This means that there is a very limited political or bureaucratic connection between the salary increases, driven by the unions, and the increase in block grants and external grants, (partly) under the control of the government, and that the gate is open for the unions organising Swedish academics to negotiate higher salaries without deeper concern for the funding available. Which they seem to have done.

To conclude, the article has presented several interesting partial answers to the research question ‘Where did all the money go?’ and also provided explanations and a thorough discussion that both assists the reader in interpreting the results and positions the study and its results in a broader context. While the article does not specifically concern the highly problematic developments in the governance, organisation and funding of academic research and teaching in Sweden and elsewhere, and does not make an explicit contribution to the lively and important debate over these issues, it contributes indirectly. The article was written with the conviction that any serious debate and scholarly investigation of the changing governance of universities and colleges, and the harmful effects of these changes, must be done with utmost attention paid to basic facts of the matter, so that problem descriptions and solutions can be drafted and discussed without being clouded by misunderstandings and erroneous beliefs.

Some basic facts have surely and evidently been provided. First, the overall resource increase documented in this article should be acknowledged and used as a counterpoint to the common lamenting among academics. In addition, it has been shown rather clearly that there is no nominal depletion of the base grant funding in Sweden in the past 20 years. The base grant for research increased more than overall income (71.7%), which means that the research base grant’s share of the total income to the universities and colleges also increased. This means that there is no basis for any claim that the base grant for research has decreased in a general or overall sense. There is also no taking over of universities and colleges by administrators, at least not in a numerary sense. The vast increase in funding for research and education in the past 20 years has, most of all, been used to hire more (permanent) teaching and research staff and to increase their salaries. That’s where not all, but most of, the money went.

Disclosure statement

No potential conflict of interest was reported by the author(s).