From Dismantling the Class Society to Investing in Human Capital: The Rise and Fall of the Selective Student Finance System in Sweden 1939–1964

The article highlights the history of the early gift-based and selective student finance system of the social democratic welfare state in Sweden, targeting students from the working classes. This lesser-known system, introduced in 1939, preceded the present loan-financed and universal system established in 1965 designed to reach students from all classes. The arguments for launching the selective system, how this system met the objective of broadening the social recruitment of students and the arguments behind the dismantling of the system are analysed. The equalising effect of the selective system was strong, but student loans were nevertheless more compatible with an emerging idea, imported from the Chicago School, that education could be considered an (loan-financed) investment in human capital, that provides future yields rather than a right. Historical institutional theory is used to analyse the shift between two diametrically opposed models that took place within the same Social Democratic regime.


Introduction
Sweden is the archetype of a universal or general social democratic welfare system that is generally contrasted with selective liberal means-testing systems. 1 This study addresses the lesser-known selective precursor of a general model for student finance in Sweden. Research on higher education funding primarily addresses changes after World War II. By starting the study in the 1930s, this article makes a contribution by examining a hitherto unexplored selective model that raises theoretically interesting questions about continuity and change. Two very different systems are contrasted: the gift-based system of in-kind scholarships for talented but poor students, introduced by a Social Democratic government in 1939, and the loan-based system for students from all social classes, established by a Social Democratic government in 1965. The Social Democrats were in power continuously between 1932 and 1976 (with the exception of a few months in 1936). Two diametrically opposed social welfare models were thus launched under the same social democratic regime.

Theoretical framework: continuity and change
Comparative research on student finance systems in different countries, which generally corresponds with the research on different types of welfare regimes (liberal, conservative and social democratic), has shown that a system, once established, remains active. 2 In the analysis of historical institutional theory this is labelled "path dependency". The example highlighted here-a cross-temporal study starting in the 1930s within one and the same country, rather than a cross-national study comparing different clusters of countries during the post-World War II era-shows the opposite to be true. The reform-orientated Social Democratic Party of the 1960s did not follow the path the party had embarked on in the 1930s. On the contrary, it seems to have completely changed track. "The borrower is slave to the lender" was the catch phrase in the 1930s. 3 Twenty years later, the Social Democrats presented a loan finance system to its traditionally supposedly loan-averse base voters. The study also shows that the objective shifted over time: the idea of education as a rightbreaking "the class-dependent education monopoly"-was toned down, while the idea that education can be considered an investment in human capital-the "human capital approach"-emerged. 4 This is clearly not a question of "path dependency", but rather of a moveable goal and a radical departure in terms of the means for achieving the political goals (loans instead of gifts). To understand how the Swedish state student finance system changed in a longer-term perspective we need to apply supplementary theoretical tools from historical institutional theory. An analysis of gradual change and more rapid upheaval must be added to the analysis of continuity.
Historical institutionalism shows that positive feedback effects have hindered changes to the student finance system. Individuals who have (or have not) themselves received state student finance will expect others to receive it (or not). Such self-reinforcing mechanisms lock countries into different path dependencies. This analysis might explain why the Swedish student finance system has remained more 2 Julian L. Garritzmann OECD Countries,1945-2015(Cham: Springer International Publishing, 2016. 3 "Den som är försatt i skuld är inte fri. " This biblical-sounding quotation, which is frequently, but incorrectly, assumed to have been uttered in 1932 by Ernst Wigforss (Minister for Finance 1925-1926, 1932-1949 captures the negative connotations of borrowing among a working-class population who had experience of being in debt to-and dependent on-both employers (loans for housing) and traders (consumer credit). The resistance to debt was also linked to an emerging culture of conscientiousness within the working-class movement. See e.g. Swedish Government Official Reports (Statens offentliga utredningar hereafter SOU) 2013:78, Överskuldsättning i kreditsamhället? Betänkande (Stockholm), 33-34, 39. Cf. e.g. Peder Aléx, Den rationella konsumenten: KF som folkuppfostrare 1899(Stockholm/Stehag: Symposion, 1994, 128, 141, 228; Göran Persson, Den som är satt i skuld är icke fri: min berättelse om hur Sverige återfick sunda statsfinanser (Stockholm: Atlas, 1997), 9-10; David Graeber, Skuld: de första 5000 åren (Göteborg: Daidalos, 2012), 365. 4 Citation no. 1 from economist and Social Democratic politician Gunnar Myrdal' s review of how Social Democratic school reforms changed over time (Gunnar Myrdal, Hur styrs landet? Del 1, Stockholm: Rabén & Sjögren, 1982, 38). Citation no. 2 from a government report on education that shows how theories on human capital and investments in education were first applied in investigations in the 1960s (SOU 1972:23, Högre utbildning: regional rekrytering och samhällsekonomiska kalkyler, Stockholm,132,139). or less intact since 1965. 5 If the period of study is extended, however, it becomes clear, as has been shown, that this is not a question of continuity. Several theoretical concepts from the tradition can be applied. Formative moments, periods of instability when actors can rewrite the rules and design future institutions, is one term used within historical institutionalism to explain the discontinuity. The changes can be abrupt, caused by economic crises (critical junctures at macro level) or perhaps a radically changed balance of power within a board of directors (critical moments in organisations at micro level). But transformations can also occur more gradually. 6 The historical institutionalist Peter Hall' s typology of policy changes can be applied to analyse the changes in Swedish student finance. Following Hall, I will note and distinguish between continual minor adjustments of the activity (first order change), the introduction of new instruments of policy but retained goals (second order change) and the introduction of both new means and goals (third order change or paradigm shift). 7 Research questions are stated after a brief background section which places the state student finance system in a wider historical context.

Historical background, previous research and research questions
In-kind scholarships were one in a series of reforms that the Social Democratic party, which came to power in Sweden in 1932, launched in the 1930s. The aim was to help talented but poor students-who were reluctant to take a loan-to study at university. The financial support was very generous. The students selected essentially had their education funded by the state. The reform was well-received in most political camps, particularly by the political Left where it was seen as a step in the effort to abolish the class society. The Communist Party's representative stated in parliament in 1939: "We are satisfied that the initiative is a first important step on the part of the government towards tearing down at least some of the barriers that the privileges of wealth have raised around the higher education institutions in this country. " 8 As far back as 1918, Social Democratic motions had resulted in interest-free state student loans "for talented, poor pupils", but as will be shown few such loans were granted, and those that were awarded lower amounts. 9 The considerably more generous reform in 1939 was seen as a paradigm shift.  1919 1922 1925 1928 1931 1934 1937 1940 1943 1946 1949 1952 1955 1958 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997  Until the 1950s Sweden had a pronounced elite system of higher education, characterised by the fact that only a small percentage of the population reached higher education. 10 In 1950, for example, newly registered students accounted for just 2 per cent of the country's 20-year-olds. 11 The expansion of higher education subsequently gathered pace (see Figure 1). By the mid-1960s Sweden had reached the threshold that was recognised as marking the entry into a mass university system, namely an inflow of students of around 15 per cent of an age cohort. 12 The in-kind scholarship system of 1939-1964 thus coincided with the final decades of the Swedish elite university. It was succeeded by the diametrically opposed student loan system in 1965, which was thus of an age with the emerging mass university. In-kind scholarships were gift-based and selective; only students with top grades whose parents had a low income and/or fortune, "the talented sons and daughters of the poor", were eligible. 13 The requirements were eased over time (the minimum grades threshold was  Börjesson, "Studentexplosionen under 1960-talet: numerär utveckling och orsaker, " Praktiske Grunde, no. 4 (2011), 20. 13 Sven Lindrot, FC minutes (1939 lowered while the maximum parental income threshold was raised). The grants were at first paid out in kind, in the form of free food and/or free accommodation, and towards the end increasingly in cash. 14 In contrast, the new system was based mainly on loans and also general. Now less talented students from wealthy families could benefit from public support: "dull children with rich parents", according to the critics. 15 Both means testing and aptitude testing were retained (an example of continuity), albeit based on new principles (an example of change within a tried and tested order). Means testing implied that it was now the student's own financial situation that was assessed, rather than that of the student's parents. Study aptitude was based on the results achieved after enrolment to higher education (students were expected to produce credits at a normal study rate), while final grades from secondary school were no longer taken into account. 16 As already stated, research on higher education funding primarily addresses changes after World War II. 17 One exception is the economic historian Anders Nilsson, who studied the variation in social recruitment to higher education during periods of different student finance systems. During the period of selective in-kind scholarships, the social imbalance in recruitment to higher education declined notably (above all during the 1950s), while social equalisation began to slow after the introduction of the general student loan system (the social imbalance increased during the 1970s). 18 Students from well-educated and high-income groups utilised student loans more than students from other social groups. 19 There are no studies on how the scholarship system worked in practice. This study contributes to earlier research by studying which social groups used the in-kind scholarship system at one of Sweden's oldest and biggest higher education institutions, Uppsala University, at the beginning and the end of the period. The rise and fall of the scholarship system is generally unexplored. The examination of the social recruitment of recipients is supplemented by a study of the national political debate in the 1930s when the scholarship system was introduced and during the 1960s when it was abolished.
14 Ringfenced support for poor groups not considered capable in the US, a "liberal welfare regime", has a rather negative association (cf. Viviana A. Rotman Zelizer, The Social Meaning of Money, Princeton, N.J.: Princeton University Press, 1997, 121). The Swedish in-kind state scholarships had, as will be shown, a more positive association. 15 Lars F. Tobisson, "Studiemedlen, " Svensk Tidskrift, no. 4 (1974), 199. 16 Bengt Gesser, Utbildning, jämlikhet, arbetsdelning (Lund: Arkiv, 1985Anders Nilsson, Studiefinansiering and social rekrytering till högre utbildning 1920-1976(Lund: Lunds universitet, 1984 The overarching theoretical question is about how the rise and fall of the early selective student finance system can be understood. To that extent, the study is more about change, the transition from one system to another, than about continuity. Should the reorganisation in 1965 be regarded as a paradigm shift or a somewhat less dramatic turning point within the Social Democratic regime? Against the background of the research situation and the theoretical framework, the following more detailed empirical research questions are also articulated: What was discussed in the debate and what positions were taken when the selective system was introduced in 1939? How was the system applied in relation to the goal of recruiting "poor talented students"? 20 Which groups received state scholarships? Was there a change in recruitment over time in line with adjustments to the means and aptitude testing (changes of the first order)? Was the system abolished because it was not considered to live up to the political objectives of broadening recruitment? What was discussed overall when the selective system was abolished in 1964 and what arguments did the government use to gain approval for a loan-based system among a supposedly loan-averse working class? How comprehensive was the 1964 reform, and did it change not only the means but also the ends of the policy (change of the second or third order)?

Method, material and research design
Two sub-studies have been carried out. The first sub-study which addresses the arguments behind the introduction, and later the dismantling, of the in-kind scholarship system is based on, among other things, public inquiries, bills, debates in the parliament's 1st and 2nd chambers and responses to proposals circulated for consideration stored in committee archives from the years around 1939 and 1964. Arguments for scholarships vis-à-vis loans are compared and how the discussion changed over time and between different groups are analysed. The second sub-study which addresses how the system worked in practice in terms of social recruitment of scholarship recipients is based on a large and to date unexplored archive from the State Scholarship Board (Statsstipendienämnden) in Uppsala. It makes use of descriptive statistics; in other respects the study is qualitatively orientated. In order to study both continuity and change in this area the study must be designed to include the 1930s, when social democracy came to power in Sweden. Starting after 1945-which has been common in many studies on the growth of the welfare system-misses a key change in direction, and over-emphasises the path dependence of politics. 21

The introduction of the state scholarship system in 1939
The class-segregated, parallel Swedish school system, with a 6-year elementary school (for the children of the working class) and a separate grammar school which was intended to lead to university studies (for the children of the resource-rich families) had gradually been broken up by reforms driven by the liberals and social democrats during the 1920s. 22 These reforms of the lower levels of education, which saw more middle class children attending elementary school and more working class children attending grammar schools, form part of the background to the basic education reforms implemented in the early 1930s. The Social Democratic government now had a central goal, which was to ensure that the continued studies of young working class pupils should not be hampered or hindered by their financial circumstances. 23 The path from elementary school to the various levels within the grammar school-a 6-year lower secondary (designed to provide a general civil education) that might lead to a 4-year upper secondary education (in preparation for university studies)-had been long and difficult; continuing from upper-secondary on to university appeared to be an even greater challenge. The government therefore had to "help the students across the threshold between grammar school and university". 24

National and international Sonderweg
The Swedish welfare policy that took shape during the 1930s came to be characterised by general reforms and monetary support (the line of policy of the Social Democratic Minister for Health and Social Affairs, Gustav Möller), rather than by selective support to specific disadvantaged groups, earmarked in kind, which was the case for students in higher education (the Social Democratic Minister for Education, Arthur Engberg's line, which is examined here). 25 In-kind scholarships were also out of step with the times internationally. As this was a new form of government support in Sweden, the student finance systems in 15 other countries were studied in 1937. 26 One observation was that the support was more extensive in other countries. Another was that it was given mainly in the form of cash grants. In-kind scholarships were not awarded in the other countries. 27 The Swedish government nonetheless set out on its own course. Previous investigations initiated by student bodies and university teachers had 22 It took a long time to abolish the parallel school system. The reform of basic education came to a standstill during the 1930s but picked up again during the 1940s. A period of experimentation during the 1950s was followed by the implementation of a compulsory elementary school in 1962, which led to the lower level of the old elementary school-primary school-being merged with lower-secondary school to become a 9-year primary-lower secondary school for all children.  SOU 1937:44, 6-19. 27 SOU 1937Engberg in Proposition (1939), 44. shown that the situation for students had deteriorated. As a result of inflation, the cash grants covered ever less of living costs. There were three problems with the student loans available. Too few loans were granted, the amounts awarded were generally low, and the loan "burdened the student's future economy". 28 Although the conditions for these interest-free loans were favourable, the money still had to be paid back. Since the purpose of the report was to make it possible to choose education without incurring a financial burden, the original brief did not include an enhanced state loan system. During the process, however, the committee found that state support in the form of grants would be most effective if it was combined with credit that was interest-free and without suretyships. The government agreed to investigate loans as a complementary form of support. 29 There was no support for loans as the main form. Of 18 consulted bodies who responded to the report's proposal in 1939, only one, The Swedish Agency for Public Management (Statskontoret), which falls under the Ministry of Finance, were in favour of loans on the grounds that grants were more costly for the state than support that must be repaid. 30 The remaining 17 bodies consulted, the committee itself and the government considered that grants were the best way to improve social recruitment and thus to be preferred. The discussion centred on whether the grants were best disbursed in cash or kind, and who should be entitled to receive them.
The majority of the authorities and bodies consulted, 16 of 18, agreed with the Minister, Arthur Engberg, that in-kind scholarships in the form of board and lodging were to be preferred. Their argument was based on control and quality. They did not fully trust young students to handle cash.
Utilising the means "in an appropriate manner" was not just about covering the most essential costs for food and accommodation-the bare necessities of life-before the money was spent on other temptations. The food must also be healthy, "sufficient and nourishing" and the housing of a good standard and "fit for purpose". 32 Under the proposed in-kind system, new local state scholarship boards would regulate the quality and allocate "the scholarship recipients among suitable eating places and housing approved by the board". 33 The few critics of the in-kind argumentthe afore-mentioned The Swedish Agency for Public Management and the teaching faculty at the medical university Karolinska institutet-believed, in contrast, that the government should trust the students to manage their accommodation themselves. This would also be a less costly administrative solution. 34 The same argument 28 SOU 1937:44, 2. 29 SOU 1937:44, 2-4, 93-94. Cf. Nilsson's (1984 analysis of motivation for the response 1937, 60. 30 Proposition (1939), 42. 31 Engberg, FC minutes (1939), 8. All quotes are translated from Swedish. 32 SOU 1937:44, 3, 112, 119. Engberg, FC minutes (1939), 8. 33 SOU 1937:44, 114. 34 Proposition (1939 for in-kind support was found in social policy (after means testing, families with children would receive shoes, clothes and such which met approved quality standards), but here the critics of this selective form of support won the day (means testing was degrading, created administrative problems and should therefore be avoided). 35 Engberg, however, did not attach any great importance to this criticism. As of the 1939/40 budget year, scholarships in the form of free accommodation and free food for a maximum of three years were introduced, and could be combined with interest-free student loans. 36

Middle class protests and expansion of the system 1950-1964
In-kind scholarships do not appear to have been stigmatising. Surveys carried out at the end of the 1940s showed that the majority of recipients questioned (75 per cent) were happy to receive the support in kind. 37 Further, those who would have preferred to see the support paid out in cash (15 per cent) had the future ahead of them. Cash payments became increasingly common during the 1950s. 38 In the final year of selective support (1964) the finance reform was even described as a "state grant in the form of free board and lodging or the equivalent cash benefit". 39 At the same time, the fact that the system achieved the intended effect-encouraging talented students from poor homes to start studying at university-gave rise to new concerns in the class society.
There was little room in the discussion at the end of the 1930s for the question of where to draw the line between groups with varying degrees of limited financial resources: those with "no means" (who were eligible for scholarships) and those with "lesser means" (who could only apply for interest-free loans). Only one body consulted, the Swedish Central Government Agency for School Issues (Skolöverstyrelsen), raised the problem. They feared that a limited group of students might benefit unduly while others with similar merits were excluded. 40 This discussion of the boundaries of the student finance system gained momentum during the 1940s. The dissatisfaction of the educated middle classes is apparent in the comments submitted to the 1948 report. Although middle class households had seen the value of their assets fall as a result of inflation and increased tax burdens, the State Scholarship Board in Uppsala wrote, they are still considered to be "of lesser means" and are therefore ineligible for in-kind scholarships. They are now less able to give their children a higher education "than those of no means". 41 Gothenburg School of Busi- 35 Rothstein (1994), 213-14. 36 Proposition (1939), 47-53. Cf. SOU 1937:44, 121, 132. 37 SOU 1948 ness (Handelshögskolan i Göteborg) stated that in addition to their own as yet unpaid student debts they were now expected to pay not only for their own children's education, but also, via income tax, for the education of the children of the working class, which was unreasonable. 42 The 1948 report captured these attitudes and revised the financial qualification requirement. State scholarships and interest-free loans should be extended to students who came from less well-off homes; that is, those "who belong to society's large middle classes". The proposal to "raise considerably" the highest parental income threshold met these demands from the dissatisfied middle class. 43 The report's other proposal, to lower the study aptitude requirement to allow students with lower grades to apply for scholarships, was less well-received by the educated middle classes. 44 The Stockholm School of Economics (Handelshögskolan in Stockholm) opposed the idea of spending huge sums on scholarships for "mediocre students from the lower classes and the large group in the middle". Student loans with a repayment obligation were a better alternative. 45 The committee ignored the criticism. The measures to support students were intended to "remove the economic differences between students from different social groups". 46 They could thus not be based on loans, as there was a widespread "psychological resistance to debt" within the very social groups for whom the government wanted to ease entry to higher education. 47 From 1950 onwards the state scholarship system underwent a series of reforms to meet the findings of the 1948 report. More higher education institutions and more students were drawn into the system. 48 The highest parental income threshold was raised successively while the student grades required were lowered in stages. 49 This influenced the number of scholarships awarded. In some years at the end of the 1940s up to 70 per cent of applicants were rejected because they did not meet the required standard of "academic ability". At the beginning of the 1960s this had fallen to around 25 per cent. 50   analysis of motivation for the response 1948, 90. 47 SOU 1948:42, 146. 48 SOU 1948:42, 94, 179. Cf. SOU 1963:53, 24-25. 49 SOU 1963:53, 25-28. 50 SOU 1948 The rising cost of scholarships highlights the strong expansion of this government gifting institution: measured in 2019 prices, the spending on state scholarships rose from around six million SEK in the 1939/40 academic year to almost three hundred million SEK at the peak in 1962/63. 51 The system grew particularly rapidly in its final years. Around 25 per cent of newly registered students in higher education in the 1957/58 academic year received a state scholarship; five years later, in 1962/63, it was almost half of new students in Sweden. 52

Social recruitment of state scholarship recipients
There is no previous research on how the state scholarship system worked in practice and changed over time. Studies at local level are completely lacking but some government reports that examined the system at national level at the end of the period suggest the support was not only in line with the general ongoing social equalisation which had gained momentum in the 1950s but also almost certainly actively contributed to a broader recruitment of students from the working class. 53 It is, however, difficult to determine exactly how much, since many factors were involved. The aim in this section is more limited: to complement earlier research through a case study of Uppsala University of which social groups received state scholarships when the system was new at the end of the 1930s and when it was at its peak at the start of the 1960s.

An unexplored but challenging archive
Uppsala University, the oldest university in the Nordic region, founded in 1477, gathered half of the country's state scholarship recipients at the start of the period studied and around a third at the end. 54 According to the Uppsala Scholarship Board's own figures, 94 applications were granted for the 1939/40 academic year, about 3 per cent of the total number of students registered (3,673). In the 1963/64 academic year 3,405 applications were successful. There were now considerably more state scholarship recipients in Uppsala; the group comprised around 26 per cent of the total number of students registered at the university (12,901). 55  information from the index cards of state scholarship recipients at Uppsala University between 1939-1964, an unexplored archive, together with birth and baptism records from the parishes in which the students were born. The name and date of birth shown on the index cards, and place of birth visible in a database of Sweden's population in 1970(Sveriges Befolkning 1970, have enabled me to find information in the church records on the parents' occupation (in practice often only the father's occupational title) at the time of the recipient's birth. The recipient's full name, the first step in the search for information via the index cards, is easily accessible in the first cross section: the names of all 1939/40 state scholarship recipients are listed in the University directory. There is no such list of named recipients for the other cross section. An explanation of how I worked with the index cards to make comparisons over time is necessary here. Briefly, there are twenty-nine boxes of records with over 17,000 index cards for named students who were state scholarship recipients and/or loan takers in Uppsala in the period 1950-1964. The cards are arranged by surname, not by the year when student support was granted. The first step was thus to arrange the material according to academic year and form of finance (grant or loan), then delimit it in order to make it manageable. I have counted all the index cards in the twenty-nine boxes (17,472 in total) by hand and systematically worked through every seventh box in this series (five boxes with 2,915 cards). 56 This information has enabled me to estimate the number of individuals who received state scholarships (and often also loans) and those who only took a loan in different academic years. A thorough examination of the five boxes of records revealed that 597 individuals received a state scholarship for the 1963/64 academic year. If I work on the assumption that the balance of state scholarship recipients and loan takers in the remaining twenty-four boxes was similar (on average, the proportion of scholarship recipients in the five examined boxes was 60 per cent) the total number of state scholarship recipients in 1963/64 comes very close to the number the scholarship board itself reports (3,567 according to my estimation; 3,405 in their figures). The list of 597 named recipients thus appears to be a stable and reasonable framework for selection. To make the work in this examination of social background manageable, 50 per cent of these, 299 individuals, are included in this sub-study. This appears to be a feasibly sized group to track down in Sveriges Befolkning 1970 and in the birth and baptism records in the parish archives. The 299 individuals were randomly selected using a function in the Excel calculation program. 57

State scholarship recipients in Uppsala 1939/40 and 1963/64
To enable a comparison between the data utilised in this paper and state investigations which examined social recruitment to higher education during this period, the same system has been used to group the occupations of students' parents. This is a classification of eight main occupational groups sub-groups. 58 The sample is quite small and should therefore be interpreted with care. A number of tentative conclusions can however be drawn.  1930-1970, " Zenit, no. 28 (1972. 60 Therborn (1972), Comment: Occupational groups are classified in line with the above-mentioned model showing eight major occupational groups A-H and a number of sub-groups (see footnote 58). The division into classes and strata is done in dialogue with Therborn (1972). Classification of the occupation of parents of scholarship recipients into classes and sub-classes is reported in more detail in Gustavsson (2017) To gain a clearer picture of the effect state scholarships had on recruitment, the social mix of recipients needs to be compared with the social mix of the student body as a whole. There is very little material on this. It is, however, possible to obtain a point of comparison with students at around the time for the first cross section. From Dismantling the Class Society to Investing in Human Capital  Table 2 reaffirms that the state scholarship system had a noticeable effect on social recruitment. A significantly higher number of recipients at Uppsala University came from the working class (in which the state investigations included tradesmen) than did other students there and at other large higher education institutions in Sweden during the 1930s (around 37 per cent, to be compared with around 13 and 7 per cent). On the other hand, few state scholarship recipients in Uppsala came from academic or senior official homes (only around 16 per cent), a social background in which studies were encouraged and which produced almost half (around 48 per cent) of students registered at various faculties in the country.
It is even more difficult to find comparable source material from the 1960s. The social mix of scholarship recipients at Uppsala University in the 1963/64 academic year cannot be compared with the social mix of the student body at this particular institution as a whole, but it can be compared with the national mix around the same time. Such a comparison (Table 3) shows that the system's social recruitment capability remained strong. Significantly more of the scholarship recipients had grown up in the more broadly defined working class (around 49 per cent) than of all students in higher education (around 16 per cent). The state scholarship system was also seen as a success at the time. According to a government report from 1963, the system attracted mainly students from working-class homes, which was its purpose. 63 The working class young generally had a "particularly strong" inclination to study. The proportion of newly registered students in this group had risen from 8 per cent in 1947 to 14 per cent in 1960 (Table  3 shows  The scholarship system helped too few students, Anderson continued, and student loans were effectively a way of financing the middle class, "and the principle of students having to pay for their own education in this way is pernicious in any case: they clearly should have wages or a salary like anybody else engaged in responsible valuable work". 67 Thus, the selective student support system worked well. The system raised moral questions (not least among the middle classes) about who should be denied support, but there was nothing in way the system worked that provides a clear explanation for why it was dismantled. What caused the Social Democratic government to suddenly dismantle the state scholarship system with its strong equalising effect-at a time when the education gap in society was still wide-and replace it with a loan-based system? To generate an answer to that question, the focus in the next section shifts from social recruitment to the ideological legitimation of different types of welfare system solutions.

Discussion on dismantling the system in 1964
The government report which paved the way for the introduction of the student loan system in 1965 proposed that the focus of the support should be redirected to the transition between junior and upper secondary school. Studies had namely shown that social selection was strongest before the final year school exams, not after. It was the "doors to upper secondary school" that must be forced, not the doors to university at which the state scholarship system had aimed in 1939. The main focus of the 1965 reform was thus on reforming upper secondary school. 68 The need for public economic efforts to help students at university and higher education institutions had not declined (social recruitment to higher education was, as has been shown, highly imbalanced), but the economic scope to direct the major share of the resources in that direction had. 69 It is clear that an analysis of the reasons for the Social Democrats' ideological change of direction in the question of student finance must include an economic dimension. There was a feeling that the financial situation of the government was coming under increasing pressure at the end of the 1950s. The welfare state was faced with many essential costs. In August 1959, the rationalisation inquiry (Besparingsutredningen), which was looking into ways of cutting costs within state activities and was working at pace with the socio-economic student inquiry (Studiesociala utredningen), suggested making "a significant shift" from grants to loans in this area. 70 Ideological arguments against grants and for loans was supported by economic arguments (to which, however, the socio-economic student inquiry initially paid no attention).
The existing research on the student finance reform, and the written notes of members of the socio-economic student inquiry, further highlight the government's financial deliberations as an explanation for the ideological u-turn in the question of student finance. As the education sector-and the entire welfare state-expanded, so the argument went, the economic scope and the arguments for particularly generous student finance reforms based on scholarships to increasing numbers 68 SOU 1963:74, Studiesociala utredningen: rätt till studiemedel (Stockholm), 22;Proposition: studiesocialt stöd till studerande ungdom (Stockholm, no. 138, 1964), 1; Olof Ruin, Studentmakt och statsmakt: tre studier i svensk politik (Stockholm: Liber, 1979), 51. The findings of the socio-economic inquiry which was appointed 20 years later to assess the effectiveness of the student finance system were similar to the analysis from the 1930s: the social selection that applied to transfers from elementary school to upper-secondary school was not enough to explain the strong social imbalance in recruitment to university during the 1980s. There was also a "significant selection process on transfer from upper-secondary school to traditional higher education". As had already been expressed in the 1930s many students needed a helping hand to cross the threshold. SOU 1987:39, Studiemedel: betänkande från Studiemedelskommittén (Stockholm), 69. Cf. Jan Johansson, Det statliga kommittéväsendet: kunskap, kontroll, konsensus (Stockholm: Stockholms universitet, 1992), 158, 164. 69 Reuterberg andSvensson (1981), 16, 18. 70 SOU 1959:28, Besparingsutredningen: besparingar inom statsverksamheten (Stockholm), 204. of students, tended to decline. 71 This is nonetheless in striking contrast to the debate on student finance at the time, which is reported below. There was surprisingly little room for budget-related questions (the socio-economic student inquiry, for example, does not refer to the rationalisation inquiry mentioned). The discussion was ideological and centred on different views of grants and loans. And although no social democratic arguments for student loans actually emerged, the ideology was nonetheless later changed and adapted to the new actual economic conditions.

Continued widespread resistance to student loans
Opposition to repayment-free student loans was strongest within the Conservative party, which had proposed a motion at the start of the 1960s to abolish state scholarship subsidies. The right wing favoured a finance system of loans. There was, however, a strong antipathy to student loans among the political majority. The most favoured form of support was still repayment-free state scholarships, now also termed "general grants" (by the Liberals) and "a student wage" (by the Social Democrats). The groups who advocated a strong expansion of state scholarships included student and teacher organisations (Sveriges Förenade Studenter SFS and Sveriges akademikers centralorganisation SACO) and most political parties (the Communists, Social Democrats and Liberals). 72 Four arguments for scholarships (and against loans) dominated the debate in the 1950s and 1960s; most are recognisable from the debate in the 1930s. Firstly, scholarships were still considered more effective in social recruitment. Even the government report that proposed the new student loan system concluded that the aim of the reform policy-to remove economic and social barriers to educational opportunity-would be fully realised through general grants. The report continued that making the grants both general and cash (student pay) would end the moralising pointers on what the recipient should purchase (approved goods) as well as the potentially degrading means testing (compare Möller's general line). 73 Secondly, a coherent argument for general grants was that they would remove the problem of debt. Grants would mean that young students did not start their professional life with "a large burden of debt hanging round their neck like a millstone", as a Conservative party member who broke rank expressed it. 74 The gains would be 71 The risk that the cost to the state of a general grant system would be unreasonably high is pointed out in Reuterberg and Svensson (1981), 19; Reuterberg and Svensson (1987), 4 and Henrik Berggren, Underbara dagar framför oss: en biografi över Olof Palme (Stockholm: Norstedt, 2010), 326. The importance of budgetary considerations is also pointed out in notes made by members of the socio-economic student inquiry:   Ruin (1979), 20-22;Reuterberg and Svensson (1981), 11-13;Lars Jonung, "Ingemar Ståhl 1939 73 SOU 1963:74, 34-35. Cf. SFS andSACO arguments in Ruin (1979), 34. 74 FC minutes, Riksdagens protokoll: studiesocialt stöd till studerande ungdom (FK, no. 27, May 26 1964), 97. Cf. SOU 1963 greatest for groups with small economic means for whom it is "an axiom, a matter of honour never to get into debt", declared a Social Democrat in the second chamber in 1964. 75 Thirdly, an argument stemmed from the Social Democratic discussions on student finance in the 1950s. This was that studies could be considered as productive work and should, therefore, as one Social Democrat expressed it in parliament in 1964, "be equated with all other paid work". 76 The inquiry's analysis pointed in the same direction. Since investing in education was expected to be highly profitable in the coming post-industrial society, the conditions should be right for such a change. "The report finds the thought of student pay attractive on a number of grounds. " 77 Fourthly, closely allied with the recruitment argument there was an argument of equity, and an understanding that equity must be allowed to cost money. 78 Against this was an argument that the cost to the state of an expanded grant system would be much too high; yet this argument had a surprisingly small part in the discussion. 79 One of the opposing side's arguments in favour of loans was that they encouraged moral discipline. Loans teach responsibility. Under this view, which was articulated primarily in the comments on the report of 1948, students in debt were forced to learn "to manage their affairs and plan for the future" while the "costs and risks for the state" were low. 80 Conversely, grants might "relax the sense of responsibility". 81 Another argument, raised by some academics and students, was that academic freedom increased when studies were financed by loans. Gifts always carried the risk of a demand for "something in return". 82 There was, however, a similar fear with regard to state loans, not least within the political Right, who wanted the loans to be managed by private banks, not the state. "Loans straight from the Treasury, " claimed a Conservative MP in parliament in 1964, gave the government "better control over students. It is then natural for the government to exercise control over the direction of study, choice of occupation and so on. " 83 However, the main argument for loans, which the political Right had defended since the 1950s, was that education should be considered in the same way as any 75 Minutes of Parliament's second chamber [hereafter SC minutes], Riksdagens protokoll: studiesocialt stöd till studerande ungdom (AK, no. 27, May 26, 1964), 131. 76 SC minutes (1964), 77. 77 SOU 1963Proposition (1964), 64 (citation). 78 Cf.  analysis of motivation in the response 1963, 128 andGesser (1985), 203. 79 The 1959 investigation directives opened up for a proposal for the socio-economic student policy that would be a greater burden on the budget than previously, but at the same time take into account the budgetary consequences. Four years later, the investigators briefly mention reduced budgetary possibilities in "the current situation". SOU 1963:74, 10-11, 38. Cf. Johansson (1992, 153-54, 173. See also Olof Palme, FC minutes (1964), 76;Proposition (1964), 64. According to Ernst Wigforss, Minister for Finance 1932-1949, student loans would not imply any greater financial strain than the proposed student finance system. Larsson (2003), 85. Economic arguments feature more prominently in retrospective analyses than in the socio-economic student inquiry and in the discussions at the time-see note 71. Given our view that studies are an investment, we believe that they should in principle be financed by students themselves. For those without means of their own, borrowing is, as with other investments-without a subsidy-a natural means of finance. 84 The idea that education could be considered an investment in human capital gained traction and legitimacy at this time, largely due to the spreading of the works by Chicago economists Theodore Schultz and Gary Becker. 85 The Swedish Conservative Party, which within this tradition rather one-sidedly attached economic profitability aspects to education, also suggested that student borrowing be combined with the right to deduct study costs. The deduction was seen as a "parallel to the depreciation, the deduction for fall in value permitted for capital investment in machinery and other items". 86 No Social Democratic arguments in favour of student loans were ever put forward.

The architect behind the reform: a market liberal Social Democrat
The political scientist Olof Ruin, who was for a time secretary of the socio-economic student committee of inquiry set up in 1959, has described in detail how the inquiry came to change direction completely, albeit without a formal decision, from having worked for an extended scholarship system to presenting a system based on loans. The differences of opinion among the committee members-above all between the student and academic representatives on the one side and the other five members on the other-grew. After some years the inquiry ground to a halt. At that point, the social debater Ingemar Ståhl, who would soon be appointed to the inquiry as an expert and later became a professor of economics, put forward a proposal to the inquiry secretariat that according to Prime Minister Tage Erlander "sounded far too complicated" but which Olof Palme, the chair of the inquiry and future Prime Minister found interesting and a possible solution. 87 It was presented as an "inverse pension system" whose working name, the General Student Loan System (Allmänt studiefinansieringssystem) (ASF), also alluded to the 1959 Swedish pension reform, the General Supplementary Pension (Allmän tilläggspension) (ATP). Under this system, students would be given money (a loan) while they studied, which would be repaid in the future when they were working and had an income. Pensions work in the opposite way: you pay in over the years and build up a sum which the system pays out in the future. 88 The thinking behind Ståhl' s proposal was-like the argument of the political Right shown above-that education is an investment in human capital which is expected to produce a future yield. A student loan system thus appeared to be the sensible solution. He had taken the argument from the Chicago School' s theory of human capital, although there is no explicit reference in the report to the technical and "complicated" system's origins in this tradition. 89 Another argument that one would not expect to come from the working class environment is Ståhl' s line of reasoning that the purpose of student finance is not distribution between individuals-a classic socialist idea-but purely for the individual over her lifecycle. 90 As a poor student she receives a loan, which she then repays when she has wealth (as long as the capital investment produces a return). Tax-funded grants/student wages, on the other hand, would lead to an unfortunate distribution of income from poorer citizens (who did not study) to-future-high earners. 91 88 The general supplementary pension (ATP) meant that the state pension which was introduced in 1948-and linked to citizenship-was supplemented with a universal earnings-related pension program to all economically active individuals, under which the benefit was related to the input into the labour market and based on the principle of loss of income: the more the employee earned, the more they received. The student finance system works inversely: the obligation to repay could be waived in cases where the recipient was unable to meet the payments. Ruin (1979), 41-43;Kruse (2018), 95; Reuterberg andSvensson (1981), 17. 89 Jonung (2018), 16-17. The term "human capital" is not mentioned in SOU 1963:74 but Appendix 2-where Ståhl tests a theoretical model argument with a calculation example-has been pointed out as one of "three examinations that can be said to be representative" of human capital investment in the early 1960s. The other two were Becker's Human Capital (1964) and Blaug's "The rate of return on investment in education in Great Britain (1965). See SOU 1972:23, 139. Ståhl himself declared that the Appendix was "the first Swedish study of the profitability of investments in education" ), 234. Cf. Bengt-Christer Ysander, "Varför högre utbildning, " Ekonomisk debatt, no. 6 (1973Gesser (1985), 69, 83, 296 and note 4 and 87 above. 90 Cf. the wider discussion on selective models of welfare which stress vertical distribution (between socio-economic classes) and universal models of welfare which stress horizontal distribution (between stages in the lifecycle of the individual). While the former prioritises income equalisation ("the Egalitarian Welfare State") the latter favours the integration of all groups in society ("the Solidaristic Welfare State"). The Swedish model, which successfully integrated the new middle classes, was later developed in the direction of the latter, according to Alexander Davidson, Two Models of Welfare: The Origins and Development of the Welfare State in Sweden and New Zealand, 1888-1988(Uppsala: Uppsala universitet, 1989. The paradox that selective welfare models that use the "Robin Hood strategy" of taking from the rich and giving to the poor in the end may still be less effective in reducing poverty and inequality than welfare models that follow the "Matthew principle" of giving more to the rich than to the poor is highlighted by Korpi and Palme (1998), 671-72. Cf. the principle of loss of income in note 88. 91 Jonung (2018), 17;Kruse (2018), 96-97. On the contrary, the aim of previous Social Democratic education reforms was to counter the ideal of "selfish private capitalists" within the field of education who invest in their own education without being trained in "collective cooperation". Alva Myrdal and Gunnar Myrdal, Kris i befolkningsfrågan (Stockholm: Bonnier, 1934) Ståhl' s line won the day. It was not, however, an easy victory, either within the inquiry or in society. It was followed by bitter struggles with some rather unexpected shifts in position. For example, in 1963 the Conservative Party anticipated the Social Democrats' 1964 proposal in a parliamentary motion on loans as a way of financing higher education studies. It was also remarkable that the Social Democrats' fundamental change of direction had taken place within the context of a public inquiry and not the party conference held every four years. The conferences in 1956 and 1960 had decided on a student wage, not loans. The new line caused some discontent within the party. A deep gulf had also arisen between SFS and SACO-the student and academic organisations respectively which were most affected by the reform-and the government. 92 There was also a widespread perception among those to whom the proposal had been sent for consideration that the new loans system would lead to those groups of students who had earlier benefitted from state scholarships being worse off. 93 Since there was still "an aversion to borrowing" among broad groups in society, the government would now have to "go out and persuade working and lower middle class families that in the future their children should borrow money to finance their studies", wrote a Social Democrat who deviated from the new party line, and that was "definitely not a task to look forward to". 94 Different verbal concepts were used to gain acceptance for the new controversial loan system. While the proposal signalled the end of the policy of scholarships, it was presented in such a way that, with a little goodwill, it could be seen as a first step towards student pay, as it included a grant portion of 25 per cent. It stressed that a future expansion of student finance towards full student pay was not being "prevented or obstructed". 95 Even if the report advocates a loan system it initially points out that, as a form of finance, loans "are more likely to hinder than help students". 96 Here was wording that would be acceptable in different camps. The report also used 92 Ruin (1979), 55, 57, 70, 78, 81. Cf. Berggren (2010 34, no. 4 (2005), 529). Their investigations showed, however, that when working class youth did begin a higher education they incurred debt to a much greater extent than students from other social groups (since less well-off families were less able to provide financial support than well-off families), but that the amounts borrowed were smaller (since working class youth were more inclined than others to choose short courses).  Quensel (1957), 156. The wish to finance living costs during their studies independently of their parents was behind the decline in the reluctance of students to take a loan, according to SOU 1963:74, 25, 27. The question at issue was whether the state should rein in the need to go into debt or whether the risk connected to debt could be considered manageable through the construction of the new state loan system (long repayment periods, waiver of charges in cases of low ability for repayment). ), 31-33, 69. 95 SOU 1963; Proposition (1964), 64; , 66. 96 SOU 1963:74, 27. a new language. Loans were not loans but "student finance with a repayment obligation". 97 This was a shift in terminology that met with criticism. It is more important to remove students' actual debt "than to abolish the terms loan and repayment", wrote SFS. SACO wrote that removing the words "debt" and "repayment" did not improve security for the borrowers. 98 But the technical vocabulary, and the fact that the "ASF" as a whole appeared complicated, in combination with a very short period of consultation, may have made protest more difficult and thereby facilitated the implementation. 99 Another course of action was to work with the shift in perspective. The focus was moved from the aspects of scholarships that had been proved positive, their equalising capability in terms of social recruitment of students with little or no means, to their possible negative aspects: Ståhl' s speech, referred to above, on an unacceptable transfer from all tax payers to a small group of academics. 100 The strong investment in supporting students at lower levels of education further moved the focus away from the continuing problem of an imbalance in recruitment to higher education and also reduced the scope for generous reforms at that level. "This strong effort clearly shows that it was not possible for purely budgetary reasons to maintain the original intention to invest heavily in the limited category of academics. " 101 Applying means testing to the economy of students and not, as previously, their parents, a liberating effort that enjoyed wide support, finally moved the focus from a social imbalance in recruitment to a possible bad investment in human capital: "Seen in this way, the most disadvantaged are not students from a certain social or economic background. Rather, the most disadvantaged are those who in their working lives are unable to repay the student loans they have received. " 102 As an unintended-we can presume-consequence of the liberation from the patriarchal family structures, the class perspective disappeared from the discussion. 103

The shift within an overarching Social Democratic regime 1932-1976
When it comes to the 1939 reform, the analysis is straightforward: it was a third order change in Hall' s sense. The 1930s were a formative period. The Social Democratic Party came to power in 1932, during an economic crisis (a macro level critical juncture) and began to introduce radical reforms in a number of areas. The student finance reform introduced both a new overarching goal (to pull down the social and economic walls around the universities) and new instruments to achieve this goal (state grants with linked interest-free loans). The system was then regularly adjusted (changes of the first order) to make the conditions increasingly generous until 1964. It is more difficult to assert what type of reform 1964 actually implied.
The Uppsala University case study showed that the state scholarship system succeeded in recruiting talented students from homes with little or no means, both at the beginning and the end of the period, which was what it had promised. The general feeling at the beginning of the 1960s, among both the committee of inquiry and the political majority, was that the system was a success. The criticism mainly concerned the financial means testing, the fact that those who did not receive support were left totally dependent on their parents. The state scholarship system should thus be extended and made general, something for all students, regardless of their parents' income. The fact that this did not happen is explained by the interplay between economic and ideological factors. This was partly due to an underlying concern that an expansion would over-burden government finances and partly the result of a changed ideological view of education and student finance. The view of education as a right, with government support in the form of gifts going to groups which were unfairly excluded, was being challenged by a view of education as an investment, with a state system of borrowing and repayment. Consideration for the budget was interwoven (but not communicated) in the ideological change of direction. For the government inquiry into student finance which was responsible for this, this was not, however, a gradual change along a set path. Rather, it was a sudden rupture, when a new member with new ideas that were better suited to a changed budgetary situation, persuaded the committee to advocate loans not grants (an organisational level critical moment). The creative rhetorical work the committee put into gaining acceptance for the view that the new loan-based system was not a true loan-based system indicates that the people behind it believed that many environments outside the inquiry, not least the traditionally loan-averse working class, were stuck in the old way of thinking. This is clearly a question of new instruments of policy (loans instead of gifts or pay) and renewed goals (human capital investment instead of class recruitment) and is thus a third order change. But at the same time it is difficult to equate this historical break within the education sector with an overarching societal paradigm shift (Hall' s own example is the transition from Keynesian to monetary macroeconomic regulation in Great Britain between 1976 and1989). 104 The goal to reduce the social imbalance in recruitment to higher education had not officially been abolished in Sweden; it remained central on a rhetorical level. The Social Democrats also remained in power until 1976, when there was an economic crisis (another macro level critical juncture). Neither was there an articulated goal that made educational policy all about producing human capital. Rather, this was the beginning of a change of direction, where the focus on "abolishing the class system"-as an intrinsic value-was de-emphasised and thoughts on education as an investment in human capital to improve workforce productivity-to meet the need for growth-gradually became more central. 105 On one level, it may also appear that the student finance reform was more of an adaption to the Social Democrats' paradigm of general reforms and monetary support (echoing the ideas of Gustav Möller). But the system of in-kind state scholarships was continually adjusted until it did pay out in cash at the end of the period if that was the wish of the recipient, and the failed proposal for a student wage was an example of a general reform. This interpretation further fails to recognise that the ideas behind the new general loan system that prevailed in 1965 came from Chicago, from an emerging, more market-oriented paradigm, although its big breakthrough in Sweden did not come until after the economic crises at the beginning of the 1970s and 1990s.

Epilogue: Formative steps towards a new order
Economic crises do not give rise to new political ideas; in Gramsci's words they can "only prepare the way" to spread certain ways of thinking, formulating problems and solving questions which can be "decisive for the continued development of the whole of state life". 106 Ingemar Ståhl, the man behind the 1960s student finance reform in Sweden, was a key importer and distributor of new market liberal ideas. In all areas of Swedish society, writes his biographer, from higher education to military defence, he tried as a government investigator from the 1960s, and as professor of economics from 1971, "to extend the territory of the market and thereby freedom of choice for the individual. " 107 While his argument finally prevailed in the socio-economic student report (SOU 1963:74) it took time before these ideas were generally accepted by society. Ståhl, who is recognised as one of the hidden powers in Sweden, 108 was ahead of his time. Many of the ideas he launched remained controversial. These include the ideas on programme budgeting, also imported from the US, that Ståhl put forward in the Defence Report (SOU 1969:25) and whose goals-means approach was first fully recognised in the 1990s (compare Management by Objectives and Results and New Public Management). 109 But a market-oriented shift had been initiated, by a key Social Democrat (who later left the party). The irony of history had, as Ståhl himself stated towards the end of the 1900s when he reflected over the socio-economic student report, led to the Social Democrats' "most important social reform" being based on thoughts and models taken from neoclassical economists such as Gary Becker and Milton Friedman, who were also "keen advocates of a loans system rather than a scholarship system". 110 The Social Democrats appropriated the idea of loans financing and gave it a new ideological connection. Tage Erlander, the Social Democratic Prime Minister 1946, has also stressed the future Prime Minister Olof Palme's "boldness" as chair of the socio-economic student inquiry, his courage in breaking completely with his own ingrained ideas of a student wage system and daring "to accept something completely new", namely a student loan system, and then furthermore successfully implementing the system despite resistance. 111 It is clear that Ståhl initiated the change work; it has, however, not been possible within the framework of this study to establish the details of what happened when the resistance within the party was overcome. That question, like the wider implications of the result for research on the emergence and administration of a universal social democratic welfare system, must be put aside to be developed and discussed further in other contexts.
The fact that the Swedish student finance system has been perceived in very different ways should nonetheless be seen against the background in the history of ideas that has been portrayed here. On the one hand, there is a more positively charged picture that the student finance reform was not simply general welfare policy but also a generous Social Democratic innovation. As the Minister for Education and Ecclesiastical Affairs stated at the time of the decision in 1964: "No other democratic country has come so far in the area of student finance as we have with this decision. " 112 In the words of a more modern assessment, the student finance system became "one of the most successful reforms in Swedish educational policy". 113 On the other hand, contrasting pictures show that, in the knowledge society, students have had to take over the state's costs for higher education. The costs of student finance, which are in the nature of loans and transfers, are included in a socio-economic calculation not as costs for the state but for the individual. 114 The contrasting view centres on that aspect. Higher education may be free of charge in Scandinavia but students carry a significant burden of debt. The general loan system implies "a transfer of costs from parents to students, not, as might have originally been intended (at least by the students and some of the political Left) to the government. " 115 Sweden was a forerunner in this. The later implementation of loan reforms in other countries often met with resistance from students, who protested against the very fact that the investment in education so essential to society was being transferred to the individual. 116