Modes of reproduction in the Swedish economic elite: education strategies of the children of the top one per cent Open Access

Within the sociology of elites, most research has defined the elite using criteria of formal power as the selection base. In the case of the economic elite, this would mean groups such as CEOs of large corporations (cf. Hartmann 2000, Ellersgaard et al. 2013). This inclination is perfectly aligned with seminal texts within the field (cf. Giddens 1974: 4, 12–13; Mills 1999 [1956]: 276). This paper, however, instead turns the attention to a group defined by distributional criteria, by focusing on the most affluent in Sweden, thereby adding sociological reasoning to a sphere hitherto dominated by economic thought (i.e. the top one per cent recently made famous in Piketty 2014). While the economic analysis stops after establishing the share of the total amount of income or wealth concentrated in the hands of the top groups, this paper utilises insights from Pierre Bourdieu among others to also shed light on their more sociological properties, such as the differences in the elite's internal capital structures – with regard to the possession of different kinds of economic capital as well as other forms of capitals. What is, the paper asks, the structure of the Swedish economic elite, when defined by distributional criteria? Although Piketty – and his collaborators – successfully have placed this group at centre stage for near global scientific attention (cf. the contributions in Atkinson and Piketty 2014), little is known about who actually populate it (although there is an emerging literature describing such top groups, see especially Keister 2014). And despite Piketty's efforts to show the re-occurrence of a ‘rentier society’, even less is known about how they have risen to their elevated economic positions (for an important exception, mapping wealth formation in Norway, see Hansen 2014).

Within sociological studies of intergenerational transmission focused on the economically privileged, the most common strategy has been to highlight things easily transmittable – i.e. money – and examine the correlation between the individual's economic level and their parents’ economic levels (cf. the contributions in Corak 2004; Björklund et al. 2012). This paper takes a different route. First, a wider set of capitals of different kinds are included, for the individuals as well as for the parents. Second, such intergenerational transmissions are not only, as most often is the case, studied ‘backwards’ – from individuals to parents – but also ‘forwards’ – from the elite individuals to their children, by an analysis of the children's education strategies. How are privileges transferred onto the coming generation? And how is the structure of the elite translated into different modes of elite reproduction? The questions and answers are restricted to the following case: households belonging to the economic elite in contemporary Sweden, a group characterised by strong intergenerational transmissions, as indicated by previous research, differentiating it from the otherwise rather mobile Swedish population (cf. Björklund et al. 2012).

The proposition that can be derived from the research strategy pursued by Piketty, that the overall structure of the distribution of (economic) power can be apprehended by zooming in on groups disproportionally privileged by the distributional pattern, fits very well with Pierre Bourdieu's notion of the social ramifications of what is at stake in what he labels the ‘field of power’ – that is the upper echelons of the class structure, or, in his terms, the ‘social space’. The outcomes of the struggles taking place within this exclusive social arena, unattainable for most people, can influence the relative values of power resources holding currency in society at large and ultimately thereby perpetuate, adjust or transform the social structure (Bourdieu 1996: 264–272).

Due to the relational configuration of Bourdieu's concepts, a comparative perspective is presupposed for grasping the struggles over ‘the dominant principle of domination’ (cf. Wacquant 1993: 20–22; Bourdieu 1996: 265) – a perspective that Piketty, giving precedence to dynamics rather than structure, does not provide. In other words, since the field of power is populated by the holders of different kinds of power (capital), accumulated in various specific fields, the relative power of one kind of elite is impossible to impute by looking at it separately. Although the approach to analyse the structure of the overall field of power never was empirically pursued by Bourdieu himself, it has been employed in subsequent studies of Norway (Hjellbrekke et al. 2007) and of France (Denord et al. 2011). In this paper, however, this comparative appeal is not fully realised, but only sketched, through a concluding comparison with a more ‘cultural’ elite, arguably inhabiting a position in opposition to the economic elite in the field of power. The main focus is, instead, to combine the study of the dynamics of power (Piketty) with an understanding of power as a structured order (Bourdieu).

This could be done by following yet another of Bourdieu's key propositions, according to which the groups populating the different sub-fields of the overall field of power should not be seen as homogeneous, but rather engaged in perpetual internal competition (cf. Bourdieu 1988, 1993).¹ The test case in this paper is the economic elite, the holders of the largest amounts of the dominant capital in the ‘objective hierarchy of forms of capital’ (Bourdieu 1996: 270, 331). How is this elite being reproduced in contemporary Sweden?

From the nineteenth century and onwards, the economic elite has attracted a large sociological interest, providing theoretical points of reference as to what kinds of internal cleavages that could be expected to characterise this elite. Thorstein Veblen, for instance, articulated an opposition that since then has been frequently restated in studies of the economically powerful: the opposition between ‘old’ and ‘new’ money, the former being more prestigious than the latter (Veblen 1994 [1899]: 17–19; cf. McNamee and Miller 1998: 196). Max Weber added invaluable insights to the understanding of the internal structure of economic power by distinguishing different kinds of proprietors: those passively living off the interest of their wealth (rentiers) and those actively working to enlarge it (entrepreneurs) (Weber 2009 [1921]: 181–182). Following Weber, yet another distinction of relevance for envisioning variants of economic elite status could be made, between proprietors (owners; rentiers or entrepreneurs) on the one hand and the top ranks of the propertyless (salaried executives and managers) on the other, an argument otherwise perhaps most attributed to James Burnham, with his writings of a ‘managerial revolution’, transforming corporate rule (Burnham 1941).

These sources of distinction use as their basis the kind of economic capital that individuals are holding (old vs. new; wealth vs. wage) and the use they make of that capital (rentier vs. entrepreneur), instead of employing positional criteria for defining the economic elite. ‘Following the money’ is a different strategy from what most sociological studies of such an elite have pursued when selecting their samples. Usually, as mentioned, the selection is made with reference to the incumbents of formal positions of power in the economic field (e.g. top CEOs). As a consequence, structures of division of the economic elite that have been established in previous studies using positional sampling are less likely to surface in this paper. Bourdieu for instance, focusing his analysis on corporate heads, saw a principal distinction, due to the volume and composition of forms of capital relevant for the group between state and family bosses (Bourdieu 1996: 300–310). When the sample instead is constructed to examine the wealthiest, this fine-grained distinction between different managers might not be as pronounced.

A common feature in elite studies is to examine the recruitment to elite positions, to discern if the incumbents are drawn from ‘all social levels and not solely from the ranks of a few privileged families’ (Baltzell 2009 [1958]: 4; cf. Clark 2014, emphasis in original). Baltzell makes an argument for the need of such an examination against the diagnosis put forth by Burnham. Since Burnham does not investigate the recruitment patterns to the new managerial positions that he highlights, the social ramifications of the process described are inevitably unclear: ‘if both the new managers and the old financiers are recruited from the same upper class, there has not been a corresponding change in the ruling class’ (Baltzell 2009 [1958]: 34, emphasis in original). In other words, the examination of the social origins of the elite is pivotal to grasp the function of the elite groups in the reproduction or refurbishing of power relations. In a study of the economic elite this perhaps becomes particularly important, since economic power to a larger extent than other power resources are inheritable and convertible (Wacquant 1993: 22–23; cf. Bourdieu 1996: 278).

In elite studies, this arrangement, although theoretically oriented towards the future (in order to secure and advance positions), has empirically mostly been studied retrospectively. The most frequent social origins of inhabitants of positions of power and/or educational trajectories leading to these positions are used to assign a modal reproduction strategy to the elite group under study (cf. Hartmann 2000; Hjellbrekke et al. 2007: 262–263; Ellersgaard et al. 2013). This constitutes a major difference in the use of the concept of reproduction strategies from how it has been empirically appropriated within the sociology of education. In such studies, the strategies are theorised prospectively, as (often middle class) plans of action devised to facilitate entry into suitable sites of learning (cf. Ball 2003; Van Zanten 2005; Poupeau et al. 2007; Waters and Brooks 2010). Moreover, in both these veins of research, the focus is mostly on between-group differences in reproduction strategies, and not on intra-elite differences (with Flemmen 2012 as an important exception).

In this paper, a stab is made at extending the interest in patterns of reproduction, recruitment or circulation, by fusing together these common research strategies from sociology of elites and sociology of education. Strategies are studied both retrospectively (the social origins of the members of the elite) and prospectively (the education strategies pursued by the children of the elite). Moreover, the patterns in the latter are analysed with reference to patterns in the former, thereby taking the specific trajectory of different fractions of the economic elite into account. To achieve this, three generations are activated in the analysis: the elite parents, the grandparents and the children of the elite. The paper is consequently a test of a research strategy in which the educational trajectories of the children of the elite are used as indicators of the position-takings made by elite families to advance, or at least preserve, their position and capital assets (cf. Melldahl 2015: 209–250). Hence, two general questions guide this study: (1) What is the structure of acquired and inherited assets within the Swedish economic elite? and (2) How is this structure translated into their children's educational trajectories?

In the following section, first the distribution of the volume and composition of capital among the Swedish economic elite – defined in the manner described below – is analysed and then the correspondence between this distribution and the dispersion of the elite children's properties is investigated.

The economic elite under study in this paper is selected in a particular way that differentiates it from previous research. First, there is a difference in how the elite is identified. Since elites do not come ready-made and are by definition too small to appear in official nomenclatures and sociological class schemes, their identification has to follow different routes than, for instance, analyses of wider social classes. In this paper, Piketty's model has been used to identify the elite. In other words, the economic elite is a group corresponding to the top one per cent in the distribution of economic capital. Hence, the individuals thereby distinguished could hold positions in the economic field in the narrow sense, but just as well be professionally active in any other field. This reflects the overall objective to tie the elite status to the possession of the dominant kind of capital rather than to the occupancy of dominant professional positions, consistent with how Bourdieu describes the prerequisites for belonging to the field of power. The individuals are selected solely since they possess ‘enough specific capital [in this case economic] […] to be able to occupy the dominant positions within their […] fields’ (Wacquant 1993: 21; Bourdieu 1996: 264). Previous studies of elites has most often used two other methods for identifying the elites, both of which characterized by the tendency to overlook individuals with large capital assets. Either to use inclusion in published collective biographies of leading individuals in different social fields – such as the economic field – as indices of elite status (e.g. Lewandowski 1974; Baltzell 2009 [1958]; Denord et al. 2011) or to identify the elite as the individuals occupying some strategically selected formal positions and gather information about them from various sources (e.g. Stanworth and Giddens 1974; Bourdieu 1996: 300–335; for a study combining positional and distributional criteria, see Flemmen 2012).

Second, there is a difference in the construction of the research object, due to the design of the study: households are examined instead of individuals and they are selected using a specific reference population. In Piketty's work, the group comprising the top one per cent is equivalent to the individuals receiving the highest incomes. Such a selection would, however, be problematic, given the sociological purposes of this study. Since the design, as mentioned above, is to analyse reproduction of the economic elite both retrospectively and prospectively, it becomes necessary to ascertain a basic comparability in these analyses. When defined in relation to the entire wage-earning or wealth-holding population, the top group can be of highly varying ages. Consequently, the adequate point in time to study their parents’ and their children's properties would differ, making the educational trajectories very difficult to disentangle, due to the ever-changing supply of positions within the educational landscape.

The solution sought for the present purposes is a way of holding certain things constant. Since the analysis of the social reproduction expressed through the children's education strategies is of vital importance, this generation has been assigned the deciding role. To facilitate comparisons, only children born in a certain year, 1983, are included, or, more precisely, households with children born in this year. This birth cohort had roughly the same supply of education to choose from, making the meaning of their respective choices more comparable. Out of all children born in Sweden in 1983, the children with the richest parents were selected. In this manoeuvre, Piketty's distributional model was followed, although slightly modified. If the parents combined (taxable) wealth or wages qualified them among the top 0.5 per cent in the distribution of either wealth or wages, the household was defined as economic elite, a measure resulting in a final population of 869 households.² For this population, key information on the individual properties of the three generations has been added from various Statistics Sweden registers, the generations being linked to each other using the Swedish identification number and the register of family relations.

The assets and characteristics of three generations are activated in the analysis: the grandparents (the first generation), the (elite) parents (the second generation) and the children (the third generation). Since the parents’ assets are the ones that qualify the households as economic elite, they are the reference point in the construction. Their own assets are defined as their acquired capital, and the assets of the first generation are defined as their inherited capital. When analysed in relation to each other, different configurations of ‘old’ and ‘new’ (in a wider sense than just restricted to money) can surface. Into the space constructed by the distribution of these acquired and inherited assets, the properties of the children (the third generation) are projected, in order to analyse how the structure is translated into different modes of reproduction.

The space is constructed using specific Multiple Correspondence Analysis (MCA) (Le Roux and Rouanet 2004, 2010) and the 16 active variables and their category values (51 active categories) are displayed in Table 1.

The first two headings in Table 1 present different forms of capital, in Bourdieu's sense. Since the group examined is defined by their amount of economic capital, several measures regarding the form of economic capital has been included – most nuanced for the second generation (the parents). A first distinction has been made between wealth and current incomes. For the first generation, it was only possible to distinguish between wealth and wages. For the second generation, however, the wealth has been further specified. Solid assets (real estate, agricultural property, land, etc.) are distinguished from liquid assets (stocks and bonds, bank holdings, taxable insurances, etc.). The solvency is also assessed, by measuring whether the liquid assets are smaller or larger than the sum of the debts – the latter signalling significant labour market independence, since all debts could be resolved using available means were the mortgage costs to become overly burdensome, without being forced to liquidise any solid assets. For this generation, the income variables also include capital incomes (the sums of these are combined in the ‘income’ variable, while the ‘income balance’ variable accounts for their internal hierarchy). For both generations, the level of education has been included, as a measure of the holdings of a different – theoretically opposing – kind of capital (cf. Wacquant 1993: 23), further nuancing their capital portfolios.³

Last, to accommodate the circumstance that in this elite, defined in the manner described above, not all members are in ‘positions of command’ (Scott 1996: 40–42) as is the case when business executives are examined (especially since whole households are highlighted in this paper), some information on the parents’ relations to the labour market has been included. Such differences in what they do, in other words, complement the information on what they have. First, their professional status – whether they are employers or employees (cf. owners vs. managers) – has been included. Second, a classification of their affiliations to different spheres of the capitalist economy – developed by Göran Therborn (1981), although somewhat modified – has been incorporated.⁴

The correspondence analysis yields three axes relevant for interpretation (see Table A1 in appendix for eigenvalues and modified rates, and Figures A1 in appendix for the clouds of individuals in the factorial plane of the first two axes). In the following, only the first two will be discussed since they are most clearly related to differences in the distribution of the children's properties. In the subsequent figures, as given in Table 1, the first set of characters in the value labels refers to the generation in focus (GP – grandparents; P – parents; some information regarding the parents is divided between F – father and M – mother) and the second set – when necessary – indicates the variable (for instance LA – liquid assets; E – education). Categories capturing information related to the second generation (parents) are black in the figures and categories regarding the first generation (grandparents) are grey.

Not surprisingly, the volume of inherited wealth to some degree corresponds to this opposition, where grandparents without any wealth are found to the right (GP-W: None) and grandparents with the top one per cent wealth are found to the left (GP-W: Top1). The pole to the left is in other words to a greater extent than the pole to the right to be seen as holding old money. Resort to education among elite members of the second generation also differs between the two poles, where mothers and fathers with low education (elementary or vocational upper secondary education – P(F)-E: Low; P(M)-E: Low) are associated with the other categories to the left and fathers with high education (longer university studies – P(F)-E: High) located to the right. Substantial educational investments are more important for entering the economic elite at the high-earning pole than the pole marked by great wealth.

The indicators of labour market positions that contribute to the structure reveal the parts of the capitalist economy that are open for – or produce – new elites, without large amounts of inherited means. Apparently, the self-made economic elite at the bottom (i.e. the entrepreneurs, actively engaged in increasing their capital) is in this respect really self-made. And the productive sector, generating surplus value, that Therborn labels the core of the capitalist economy, is according to the pattern also related to the bottom half (P(M): Prod.sect; P(F): Prod.sect), marked by smaller volumes of inherited and acquired capital, which puts the sector in a clear opposition to the reproductive sector (P(F): Repr.sect) that seems to harbour the established elite (a circumstance that could constitute a legitimacy problem for a state that for the most part of the twentieth century was ruled by social democrats, since the state during this century appropriated most of the reproductive functions). The structure of the space in these two axes bears a close resemblance to that recently disclosed for wealthy business executives in Norway, although the relation between the axes are inversed in that analysis with inherited capital primarily related to the first axis and the composition of capital related to the second (Flemmen 2012).

While the relation between the two elite fractions on the right in the figure expresses hierarchical distances in the labour market, the two fractions to the left are clearly the two propertied groups described by Weber: the rentiers (the old wealth elite – in this group the largest incomes are to be found when wages and capital incomes are combined) and the entrepreneurs (the new wealth elite). At the top right, a fraction much resembling the State nobility described by Pierre Bourdieu is located: Highly educated, receiving high incomes at – judging by their incomes – dominant positions predominantly within the reproductive sector working through, or sanctioned by, the state (cf. Bourdieu 1996: 116–117). The bottom right quadrant is populated by a fraction less easily labelled, built up by similarly leading functionaries, but in this case employed in the private sector, originating from less privileged backgrounds. In the two fractions located in the top of Figure 3 the tendency of social reproduction is clear: the acquired capital defining the households of the second generation very much resembles the inherited capital, held by the first generation.

Then what about the children, the third generation in the analysis? In what way is the distribution of their properties related to the internal differentiations of the Swedish economic elite, described above? In the following, this will be discussed. First, their basic properties is described, displaying their educational capital accumulated in upper secondary education and their own financial means, and, second, their major investments in Swedish higher education is revealed (i.e. the educational programme where they spent the most time).

Three observations regarding the distribution of educational choices should be made. First, the educational programmes located at the centre, or near the centre of the space are either typical for the entire group or non-distinctive for any specific sub-group – in the plane of the first two axes. The least distinctive educational choice in this population is Economics (in which all economic tracks, such as business studies, economics, finance and accounting have been included – at all Swedish universities except the Stockholm School of Economics (SSE), discussed separately in the following). This category is also the singular most common choice made by the children of the economic elite (a total of 147 – or 16.9 per cent – of the children of the elite choose this field of study, see Table A4 in appendix for the absolute and relative frequencies of the educational choices plotted in the figure). Although Economics in this wide sense is a frequent choice also among other students within Swedish higher education, the corresponding representation at these kinds of educations among individuals born the same year as the elite children (1983) is considerably lower (5.2 per cent). This heightened and unison interest for studies towards the economic domains thus points to the fact that the well-endowed households under study have vested interests in the functioning of the economic field.

Second, the choice to altogether abstain from entering Swedish higher education (at least before 27 years of age) does not occupy a similarly neutral position in this constructed space. Instead, it is a token of a relative lack of seniority in the elite (Not in HE is clearly positioned downwards in Figure 5, among the newcomers – an urgent question for future research is hence to discern what these children do instead of seeking formal higher education in Sweden and what kinds of positions they eventually secure without such education). In other words, higher education does not serve a compensatory function in this group (as when less privileged individuals intensively invest in an activity to catch up in the competition). Third, some specific – more active – educational choices are related to certain configurations of the households’ capital structure.

The dispersion of educational choices over the second axis, mapping seniority in the elite, roughly corresponds to the social hierarchies in the Swedish field of higher education in general, with highly selective and prestigious education such as the SSE and the medical programmes (Med. School) at the top – together with studies in Art/Culture – and less selective education such as teacher training (Teaching) is located further down in the space (cf. Börjesson 2005: 480–486). In this matter, the children of the economic elite are not that different from less wealthy children. However, along the first axis that separates households with large wealth from households dependent on wage incomes, a more elite-specific pattern emerges.

The dispersion over the first axis shows the distance in this space between on the one hand highly selective professional programmes (the mentioned SSE and Med. School but also Law and Journalism) and substantially less selective but traditional academic studies (the social and natural sciences – SocSci and NatSci) (to the right) and on the other hand both one particular professional programme (Agriculture) and also to some extent less selective programmes cultivating a more general cultural capital (history and philosophy – Hist./Phil. – and Teaching).

The pattern in fact only assigns one out of the four labelled fractions a distinctive education strategy: the children of the State Nobility overly invest at highly selective and competitive programmes, especially the Medical Schools, but also the SSE. This is illustrative of their relative position in the space of the economic elite, a position that appears to be populated by dynasties of high-earning professionals that owe their position to substantial acquired educational merits. The extreme position of the Agricult. category on the other side of the axis could also be seen as distinctive – although non-assignable to either of the two propertied groups – if this realised strategy is interpreted not as a school-mediated reproduction strategy, but rather as an expression of a ‘family mode of reproduction’, in which all strategies are ‘subordinated to the strictly economic strategies aimed at ensuring the reproduction of its economic capital’ (Bourdieu 1996: 279). The investment in studies in agricultural (and forestry) science could be necessary for acquiring suitable competencies to successfully manage inherited landed estates – and hence preserve them for future generations (cf. af Kleen 2009). Beside this partial anomaly, the lack of any ‘rational’ investment in the left part of the space – especially in the richest upper left quadrant – comparative to the ones made in the upper right quadrant, reveals a great deal about the ‘freedom from economic necessity’ that large fortunes bequeaths to its holders. Their wealth is of such a magnitude that any modal (education) strategies are neither necessary nor realised. The hope assigned to principles of meritocracy among the children of the State nobility is contradicted by the freedom among the children of the rentiers that is awarded by their patrimony.

Altogether, this shows that the education strategies (or the lack thereof) pursued by the children of the elite are clearly related to their origin in different segments of this elite. No indicators of formal qualifications of the children were utilised as active elements in the analysis. Although acquired competences and not inherited assets should decide the entry to formally meritocratic institutions, their positions in higher education are not distributed randomly in this constructed space, emphasising the validity of treating these education strategies as related to the family and not to the individual.

Different kinds of resources should give rise to different ways in which their holders seek to transmit them (cf. Bourdieu 1996: 272). So far, this hypothesis has been tested solely within a delimited elite group. However, Bourdieu's concept of the field of power is, as mentioned, relational. The education strategies of the economic elite should, therefore, be contrasted with the strategies pursued by different, relationally opposed elites.

To test what such an approach could generate, Table 2 compares – in a crude manner – the economic elite's relation to higher education with that of an academic elite's (composed of all children, born the same year as the children of the economic elite into families where at least one parent held a PhD degree – resulting in a group of roughly the same size as the discussed economic elite, 862 households) and with all children born in the same year (Melldahl 2015: 211–213). Three sets of figures are compared: the proportion of the children enrolling in higher education (before 27 years of age) and the enrolment rates at the most selective educational institutions, the SSE and the Medical programmes (the rates at these selective institutions standardised to per thousand, to enhance the comparability between the groups).

As seen in the table, children from both elites are vastly overrepresented at both these institutions, but in a diverging manner. Even though the children of the academic elite are a rather common sight at the SSE, the children of the economic elite far exceed their representation by being more than 10 times overrepresented (1105 per cent). In the same way, the children of the academic elite at the SSE outnumber the children of the economic elite at the medical schools: 1 in 10 end up at such an institution, as compared to 1 in 80 among all children born the same year (an overrepresentation equalling almost 750 per cent). There are, apparently, differences in how different elites make use of the education system, related to their principal source of social power.

But the table also provides a certain urgent sense of proportions. Children of the elites evidently enter higher education in far greater proportion than ‘average’ children of the same age. For the academic elite, this eagerness is understandable. Bourdieu argues that an agent's eagerness to invest in education is related to ‘the degree to which her social success depends on her academic success’ (Bourdieu 1996: 276), which would be especially true for a group highly dependent on the education system. But it is noteworthy that children from households extremely rich in the dominant form of capital in capitalist societies (the children of the economic elite) also deem it necessary, or relevant, to engage in higher studies in such great numbers.

The dramatic overrepresentation of the elite children at the two institutions are safeguarded by their numerus clausus. Each year a restricted number of places open at these institutions, something that further accentuates their substantial relative presence. This can be illustrated by the following: Out of all individuals born in 1983, only 251 of them made their principal educational investment at the SSE. Twenty six of these were children of the economic elite (see Table A4 in appendix).

This paper has sought to combine elements from Piketty's analysis of economic power with Bourdieu's analysis of social reproduction, in a study of the social trajectories of different fractions of the Swedish economic elite, composed by the households richest in economic capital. Although Piketty's analysis is unsatisfactory from a sociological perspective, as economic power is treated solely as a distributional entity, the model he has devised for locating the holders of such power solves a practical problem in identifying the relevant agents in what Bourdieu labels ‘the field of power’. Through this model, power is untied from the occupancy of certain formal positions and instead defined by the possession of ‘enough specific capital’. In this way, a sociological perspective is added to Piketty's economic analysis and an economic dimension is added to Bourdieu's sociological concepts.

The correspondence analysis of the distribution of capitals and properties within the top wealthiest one per cent in Sweden showed that this distributional group is marked by strong internal differences. The analysis revealed a structure of oppositions along two main dimensions. The first was related to differences in the composition of economic capital. Here, an opposition between holders of great wealth and receivers of large wage emerged, the latter more indebted both to the financial market and to the education system and hence much more dependent on outcomes from the labour market for maintaining their position. This effectively illustrates the power balance between the propertied and the propertyless, as articulated by Weber.

By combining a sociology of elites with a sociology of education, the intergenerational transfer of privileges was examined over three generations. If such privileges are transferred, the elite could over time arguably develop into a class (cf. Baltzell 2009 [1958]: 6–7). The results indicate that there is a substantial element of intergenerational transmission to – and from – the economic elite (as hinted at by Björklund et al. 2012). However, this transmission followed certain patterns. Four different fractions of the elite were identified: Rentiers, Entrepreneurs, State nobility and Salaried managers. The last two groups depended on incomes and the first two held greater wealth. Both the Rentiers and the State Nobility were to a large extent populated by households characterised by ‘old’ capital, originating from very prosperous backgrounds, but separated by the principal channel of social reproduction – inherited wealth in the first case, intense educational investments in the latter.

The distribution of capital in the third generation indicated a large degree of elite-specific social reproduction. The children of the State Nobility were on average more highly appreciated by the evaluative functions of the education system and the children of the Rentiers were the richest in monetary terms. When looking at this group, the ‘rentier society’ has not returned but has rather been in place for some time, effectively transferring privileges across generations, reproducing both ‘economic’ and ‘educational’ dynasties.

That there are strong ties between the economic elite and the economic field was made evident by the fact that studies in Economics was the most common education strategy pursued by the children of this elite and by their massive overrepresentation at the highly selective SSE. This pattern is consistent with much research in sociology of education investigating differences in choices of field of study that repeatedly has identified similarities between the choices made by the individuals and by their parents (cf. van de Werfhorst et al. 2001; Helland 2006; Hällsten 2010).

When focusing elites instead of representative samples, this adds a valuable explanatory component to the somewhat ambiguous concept of ‘elite education’. By entrusting certain educational institutions with the erudition of their children, the elite – in this case the economic elite – confirm the institutions’ positions within the hierarchy of competing educational destinations. This provides the institutions with symbolic capital, further enhancing their reputation and sorting function. Yet, the pursuit of such elite education among the children of the established economic elite was evident for the children of the fractions dependent on educational merits for their position. The children of the dominant fraction of the largest holders of the dominant form of capital – the Rentiers – do not need to follow in such augmented numbers this scholastically competitive path to reproduce their elevated social position. Their freedom from economic necessity bequeaths them with a greater freedom of choice within the supply of Swedish higher education.

No potential conflict of interest was reported by the author.

Andreas Melldahl is a senior lecturer in Sociology of Education at the Department of Education, Uppsala University, and a researcher at Stockholm School of Economics Institute for Research and at Sociology of Education and Culture (SEC), Uppsala University. He has previously published research on contemporary and historical elites, especially cultural and economic elites. His doctoral thesis (2015) was devoted to changes in the distribution of education in Sweden during the twentieth century and the consequences for different social generations and social groups.

Andreas Melldahlhttp://orcid.org/0000-0002-0100-9934