Table of Contents
Table of Figures
Chapter 1 : Introduction
Chapter 2 : Human Capital, Social Capital and Cultural Capital – A Theoretical Discussion and Integration of the Different Concepts
2.1 The Human Capital Concept of Gary Becker
2.2 Pierre Bourdieu’s “Cultural Capital”
2.3 Pierre Bourdieu’s “Social Capital”
2.4 “Social Capital” in the Sociology of Gary Becker and James Coleman
2.5 Integration of the Concepts of Bourdieu, Becker and Coleman – The “Resource Base”
Chapter 3 : The Resource Base and Youth Transitions of Offspring of Migrants and Natives in Germany – An Empirical Comparison
3.1 Methodological Preliminaries
3.1.1 Description of the Data Set Used
3.1.2 Description of the Sample
3.2 Setting Up The Empirical Hypotheses
3.3 Comparing Transitional Patterns Between Groups With Different Parental Migrant Backgrounds.
3.4 Defining Transitional Clusters
3.5 Some Interim Conclusions
3.6 Conceptualising Labour Market Success
3.6.1 ‘Linearity’ of Transitions
3.6.2 Some More Interim Conclusions
3.6.3 Job Security and Satisfaction as Measures of Success
3.7 Youth Transitions and the Composition of the Resource Base
3.7.1 Personal Capital I – Educational Qualifications and Job Experience
3.7.2 Personal Capital II – Personal Agency and Language Proficiency
3.7.3 Social Capital
Chapter 4 : Conclusion: How the Resource Base Model Can Help Identify Discrimination
Appendix A – Concurrent Statuses at Equal Time Points
Appendix B – Main Statuses in May
Appendix C – Educational Achievement at Age 23
Appendix D – Membership of Transition Clusters by Gender and Migrant Group
Appendix E – Lifecourse Charts by Group and Cluster
Appendix F – Logistic Regression on Labour Market Success – Model 2
Appendix G – Transition Cluster and Language Proficiency
Table of Figures
Figure 1 The Resource Base Model
Figure 2 Conceptualising Linear and Non-linear Transitions
Table 1 Main Status by Average Age and Migrant Background of Parents
Table 2 Mean Number of Months Spent in Each Status by Cluster
Figure 3 Clusters with 'Track Relation' by Group (Without 'Domestic' Cluster)
Table 3 Linearity of Transition by Group
Table 4 People Having Made Successful Transitions by Age 23
Table 5 Odd Ratios of Logistic Regressions on Labour Market Success
Table 6 Formal Educational Qualifications by Cluster
Table A-1 Frequencies of all Existing Patterns of Status Concurrences.
Table A-2 Priorities of Statuses in the Case of Concurrent Events
Figure B-1 Main Status by Average Age and Migrant Background of Parents
Table C-1 Highest Educational Achievement at Age 23 (CASMIN) by Group
Table D-1 Membership of Transition Clusters by Migrant Background (all)
Table D-2 Membership of Transition Clusters by Migrant Background (females)
Table D-3 Membership of Transition Clusters by Migrant Background (males)
Figure E-1 Lifecourse Chart by Cluster (Descendants of Germans)
Figure E-2 Lifecourse Chart by Cluster (Descendants of Turkish Migrants)
Figure E-3 Lifecourse Chart by Cluster (Descendants of 'Other' Migrants)
Table F-1 Odd Ratios of Logistic Regressions on Labour Market Success
Table G-1 Spoken Language Proficiency by Cluster
Table G-2 Written Language Proficiency by Cluster
Based on results from previous empirical studies, which highlighted the relatively disadvantaged situation of descendents of Turkish migrants on the labour market in Germany (compared to descendents of natives and migrants from other countries of origin), this dissertation aims to gain further insights in the underlying reasons for these disadvantages. It aims to do so by first introducing a new concept - called the 'Resource Base' - which is derived from elements of Bourdieu's work on cultural and social capital as well as Coleman's and Becker's notes on human capital and social capital.
This concept is then used to investigate potential causes for the disadvantageous labour market positions of descendents of Turkish migrants by focusing on their transitions from school to the labour market and the impact of these transitions on their individual Resource Base, as well as the impact of the Resource Base on labour market success. This investigation employs quantitative methods (logistic regression, cluster analysis and chi square tests) and a quantitative data set - the German Social Economic Panel survey.
Results include detailed insights into the differential acquisition of relevant resources by each group (descendents of migrants from Turkey and other countries, as well as native Germans) and the negative impact of the lack of these resources on labour market success. It is further suggested that the Resource Base model can be used to identify discrimination regulations and practices, advancing the model as a useful theoretical and analytical tool that justifies further research in this area.
First, I want to express my sincere gratitude to Dr. Robert Gibb whose supervision of this dissertation greatly contributed to the success of this endeavour.
Furthermore, I received invaluable help from Rena Kim Bivens, who (being a native of a language closely related to English) contributed greatly to making the English of this dissertation more readable (a task that sometimes - I am sure - dwarfed the labours of Hercules).
Finally, I want to thank the country of Scotland, the people I met here and the City and University of Glasgow for providing me with a dear and friendly home and some intellectual challenges during the last year.
Chapter 1 : Introduction
Migrants in Germany are regarded as a problematic group in the labour market - unemployment figures are much higher for migrants than 'native' Germans (Bundesamt 2004:115). These groups are also disadvantaged by their greater likelihood of achieving lower educational qualifications, receiving lower wages and working in unhealthy job environments (Hradil 2001: 345). Not surprisingly, their children also face difficulties - they drop out of school and vocational training twice as often as their German peers (Solga 2002: 6) and are more often pupils of special schools (Powell and Wagner 2001). These children find themselves in a challenging situation since they are simultaneously members of two of the most problematic segments of the labour market - they are both 'foreigners' and young people (Bundesamt 2004:115).
Being a migrant or having migrant parents is, therefore, an important determinant of social inequality in Germany. A rich volume of literature investigates the disadvantaged labour market situations of migrants (Konietzka and Seibert 2003), as well as issues of assimilation and integration (Kalter and Granato 2002). In recent years, the specific issue of 'the second generation' - individuals born in Germany as children of migrant parents - was increasingly addressed as a means of assessing the degree to which descendants of migrants are able to integrate into German society. Moreover, these investigations have attempted to seek out explanations for the phenomenon that 'second generation migrants', despite being born, socialised and educated in Germany, were usually less successful than their German peers in school, vocational training and the labour market (Konietzka and Seibert 2003; Worbs 2003).
Previous research concluded that the educational and social background of these children's parents, who largely arrived in Germany as guest workers from rural areas with only very basic educational opportunities (Crul and Vermeulen 2003), was the main reason for their comparatively worse situation1. While these insights have explained much of the relative disadvantages experienced by offspring of migrants from some countries of origin, these explanations are simply inadequate for our understanding of how the largest group in Germany is affected - the descendants of migrants from Turkey (Granato and Kalter 2001; Worbs 2003).
Although there have been a few further attempts at uncovering the reasons that lead to this exceptional situation experienced by descendants from Turkish migrants (particularly in comparison to offspring of migrants from other countries) (Haug 2002; Kalter 2004), satisfactory answers are still missing.
This dissertation will contribute to filling this gap, advancing current understanding of this phenomenon. To achieve this aim, a theoretical framework will be outlined - more specifically, a multidimensional Resource Base consisting of educational, motivational, economic, social and other resources will be presented. This model attempts to 'synthesise' some components of Bourdieu's concepts of cultural and social capital with the notion of human capital and social capital in the work of Becker and Coleman.
This theoretical foundation is subsequently employed in order to assess the usefulness of this strategy when attempting to gain further insight into the disadvantaged situation of descendants of Turkish migrants. This investigation will make use of a panel data set (the German Socio-Economic Panel Study) and quantitative methodological analysis as a means of following youth transitions from school to the labour market (age 16 to 23) for three different groups: descendants of Germans, of Turkish parents and of parents from 'other' countries of origin.
The studies mentioned above attempt to explain differential labour market success with differences in the individual's educational or social status. The following analysis is distinct from these approaches since it focuses instead on the processes (youth transitions) within which these differences arise. Using this alternative approach is expected to achieve new insights into this area of research while the multidimensionality of the Resource Base model is expected to foster a more holistic perspective of disadvantageous positions.
Chapter 2 : Human Capital, Social Capital and Cultural Capital - A Theoretical Discussion and Integration of the Different Concepts
When analysing and attempting to comprehend inequalities in the labour market, a range of different theories are at one's disposal2.
One of these theories is Becker's (1964: 244) human capital theory. In the decades that have passed between the first publication of his ideas and today, his theory has definitely proved valuable when attempting to explain certain types of inequality, but it has also been criticised for not taking into account many important causal factors. Empirical material quoted in the previous chapter supports both the usefulness of Becker's approach and these valid criticisms.
Hence, it is worthwhile to provide an introduction to Becker's theory in this chapter, paying particular attention to its weaknesses. In an attempt to solve (at least partially) some of these weaknesses that exist when using the concept of human capital to explain inequalities within the labour market, Becker introduced the concept of social capital in his later work (e.g. Becker 1996). A concept with the same name and some common features had already been put forward by Pierre Bourdieu (1983)3 and Becker's colleague James Coleman (1988). Other authors (e.g. Putnam 1995) and institutions (e.g. The World Bank; cf., for example, Grootaert et al. 2004) also later made use of the social capital concept for their own purposes. This multitude of different sources and their very different ideological background suggests that the concept has been utilised in very different ways. Following this introduction, the next step for this chapter will therefore be to briefly examine and critically discuss the differences of the aforementioned concepts of social capital after having compared Becker's concept of human capital with the 'cultural capital' of Bourdieu. A modified concept of social capital, which incorporates the strengths of each different approach and abandons their weaknesses, will then be proposed in order to combine it with the other forms of capital suggested by the authors mentioned, to model a concept of a multidimensional resource base. This model will be used in the subsequent chapters to assess its feasibility and explanatory potential for ethnic disadvantages on the German labour market.
2.1 The Human Capital Concept of Gary Becker
The human capital theory was first established in the early sixties4 as a result of the Sputnik crisis. It centres around the idea that economic growth should involve not only investments in real goods (i.e. real means of production), but also investments in human capital - that is, the general education and the specific skills of the labour force (Hradil 2001: 74). However, this idea was later extended to include more general factors that influence the contributions humans can make to the production process - for instance, their general health, and values such as punctuality and honesty etc. (Becker 2002). The name human capital originates from this investment point of view.
The human capital theory assumes, akin to any other theory which utilises an economic approach to analyse human behaviour, that individuals choose from a set of alternative actions at their disposal based on quasi-rational decisions5 about their investments in their own human capital. Decisions that consider 'investment' include choices surrounding (amongst others) education issues (such as continuing education in order to obtain better credentials as opposed to leaving school with only a basic formal qualification or going to university instead of entering the labour force directly after school) as well as health issues (considering investments of time or money in health). According to human capital theory, these decisions carry with them on one hand certain costs and on the other certain gains or utility, and the result of quasiweighing these two against each other leads to the actual decision - the alternative for which the gain is highest in relation to the loss will be chosen.
The costs can involve the direct costs of university education in the form of tuition fees or the bill for dental treatment, for example, but, crucially, non-monetary costs can also exist6. Within the realm of human capital theory, one of the most important examples used is the decision to extend educational participation. In these scenarios, the opportunity costs people take into consideration when making human capital investment decisions are first of all the wages the individual would expect to earn if they joined the labour force instead7.
The gains of such decisions are mainly described as the chances of higher wages and a higher probability of finding a job in the first place. In fact, rich empirical evidence supports a strong relationship between (formal) educational achievement and income8, but it is also commonly stressed that not only do the chances of higher monetary returns influence the decision to extend education but non-monetary and cultural returns are also important factors; for example, the belief that it would provide access to more interesting and more creative work or some idea of self-realisation. These reasons, however, prove much more complicated to measure and operationalise than monetary ones; but this is not to suggest that they are less important - on the contrary, as will be argued later, it can be assumed that these non-monetary returns are very important indeed and depend very much on socialisation and family background and are, therefore, likely to be one factor that makes "class differentials in educational attainment [...] display a high degree of stability" (Breen and Goldthorpe 2000:182).
Another important characteristic of human capital (and a further reason for the name of the term) is that it cannot be separated from its bearer. This has two important implications. First of all, the 'production' of human capital (in an individual sense) cannot be delegated to someone else; the improvement of one's own health and education, for example, inevitably involves one's own effort. Secondly, due to the fact that it cannot be separated from its bearer, companies that want to use the skills of potential employees cannot just 'buy' the human capital from them; individuals who want to share the human capital of another person have to make an effort themselves when learning the other's skills.
In addition to the described notion that centres around single individuals' investment decisions and human capital accumulation, Becker, and others who also make use of the concept, use the term to refer to a characteristic of groups as nations or companies - Becker, for example, writes: "better accounting methods would include the specific human capital of employees among the principle assets of most companies" (Becker 1997: 44). This view of the concept appears to be the main reason for criticism of the term, as many argue that this reduces people to the level of machines and to a monetary and economic quantity - a view which indeed could be regarded to a certain degree as misanthropic and ethically questionable9. For economists like Becker, the process involved in the individual human capital investment decision is an analytical step that explains why certain groups (developing vs. developed countries; different ethnic groups within the same society) possess more or less human capital. However, since a further discussion of these objections would exceed the scope of this paper, it should instead be stressed that for the purpose of this paper this aspect of the human capital theory will be neglected and only the individual human capital will be the focus of the analysis.
One of the critics of the economists' human capital theory was French sociologist Pierre Bourdieu, who can conceivably fall under a Marxist tradition10 and therefore tends to use differences in possessions to explain the differences between the classes. In his analysis, he introduces the concept of 'cultural capital' - a concept that is, soon to be argued, not too different from the human capital concept outlined above.
2.2 Pierre Bourdieu's "Cultural Capital"
For Pierre Bourdieu (1986: 242), "the structure of the distribution of ... capital ... represents the immanent structure of the social world". This view appears closely related to the Marxist view that the possession of the means of production (in other words, economic capital) determines the class situation of individuals. However, Bourdieu extended this Marxist concept by introducing another dimension since he realised that the use of economic capital, which was sufficient in the early stages of industrialisation, did not sufficiently explain differences of the 20th century anymore. The highly skilled managers or professionals, for example, did indeed pose a problem for the classical Marxist class theory as it seems implausible to classify them as belonging to the proletariat since the exploitation assumption did not seem to apply for those groups; but they were in fact mostly dependent employees without possession of means of production and, hence, not capitalists either.
Bourdieu (first in Bourdieu 1984)11 therefore introduced the concept of cultural capital. In a later publication he describes his main concept as follows:
Cultural capital can exist in three forms: in the embodied state, i.e., in the form of longlasting dispositions of the mind and body; in the objectified state, in the form of cultural goods (pictures, books, dictionaries, instruments, machines, etc.), which are the trace or realization of theories or critiques of these theories, problematics, etc.; and in the institutionalized state, a form of objectification which must be set apart because, as will be seen in the case of educational qualifications, it confers entirely original properties on the cultural capital which it is presumed to guarantee. (Bourdieu 1986: 243, his italics)
By adding this capital form to the class concept, Bourdieu no longer encounters major difficulties when explaining why the situation of highly skilled managers or professionals is indeed very different from the proletarians, on one hand, and the capitalists as well; the concept of objectified cultural capital (which will not be addressed in further detail here) even allows Bourdieu to account for the very special relationship of mutual dependence between the capitalists (who have the economic capital to buy the cultural goods, like very specialised instruments and machines, but not the skills to use them) and the highly skilled professionals (who have the skills, but not the economic means to purchase the cultural goods).12
The common feature of Bourdieu's cultural capital13 and the human capital of the economists is the fact that it cannot be separated from its bearer, an idea which Bourdieu expresses by describing it as 'embodied'. He also stresses the fact that investments in this type of capital can be regarded as sd/-improvement and, hence, "the work of acquisition is ... an effort that presupposes a personal cost" (Bourdieu 1986: 244). Bourdieu also tends to agree with the economists' claim that an important part of these costs are opportunity costs: "It follows that the least inexact of all the measurements of cultural capital are those which take as their standard the length of acquisition" (ibid.).
However, he stresses that this measure will fail if it only takes into account the time spent in formal education. He argues that the acquisition of cultural capital has often already started in the first years of life. This contributes to the relative disadvantage of the offspring of the lower classes since the "accumulation of ... cultural capital starts without delay ... only for the offspring of families endowed with strong cultural capital" (Bourdieu 1986: 246). This mechanism leads to early advantages which can hardly be made up for during the time of formal education and therefore further entrench societal class structures. Another mechanism which has a similar effect stems from the economic capital of the family: in line with the argument of the economists, Bourdieu states that, of course, only families with sufficient economic capital can provide their offspring with enough time to prolong their (formal) "education beyond the minimum necessary for the reproduction of the laborpower least valorized at a given moment" (Bourdieu 1986: 245).
Bourdieu's main criticism of the economic view of human capital is that it only accounts for the direct human capital investment decision and fails to take systematically into account the differences in "domestic transmission of cultural capital" between classes. Bourdieu argues that the offspring of higher classes will have a head start in terms of cultural/human capital acquisition since it is more likely that their acquisition process started very early, at the beginning of their socialisation process. For Bourdieu this is the main mechanism behind the intergenerational stability of social structure14.
In addition to the embodied cultural capital, Bourdieu regards the institutionalised cultural capital as important. Here he is mainly referring to cultural capital in the form of certificates of educational achievement that can generally be regarded as certifying a certain amount of a specific type of cultural capital. Bourdieu sees two main problems with this. First, he stresses that the institutionalised cultural capital remains stable at all times, although the corresponding embodied cultural capital might decrease due to the biological limits of its bearer (for example, memory loss). And secondly he points out that the "performative magic of the power of instituting" produces out of "infinitesimal differences ... lasting differences, such as that which separates the last successful candidate from the first unsuccessful one" (Bourdieu 1986: 248). For empirical analysis, however, institutionalised cultural capital offers an easy and straightforward means of measuring the amount of cultural capital a person has at their disposal. The problem with this approach to measure cultural capital is that it only measures a small share of total cultural capital. Therefore, if cultural capital is to be used as an explanatory factor in the analysis of differences in, for example, the labour market, some further measures must be found to account for the share of cultural capital which is not guaranteed through formalised educational qualifications.
In Distinction, Bourdieu (1984) gives rich and vivid examples of how such measurement could be achieved.
2.3 Pierre Bourdieu's "Social Capital"
In addition to economic and cultural capital, Bourdieu attaches much importance to social capital as well, extending the analytical instrument used to investigate class differences to a space, which is three-dimensional in two senses. In the first sense it is, of course, identified by the three forms of capital and in the second, "by volume of capital, composition of capital, and changes in these two properties over time" (Bourdieu 1984: 114).
Social capital is the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance and recognition - or in other words, to membership in a group - which provides each of its members with the backing of the collectivity- owned capital, a "credential" which entitles them to credit, in the various senses of the word. (Bourdieu 1986: 248f.)
This definition strongly emphasises the aspect of group membership and in fact, all the examples Bourdieu discusses in this text relate to the membership of some more or less formal group. It remains, however, not entirely clear whether he conceptualises social capital as the capital of an individual or that of a group. The above definition supports both views: in the first half, he equates social capital with the value of the capital that agents can use through their social network; in the second half he speaks about "collectivity-owned capital". In his subsequent discussion, he often refers to social capital as group-owned, although his argument that differences in social capital can explain the differences in life chances of members of different classes suggests that the social capital must be regarded as the property of individuals.
In general, social capital has some characteristics in common with cultural capital: it cannot be separated from its bearer and the acquisition (i.e. the establishment of ties within the network in question) must be carried out by the bearer themselves. Furthermore, there is one other important way that class structure remains stable over time: since very important 'knots' in an individual's network certainly consist of their close relatives, the networks of, say, one's parents automatically represent a part of the networks of their children, who, thus, as it were, inherit the social capital.
The measurement of social capital in that sense is a very difficult exercise as it - according to Bourdieu - depends on both the width of the network and of the 'value' of each member of the network. The value of a member of the network is a function of the extent to which it is possible to achieve an aim with the use of his or her economic or cultural capital (and, thus, one's own social capital) which would not have been possible to achieve with the sole use of one's own economic and cultural capital.
Investments in social capital are made in such a way that ties are established based on "obligations subjectively felt (feelings of gratitude, respect, friendship etc.) or institutionally guaranteed" (Bourdieu 1986: 249f). Often these ties are reciprocal, in which case the very use of the social capital might increase their strength, as the ties grow tighter.
Some authors (e.g. Swain 2003: 180) claim that for Bourdieu it is important that only the dominant classes possess social capital, and in fact - recalling the previously quoted concept of the three dimensional space - his analytical interest supports that claim, as he wants to use the three forms of capital to assess the differences in chances between the classes. And indeed, his discussion particularly emphasises the advantages which members of higher classes have from their possession of larger quantities of social capital. However, it will later be demonstrated (see paragraph 2.5) that, especially with respect to the labour market, lower classes can also possess or make available considerable amounts of social capital. Significantly, it will be shown that conceptualising social capital only as "contacts with influential people" (Swain 2003: 189), a "social capital [...] that is [...] winning and keeping the confidence of high society" (Bourdieu 1984:122), does not fulfil the entire potential of the concept.
2.4 "Social Capital" in the Sociology of Gary Becker and James Coleman
The views of Becker and Coleman15 are in general not too far away from Bourdieu's concept of social capital. Both also conceptualise social capital as resources that enable a social network an individual is embedded in to aid in the successful implementation of certain goals.
Becker (1996: 4) sees social capital as an element of human capital. Therefore human capital consists of personal and social capital. He claims that the assumptions and mechanisms laid out by human capital theory with respect to the impact of investments in educational achievements on earnings can also be applied in the more general case to social capital, although here the "rates of return ... cannot be directly measured since utilities cannot be observed" (Becker 1996: 5). Becker stresses that the value of social capital is a function of the decisions of the peers in one's given network and that people have little control over its production and retain little direct influence. The only influence is an "indirect influence over it, since they try to become part of social networks that benefit rather than hurt them" (Becker 1996: 13). His description suggests viewing social capital as some sort of public good that is possessed by the group or network in question16. An individual's maximum social capital can therefore be seen as the sum of the investments that other group members have made in the social capital of the group, less the social capital that has already been consumed by other members of the groups17.
An important difference between Becker's view and Bourdieu's ideas about social capital is the fact that for Becker social capital is not only a property of the dominant classes - everyone possesses social capital; he is even aware of the fact that an increase in social capital can both lower and raise the utility of the individual18. Hence, Becker's conception does not seem as restricted to certain aims as Bourdieu's view and therefore becomes more amenable to different situations.
Coleman (Coleman 1988; Coleman 1990: 300ff) further develops Becker's adoption of social capital for a theory of rational action. He provides a very useful definition of social capital as "productive, making possible the achievement of certain ends that in its absence would not be possible" (Coleman 1988: S98). In this respect he is closer to Bourdieu than Becker, as he stresses the property of social capital that makes it a production factor of "certain ends" which makes his social capital concept more specific and in the end more empirically functional and measurable than Becker's. On the other hand, Coleman still acknowledges the possibility that the use of social capital can also have a negative effect, which has been stressed by Becker, but rather neglected by Bourdieu, who seemed to conceptualise social capital as a solely positive condition. Coleman, however, stresses that "a given form of social capital that is valuable in facilitating certain actions may be useless or even harmful to others" (Coleman 1988: S98).
Some authors (e.g. Swain 2003) claim that Coleman developed the public good properties of social capital that have been described in Becker's writing "to the point where it becomes a social fact ... in a more Durkheimian direction" (Swain 2003: 191). Although Coleman stresses very much the public good aspect of the social capital, the "social fact" claim cannot be supported from his work. On the contrary, Coleman stresses that social capital is mainly a resource for persons, although he states that it "inheres in the structure of relations between actors and [...] is not lodged [...] in the actors themselves" (Coleman 1988: S98).
The above statement might have informed Swain's misinterpretation. If Coleman would conceptualise social capital as a social fact in the Durkheimian sense19, it would be independent of individual action. Far from that, Coleman stresses the fact (even more than Becker) that social capital is to a very large extent dependent on individual action (e.g. when he discusses the effects of an individual leaving the group, for example in Coleman 1988: S117; Coleman 1990: 316). Another statement that could support the social fact claim is that some "social capital arises or disappears without anyone's willing" (Coleman 1988: S118). This statement however does not mean that social capital is independent of individuals' actions (as the notion of social fact in a Durkheimian sense would suggest), but can rather be seen as an example of the well- known effect that rational action of actors on the micro level may have completely unintended (positive and negative) effects on the macro level.
Although it appears to be quite reasonable to reject the claim that Coleman constructs his concept of social capital in a way that would relegate it to the realm of social facts, it is important to keep in mind that he nevertheless mostly stresses the public good character of social capital. Therefore, according to his theory, individuals who use their social capital to make "possible the achievement of certain ends" in actuality are using their social capital in the sense of the characteristic of a group they belong to as opposed to something they possess themselves. This is obviously very far from Bourdieu's social capital understanding - who conceptualises it mainly as the property of individual actors. This view is also shared by some other researchers who belong to the Rational Action tradition as well20 and it will be argued in the next section that both dimensions of social capital are important for an adequate account of the concept.
2.5 Integration of the Concepts of Bourdieu, Becker and Coleman - The "Resource Base"
Before a synthesis of the most useful parts of the concepts laid out above will be undertaken to derive an integrative model of 'The Resource Base', which will for the basis of the empirical analysis, it is worthwhile to provide some consideration of Bourdieu's likely objections to this venture. Obviously this paper has only focused on one aspect of Bourdieu's work - that of cultural and social capital - and set it out in isolation from the remainder of his work. In particular, the theory of the habitus (e.g. in Bourdieu 1984:169ff) is very closely related to both forms of capital - both of which can not only be seen as part of the habitus but are also mutually interwoven with the habitus in a way that they influence the habitus and are influenced by it at the same time. Therefore one could argue that it is unfair and inappropriate to isolate the concepts of social and cultural capital from the notion of habitus, especially since Bourdieu designed the latter as an "opposition [to] the paradigm of the Rational Action Theory (RAT), as its defenders call it" (Bourdieu 1990: 47).
However, here a concept of social and cultural/human capital will be brought forward that does not ask why actors decide to invest or accumulate (a question that the concepts of the habitus and the RAT respectively aim to answer) but is rather used to investigate which effects varying amounts of different types of capital might have on labour market performance of the actors. If - in that current investigation - the concept proves useful, in later work the reason for these differences in capital that actors have at their disposal might be necessary to focus on21 in an attempt to identify the mechanisms that lead to certain actors and groups of actors finding themselves in relatively better positions than others.
This idea of social and cultural/human capital, however, is only a small portion of a broader concept of "The Resource Base"22 (shown in figure 1.). The Resource Base suggests that individuals who seek to realise certain aims (for example labour market success) are not only using some specific kind of resource, as empirical investigations that focus on single, isolated types of resources might suggest, but rather have a very specific base of quite distinct resources at their disposal. It is important to note that the lack of one resource (or capital) type can to a certain degree be made up of another type, or that resource types can be converted into one another to a certain degree (the reverse arrows represent this relationship).
illustration not visible in this excerpt
In the model, three types of capital23 are distinguished. The economic capital is the stock of all material assets an individual can use to realise any specific aim. These can be, initially, any monies individuals have at their disposal, but also include all other material goods that can help them to achieve their goals. It is important to note that not all forms of economic capital are equally suitable for all aims. This is one of the main characteristics of the model presented here - that a specific form of capital might be helpful to realise some aims but might be of no value at all to reach other goals. For example the possession of a large amount of money might be useful to gain a certain degree of education that (ceteris paribus) would not have been possible to achieve - the money can, for example, grant access to better schools or can be used to hire private tutors - but for the achievement of other goals money may be completely useless.
Analogously, this mechanism is also important to consider for all other types of capital and it is necessary to keep this point in mind in order to allow for a more general application of the concept. The second type is called here the personal capital and should be seen as the sum of the outcomes of attempts for self-improvements24 that individuals have made during their lifetimes. The term 'personal capital' is used to ensure that it is confused neither with Bourdieu's cultural capital nor with the human capital laid out by the theorists of the Rational Action Theory. The idea of personal capital as it is used here, however, borrows elements and ideas from both directions. First, it is Bourdieu's notion of "self-improvement" (Bourdieu 1986: 244) that originates within his cultural capital concept. Secondly, it is acknowledged that the acquisition of personal capital begins very early in a person's life, probably very shortly after birth and continues throughout most of the life. Particularly in the stage that social psychologists call primary socialisation, many important self-improvement acts25 are carried out that are described by Bourdieu as "best hidden and most determinant [...], namely, the domestic transmission of cultural capital" (Bourdieu 1986: 244).
As mentioned earlier, Bourdieu mainly focuses on the dyad possession vs. nonpossession of cultural capital. According to the idea of personal capital in the sense used here, however, this two-dimensional differentiation is not appropriate. In this conceptualisation, every individual is in possession of personal capital, which has - as already elaborated in the case of economic capital - a value relative to the specific aims an individual seeks to achieve in a given situation. That said, this is not meant to suggest that there are no differences in the equipment of individuals with personal capital. These differences, which indisputably exist, show their effects when changing the focus from specific to more general aims. In that sense, a specific aim could be securing a certain job - here it is plausible that there are specific types of personal capital that help to secure one kind of job but might prove to be a hindrance when seeking to secure a different one26. When widening that view, however, to more general aims - say, securing a job instead of being unemployed - it becomes clear that there are specific types of personal capital that are more likely to be helpful than others; this is simply because they are, first, compatible with many specific sub-aims of the general aim (i.e., in the example, many different types of jobs) and, secondly, because of differences in the distribution of the specific type of personal capital in the population (i.e., in the example, it might be the case that for some jobs there is less competition than for others - accordingly, individuals who have the personal capital equipment to secure the former type, will more likely succeed in realising the general aim of having a job).27
From a labour market point of view, for instance, this concept of personal capital claims that, rather than having more or less personal capital individuals have personal capital that is more or less appropriate for achieving certain labour-market-specific aims (for example securing jobs in a certain labour market segment). For practical analysis this concept results in a number of conclusions. For example, investigating which specific kinds of personal capital would lead individuals to focus on certain labour market segments and not others (identifying those segments as being most favourable for their specific personal capital resources) and how the structure28 of the labour market (e.g. more competition in some segments when compared to one another) would cause relative discrimination to these individuals, compared to those that have personal capital at their disposal that is more appropriate for less competitive labour market segments and therefore causing, for example, relatively more unemployment for certain types of personal capital29.
Furthermore, some effort should be put towards the development of further measures or proxies for personal capital. Admittedly, educational achievement is the easiest and most straightforward proxy (this is due to the mechanisms that Bourdieu (1986: 247f) describes when discussing the institutionalised state of cultural capital), but it is not completely appropriate for the concept of personal capital that has been laid out here. It might be argued that a personal capital model that suggests focusing on specific aims and resources is not useful and practical for sociological analysis as it commits itself too wholeheartedly to the micro level, but on the other hand, empirical analysis must always at least partially abandon the clarity of theoretical concepts in favour of empirical practicability - and still, the underlying theory provides the basis for both the testing of the empirical hypothesis and the drawing of conclusions.
For social capital, however, the definitions of Bourdieu and Coleman (both quoted above in paragraph 2.2 and 2.3 respectively) provide important characteristics of social capital, particularly with respect to the way in which this concept is conceptualised here. Coleman stresses that with social capital, individuals can achieve "certain ends that in its absence would not be possible" (Coleman 1988: S98), and Bourdieu stresses the fact that social capital is the "aggregate of the actual or potential resources" in a person's social network (Bourdieu 1986: 248f.).
Again, it is important to stress that the value of the social capital of an individual should be seen in context with the realisation of a specific aim. It is quite obvious, for example, that contacts employed (and, hence, provide social capital in respect to that aim) in the hopes of obtaining information about the behaviour of one's own children in school (one of Coleman's examples) are not necessarily valuable when the aim is instead to secure a job (an example discussed to some length in Granovetter 1973).
The value of social capital, however, is not only dependent on the number of contacts an individual possesses and the potential of those contacts to aid in the realisation of that individual's aims; it is also dependent on the willingness of the contacts to provide help within the context of a certain situation. This is mainly what Coleman describes as the public good aspect of social capital and it can be understood as an interaction with the sum of the resources of all contacts in an individual's network: in a group where there is a low willingness to provide help, even vast amounts of potential social capital within the group are not of any use for the realisation of one's aims whilst, on the other hand, in a group with a strong willingness to help30 even lower absolute amounts can be of much use for a person in that group.
In terms of empirical analysis, this means that an appropriate account for the social capital concept suggested here includes both i. measurements of the social network a person is embedded in (i.e. the width and strength of the network) as measures of the amount of social capital the person could potentially activate for the realisation of aims and, ii. some measure to assess the level of mutual trust or willingness to help in the person's network. The large social science survey studies31 seem to fall short in this respect. Clearly, individuals tend to be members of more than one single social group and, hence, the investigations would have to analyse all three dimensions mentioned (width, strength and willingness) for each group the person is a member of. Although instruments used for the measurement of social capital for each of these dimensions already exist and initial empirical tests with these instruments have been undertaken (cf., for example, Grootaert et al. 2004: 25ff.), the implementation in standard surveys is still to be made.
However, with respect to the form of social capital that is a part of the individual Resource Base for the realisation of a specific aim, the analytical problem which has already been mentioned in the discussion of personal capital remains and as such the analysis would have to take both into account: the aim and the relative value of the individual's Resource Base for these aims.
Chapter 3: The Resource Base and Youth Transitions of Offspring of Migrants and Natives in Germany - An Empirical Comparison
3.1 Methodological Preliminaries
The following chapter empirically assesses the analytical value of the Resource Base model that was presented in chapter two. Empirical problems result from the use of a concept that proposes a rather holistic view of individual resources and goals, and some of these issues were identified in the previous chapter. In addition to these problems, the following analysis is inevitably limited by the use of secondary data, as opposed to primary data, and the inaccessibility of qualitative data that would otherwise have added to this quantitative assessment of the Resource Base model.
The use of secondary data32 has some obvious advantages. First, the collection of primary data is restricted by the availability of appropriate time and financial resources - an issue which dictated the use of secondary data in the present case. The second and foremost reason, which is also strongly linked to the time restrictions mentioned, is that the following analysis applies the concept of Resource Base to youth transitions - that is, a temporal process, which takes place in this case between the ages of 15 and 23 - which requires the collection of data from the same individuals over a period of 8 years. Such a venture can take two forms: retrospective collection of data from 23 year olds or longitudinal collection over the specified period of time. The former suffers from the fact that the data merely represents the respondent's perception or, rather, interpretation of the last ten years. The methodological literature commonly states that this approach will be strongly biased due to both deliberate and unconscious errors in the person's memory33 and therefore retrospective survey design should be avoided whenever possible. The latter option would not be feasible since the data could only be analysed after the ten year collection period, which would fall far beyond the scope of this dissertation. Hence, the use of secondary panel data - that is, data that are collected with the same sample at different points in time (e.g. yearly) - is an alternative that is appropriate for this analysis of the Resource Base.
The use of secondary data of course involves disadvantages as well. While the advantages outweigh the accompanying disadvantages, they are nonetheless important to note as they exert constraints on both the analysis and interpretation of results. Basically, all disadvantages stem from the fact that someone else collected the data. First, this means that theoretical assumptions undertaken in the construction of items, scales and questionnaires are often not readily apparent for the secondary analyst. These assumptions likely informed the wording of questions, the training of interviewers, the coding of results and many other factors upon which the secondary analyst therefore has no influence and, maybe even more seriously, often has at the very most limited insight into these processes.
1 For example, Breen (2004) demonstrates, by way of a European comparative perspective, that in the last 30 years Germany maintained the least amount of upward social mobility or, in other words, in Germany social origins and the educational background of one's parents were the best determinants of one's own social position and educational achievements amongst all European countries.
2 For a brief overview see, for example, Hradil (2001: 47ff.)
3 This text, originally published in German, is used here in its English translation (Bourdieu 1986).
4 This is not entirely true. Becker himself stresses the fact that already Adam Smith, Alfred Marshall, and Milton Friedman provided valuable "analyses of investments in education and other training" which were, however, "not integrated into discussions of productivity" (Becker 1997: 43). This remained the case until the early 1960s when T. W. Schultz, among others, "began to pioneer the exploration of the implications of human capital investments" (ibid.).
5 The term 'quasi-rational' has been chosen here as an attempt to avoid one of the main misunderstandings of economy-related approaches in the social sciences: the economic approach to the analysis of human behaviour assumes that individuals act as if they were seeking to maximise their welfare as they conceive it. Therefore this approach does not claim that individuals literally calculate their situations - in most instances the assessment(s) used to identify the preferred choice is in fact un- or sub-conscious, or even made merely because the act falls within a set of everyday activities or customs belonging to the individual. Furthermore, welfare as they conceive it may also include the satisfaction of altruistic, selfish, loyal or even masochistic needs. The decision process might also be influenced by an 'inaccurate' perception of the actual situation, which is largely the consequence of incomplete information. For example, the probability of the occurrence of costs (i.e. situations to be avoided) and gains (i.e. outcomes sought after) or the complete ignorance of some potential outcomes of a certain action can all interfere in one's perceptions and therefore actions. Hence, the economic approach does not make any assumptions about motivations for human behaviour, especially the assumption regarding the maximisation of welfare, which does not (a common reproach of its critics) assume that individuals mainly follow selfish motivations.
6 These are, above all, the so-called opportunity costs of the 'missed opportunity', which result from the fact that time which is used for carrying out one decision cannot be used to do something else
7 Certainly in many cases this is not only an individual decision since the stage of life in which such decisions about school and university have to be made tends to coincide with a high dependency on one's family and as such the decision can often be regarded as a family decision. Thus, the decision largely depends on how long a family can provide freedom from the economic necessity of joining the labour force. This clearly provides an opportunity to model differences between children from different social strata or classes. In general, however, to what degree the investment decision is the decision of the children themselves or that of the rest of the family is not important. One can treat children and parents as a single "decision-making entity" with no need to look further into the "intrafamilial disagreements, bargaining, compromises etc." (Breen and Goldthorpe 2000:184). The rational choice approach also provides a platform for identifying discriminating policies. Becker (2002) points to the American case where white college students pay tuition fees that are only about 20 per cent of the total costs of a college education (most of the remainder being aforementioned opportunity costs), whereas most black students (more likely coming from lower income families) pay tuition fees that account for a larger share of the total college costs (i.e. their families have more problems to provide them with time free from the necessity to participate in the labour market). Hence, when the tuition fees rose by 39 per cent from 1980 to 1986, this rise, which affected all students in the same way when only taking into account the monetary increase, did affect the relative costs of black students more than those of their white contemporaries.
8 This does not necessarily only refer to education in school and university. On-the-job training and working experience can also be seen as investments in human capital and have a strong impact on the income of individuals as, for example, Jacob Mincer (1974) already showed for the American case in the 1970s.
9 That is also the reason why the German Language Society (GfdS) proclaimed the term the "lexical monstrosity" (Unwort) of the year in 2004. As the reason for their decision, the jury claimed the term "not only reduces employees, but in general all people to quantities that are only economically interesting" (GfdS 2005, own translation).
10 In fact, Bourdieu responded to his categorisation as a Marxist with the words "to be or not to be a Marxist is a religious alternative and not at all scientific" (Bourdieu 1990: 49). Arguments could also be brought forward to classify him as Weberian, which would potentially have led him to a quite similar statement. However, in his work, the notion of class is a predominant theme and to use the term 'Marxist' here is merely to appreciate the roots and interests of his work; although Bourdieu aims "to surpass what he [Marx] thought he did" since he, Bourdieu, "won't go as far as to say that Marx's theory of classes satisfies" him (ibid.).
11 The French original "La Distinction" was published first - five years earlier - in 1979.
12 Erik Olin Wright (1985), a theorist in a Marxist tradition who shared with Bourdieu the interest in analysis of social classes, added a further dimension: authority. He argues that individuals that possess too much of either i. means of production, ii. skills or iii. authority belong to the exploiting class in respect of this dimension and those who have a lack in one of these dimensions belong to the exploited class.
The skill dimension of Wright's approach is comparable with (some aspects of) Bourdieu's cultural capital concept, and with Wright's model the aforementioned mutual dependence between capitalist and highly skilled employee can be explained very well: both belong to the exploiting class although with conflicting interests.
13 In the following the term 'cultural capital' shall always refer to what Bourdieu calls 'embodied cultural capital'. When referring to 'institutionalised cultural capital' or 'objectified cultural capital', these phrases will be used explicitly.
14 It can, however, be argued that the economists' human capital theory also accounts for the effects of family background on the acquisition of cultural capital and that the usual ignorance of the differences in early, domestically acquired human capital is mostly due to problems in measuring this capital.
Rational Choice Theories could account for the impact of family background, that goes beyond the aforementioned economic reasoning regarding the possibility of providing time for extended education that is free from economic demands, in two different ways (although unfortunately, the measurement problem remains in both cases). First, it could be argued that preferences for extended education are different among classes. In many cases children from, for example, a working class background may not perceive extended or higher education as a relevant option that is open to them. On the other hand, offspring from, what Bourdieu (1984: 114) calls the "dominant class", may not perceive the option of leaving school with minimal qualifications as a relevant option. In addition to these differences in perceived opportunities, different perceptions of children and their parents regarding the probability of success in gaining further or higher education can be assumed, which would result in a lower expected outcome of educational investment.
15 Who are only chosen for the purpose of representing the greater number of authors with a rational choice background that deal with social capital or social network issues.
16 His comment that "if only one person in a social network is given an incentive to change the amount invested in her social capital, the capital of the others [and the total capital stock in the network, C.H.] in a large network is only slightly affected" (Becker 1996: 13f.) can be interpreted as a description of the production function of public goods. Social capital, however, can be regarded as a very special case of a public good. One of the main problems in the production of public goods is the so-called free rider problem. The free rider problem stems from the fact that nobody can be excluded from the consumption of the public good, even if they have not contributed to its production. In the case of social capital, however, the free rider problem is solved (at least partially) as every actor who contributes to the production of the social capital of the group does so as a by-product of an action that mainly aims at something else (i.e. the production of a private good).
17 It is worth noting that this does not imply that this kind of calculation would indeed be possible. This 'equation' should rather be seen as an analytical tool which allows for the consideration of the implications of certain types of 'changes in its parameters'; for example, what would happen if the 'maximum amount of social capital consumed' would exceed the amount invested and vice versa. In fact it has been stated explicitly that the actual measurement of social capital in this sense is hardly possible. However, research has been undertaken (Grootaert et al. 2004) to find at least close proxies in order to conduct an empirical scrutiny of the theoretical considerations regarding social capital.
18 The utility of the individual lowers, for example, when peer pressure leads teenagers to start smoking, join violent gangs, neglect school and suchlike.
19 That is, "(i) exterior to each individual [...] and (ii) capable of exercising a 'coercive' action on individual consciousness" (Gane 1988:13) and, thus, independent of individual action.
20 Namely the studies of Granovetter (e.g. 1973) and Lin (e.g. Lin and Dumin 1986), who investigate the use of social ties in one's social network to find a job are useful in this respect. Much research has been done in that field in the last few years; more recently, for example, Mouw (2003) reveals some causal mechanisms that lie behind the effects that have been found in earlier studies.
21 In that venture as well, blending Bourdieu's habitus with Rational Choice Theory (even if he would potentially object - although he states himself that "science is there to be surpassed", so it would be "the only fitting homage [...] to use what he did [...] to surpass what he thought he did" (Bourdieu 1990: 49)) might prove valuable: the Rational Choice Theory claims that each individual actor seeks to maximise their utility according to their individual choice preferences. The RCT, however, lacks explanation of where these preferences come from and why they are inter-individually different. Here Bourdieu's habitus concept could fill the gap by providing a theoretical framework for explaining why "maximisation of utility" might consist of very different things within different classes or ethnic groups.
22 This concept (and especially its name) has been inspired by the discussion of a reflexive model of youth transition (Furlong et al. 2003: 5ff, especially figure 1:3 on p.7), although, at first glance, it has not very much in common with that model.
23 The term 'capital' is used to clarify that it is adopted in the sense of a quantity, of which individuals can have or possess a certain amount of, and use in an attempt to realise their specific aims. The use of the term shall, however, not imply that the actual amount of capital is always measurable; this is rather in most cases impossible.
24 More generally, the term 'self-modification' could be used since it might be argued that not all attempts are in fact improvements and may not initially have been thought and planned as improvements either. The outcomes of those modifications that have no improving character, however, might still be important here since they might potentially have an impact on the achievement of specific aims. For example, one can think of the loss of physical or mental capacities as an accepted outcome of unhealthy activities, which make more difficult or impossible the achievement of certain goals.
25 Here it becomes obvious that these self-improvement acts do not necessarily mean that they are the result only of decisions that have been made specifically for the purpose of achieving self-improvement. These ' acts' are also events that just ' happen' to the individual (e.g. the more or less unconscious everyday 'transmission' of personal capital that happens due to the living with one's parents, etc.)
26 For example it is plausible that in a hospital the management will not employ a person as a cleaner whose personal capital resources are perceived by the management as more appropriate for the position of medical consultant and vice versa.
27 Of course, the property of a specific type of personal capital that enables its bearer to achieve one general aim and hinder another general aim is still valid accordingly.
28 It is worth noting that the structural component has a threefold effect on the model presented here. First of all, there are effects on the resource base of the individuals; the foremost of which involve the effects of 'heritage' of social and cultural capital that are discussed by Bourdieu (1986). Secondly, there are structural influences on the perception of aims - that is partly due to the same 'hereditary' effects (it can be argued that certain aims are more or less transmitted from the parents to the children and other aims are inherent in the domestically transmitted cultural capital itself) and partly due to the individual's assessment of the labour market structure: aims that seem to be hopeless to achieve might not even be taken into consideration. The third way in which the structure comes back in is on the process of resource use for aim achievement: it can be argued, and is indeed quite obvious (and empirical evidence of changing labour markets - for example, in connection with the decline of the secondary sector and the incline of the tertiary sector - might support this claim as well), that the same personal capital resources might be more or less appropriate for the achievement of the same specific aim under different labour market conditions, i.e. under different structural circumstances. These important structural components will only play implicit roles in the empirical analysis of this dissertation. The empirical analysis will focus on individual actors and their experiences in the labour market. When interpreting the results, structural components (for example the effects that the structure of the schooling and apprenticeship system in Germany might have) will be considered as well and chapter 4 will suggest some ways in which structural components can be addressed more systematically in conjunction with the Resource Base model.
29 Another practical conclusion would be, for example, to use educational achievement (probably the most often used proxy for personal capital) with a nominal rather than ordinal (or even interval, as some studies do) level of measurement - simply because the latter would be based on the more vs. less personal capital assumption.
30 That is, in Bourdieu's (Bourdieu 1986: 248) words, a "durable network of [...] relationships of mutual acquaintance and recognition". And Coleman (1988: S101) states in that context that a "group within which there is extensive trustworthiness and extensive trust is able to accomplish much more than a comparable group without that trustworthiness and trust."
31 An initial, rough overview, or 'scan', of the items in the British Household Panel Survey, the German ALLBUS, and the German Socio Economic Panel Survey has been undertaken.
32 The following remarks generally apply only to the secondary analysis of quantitative data sets. For the secondary analysis of qualitative data, the comments are still applicable but only to a limited degree since analysis of qualitative data requires consideration of a separate set of issues; nevertheless, this discussion is not relevant for this chapter.
33 The Twenty-07 study of Furlong and colleagues (2003), for example, used this retrospective approach to some extent. Here, the researchers conducted interviews that covered the last two to three years of the lives of their respondents. Since the study utilised a longitudinal design and had been conducted with the same sample every two to three years, in some cases there were short overlaps with periods that had been covered by two interviews with a distance of about two years. It is important to note that in some cases the interviewees reported different information for the same period; for example, they claimed that they had been housewives in a period which they indicated some years early earlier as a period of unemployment or remembered that they were full- or half-time employees in a period for which they indicated in earlier interviews that they spent it in a Youth Training Scheme (based on personal communication with one of the authors, Fred Cartmel). It can be argued that these contradictions are a result of the processes involved in reconstructing one's own memory, which are not favourable for retrospective work.