Navigation – Plan du site

Catch me if you can

Configurations of atypical and precarious employment in Switzerland
Félix Bühlmann
p. 73-90


A typical and precarious employment has become an increasingly important issue in the last 30 years. However, atypical employment includes very heterogeneous forms of work and is difficult to capture empirically. In this paper we propose a configurational approach in order to overcome these difficulties. We conceptualise the labour market as a field and investigate the data of the 2004 sweep of the Swiss Household Panel with a multiple correspondence analysis. We discover four configurations of atypical work and then relate these configurations to possible explanatory factors and possible consequences. Our findings show that in Switzerland only one, very specific configuration of atypical employment leads to social suffering and vulnerability: jobs that combine the absence of biographical prospects, forced flexibility and low wages. Other atypically employed fractions use their specific resources to deploy compensation strategies which render atypical employment socially bearable.

L’emploi atypique et précaire est devenu une problématique de plus en plus importante ces 30 dernières années. Or, ce qu’on appelle emploi atypique englobe des formes très hétérogènes de travail et est difficile à capter empiriquement. Dans cet article nous proposons une approche configurationnelle afin de surmonter cette difficulté. Nous conceptualisons le marché de l’emploi comme un champ et étudions les données de la vague 2004 du Panel suisse des ménages grâce à l’analyse des correspondances multiples. Nous découvrons quatre configurations d’emploi atypique et les mettons en lien avec des facteurs explicatifs et des conséquences potentielles. Nos résultats montrent qu’en Suisse seulement une configuration d’emploi atypique très spécifique mène à la souffrance sociale et la vulnérabilité : des emplois qui combinent l’absence de promesses biographiques, la flexibilité forcée et des salaires bas. D’autres fractions de personnes employées de manière atypique utilisent des stratégies de compensation qui rendent l’emploi atypique socialement tolérable.

Haut de page

Texte intégral

1. The mercurial character of precariousness

1In the last thirty years, employment relationships have undergone deep changes. The continent of standard employment relationships—full-time, nine-to-five on weekdays, permanent, at the employers place of business, and under the employer’s direction—has shrunk, and an archipelago of atypical and precarious employment has emerged and gained land [Rodgers, Rodgers, 1989; Castel, 1995; Gallie, 1998; for Switzerland: Prodolliet, 2000; Magnin, 2005; Pelizzari, 2009; Walker et al., 2010].

2Schematically, the literature explains this evolution as follows: subsequent to the end of the BrettonWoods system in 1971 and the oil shock in 1975, the economies of the Western world slid into a crisis which forced the shareholders and managers of the leading corporations to rethink the Fordist organisation of employment [Gallie, 1998; Kalleberg, 2008]. The increased international competition (and/or the desire to redistribute revenues differently) led the economic elites to reorganize corporate structures, to introduce more flexible forms of employment, and to outsource and shift labour-force-intensive work to countries with lower average wages [Boltanski, Chiapello, 1999]. This was made possible by technological advances in work organisation and also profited from changes in labour legislation: in many countries the standards of labour market regulation eroded, the influence and the membership numbers of unions waned, and the politics of workfare forced people to accept poorly paid jobs or precarious work conditions [Kalleberg, 2000; 2008]. These structural changes were backed by ideological and discursive evolutions. Flexibility and individual responsibility were propagated as the future key virtues and the horizontal and versatile network was praised as an important prospective management device [Boltanski, Chiapello, 1999]. This ideological shift encouraged (and sometimes compelled) people to accept flexible jobs, eroded the boundary between self-determined and forced forms of flexibility, and favoured the short over the long term [Sennett, 1999].

3Atypical employment is defined negatively against standard employment. Therefore, it can consist of part-time work, fixed-term and otherwise temporally limited contracts, work at home, work in pseudo self-employment, work on call (and all other forms of forced flexibility), night and weekend work, multiple employment, poorly paid jobs or underemployment, and more. When these insecure and flexible forms of employment lead to psychological suffering, social isolation, or political disillusionment, we may speak of precarious employment. For three reasons, atypical and precarious employment is difficult to grasp:

  1. Not only are these atypical and precarious employment situations very different, but they can have completely different explanations and different consequences in the lives of those who are in such situations. For example, on-call work renders it difficult to maintain a social life (to meet friends, organise a family, or participate in a club), a fixed-term contract raises problems with respect to the planning and foreeseability of life (such as starting a family, having children, or buying a house). In addition, certain forms of atypical employment entail a lack or a lower standard of social security (pension funds, unemployment insurance, and so on), whereas others do not.

  2. It is often not a single form of atypical employment, but the combination of two or more of them which makes them problematic. Most of these forms of atypical employment are also part of arrangements which are not considered precarious. Part-time employment, for example, can give people more freedom and flexibility. Only when it is involuntary it becomes a problem—because the wages are not high enough or because working hours are more easily shuffled around during the day. The same is true for independent contractors, for work at home or weekend work. All these forms are part of employment situations in which employees feel perfectly at ease and enjoy a great deal of freedom. But they are also part of jobs in which self-employed people have only one client (which in fact is their real employer), work at home to save infrastructure costs, or are forced to work shifts only at nights and weekends.

  3. Precariousness cannot be understood in itself, but must be systematically related to the proportion of those who are in a) stable forms of employment and b) in unemployment. What defines precarious work positions is not (only) the nature of the employment relationships, but their link to stable positions and situations of exclusion [Castel, 1995]. This is particularly important when it comes to understanding precariousness in comparative terms [Barbier, 2005]. The relative status and meaning of employment with a temporary agency or a fixed-term contract, for instance, varies widely according to the protection enjoyed by those in a stable, non-fixed-term contract. If the labour laws facilitate the layoff of workers, then employment with a temporary agency might not be too problematic. However, if there is a strict separation between well-protected core employments and very contingent work, the meaning of being employed by an agency is different [Rodgers, 1989].

4As a consequence, this paper’s goal is twofold: we try to develop a configurational approach that is able to grasp atypical employment despite its mercurial nature, and we seek to demonstrate its usefulness on the basis of the Swiss case. From this perspective, we first outline this configurational approach and discuss it in the light of existing literature on precariousness. Second, we will analyse the social characteristics—individual and structural—related to these configurations. Next we will analyse how these configurations are related to the feelings, attitudes, and practices of the actors. To conclude we will summarise the results, sharpen the distinction between atypical and precarious employment, and argue against the careless use of dual metaphors of the labour market.

2. A configurational approach?

5Mercurial phenomena such as atypical or precarious employment, which include a large number of potential situations and which change their social meaning with respect to their specific configuration, require particular conceptual tools and methods. Our argument combines theoretical and methodological elements.

6Theoretically we argue against the a priori use of dualistic conceptualisation of the labour market, as is the custom in dual-labour market theories [Piore, 1970], segmented labour market theories [Doeringer, Piore, 1971], or more recently, theories of the insider-outsider divide [Palier et al., 2012]. These theories deem that all deviations from the hitherto standard male employment relationship can be pooled together and be labelled with words such as “peripheral”, “external”, or “secondary” because they are different from the standard benchmark. In doing so, they potentially overlook differences and grey zones that are crucial to understanding the labour markets.

7Also, we are sceptical with respect to theories that construct additive thresholds of atypical aspects above which an employment can be considered precarious [Rodgers, 1989; Walker, 2010]. These theories state, for instance, that the combination of two forms of atypical employment (such as part-time and night work) is still bearable, whereas from the accumulation of three forms of atypical employment on (part-time, night work, and fixed-term contract) the situation must be considered precarious. This conceptualisation is unable to capture the meaning of specific combinations of atypical employment and therefore might consider combinations that are problematic as unproblematic (and conversely combinations which are unproblematic as problematic). Thus, we do not believe in labour market theories that consider atypical employment forms as essential attributes and do not relate them to each other or their context.

  • 1 Which is useful for the analysis of other phenomena; we are not making a general, but a specific ar (...)
  • 2 See, for example: Bourdieu et de St. Martin, 1978; Bourdieu, 1979; Bourdieu, 1989. For an external (...)

8This is linked to a parallel methodological argument. We think the success of such simple binary conceptualisations of the labour market can partially be explained by their convenience with respect to mainstream methods such as regression modelling. Such methods aim at analytically separating the independent variables to identify the direction and force of their effects, and often require a relatively simple operationalization of the dependent variables. Hence, as shown above, atypical employment is not a simple, but a very heterogeneous and interrelated bag of phenomena. As regression analysis with its emphasis on analytical separation struggles to account for the variety and the relational qualities of atypical employment, we suggest in the present contribution to study the labour market with the concept of field and the method of multiple correspondence analysis1. How can these two tools, which, for instance, in a Bourdieusian framework have traditionally been used together, be characterised?2

9A field is a historically grown space, defined as the system of relations between positions unequally endowed with specific resources and capital [Bourdieu, 1984; 1989; Fligstein, 2001]. Although, like in a sports game, the structures are historically moving, a field functions along certain inert principles. The labour market is, for example, characterised by rather stable recruitment mechanisms or by a hierarchy between occupations and positions.

10Secondly, the structure of the field is the state of the relations of force between agents and institutions involved [Bourdieu, 1984; Fligstein, 2002]. To be able to accede to a specific position, one needs to possess a volume and a combination of resources or capital which—comparatively—the competitors lack. This capital is the object of dispute in a field; in the labour market this can be educational capital, cultural capital, gender, or employment experience. The endowment with capital and resources tells us also about the location of the precariously employed in the social space [Marchart, 2010]; we can study the amount of capital possessed by those who occupy an atypical post. However, certain configurations may also be more common to specific economic sectors or occupations, independently of the individual capital endowments of those who occupy the position.

11A specific position in a field is always relationally defined. As in a magnetic field, each particle behaves (or takes its meaning) in relation to the totality of forces at work in the field; in a social field, each position is defined by its relations to all the other positions surrounding it [Martin, 2003]. We assume that no single group accumulates all forms of atypical employment and that, on the other hand, none is completely spared. Rather, we think certain forms of atypical employment are systematically coupled, whereas others might be mutually exclusive. To give an example: whereas those who work on call might also have a tendency to work at night, those who work full-time probably rarely have more than one job (although this is not excluded). In the following sections, we speak of the bundle of relations typical for a group of positions as of configurations.

12According to Bourdieu, people who occupy certain configurations not only share social characteristics, but tend to think and act in a similar way [Bourdieu, 1979]. Applied to the labour market, we can assume that atypical employment is psychologically difficult to bear; that feelings of insecurity translate themselves into work dissatisfaction [Gallie, 1998: 16] or depression [Sennett, 1998]. Certain researchers even assume that people in atypical positions embrace specific political attitudes and/or become disillusioned with politics altogether and thus retire from political and social activity [Paugam, 2009]. We will therefore analyse indicators of employment satisfaction, psychological wellbeing, and political attitudes, and relate them to the configurations of the field of employment.

  • 3 In Switzerland, this concerns, for example, eligibility for unemployment benefits.
  • 4 In Switzerland, to be included in the “second pillar” of the pension fund system, employees must ea (...)

13Four configurations of the Swiss labour market have the potential—alone or in combination—to be prone to precariousness. First, employment contracts which are limited in time and do not allow for long-term plans or creating effective professional, political, or associative projects [Bourdieu, 1989; Magnin, 2005]. A second aspect refers to compatibility with daily and weekly social rhythms. Shift work, work on call, evening or night employment, irregular overtime, or weekend work are all forms of work which potentially hinder the organisation of everyday life and make it difficult to synchronise one’s own rhythm with that of other people. This potentially causes issues for the organisation of family life, for friendship relations, and also represents a potential obstacle to religious, political, and social participation. Unusually low or unusually high amounts of work are a third form of atypical employment. A very small amount of work can create legal and financial problems, as people with marginal employment lack certain social protection3. Too much work, even when it is “chosen”, can create problems in the social organisation of daily life and leads to stress, health problems, and psychological difficulties. Small wages are a fourth dimension of atypical and precarious employment: following Walker et al. in this point, we think those whose household income lies below a certain threshold are atypically employed [Walker et al., 2010]. The lack of a minimum household income not only renders it difficult to participate socially and politically but prevents individuals from being eligible for the pension fund system4.

14A method particularly well suited to translate the field concept into an empirical investigation is multiple correspondence analysis (MCA). MCA has been used to study the fields of power [Bourdieu and de St. Martin, 1978], science [Bourdieu, 1984], and education [Bourdieu, 1989]. Applied to the labour market, this method allows us to understand the field of precarious employment as system of relations between positions characterised by specific configurations of employment conditions. MCA is a graphical multivariate method which makes it possible to integrate a large number of varied variables. It “provides a visualisation of the clustering of complex datasets, where properties that frequently occur together with the same respondents, or with respondents who in other respects resemble each other, will also figure close to each other in the map, while properties that rarely occur together will be situated far from each other” [Savage, Prieur, 2011]. The different forms of atypical employment, therefore, have not to be separated analytically but can be investigated jointly. This makes it possible to identify typical combinations and configurations of the forms of atypical employment.

15What is more, MCA is an intrinsically relational method [Bourdieu, 1984; de Nooy, 2003]. It is therefore better suited to help researchers understand the relational nature of atypical and precarious employment situations. By drawing the different configurations of atypical employment in a graph we will be able to identify some of the relations between the configurations. For example, we will identify relations between more secure and less secure configurations, but also potential complementary relations between employment situations within the couple.

16Finally, MCA allows the researcher, by the use of passive variables projected into the space without contributing to its construction, to relate the configurations of atypical employment to potential causes and consequences [Le Roux, Rouanet, 2010]. According to the concept of homology, the social space, the space of employment, and the space of representation can graphically be laid on top of each other. Thereby, certain configurations of atypical employment can be related to specific causes and consequences [Bourdieu, 1984].

3. Data and methods

  • 5 Certain researchers doubt that quantitative surveys can give answers to questions of work precariou (...)
  • 6 Of course, the concentration on a single year can also be problematic, as the meaning of atypical j (...)
  • 7 We also carried out MCA’s that included the whole active population. These analyses revealed rather (...)

17The Swiss Household Panel (SHP) is, in our eyes, the most adequate survey for our endeavour5. Compared to the Swiss Labour Force Survey (SLFS), the other potentially suitable survey for studying precariousness, its information on employment conditions is more detailed and also features subjective indicators [Voorpostel et al., 2011]. This is pivotal because the central theoretical advancements on precarious employment [Paugam, 2009] deem the feeling of insecurity a central component of precarious employment. In addition, the SHP includes also information on other aspects than work and employment, such as political participation and membership in associations. In addition, the information on the household structure and the employment situation of a potential partner or the number of children is better developed in the SHP than in the SLFS. On all these points, the Swiss Household Panel is better suited to examine precariousness: its only flaws are its comparatively small sample; the lack of an oversampling of those potentially exposed to atypical employment; and the attrition rate, which is particularly high for the vulnerable [Voorpostel, 2010]. To respond to these weaknesses we will use the 2004 wave of the SHP. In this year, the sample has been augmented with a second cohort and is not yet touched by the attrition of the following years (n=14,086)6. We reduced the sample in a first step to the active population (n=5,565) and included in a second step only those exposed to at least one form of atypical employment7. This leaves us with 4,410 individuals.

18According to the proposed theory, we will deal with three sets of variables. The first and pivotal of these sets addresses employment conditions. It includes seven active variables, which for illustrative purposes have been completed by a number of partially redundant passive variables.

    • 8 Pseudo-self-employment remains important because it has a long series of consequences in terms of s (...)

    First, we divide the number of effective hours worked weekly (not the employment rate, which is only added as an illustrative, passive variable) into the categories 1-15 hours, 16-30 hours, 30-45 hours and 46 + hours. This renders it possible to capture those who only work a few hours, but also those who work very long hours. Second, we have chosen the variable “flexibility of work”. Crucially, this variable takes into account work flexibility and its forced or self-determined character. This distinction is important because it distinguishes between those who are forced to be constantly ready to work without knowing if and how much they will work, and those who can autonomously choose their flexibility. The modality “shift work” has been excluded as redundant with the variable “night and weekend work”. We have distinguished night and weekend work as “yes or no”. The next variable concerns homework and pseudo-self-employment. Tests revealed that pseudo-self-employment and homework were redundant, but that the second variable was able to reflect a wider variety of situations. Therefore, WE use homework as an active and pseudo-self-employment as a supplementary passive variable8. The answer “office at home” is only used as a passive answer, as this proportion was under 5%. The fifth variable used was the number of jobs, which differentiates between those with one job from those with more than one. Sixth, we used the question of the duration of the current employment contract. As the SHP in this wave features no specific question about the employment by a temporary agency, we have chosen to contrast those with a temporal limitation of contract (of different forms) to those with a temporally unlimited contract. In addition, we introduced fixed-term contracts as a passive control variable. The seventh variable concerns the household equivalent income. This measure indicates the income per person in the household and can help distinguish between precarious employment and a precarious situation in general [Marti et al, 2003; see also Kraemer, 2008]. The variable was divided into the categories 0 to 42,000, 42,001 to 66,000, 66,001 to 90,000, 90,001 to 120,000, and 120,001 and up; the last category was only used passively.

    • 9 Using the occupational typology as a passive variable, we are not compelled to limit the number of (...)

    We distinguished between individual and structural factors in the social recruiting space of atypical employment. Among the individual variables, we introduced sex (men vs. women), cohorts (15 to 24 years, 25 to 39 years, 40 to 54 years, and 55 to 64 years), and educational capital (compulsory education, apprenticeship/high school, higher vocational education, and university/applied university). Another important individual factor for precariousness is nationality [Pelizzari, 2007; Marti et al., 2005]. In recent years, the working-class migration from southern countries in the 1960s, 1970s, and 1980s has increasingly been completed by a group of better educated immigrants, often from northern European countries. We have distinguished between Swiss, southern countries (Italy, Portugal, Ex-Yugoslavia, and so on) and rich countries (Germany, France, and others). The structural factors are themselves divided into two groups: first we examine the economic sector according to a detailed version of the Swiss NOGA nomenclature, which distinguishes between sixteen categories: agriculture and forestry, education, real estate, finance and insurance, public administration, social and personal services, extraterritorial activities, health and social work, electricity gas and water, construction, manufacturing, wholesale and retail, hotels and restaurants and transport and communication. Second, we use a fine-grained typology of occupations according to ISCO9. It includes about 129 different occupations and allows us to make very fine distinction between different levels of occupations. Any of the occupations with an n > 10 were included in the graph.

  1. To understand the psychological, social, and political repercussions of employment configurations, we introduced a series of passive variables, which are mainly based on standardized questions on well-being and politics. The first question deals with general job satisfaction, a variable measured by eleven answer items going from “not at all satisfied” to “completely satisfied”. In the graph, only the two opposed categories are represented. The second type of variable concerns psychological wellbeing. We concentrate here on the frequency of depression. It asks how frequently the respondent faced “depression, blues, and anxiety” and goes from “never” to “always”, through eleven answer items. Finally we use a question about the respondents’ interest in politics (from 0, “not at all interested in politics” to 10, “very interested in politics”).

4. The space of atypical employment in Switzerland

  • 10 These numbers draw on the Swiss Labour Force Survey. See also Pelizzari (2009) for the period 2001- (...)

19Compared to most European countries Switzerland’s economic situation, and in particular the Swiss labour market, remained for a long time spared from major crisis in the form of mass unemployment. Even in the early 1990, when the Swiss unemployment rate rose, its increase was gentler than in many other Western countries. This has specific reasons: unemployment was for a long time buffered by the periodic pushing out of the labour market of the foreign workforce and women [Streckeisen, 2012]. Only in the early 1990s, when women increasingly crowded into the labour market and the regulation of labour immigration was modified, did the economic crisis increase the unemployment rate from about 1% to 4% and higher. In addition to the increase of the share of people who are durably excluded from the labour market, we can also trace the rise of precarious and unstable jobs through the same period. An increasing proportion of the workforce labours in fixed-term contracts, with temporary agencies, on call or in part-time arrangements. Whereas, for instance, in 1993 only about 4,500 people were working for a temporary agency, this number rose to 46,000 in 1997 and to 201,000 in 2003. Also, the proportion of employees in fixed-term contracts and in employment on call increased about 35% and 38%, respectively, between 2001 and 2009. Part-time work increased more modestly, about 20% in the same period.10

20This situation can now be analysed in field theoretic terms. As a first step, we display and interpret the space of atypical employment in Switzerland, constructed on the basis of the active variables describing the employment situation. The eigenvalue of the first axis is 0.23; that of the second axis is 0.19. We restrain ourselves to two dimensions; although though the explanatory force of the third axis is not dramatically lower, it does mainly highlight the difference according to income and provides little theoretical added value to our argumentation. Figure 1 represents simultaneously a cloud of individuals and a cloud of categories and constructs a “field” or a “space” illustrating the system of relationships among them. The cloud of individuals is represented in light grey, the categories in black. The more closely individuals are situated to each other, the more categories they share. Inversely, spatial proximity between two categories indicates these categories are shared by a larger number of individuals. The average contribution of each category to the construction of the space is 5%. Categories which contribute above this average to the first axis are indicated in bold. Those which contribute more than average to the second axis are in italics. Categories that contribute above the average to both axes are in bold italics.

Figure 1. The space of a-typical employment in Switzerland (2004)

Figure 1. The space of a-typical employment in Switzerland (2004)

Legend: categories contributing above average to the construction of: horizontal
axis are
bold; vertical axis are in italics; to both axes are in bold italics.

21The following three questions contribute above the average to the variance of the first axis, in this hierarchical order: number of weekly hours (28.8%), form of flexibility (20.9%) and homework (18.8%). Together they add up to 68.5% of the explained variance of the first axis. The eight categories for the interpretation of axis 1 (in bold) amount to 81.1% of its variance. On the right side can be found: 46 and more hours (21.1%), autonomous flexibility (12.2%), occasional homework (8.4%) and household income 90,000 to 120,000 (5.5%). On the left side can be found: no homework (9.3%), same hours every day (8.7%), no night or weekend work (8.2%) and temporally limited employment (7.7%).

22With 86.0%, four questions contribute above the average to the second axis: night and weekend work (29.2%), number of weekly hours (23.5%), homework (18.0%) and form of flexibility (15.7%). The following nine categories have been retained to interpret the second axis, adding up to 86.2% of the explained variance. We present them in a hierarchical order, depending on the force of their contribution. In the upper half are situated: no night or weekend work (18.6%), more than one job (8.9%), 16-30 weekly hours (7.6%), 1-15 weekly hours (6.7%) and occasional homework (6.1%). In the lower half can be found: night and weekend work (10.7%), forced flexibility (10.3%), 31-45 weekly hours (9.1%) and homework no (8.2%).

23This means the horizontal axis distinguishes a group of hard-working and well-earning actors in the upper-right quadrant. For them, atypical work has a sporadic and self-determined character. When this group works overtime, at home, at night, or at the weekend, it seems to be a more or less a consenting choice; its members have a certain mastery over when and how they work. In addition, this group’s willingness to accept atypical employment conditions is rewarded with a relatively high salary.

24In the upper-left quadrant can be found a group of people who works few hours (and in fact is employed up to 80%) and occasionally hold several jobs simultaneously. At the same time, this group has regular working hours and does not work on nights and weekends. Even though household income is rather low in this group, it enjoys stability both biographically and in its weekly social rhythms. This group is hardly exposed to fixed-term employment or flexible work.

25In the lower-left quadrant we can identify a fraction that is touched by contracts limited in time (mostly fixed-term), forced flexibility (shift work, work on call, and the like) and poor pay (household income 0 to 42,000). This accumulation of three forms of atypical employment may be a hint that this configuration is particularly difficult to endure. However, it is noteworthy that this group does not work at home, hardly ever has more than one job and usually works between 31 and 45 hours per week (no overtime, no marginal employment).

26Finally, in the lower-right quadrant we find a configuration that is characterised by a combination of forced flexibility and night or weekend work. As shift work (as a passive category) is also located in this area, we can make the assumption that this fraction works regularly at marginal hours. At the same time, these positions are biographically more stable, better remunerated, and involve neither multiple jobs nor homework.

27Overall, we can note that no single group is touched cumulatively by all forms of atypical employment. In particular, it seems the exposition to forced flexibility is bound up with otherwise long-run contracts, reasonable pay, and a single employment. Those who have shorter hours simultaneously enjoy regular work hours and are not concerned by forced flexibility or night work. The only group that accumulates several relevant forms of atypical employment works in positions that are simultaneously limited in time, characterised by forced flexibility, and poorly paid.

28Furthermore, we are surprised about the wide spread of atypical employment. Atypical employment is indeed everywhere [Bourdieu, 1998). Even well-earning and well-educated labourers are affected by atypical forms of employment such as a large amount of work, work at home, flexible work, or work at night and on weekends. With respect to dualistic conceptions of the labour market, certain nuances must be introduced: while one part of the outsiders seem to be confronted by cumulative disadvantages, the landscape of the insiders is more varied and includes groups that are disadvantaged in some regards, but not in others. Even the otherwise most privileged group is not spared certain forms of atypical employment.

5. Resources and institution

29In this next step of analysis, we project a series of individual and structural factors as passive variables into the space constructed above: individual resources on the one hand and structural factors (i.e. economic sector and occupations) on the other. We will only display the cloud of individuals and the cloud of the selected passive variables. Again, the more closely individuals are situated to each other, the more categories they share. The aim of this section is to investigate if certain of the identified configurations of atypical employment can be explained by specific variables. According to Le Roux and Rouanet (2010, p. 59) in this space of individuals “distances greater than 0.5 will be deemed to be “notable”; a deviation greater than 1, definitely “large”.

30First, we introduce different resources and capitals into the space with which usually precarious employment is explained: age (as a measure of experience), gender and the family constellation, educational capital, and nationality (as a form of cultural capital). In the case where these resources are organised in a gradual way (age, educational level) we connect the categories of the variable with a line.

Figure 2. Individual Resources and Capital (2004)

Figure 2. Individual Resources and Capital (2004)

31Age, educational resources, and—to a certain degree—nationality form a diagonal axis from the lower-left to the upper-right quadrant. Poorly educated young adults stemming from traditional countries of working-class immigration are situated in the lower-left quadrant, whereas older employees with a university degree coming from northern countries are situated in the upper-right quadrant. These variables explain the difference between a configuration of temporal limited contracts, forced flexibility, and low income on the one side and a group of well-earning, long–working, and autonomously flexible fraction on the other.

32Only men and women are distributed according to a different pattern: women seem to be closer to part-time occupations in the left-upper quadrant and men can be found in the lower-right quadrant, characterised by full-time employment with night and weekend work. In particular, married (older) women with children seem to rely on part-time employment. Single (younger) women work to a higher percentage, but more often are time-limited or within flexible contracts.

33What is conspicuous when we examine the whole space is that certain zones are hardly related to this distribution of individual characteristics: in particular the distinction of the vertical axis, night and weekend work, forced flexibility, and more than one job. These configurations seem to have no common individual characteristics. People occupying them are not particularly young or old, neither poorly nor well-educated, and neither are they typically female or male. To explain them, we might turn to structural factors such as the economic sector or the occupation.

34For this purpose, we plot the Swiss typology of economic sectors into the space constructed by variables of employment. As they have no specific gradual order, they are not connected by a line.

Figure 3. Institutions (1): Economic Sectors

Figure 3. Institutions (1): Economic Sectors

35The variables of the economic sector are distributed along the vertical axis. The following sectors are particularly notorious for forced flexibility and night and weekend work: restaurants and hotels, transport, storage, communication, and—to a lesser degree—wholesale and retail. This means that in a sector like hotels and catering, even well-educated employees are forced to work flexibly or at hours incompatible with normal social rhythms. Multiple jobs or marginal employment are particularly frequent in the education sector. But in general, this upper half of the space is not particularly well explained by the economic sector. The same is true for the horizontal axis, defined by an opposition between well-earning and autonomous employees and the temporally limited employed.

36Very possibly, it is not whole sectors that offer similar employment conditions, but specific occupations (which are evenly spread among all sectors) that are more frequently touched by atypical employment. To test this assumption, we project the occupations of a detailed ISCO classification as a passive variable on the space of atypical employment.

Figure 4. Institutions (2): Occupations

Figure 4. Institutions (2): Occupations

37The inclusion of occupations complementary to economic sectors sheds lights also on zones that have so far not been well elucidated. Forced flexibility is a common problem of drivers and travel professionals of all sorts: train drivers, travel attendants, lorry drivers or controllers on ships and planes are the most concerned. Also, midwives and nurses, people working in security jobs, and special education professionals are confronted by specific forms of forced flexibility. The group in the lower-left quadrant that is both touched by forced flexibility and temporally limited contracts is much more heterogeneous: it includes both non-manual service workers (cashiers, personal care, restaurant services, messengers, and porters) and manual industrial workers (machine operators, mechanics, printing workers, and so on). It comprises both occupations of the traditional industrial male working class and those working in new routine service occupations [Oesch, 2006]. The analysis according to occupations confirms that in the upper-left quadrant we mainly find so-called female occupations: secretaries, clerks, and administrative professionals; primary school teachers, archivists and librarians, and associate teaching professionals; and social workers, associate health professionals, domestic helpers, and other personal services. The configuration characterised by small but regular employment is dominated by female occupations and professions, not by certain sectors.

6. Consequences and repercussions of precariousness

38In the next section, we will investigate how specific configurations of atypical employment are related to work satisfaction, feelings of depression, and political interest. The question will be, for instance, whether work dissatisfaction is widely spread across the whole space or if it is concentrated in specific configurations of atypical employment. Addressing issues raised by Paugam (2009) about the link between atypical employment and political attitudes, we hope also to get some tentative answers about political attitudes in different configurations of atypical employment.

Figure 5. Consequences of atypical employment: Work Satisfaction, Depression and Political Interest

Figure 5. Consequences of atypical employment: Work Satisfaction, Depression and Political Interest

39When we examine the distribution of those who are satisfied with their work, it appears that a high salary or autonomous flexibility is not positively related with a higher work satisfaction. Also, these good jobs, we could conclude, seem to have certain downsides, notably that they are chronically linked with a high work load, some night and weekend work, and occasional work at home. On the other hand, there is clear relation between certain, but not all, forms of atypical employment and job dissatisfaction. In general, there seems to be a link with forced flexibility and temporally limited employment contracts. Opposed to this, configurations of small and regular employment in the upper-right quadrant, as well as configurations of forced flexibility in the lower right quadrant, are not particularly affected by work dissatisfaction.

40How is the feeling of depression distributed? The situation is similar to work satisfaction when it comes to the absence of depression. People on the right side of the plane are not exceptionally rarely afflicted by depression. On the other hand, those who are employed in situations where biographical insecurity is coupled with forced flexibility are more often depressed. All the other configurations of atypical employment are not particularly linked to depression.

41As a last dimension, we consider the interest in politics. This aspect is part of Paugam’s hypothesis that certain employment conditions can also contribute to political disillusionment or hinder political organisation [Paugam, 2000]. Again, those in the lower-left quadrant, characterised by temporally limited contracts and forced flexibility, lack in political interest. Those on the right side brim with interest in politics.

7. A nuanced picture of atypical and precarious employment

42Starting from the assumption that scholars frequently struggle to grasp atypical and precarious employment empirically and to describe it theoretically, we tried in this paper to capture it with a configurational approach. Based on the data of the 2004 SHP, the concept of field, and a multiple correspondence analysis, we attempted to identify relationships between certain configurations of atypical employment, endowment with capital, and the space of attitudes and practices.

43We show that in Switzerland, atypical employment in one of its numerous variants is rather widespread and also touches actors who are well endowed with educational or economic capital. Today, atypical employment touches a wide proportion of the employed population and is not confined to a small group of excluded workers or to specific zones of the labour market [Marchart, 2010]. This dispersion entails that there is no dual relationship between a pole that reunites all disadvantages of atypical employment and a pole which is spared from all disadvantages. This is theoretically important because it shows that both rigidly dualistic conceptions of the labour market [Doeringer, Piore, 1971; Palier, 2012] and cumulative threshold theories [Rodgers, 1989; Walker et al., 2010] fail to capture the nuances of the Swiss labour market. Our analysis revealed four configurations of atypical employment:

  • The first of these configurations is defined by forced flexibility, a temporally limited engagement, and material insecurity. It accumulates difficulty in planning for the future, discordant weekly social rhythms, and lower income. However, these people do not work at night or on weekends, work away from home, and do not have several jobs.

  • A second form of atypical employment seems to combine night work and forced flexibility, in most of the cases in the form of shift work. Here the income situation is better and the long-term security is assured as well. These workers have one main job apiece and do not work at home.

  • The third form of atypical employment could be called marginal employment, which sometimes also involves several jobs. In these situations, the labourers are not required to

44It is important not to level out these configurational nuances with dualistic or cumulative conceptions because of all the different explanations and consequences that are bound to these configurations. Certain configurations are populated by individuals with a specific amount and composition of capital: the lack of experience in the labour market (as expressed by a young age), poor educational resources, and an origin from a poor country are related to the configurations characterised by a combination of forced flexibility, biographical insecurity, and low wages. A university degree, an origin from a northern country, and—to a certain degree—a higher age (and therefore longer work experience), explain the combination of high income, overtime, autonomous flexibility, and occasional homework. Marginal but regular employment can be related to “female” occupations (such as clerks, teachers, assistant teachers, and other service occupations) and the family constellation. It seems this configuration is typical for women who work in often subordinate female occupations that are often linked to part-time and that offer only few career opportunities [Wetterer, 1992]. These jobs are biographically relatively stable, require no night work, and often have regular hours. This could be one reason why these configurations are frequently associated with married women with children: in Switzerland, having children is still often linked to a reduction of the employment rate of women (whereas men stay in full-time employment) [Bühlmann et al, 2010].

45Not all configurations can be related to the amount and composition of individual resources. The best explanation for the combination of forced flexibility and night and weekend work is the fact that these jobs are in the sectors of transport and communication or agriculture. It is in occupations such as train and lorry drivers, travel attendants, protective services, or special education occupations that we find these constellations.

46Even though atypical employment is widespread and multi-faceted, our analysis allows a distinction between atypical and precarious employment. Of all the configurations of atypical employment, only one particular is related to dissatisfaction, psychological problems, and political disillusionment. In a cumulative and coherent way, these attributes are concentrated in the lower-left quadrant, which is characterised by a combination of poor pay, contracts limited in time, and forced flexibility. These jobs are situated in the gastronomic, wholesale, and retail sectors. They include occupations such as housekeeping and restaurant services, messengers and porters, cashiers and tellers, client information clerks, and personal care occupations. They also include industrial occupations such as mechanics, plant machine operators, craft printing, or elementary occupations. This configuration is a blend of the new service proletariat and the traditional industrial proletariat [Oesch, 2006]. All other configurations are, despite their atypical character, not affected by low work satisfaction, the absence of psychological well-being, and the lack of political involvement. Thus, certain configurations that have traditionally been considered “secondary” or “peripheral” are obviously not that unfavorable, per se, and therefore challenge the strict divide into insiders and outsiders.

47These atypical forms are characterised by several elements. The group that suffers from precarious conditions is devoid of all possible resources and capital. They possess no educational resources, labour market experience, or social capital (in form of household members who would complete their income). Finding themselves in these positions is, in a way, simply the result of a negative allocation process. The composition of their capital is such that they are not able to compensate for the lack of one sort of capital with another. People in other atypical but not precarious configurations have at least one compensation mechanism at their disposal and use a part of their resources strategically: women working in part-time positions, especially when it concerns a low percentage of work, are also at risk of poverty. However, they can rely on the social capital that links them to other members of the household and completes their salary so that the household equivalence income is more than 42,000 Swiss francs. The majority of men who work in occupations with unusual hours and shift work in the lower-right quadrant seem to trade working hours that are hardly compatible with the social rhythms of others against a slightly better salary and biographically higher security. In addition, as their situation is shared collectively with many others in the same sector or occupation, it may be psychologically easier to bear than other situations of atypical employment. Therefore, the relational attributes of employment situations and the potential compensation strategies that are at the disposal of employees must be taken into account more systematically to understand the structure of contemporary labour markets.

Haut de page


BARBIER J.-C. (2005), La précarité, une catégorie française à l’épreuve de la comparaison internationale, Revue française de sociologie, vol. 46, pp. 351-371.

BOLTANSKI L., CHIAPELLO E. (1999), Le nouvel esprit du capitalisme, Paris, Gallimard.

BOURDIEU P., DE ST. MARTIN M (1978), Le patronat, Actes de la recherche en sciences sociales, vol. 20-21, pp. 3-82.

BOURDIEU P. (1979), La Distinction, Paris, Éditions de Minuit.

BOURDIEU P. (1984), « Quelques propriétés des champs », in P. Bourdieu, Questions de sociologie, Paris, Éditions de Minuit, pp. 113-120.

BOURDIEU P. (1984), Homo academicus, Paris, Éditions de Minuit.

BOURDIEU P. (1989). La noblesse d’État. Grandes écoles et esprit de corps, Paris, Éditions de Minuit.

BOURDIEU P. (1998), Contre-feux, Paris, Liber, Raisons d’agir.

BÜHLMANN F., ELCHEROTH G., TETTAMANTI M. (2010), The Division of Labour Among European Couples: The Effects of Life Course and Welfare Policy on Value–Practice Configurations, European Sociological Review vol. 26, pp. 49-66.

CASTEL R. (1995), Les Métamorphoses de la question sociale, une chronique du salariat, Paris, Fayard.

CASTEL R. (2003), L’insécurité sociale. Qu’est-ce qu’être protégé?, Paris, Seuil.

CASTEL R., DÖRRE K. [eds] (2009), Prekarität, Abstieg, Ausgrenzung. Die Soziale Frage am Beginn des 21. Jahrhunderts, Frankfurt, Campus.

CROUCH C. (2010), Flexibility and security in the labour market: An analysis of the governance of inequality, Zeitschrift für Arbeitsmarktforschung, vol. 43, pp. 17-38.

DE NOOY W. (2003), Fields and Network Analysis: Correspondence Analysis and Social Network Analysis in the Framework of Field Theory, Poetics, vol. 31, pp. 305-27.

DESROSIÈRES A. (2003). « Bourdieu et les statisticiens: une rencontre improbable et ses deux héritages », in  P. Encrevé , R.M. Lagrave, Travailler avec Bourdieu, Paris, Flammarion, pp.209-218

DOERINGER P., PIORE M. J. (1971), Internal Labor Markets and Manpower Analysis, Lexington, Heath.

DIPRETE T. A. (2002), Life Course Risks, Mobility Regimes, and Mobility Consequences: A Comparison of Sweden, Germany, and the United States, American Journal of Sociology, vol. 108, pp. 267-309.

DÖRRE K (2006), Prekäre Arbeit. Unsichere Beschäftigungsverhältnisse und ihre sozialen Folgen, Arbeit. Zeitschrift für Arbeitsforschung, Arbeitsgestaltung und Arbeitspolitik, 15, pp. 181-193.

FLIGSTEIN N (2001), Social Skill and the Theory of Fields, Sociological Theory, vol. 19, pp. 105-125.

FLIGSTEIN N (2002), The Architecture of Markets. An Economic Sociology of Twenty-first-century Capitalism, Princeton, Princeton University Press.

GALLIE D. (1998), Restructuring the Employment Relationship, Oxford, Oxford University Press.

GALLIE D., PAUGAM S (2000), Welfare regimes and the experience of unemployment in Europe, Oxford, Oxford University Press.

KALLEBERG A. L. (2000), Nonstandard employment relations: Part-time, temporary and contract work, Annual review of sociology, vol. 26, pp. 341-365.

KALLEBERG A. L. (2009), Precarious work, insecure workers: Employment relations in transition, American Sociological Review, vol. 74, pp. 1-22.

KRAEMER K. (2008), Prekarität–was ist das?, Arbeit, 17, pp. 77-90.

LE ROUX B., ROUANET H. (2010), Multiple Correspondence Analysis, London, Sage.

MAGNIN C. (2005), Prekäre Integration. Die Folgen unsicherer Beschäftigungsverhältnisse, Wien, IHS.

Marchart O. (2010),  Auf dem Weg in die Prekarisierungsgeselischaft. Zur Analyse des Definitionskampfs um die zunehmende Prekarisierung von Arbeit und Leben, Schweizerische Zeitschrift für Soziologie, vol. 36, pp. 413-429.

MARTIN J. L. (2003), What Is Field Theory?, American Journal of Sociology, vol. 109, pp. 1-49.

OESCH D. (2006), Coming to Grips with a Changing Class Structure. An Analysis of Employment Stratification in Britain, Germany, Sweden and Switzerland, International Sociology, vol. 21, pp. 263-288.

PALIER B., EMMENEGGER P., HÄUSERMANN S., SEELEIB-KAISER M. [eds] (2012), The age of dualization, the changing face of inequality in deindustrializing societies, Oxford, Oxford University Press.

PAUGAM S. (2000), Le salarié de la précarité. Les nouvelles formes de l’intégration professionnelle, Paris, PUF.

PAUGAM S. (2009), Die Herausforderung der organischen Solidarität durch die Prekarisierung von Arbeit und Beschäftigung, In: Castel R, Dörre K (eds) Prekarität, Abstieg, Ausgrenzung. Die Soziale Frage am Beginn des 21. Jahrhunderts, Frankfurt, Campus.

PELIZZARI A. (2009), Dynamiken der Prekarisierung. Atypische Erwerbsverhältnisse und milieuspezifische Unsicherheitsbewältigung, Konstanz, UVK.

PIORE M. J. (1970), “The Dual Labor Market. Theory and Implications”, in S-H Beer, R.E. Barringer (eds), The State and the Poor, Cambridge, Winthrop Publishers.

PRODOLLIET S. (2000), Arbeit in der Grauzone. Die neuen prekären Arbeitsverhältnisse, in Caritas Schweiz (ed), Sozialalmanach 2000. Sozialrechte und Chancengleichheit in der Schweiz, Luzern, Caritas Verlag, pp. 115-131.

RODGERS G., RODGERS J. (1989), Precarious jobs in labour market regulation: The growth of atypical employment in Western Europe, Geneva, International Institute for labour studies.

SAVAGE M., PRIEUR A. (2011), Updating cultural capital theory: A discussion based on studies in Denmark and Britain, Poetics, vol. 39, pp. 566-580.

SENNETT R. (1998), The corrosion of character: The personal consequences of work in the new capitalism, New York, Norton & Company.

STRECKEISEN P. (2012), “Steigende Erwerbslosigkeit und Prekarität in der Schweiz: Das Ende eines Sonderfalls”. in K. Scherschel, P. Streckeisen, M. Krenn, Neue Prekarität. Die Folgen aktivierender Arbeitsmarktpolitik – europäische Länder im Vergleich, Frankfurt, Campus, pp. 47-73.

VOORPOSTEL M. (2010), Attrition Patterns in the Swiss Household Panel: An Analysis of Demographic Characteristics and Social Involvment, Swiss Journal of Sociology, vol.36, pp. 359-377.

VOORPOSTEL M. et al. (2011), Swiss Household Panel Userguide (1999-2010), Wave 12, October 2011, Lausanne, FORS.

WALKER P., MARTI M., BERTSCHY K. (2010), Die Entwicklung atypisch-prekärer Arbeitsverhältnisse in der Schweiz, Bern, Seco.

WETTERER A. [ed] (1992), Profession und Geschlecht. Über die Marginalität von Frauen in hochqualifizierten Berufen, Frankfurt, Campus.

Haut de page


1 Which is useful for the analysis of other phenomena; we are not making a general, but a specific argument against regression modelling.

2 See, for example: Bourdieu et de St. Martin, 1978; Bourdieu, 1979; Bourdieu, 1989. For an external view on the link between field theory and multiple correspondance analysis: Desrosières, 2003.

3 In Switzerland, this concerns, for example, eligibility for unemployment benefits.

4 In Switzerland, to be included in the “second pillar” of the pension fund system, employees must earn at least 20,880 francs per year.

5 Certain researchers doubt that quantitative surveys can give answers to questions of work precariousness (Pelizzari, 2009). The doubts about the ability of these surveys to capture those who are the most unprotected and whose lives are so volatile that they are hardly reachable by telephone surveys must be taken very seriously.

6 Of course, the concentration on a single year can also be problematic, as the meaning of atypical jobs may also change with its duration. One month of work on call might not be a problem, but ten years probably are.

7 We also carried out MCA’s that included the whole active population. These analyses revealed rather comparable structures, but gave us less detailed information concerning the different configurations of atypical employment.

8 Pseudo-self-employment remains important because it has a long series of consequences in terms of social protection: there is no compulsory pension fund system, the self-employed are not insured against unemployment, and more.

9 Using the occupational typology as a passive variable, we are not compelled to limit the number of categories. However, in the graph we only display the categories with a number of members > 10.

10 These numbers draw on the Swiss Labour Force Survey. See also Pelizzari (2009) for the period 2001-2006.

Haut de page

Table des illustrations

Titre Figure 1. The space of a-typical employment in Switzerland (2004)
Légende Legend: categories contributing above average to the construction of: horizontalaxis are bold; vertical axis are in italics; to both axes are in bold italics.
Fichier image/jpeg, 176k
Titre Figure 2. Individual Resources and Capital (2004)
Fichier image/jpeg, 124k
Titre Figure 3. Institutions (1): Economic Sectors
Fichier image/jpeg, 116k
Titre Figure 4. Institutions (2): Occupations
Fichier image/jpeg, 196k
Titre Figure 5. Consequences of atypical employment: Work Satisfaction, Depression and Political Interest
Fichier image/jpeg, 106k
Haut de page

Pour citer cet article

Référence papier

Félix Bühlmann, « Catch me if you can », Espace populations sociétés, 2013/3 | 2013, 73-90.

Référence électronique

Félix Bühlmann, « Catch me if you can », Espace populations sociétés [En ligne], 2013/3 | 2013, mis en ligne le 31 décembre 2015, consulté le 28 mars 2017. URL : ; DOI : 10.4000/eps.5554

Haut de page


Félix Bühlmann

Bâtiment Geopolis
Université de Lausanne
CH. 1015 Lausanne

Haut de page

Droits d’auteur

Licence Creative Commons
Espace Populations Sociétés est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Université de Lille 1 - Sciences et technologies
  • Logo CNRS - Institut des sciences humaines et sociales