Institutional and Individual Constraints of an “Un- gendered”

Order of Professions

Kathrin Leuze∗ and Alessandra Rusconi, University of Bremen, Germany

- First draft. Please do not quote without authors’ permission. -

Abstract

Occupational gender segregation is a persistent source of social inequalities. However, the

increasing participation of women in higher education has given hope that among the highly

qualified gender inequalities will diminish. Optimists believe that particularly the rise of a

service economy increases the chances for women of working in high-skill occupations, such

as professions. The paper asks whether such optimistic accounts are justified by comparing

male and female professional career trajectories in Germany. Our main assumption holds that

strong gender differences persist between public and private sector professions and is further

aggravated by different forms of family commitment.

Key words: professions, gender segregation, labor market outcomes, higher education, family

formation

∗ Corresponding author: kathrin.leuze@sfb597.uni-bremen.de

1

Introduction

One of the most striking features of recent decades has been the persistent upward trend in

female employment across Europe. Also in Germany there has been a continuous rise of

female labor force participation rates: from ca. 30 per cent in 1970 to 41 per cent in 2001

(Statistisches Bundesamt 2002: 89). When looking for explanations of increasing female

employment rates, rising educational levels among women are often considered one of the

main factors (Charles 2005; Estevez-Abe 2005; 2006; Gottschall and Shire 2006; Rubery and

Grimshaw 2003). In Germany, for example, the share of women with a higher education

degree rose from 2 per cent in 1971 to 14.5 per cent in 2004 (Rusconi and Solga

forthcoming). Among highly educated women there is a common pattern of relatively high

labor force participation across countries. This is particularly the case in service sector

occupations (European Commission 2006: 62). The rising expansion of the service sector in

all Western societies has lead to an increase of high-skill/high-pay jobs, such as professionals.

In Germany, for example, the proportion of people employed as service sector professionals

or managers increased from 25 per cent 1984 to 37 per cent in 2002. High-skill service sector

employment thus constitutes the sector with the highest growth rates during this time period

in Germany (Fagan et al. 2005: 10).

The effects of rising female graduation rates are only just beginning to be felt and often lead

to the positive evaluation that equal participation of women and men in higher education will

eventually lead to strong decline of gender inequalities in the labor market. Also, the rise of

high-skill service-sector occupations, such as professions, should strongly increase the

chances for highly-qualified women of facing similar career prospects as their male

counterparts. This paper asks whether such optimist claims are justified by analyzing patterns

of sex segregation among professionals. Can we observe increased gender equality in the

career trajectories of female and male professionals? There are already some empirical hints

that this might not be the case. Particularly the fact that the professions were and still are

highly segregated horizontally into public and private sector professions (Leuze 2007) should

have implications for the gendered development of career trajectories. At the same time, the

choice of fields of study in higher education is still highly gender segregated and should also

play a role for gendered job allocation processes in the professional sphere (Jacobs 1995;

Smyth 2002). In the following, we develop a theoretical framework based on the theory of

labor market segmentation that helps to explain differences in the career perspectives of

2

public and private sector professions. By making reference to the societal roles of caretaking

and moneymaking, we will show how institutional and individual constraints might carry

forward patterns of gender segregation between public and private spheres. Our main

assumption holds that strong gender differences persist between public and private sector

employment and is further aggravated by different forms of family commitment. Given the

scarcity of available data for this endeavor, the derived hypotheses will be tested by

combining a longitudinal analysis of the transition from higher education to a first

professional placement using the German Socio-Economic Panel (GSOEP) with a cross-

sectional analysis of the impact of family foundation on professional career outcomes using

the German Microcensus (2000).

Theoretical Framework

Gendered occupational segregation has received much attention in recent years in order to

account for the persistence of gender inequalities in the labor market. It has become common

to focus on two kinds of occupational segregation: vertical and horizontal segregation.

Vertical segregation refers to the under-representation of women in high-status occupations

and their overrepresentation in low-status occupations. Horizontal segregation, in turn, refers

to the under-representation of women in particular occupational fields or sectors of the

economy and their overrepresentation in others, constituting typical male and female

occupations. Today, both forms of sex segregation persist in virtually all countries, as the

United Nations measure of Gender Empowerment (GEM) tells. The various components of

GEM indicate that everywhere women are underrepresented in seats in parliament, in

administrative, managerial, professional and technical occupations and generally have lower

levels of income (UN 2001).

Particularly the changing composition and structure of employment in advanced industrial

countries following from service sector expansion has given rise to new forms of gender

segregation. Even though the level of service sector growth varies across countries, it is a

common phenomenon that the expansion of service sector jobs has occurred at both ends of

the job hierarchy: high-skilled/high-wage and low-skilled/low wage. So far, empirical

literature has mainly dealt with the vertical segregation following from this divide, since

female employment is in general more heavily concentrated in low-skill service sector

occupations and among the low pay (Fagan et al. 2005; Rubery and Grimshaw 2003; Rubery

et al. 1999). But also on the level of high-skill/high-pay occupations, gender segregation

3

persists, even though mainly in form of horizontal segregation. On the one hand, women are

now increasingly working in particular professional areas associated with education, health,

social sciences and some business-related professions. However, they continue to be

underrepresented among managers in general and in many professional areas such as

engineering and ICT (Fagan et al. 2005). The system of professions therefore provides a good

example for analyzing how gender segregation nowadays develops and manifests itself among

highly qualified women and men.

Labor Market Segmentation and the Public-Private Order of Professions

Professions can be defined as occupations with a high exclusiveness of their knowledge base

systematically delineating a specific occupational domain (Abbott 1988; Brater and Beck

1981; Heidenreich 1999). Structured professional training forms the prerequisite for entry into

the profession, based on which the specific fields of activity become exclusively reserved for

the members of a professional group. Further career development thereafter strongly depends

on the status interest of professional bodies. The social closure of career lines takes place

particularly by the official acknowledgement of professional training and furthermore, by the

legal protection of professional titles (Heidenreich 1999). Prestige as much as organizational

autonomy or material position can thus be used by professional groups in maintaining their

exclusive role and jurisdiction within the labor market.

However, the labor market segment of professions is not a homogenous one, but is segmented

into private and public sector professions. The theoretical notion of labor market segmentation

implies that the labor market is divided in several segments, all of which offer specific career

prospects and are characterized by a high degree of social closure. This closure is based on

specific certificates serving as a necessary precondition for entering a particular segment as

well as specific allocation principles and career structures serving as a means to maintain

segmental closure and to prohibit mobility between the segments (Kalleberg and Sorensen

1979). A segment may be defined by occupations, industries, organizational characteristics,

or, as in this paper, by public and private sectors.

The main differences found between public and private sector professions can be related to

the theoretical differentiation between internal and external labor markets. External labor

markets function in line with the “pure” market logic, where allocation and mobility decisions

are controlled directly by mechanisms of labor demand and supply (Kerr 1954). Internal labor

4

markets, on the other hand, are according to Doeringer (1967) defined as “an administrative

unit within which the market functions of pricing, allocation, and often training are

performed. It is governed by a set of institutional rules which delineate the boundaries of the

internal market and determine its internal structure” (Doeringer 1967: 207). In an internal

labor market, employees are assumed to enter at specific entry positions and thereafter pursue

their careers at least partly protected from market competition by following particular career

ladders and chances for promotion. Recruitment from the external labor market ideally takes

place only once, when external applicants are employed for a restricted number of specific

“entry-jobs”.

There are a number of reasons why employers and employees are keen on establishing

internal labor market career arrangements. For employers, internal markets are profitable

since they reduce costs for education, training and personnel selection. Screening and training

investments occur only once at the beginning of an internal career, while the process for

promotion follows institutionalized rules which are easy to assess. At the same time, internal

labor markets keep personnel fluctuation low and ensure that initial investments in employees,

such as on-the-job training or further training, pay off in the long run. For employees, internal

labor markets offer long-term employment security through stable positions and foreseeable

career prospects. They can make sure that training investments in firm-specific skills are not

lost through job shifts. Overall, internal labor markets guarantee a lasting planning security

and protect the investments in human capital for both employers and employees (Doeringer

1967; Doeringer and Piore 1971; Kalleberg and Sorensen 1979; Rosenfeld 1992).

In the literature, public sector employment has been identified as the prototype of firm-

internal labor markets (Becker 1993). The employment in state administration is strongly

associated with highly protected labor arrangements within the public enterprise, i.e. with

explicitly defined “ports of entry” at the lower end of the job hierarchy, stable employment

relationships, calculable promotion schemes often based on seniority entitlements, and an

almost complete closure of higher level positions from the external labor market. The

strongest form of such an internal labor market career protection in the German public service

is given to state civil servants (Beamte), who are guaranteed life time employment, stable

career progression and high pensions after retirement (Rothenbacher 1999; 2004). Based on

these characteristics, employment in the public service constitutes the purest form of firm-

internal mobility since it is much more strongly protected from market competition than firm-

internal labor markets in the private sector.

5

Professions in the public sector therefore should exhibit a higher institutional protection of

career trajectories than professions in the private sector. Competition is likely to occur only at

entry-ports, while latter mobility mainly depends on internal rules, such as seniority

entitlements. This should in general guarantee steady and calculable career progression and a

high protection against unemployment. Opting out of the labor market in times of parental

leave should delay, but not hinder further career development thereafter due to the formalized

nature of rules of progression. At the same time, the highly regulated career progression does

not allow for high upward mobility shifts even in times of economic upturns. These lower

chances of upward mobility set an upper limit to monetary and status returns for life-time

employment.

Compared to internal labor market arrangements, private sector professions offer a lower

institutional protection of career trajectories. Even though the high level of professional

qualification also constitute a means for social closure, professionals in the private labor

market segments have to cope with higher risks than those in the public sector. First of all

they cannot rely on the existence of entry-port occupations, where competition for

employment ideally takes place only once. Private sector professionals face competition for

recruitment and career progression throughout their careers. In this regard, they not only

compete with colleagues from within their own company, but also with applicants from the

external labor market. If professionals are self-employed, the risks of market competition are

even aggravated. Consequently, private sector professionals face a greater risk of “turbulent”

career development, since they have lower institutional safety nets against unemployment or

bankruptcy. This is particularly the case for career interruptions, such as child birth, where re-

entering parents have to compete with colleagues holding “gap-less” career trajectories. The

positive sides of less institutionalized career trajectories are the higher chances of upward

mobility, both monetary and status. The less formalized structure of promotion and payment

can much more flexible react to economic booms, often leading to faster upward career and

income shifts.

On first sight, these institutional differences between professional career structures in the

public and private sector can be considered gender-neutral. However, as the short reference to

career interruption has already hinted to, the public-private segmentation of professional

occupations should provide varying incentives for men and women and therefore lead to

horizontal sex segregation in the professional labor market. The following section will explore

6

the individual and institutional basis for such gendered career perspectives by making special

reference to the welfare state and work/life strategies.

Institutional and Individual Constraints of Un-gendered Professional Careers

Institutional Contraints

On the institutional side, previous research shows the existence of a horizontal segregation

among field of studies, whereby ‘female’ subjects are less rewarded on the labor market

(Allmendinger and Podsiadlowski 2001; Dressel 2005; England 2005; Reskin and Padavic

1994). The share of men and women enrolled and achieving a university degree in Germany

varies considerably in different fields of study: in 2005, 70 per cent of the students enrolled in

linguistic and humanities were women, while their share was only 37 per cent in natural

sciences and 20 per cent in engineering (Statistisches Bundesamt 2007: 27). Most

importantly, different fields of study offer “gendered” labor market prospects as regards the

work/life balance. This is particularly the case for typical ‘male’ fields of studies and

occupations such as technical and natural sciences. While generally graduates from these

fields of study have better chances on the labor market than those holding humanities or social

sciences degrees, female graduates in ‘male’ subjects face greater difficulties in the transition

to an occupation (Janshen and Rudolph 1987; Minks and Filaretow 1996; Schreyer 2000).

The majority of women with an engineering degree do not work as an engineer or in a

technical occupation (Haffner et al. 2006: 8). Professional success in technical fields is less

related to objective criteria of achievement (such as professional mobility, international

experience and further training), but rather to more informal, unspoken rules. These refer to

the private life situation and are linked to normative expectations on professional commitment

that demand an exclusive identification with the occupation; particularly with regard to (long

and unpredictable) working hours and frequent absences from home (Haffner et al. 2006).

Such working culture hinders any engagement outside the occupation and endorses the

traditional male breadwinner model (Haffner et al. 2006: 6).

Yet, among typical male field of studies differences between public and private labor market

segments can be observed, too: over half of physics professionals are employed in the public

sector (especially in universities and research institutions), while the grand majority of

chemistry and engineering graduates work in private enterprises (Haffner et al. 2006: 14-15).

Particularly in the private sector, professional success of female and male graduates diverges

7

significantly, because success is more closely related to the amount of working hours. In

addition, in the private economy the actual presence at the working place (firm) appears to be

important: among professionals with same amount of weekly working hours those who

regularly work at home are less successful (Haffner et al. 2006: 40). Thus, professional

careers in ‘male’ subjects/occupations expect and entail a ‘male’ work-centered biography –

particularly regarding work time availability. Individuals with other (female) types of

biographies who are not able or willing to follow such work-dominated lifestyles – e.g.,

because of childcare duties – face the risk to be excluded or hindered in their career

development.

A second institutional source of gender inequality in professional careers is the so-called

‘statistical discrimination’. This refers to employer behavior that discriminates against women

on the basis of gender-typical expectations regardless of the actual (family) commitments and

work/life arrangements of individuals. Previous research shows that employers expect also

female graduates to be less career-oriented and productive than men and assume that these

women would reduce (or even quit) their professional engagement for the benefit of their

family (England 2005; Konrad and Cannings 1997; Reskin and Padavic 1994; Stroh and

Reilly 1999). On the basis of such traditional gender-roles expectations, employers are

uneager to hire and promote women. This results not only in horizontal, but also in

contractual and vertical segregation. With regard to the former, women are offered more

frequently than men untypical and precarious employment forms such as fixed-term and part-

time contracts (Dressel 2005; for German academia see e.g., Fuchs et al. 2001; Krimmer et al.

2003). Concerning vertical segregation, men and women occupy different hierarchical levels

and functions, and the proportion of women decreases at every step up the career ladder.

Employer’s traditional gender-roles expectations, formal promotion criteria such as seniority

in a company or firm, and wage agreements often work in favor of men resulting in a higher

share of men in higher and leadership positions (Dressel 2005: 133).

A third source of gender inequality in professional chances is the presence and availability of

childcare facilities as institutional means to enhance (female) participation on the labor

market. Previous research shows that women can realize and negotiate with their partners an

own employment and career only if they are able to externalize childcare and housework

(Kirner and Schulz 1992; Stephens 1999; Swiss and Walker 1993). Thus, public or private

childcare is often regarded as paid substitution for women. Consequently women’s

employment outside home is challenged when the price to be paid is too high, or the presence

8

or opening hours are inadequate (Hertz 1986). Up to now childcare facilities for children

under three are insufficient, especially in West Germany: here only 10 per cent of the children

younger than three attend childcare (the corresponding proportion in East Germany is 40 per

cent) (Bien et al. 2007: 6). Due to a lack of childcare services 80 per cent of dual-earner

families and 70 per cent of single-parent families with children under three must rely upon

alternative childcare solutions (Bien et al. 2007: 7).1 Yet, not only the number is insufficient,

childcare often does not meet the needs of parents. For every fifth child under three years and

every sixth child aged three to six the childcare opening hours do not (or only barely) cover

parents’ working hours (Bien et al. 2007: 12). If at all, childcare opening hours conform only

to ‘traditional’ nine-to-five working days and often do not meet the needs and expectations of

many highly qualified professionals. The lack and inadequacy of childcare services indicates

an institutionalized ‘traditional’ gender-roles ideology that conceives and depends upon

mothers as primary caregivers. As result, women reduce – at least temporary – their

employment and careers. Yet, any (even only temporary) work interruption or working hours

reduction entails the risk of a more or less permanent professional setback since career

requirements, such as age standards, are often based on male (full-time) continuous careers

and biographies (Born 2001; for German academia see e.g., Fuchs et al. 2001; Geenen 1993;

Vogel and Hinz 2003).

Individual Constraints

On the individual side, gender roles shape individual choices as well. Previous research shows

that gender-specific socialization processes influence the choice of fields of study (England

2005; Jacobs 1989; Lueptow et al. 2001). Gender-typical socialization leads individuals to

choose gender-appropriate fields of studies; i.e. subjects that are regarded being socially

acceptable for a person of one’s own sex. On the one hand, appropriateness might refer to

stereotypes of what is typically masculine (e.g. analytic thinking and accordingly mathematics

or life sciences), and what is typically feminine (e.g. nurturing and thus human sciences or

education) (Jacobs 1995). On the other hand, it might concern gender roles expectations

regarding the division of labor within the family. Anticipating their role as primary caregivers

women avoid enrolling in fields of study that lead to professions perceived being

incompatible with a family, whereas men foresee their role as primary breadwinner and thus

are not hindered in their choices by potential family/career conflicts (Ware and Lee 1988).

1 According to the same research, 84 per cent of the non-employed mothers of children under three years report that they aspire to employment; for 55 per cent of these women missing or inadequate childcare facilities were one of the reasons for being out of the labor force (Bien, Rauschenbach and Riedel 2007: 8).

9

Rather men avoid enrolling into female subjects as there are only lower incentives to enter

devaluated (i.e. under-rewarded) female spheres (England and Li 2006). As German

researchers observe, the definition of professional success might vary between the gender too,

and more women than men define success by the compatibility of a profession and a family

(Hoff et al. 2002). However, socialization processes are not limited to the early (pre-

university) life, but ‘social pressures throughout the life cycle continually produce and

reproduce the sexual division of labor’ (Jacobs 1995: 83). And education institutions

themselves appear to reinforce gender segregation, as many women initially enrolled in male

fields of studies switch to female ones during the course of their studies (Jacobs 1989; 1995).

A second individual source of gender inequality is the type of coordination arrangement

between the partners. Due to education expansion the number of couples in which both

partners hold an academic degree has increased; particularly for female university graduates it

is common to live with an equally high qualified partner (Blossfeld and Timm 2003; Rusconi

and Solga forthcoming).2 Since both partners have made substantial training investments,

academic couples often show a pronounced interest in both partners’ professional careers, but

they face as well specific constraints in transposing their potential into dual-careers.

Particularly the temporal and geographical coordination of two careers while simultaneously

taking care of a partner and eventually children constitute the main challenges for couples in

which both partners purse a professional job.

Previous research shows three employment arrangements by which couples coordinate their

family and career(s). The most common is the hierarchical model in which only one partner –

often the man – has the career role, while the other partner supports this leading career

through a primary responsibility for ‘family matters’. Even among academic couples, a

common strategy is to follow the (male) partner with the better career prospects and

opportunities; often resulting in a ‘leading’ and a ‘following’ career. The ‘following’ partner

purses an own occupation only within the context of the leading career’s commitments and

requirements; e.g. women often interrupt their employment, accept jobs with a lower income

or reject job offers which would require relocation (Becker and Moen 1999; Boyle et al. 2001;

Deitch and Sanderson 1987; Marwell et al. 1979). Once women have scaled back and male

careers have become predominant it is quite difficult reverse this pattern, since in order to

revive their careers women have to start at lower (usually precarious and less paid) positions.

2 In 2004, one third of German male academics and almost half of female academics live together with an equally academic trained partner (Rusconi and Solga forthcoming).

10

Thus, the strategy of following the ‘leading’ career prospects has self-reinforcing effects on

reproducing gender differences in professional careers.

A different coordination strategy is the individualistic model, in which both partners

independently pursue their own careers and the partnership itself plays a secondary role. This

arrangement is frequently linked with long-distance or commuter relationships, with the goal

of optimally structuring the career chances of both partners (Kilpatrick 1982). However, the

birth of a child challenges this individualistic arrangement, because it rests on the male

partner’s acceptance of a female career – and this only as long as his own career is not

jeopardized by it (Hertz 1986; Levy and Ernst 2002). Thus, as soon as career/family conflicts

arise traditional gender roles expectation might be revitalized and often couples who initially

followed an individualistic arrangement turn to more traditional hierarchical models (Levy

and Ernst 2002; Schulz and Blossfeld 2006).

A third, albeit less common, arrangement is the egalitarian model, in which partners give an

equal importance to their professional careers and their family. Both partners are willing to

make compromises in their careers for the benefit of their family or for an optimal

combination of career opportunities for both partners (Becker and Moen 1999; Behnke and

Meuser 2005; Costa and Kahn 2000; Dettmer and Hoff 2005; Hardill et al. 1999). Couples

who follow such arrangement do not fully take advantage of their career potentials and

possibly have put up with restrictions in their professional development. Noteworthy, given

employers’ expectation of gender roles congruence, men who do not follow work-dominated

lifestyles but reduce their professional commitment in order to accommodate family demands

might be more strongly penalized than women (as the latter are to some extent expected to do

so) (Konrad and Cannings 1997).

Finally, the presence of children is known to influence females’ professional careers. Several

scholars note that many couples shift to a more traditional division of labor within the family

after the birth of a child (see above). This shift reflects still prevalent gender roles (applying

also to academic women) according to which, mothers should accommodate family needs,

whereas fathers should ensure the financial resources of the family. In consequence, women’s

career chances diminish, since even temporarily compromises often have enduring negative

consequences for professional careers. The impact of children might vary, however, with

employment-family-coordination patterns of couples as well as professional fields (see

above). In order to avoid career/family conflicts, another strategy is to renounce to or

postpone having children (Hoff et al. 2002; Swiss and Walker 1993). In Germany this appears

11

to be a strategy more frequently followed by highly qualified women, who are to a much

higher extent childless than less qualified women, but also than highly qualified men (Huinink

1995; for differences among German professors see Krimmer et al. 2003).

Hypotheses

As the previous sections have shown, there are a number of reasons why women and men face

differing constraints and incentives to enter public or private sector professions. Due to the

more protected career prospects offered by public sector professions, we assume that this

labor market segment is particularly attractive for women. Since German women are still

primarily responsible for family matters, we expect that after graduation women will seek

employment more often in public professions. Simply the fact of being female, with its

stereotypical socialization processes and anticipated labor market risks following from

childbirth, should make the career perspective in the more sheltered, but less upwardly mobile

public sector more attractive. Taken together, internal labor market structures and gender-

stereotypical socialization provide strong incentives for women to work in public sector

professions.

During the family-intensive phase, the higher degree of institutional protection should lead to

more favorable employment chances for women in public sector professions. Family

foundation and child birth should not be detrimental for re-entering the labor market at the

high-skill level, since the structure of international labor markets ensures that skill

investments are not lost, even after phases of economic inactivity. Also in the case of public

sector professionals, the lack and inadequacy of childcare services should make women

reduce their working hours or make them opt out of the labor market – at least temporary, i.e.

while their children are small. However, the highly standardized and sheltered career structure

in public sector professions should reduce the risks associated with work interruption or

working hour reduction. Therefore, during the family intensive life phase, female employment

chances in public professions are expected to be higher when compared to private-sector

professionals.

Private sector professions are less institutionalized and therefore offer less stable career

perspectives. They are associated with higher risks, be it in form of external competition,

unemployment, or bankruptcy in the case of self-employed professionals. At the same, the

lower degree of regulation offers higher profit margins and returns to human capital

12

investment, since wages higher than those set by bargaining are often paid, particularly in

times of economic prosperity. Due to the higher chances of career mobility, we expect male

graduates to seek their first employment more often in private sector professions. This

propensity should be supported by the fact that men anticipate their primary responsibility for

the financial resources of the families. Therefore they will try to maximize the monetary

returns of their human capital investment by seeking employment in higher paid professions

in the private sector.

During the family intensive life phase, we assume that male employment chances in private

professions is unaffected by their family situation. This is particularly due to the higher risks

associated with the private sector, which bear negative consequences for any labor market

interruption and therefore hinder men to opt out of the labor market, even if children were

born. Since professional careers in the private sector often resemble ‘male’

subjects/occupations, they are often based on a ‘male’ work-centered biography. Individuals

with (female) types of biographies who are not able or willing to follow such work-dominated

lifestyles – e.g., because of childcare duties – face the risk to be excluded or hindered in their

career development. The common strategy for couples in which men pursues a career in the

private sector should be to follow the male partner with the better career prospects and

opportunities, while the female ‘following’ partner purses an own occupation only within the

context of the leading career’s commitments and requirements.

Overall, we can conclude that the public-private order of professions interacts with

institutional and individual constraints of un-gendered career prospects and therefore is likely

to result in a horizontal segregation between female public professionals and male private

professionals. This horizontal differentiation does not necessarily result in gender inequalities,

since the definition of professional success varies between the sexes, too, and more women

than men define success by the compatibility of a profession and a family. However, the

indication of restricted upward mobility prospects and lower lifetime incomes in the public

sector implies as well that an initially only horizontal segregation might aggravate over the

life course into vertical segregation, with men obtaining higher status positions and wages

than women.

13

Data, Variables and Methods

In order to examine the importance of public and private sector professions for gendered

career trajectories, female and male employment outcomes are examined for Germany.

Whether and how labor market segmentation translates into career outcomes is tested by

analyzing graduate career mobility during the first five years after graduation as well as by

examining labor market outcomes of 30-49 year-old academics. The analysis of the transition

from higher education to work is carried out with the German Socioeconomic Panel

(GSOEP). The GSOEP is a longitudinal survey of private households in Germany (Haisken-

DeNew and Frick 2005). It is conducted as a panel survey and includes a large variety of

information on labor market positions, educational attainment, attitudes, or family status. For

purposes of this research we have selected all respondents who graduated from higher

education institutions in the years 1984 – 2001 while surveyed by the GSOEP, meaning that a

total number of 878 graduates were included in the calculations. For these respondents, the

first five years after graduation are analyzed on a monthly basis (i.e. waves 1984 – 2005),

taking into account the duration and type of labor market events occurring during this time

interval. The GSOEP can unfortunately not be used to examine further career process due to

the low number of cases in the family intensive age group (30-49 years old). Statistical

modeling of latter career development can therefore unfortunately not be carried out by a

longitudinal design, but has to use cross-sectional data.

For analyzing the impact of family formation on latter labor market outcomes, we make use

of the German Microcensus 2000. The German Microcensus is the official representative

census on the population living in Germany, in which 1 per cent of all households participate.

Each year data from approximately 370,000 households with 820,000 persons are gathered.

For the analyses at hand, the German Microcensus from 2000 has been used, since it contains

detailed information on the field of study, which is only available every four to five waves of

the Microcensus. From the available data pool, the unit of analysis is every respondent

holding a higher education degree, who is between 30 and 49 years old; a total of 9168

individuals have been included in the analyses. This age restriction allows us to exclude

graduates immediately after finishing higher education as well as higher education degree

holders after the family intensive life phase. In view of the examination of professional

careers, the “prime age” in career and family life is taken into account, i.e. labor market entry

processes should already have ensued and processes of exit from the labor market (due to

retirement) are ruled out.

14

For the analysis of graduate career mobility and labor market outcomes several core variables

have been considered (see Appendix A for the distribution of the most important variables in

both samples). In the GSOEP, work histories are observed on a monthly basis during the first

five years after graduation, starting directly after leaving higher education for the first time,

while jobs before graduation are not taken into account. In the Microcensus, the currently held

job or labor market status is taken into account. The operationalization of the professional

labor market segment was based on the ISCO88 3-digit category 200 "Professionals",

consisting of the sub-categories 210 "Physical, mathematical and engineering science

professionals", 220 "Life science and health professionals", 230 "Teaching professionals", and

240 "Other professionals", such as business professionals (241) or legal professionals (242)

(see Appendix B). Measurement of public sector employment was based on the variables on

public and private sectors provided in both data-sets.

Fields of study are coded in six categories of the higher education subject closely representing

the classification used the OECD publications (OECD 2004): Education; Humanities and Arts

(including Services); Social Sciences, Business and Law; Sciences (including Agriculture);

Engineering, Manufacturing and Construction; Health and Welfare. They are included into the

models in form of male-dominated (more than 60 per cent male graduates), gender-mixed

(between 40 and 60 per cent male graduates) and female-dominated (less than 40 per cent

male graduates) fields of study (Smyth 2002), which have been estimated on bases of the

weighted Microcensus 2000 subject distribution. Based on this calculation, Engineering,

Science as well as Social Science/Business/Law can be considered male intensive,

Health/Welfare and Humanities/Arts as gender-mixed, and education as female-dominated

(see Appendix A). To measure the higher education attainment, the CASMIN educational

classification is applied (see Brauns and Steinmann 1997; König et al. 1988; Shavit and

Müller 1998 for more details). In Germany, the CASMIN level of lower tertiary education 3a

refers to respondents holding a technical college degree (Fachhochschul-Diplom), while the

upper tertiary level 3b includes all kinds of university degrees (Diplom, Magister,

Staatsexamen, Promotion).

On the individual level, the main focus is on gender and its interaction effects, as well as the

family structure related to the presence of a partner and his/her qualification, presence and age

of dependent children. Control variables on the micro-level consist of general socio-structural

indicators such as the parental socio-economic status/education, and nationality/ethnicity. In

addition, further indicators related to human capital and indicators of labor market experience

15

were included in the models. Some of them are more general in nature, for example

vocational training in addition to higher education. Others are related to higher education, but

are not specifically captured by variables introduced above. For example, a variable

measuring graduation (GSOEP) or residence (Microcensus) in East Germany controls for

differences between West and East Germany after re-unification. Also, in the GSOEP the age

of graduation was controlled for and included in form of age intervals (below 24 years, 24 to

29 years, and 30 years and over) to allow for curvilinear age effects. The Microcensus entails

too many missing information for this variable, but we control for the age of individuals (30

to 39 years old and 40 to 49 years old).

The analysis of the transition from higher education to work is carried out by estimating

discrete time piecewise constant exponential models with event history analysis (Blossfeld

and Rohwer 1995; Jenkins 2004).3 By means of event history analysis it can be shown how

the length of transition periods varies according to relevant covariates. For the multivariate

analysis of family formation and latter employment, we present results of multinomial logistic

regressions, as each of the dependent variables has several categories. Applying a maximum

likelihood estimation, logistic regressions estimate the probability of a certain event

occurring.4

The Importance of Labor Market Segmentation for Sex Segregation among

Professionals

Becoming a Professional after Graduation in Germany

In order to answer our questions, we first look at the proportion of graduates who have

obtained a professional job within the five year after graduation. In a second step, we

distinguish professional jobs according to the public or private sectors. Furthermore we

compare the proportion of newly graduates in professional positions with those of academics

who are in the family-intensive life phase (i.e. 30 to 49 years old academics).

3 The piecewise constant exponential model does not impose too many restrictions on the shape of the hazard function and furthermore has already proven its validity for studying education to work transitions (Falk et al. 2000; Hillmert 2001). Its flexibility stems from the possibility to allow hazard rates to vary between different time periods (Blossfeld and Rohwer 1995). 4 Logistic regressions in general have less stringent requirements: do not assume a linear relationship between the independent variables and the dependent, do not require normally distributed variables, do not assume homoscedasticity.

16

A rather straightforward representation of the duration it takes to obtain a profession

placement after graduation is provided by the survivor function, which indicates the share of

persons that have not yet made the transition to a first job at any given point of time (Allison

1984; Blossfeld and Rohwer 1995; Jenkins 2004). A graphic comparison of the survivor

functions of women and men will give first insights in the gendered nature of the transition

process. Figure 1 displays the Kaplan-Meier survivor functions of obtaining a professional

position as first employment after graduation for men and women. Both curves indicate that

entry into a profession takes place at a fast pace. Around 30 per cent of all graduates have

found employment as a professional already during the first month after graduation, and 50

percent enter during the first half year. By the end of five years roughly 60 per cent of all

graduates have made the transition, which indicates that professional employment constitutes

one of the most important destinations for German higher education graduates. Most

importantly, no significant differences between female and male survivor functions exist,

which is also indicated by the Log-rank test.5

Figure 1: Becoming a professional after graduation in Germany

0.00

0.25

0.50

0.75

1.00

0 10 20 30 40 50 60Analysis time in months

sex = male sex = female

Kaplan-Meier survivor function for obtaining a first job as a professional after graduation

0.00

0.25

0.50

0.75

1.00

0 10 20 30 40 50 60Analysis time in months

sex = male sex = female

Kaplan-Meier survivor function for obtaining a first job as a professional after graduation

Log-rank test for equality of survivor functions: chi2 = 0.08, Pr>chi2 = 0.7714

Source: GSOEP, authors’ estimations

Thus, initially there appears to be no gender inequality in the chances of obtaining a

professional position after graduation. Yet, as discussed in the previous sections, professions

can take place in different segments of the economy and our analyses show considerable

5 The statistical difference of both curves is expressed by the Log-Rank Test (KM), which indicates whether transition rates between women and men differ significantly, particularly with increasing time (Blossfeld and Rohwer 1995).

17

differences in the proportion of female and male academics employed as professional in the

private and public sector. What is more, gender differences amplify for academics in the

family-intensive life phase (see Figure 2).

Albeit 60 per cent of both female and male graduates have achieved a first job in a

professional position within five years after graduation, figure 2 shows that more men than

women have managed to do so in the private sector (36% of men versus 23% of women).

Conversely, more women than men achieved a profession in the public sector after graduation

(37% of women versus 25% of men). The proportion of graduates who did not enter the labor

market (neither as a professional or non-professional) is very small for both genders, although

the share of female non-employed is twice as much as the male one (5% of women versus

2.5% of men). About one third of both female and male graduates is employed in a non-

professional occupation.

Figure 2: Type of first employment after graduation and of prime age academics, by professional sector

Source: GSOEP, author’s calculations Source: Micro-census 2000, authors’ calculations

The situation appears to be quite different for academics in the family intensive life phase:

gender differences regarding both labor market participation and professions in the public or

private sector are considerably wider than those found among newly graduates. First of all,

considerably more female than male academics are not employed: in 2004, one sixth of the 30

to 49 years old women, but only 5 per cent of the men do not have job. Secondly, whereas

male academics pursue most frequently a professional employment in the private sector

First employment after graduation

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

total male female

Employment of 30-49 academics

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

total male female

non-employed

non-professional

Prof. publicsector

Prof. privatesector

18

(42%), only one fourth of the women do so. Almost a third of female academics holds a

professional job in the public sector and another 30 per cent pursues a non-professional

employment. Differently, for men is more common to be a non-professional (33%) than

practicing a profession in the public sector (19%). These descriptive results show that from

the very beginning male and female access to professional positions in the private and public

sector of the economy differ. Furthermore, gender differences in professional positions

augment in the family-intensive life phase: women are still underrepresented in private sector

professions, and they are not employed to a much higher extent than men.

Individual and Institutional Influences on Professional Sector Placement after

Graduation

In order to examine the importance of gender, family structure and higher education

attainment on the transition to public and private sector professions after graduation, two

separate single event piecewise constant hazard models were estimated. Taken together, the

two separate sets of estimates can be interpreted like a competing risk model, where the

transitions to non-employment or non-professional employment constitute the reference

category.6 Following from the Kaplan-Mayer survivor functions, transition rates to a first

professional placement differ most strongly during the first year after graduation, while later

on they level off. Therefore, the chosen bands of piecewise constant time intervals are narrow

during the first year and are wide thereafter.7 For each transition process, we estimated two

models, the first containing the main effect, while the second also reports important

interaction effects between gender and family variables or field of study respectively. Since

model coefficients are reported as odds ratios, they can be interpreted as relative transition

rates to employment, being higher for values above 1 and lower for values between 0 and 1.

In terms of key variables of interest, the estimates show clear evidence of gendered transition

processes. All models confirm the descriptive results in that women have higher transition

rates to public sector professions, while men are more likely to obtain a first professional job

in the private sector. However, the main gender effect is only significant for the transition to

6 These categories were taken together as reference category, since the number of graduates not finding employment during the first five years after graduation was too low (< 5 per cent) to constitute a separate category. 7 In total, six different time intervals are differentiated: entry in the first month after graduation, entry in the second or third month after graduation, entry between month four and month six, entry in the second half of the first year, entry in the second year, and entry thereafter.

19

private sector professions (model 1). There, women exhibit around 30 per cent lower

transition rates than their male counterparts. Obviously, the more risky, but also more

profitable career structure sets clear incentives for both sexes either for or against seeking a

job in the private sector. As expected, women in general opt against seeking employment in

this sector due to the higher anticipated risks. This propensity of avoiding private sector

employment remains significant irrespective of type of higher education degree obtained, the

subject studied, or the partnership and family constellation.

Table 1: Transition to first employment as a professional in the public or the private sector

Transition to private sector profession

Transition to public sector profession

Model 1 Model 2 Model 1 Model 2 Base line (Ref: Entry 1st month) Entry > 1 month 0.109 *** 0.109 *** 0.100 *** 0.100 *** Entry > 3 months 0.035 *** 0.035 *** 0.040 *** 0.040 *** Entry > 6 months 0.016 *** 0.016 *** 0.013 *** 0.013 *** Entry > 12 months 0.004 *** 0.004 *** 0.009 *** 0.009 *** Entry > 24 months 0.000 *** 0.000 *** 0.001 *** 0.001 *** Female 0.718 ** 0.567 * 1.189 0.895 Child < 6 years 1.019 1.190 1.147 0.764 Married 0.705 ** 0.678 ** 1.140 1.122 Non-German 0.799 0.789 0.593 ** 0.614 * Father with higher education 0.775 * 0.769 * 1.214 1.212 Vocational education 1.156 1.163 0.581 *** 0.580 *** Yearly unemployment rate 0.974 0.973 0.955 0.957 Graduation age (Ref.: 24-29 years) younger than 24 years 0.467 ** 0.450 ** 0.612 0.615 older than 29 years 0.875 0.861 1.017 1.024 Degree of East Germany 1.189 1.200 0.951 0.942 Degree (Ref.: Casmin 3a) Casmin 3b 0.713 ** 0.715 ** 2.987 *** 2.903 *** Field of study (Ref.: gender-mixed) Male-dominated 1.119 0.926 0.676 *** 0.561 ** Female-dominated 0.461 * 0.397 1.557 ** 1.864 ** Interaction terms Female * child < 6 years old 0.469 2.272 ** Female * married 1.095 1.076 Female * male-dominated 1.388 0.775 Female * female-dominated 1.262 1.367 N 43572 43572 43572 43572 Log likelihood null model -994.820 -994.820 -1056.208 -1056.2084 Log likelihood end model -973.542 -972.539 -1001.516 -999.02357 Chi2 Likelyhood ratio test 42.56 44.56 109.38 114.37

Non-professional employment and non-employment constitute the baseline category. Coefficients are significant: * p<0.1, ** p<0.05, ***p<0.01, Source: GSOEP, authors’ estimations

Men, on the other hand, are apparently more often attracted by the higher profit margins

available in the private sector and therefore take the risks of higher working hours, less

20

institutional protection against career interruptions and generally more unstable career

prospects. This assumption is supported by the influence partnership has on career decisions

after graduation (model 2). The interaction terms indicate that particularly non-married men

will gain their first employment as private sector professionals, while marriage apparently

increases risk aversion and accordingly reduces transition rates among men substantively.

Having young dependent children does not play a role immediately after graduation, so we

can only speculate that it is more the anticipation of family foundation that influences the

male transition pattern.

But not only gender and family structure are important for explaining the gendered transition

process among professionals. Also type of degree obtained and the gender-typing of fields of

study apparently structures entry into the private sector. Most importantly, graduates holding

a university degree are considerably less likely to work as private sector professionals (models

1 and 2). This should be due to the fact that some German university degrees, such as the

Staatsexamen (state examination), constitute exclusive entry certificates into the classic public

sector professions such as medicine, law or teaching. On the other hand, the gender-typing of

fields of study has only weak influences, since only graduates of female-dominated subjects

have lower transition rates to private sector professions when compared to graduates of

gender-mixed subjects (model 1).

The transition to the public sector provides a kind of mirror image as regards explanatory

variables. An important difference, however, lies in the fact that the main gender effect is not

significant (model 1). Even though Figure 2 has shown that women exhibit higher

participation rates in the public sector than men, the multivariate analysis shows that this is

not the case for all women. This, on first sight, contradicts our hypothesis that women in

general opt for public sector employment due to the more sheltered career arrangements there.

Yet, we find that women with young dependent children have twice as high transition rates

when compared to women without children (model 2). This finding again stays in line with

the argument related to internal labor markets. Obviously, women in general avoid working in

the private sector professions, but do not significantly prefer the public sector as compared to

non-professional careers, if there is no need arising from their family structure.

In addition, we find that both men and women seek employment in the public sector if it is an

option with their higher education degree. Particularly university graduates have much higher

transition rates in this regard (models 1 and 2). Most interestingly, though, is the fact that

male-dominated subjects constitute a clear disincentive for the public sector. While graduates

21

from male-dominated subjects such as engineering or science are much less likely to enter this

segment of the labor market, the female-dominated subject of education constitutes clearly an

important entry certificate. This relationship is most obvious for men (model 2), again

supporting our claim that male graduates seek their first employment mainly on basis of their

education credential. But it holds also true for women, even though the coefficients are not

significant.

Taken the results presented so far together we can conclude that after graduation no gender

differences exist as regards obtaining professional employment in general. However, already

the first job placement exhibits clear horizontal sex segregation patterns among professionals

in the public and private sector. The internal labor market arrangements found in public sector

professions predominantly attracts women, particularly with young dependent children, but is

also attractive for men with university degrees and female-dominated fields of study. The

higher risks associated with a more external labor market arrangement in private sector

professions attracts more men, if it is an option with their higher education degree, while

women generally avoid working in this environment, irrespective of the degree obtained or

their family structure. The following analysis will show whether such gendered career

arrangements are observable in the family-intensive phase as well. Of major interest is in this

regard whether the more sheltered labor market environment indeed increases female labor

market chances despite career interruptions due to child birth.

Individual and Institutional Influences on Professional Sector Placement during the

Family-intensive Phase

In order to examine the importance of gender, family structure and higher education

attainment for public or private sector professions during the family intensive life phase, two

multinomial logistic regressions were estimated. Given that our descriptive results have

shown that among 30 to 49 years old academics a gender dive exists not only between these

two sectors of the economy, but also regarding labor market participation, the models

presented will have four categories. We estimate the chance/risk of being non-employed,

holding a profession in the private sector or being in a public sector profession in comparison

to being in employment as a non-professional (reference category). We estimated two models,

the first containing only the main effects, while the second also includes important interaction

effects. The model coefficients are reported as odds ratios that can be interpreted as relative

22

likelihood of employment type, being higher for values above 1 and lower for values between

0 and 1.

Table 2: Multinomial logistic regressions on employment status of 30-49 years old academics (Reference

category: employed in non-profession) (odds ratios, N=9168)

Not employed Profession in private sector

Profession in public sector

Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 Female 3,452*** 0,933 0,630*** 0,704** 1,031 0,624*** Partner (Ref: no partner) Non-academic partner 0,579*** 0,380*** 0,780*** 0,816** 0,750*** 0,726** Academic partner 1,064 0,502*** 0,997 0,986 1,115 0,971 Underage children (Ref: no children) Youngest Child < 3 yrs 3,102*** 1,775** 1,309*** 1,296** 0,989 1,011 Youngest Child 3-9 yrs 2,079*** 1,347 1,289*** 1,394*** 1,086 1,194 Youngest Child 10-17 yrs 0,987 0,972 0,859* 0,923 0,983 1,004 Degree (Ref: Casmin 3a) Casmin 3b 1,323*** 1,323*** 1,575*** 1,581*** 2,151*** 2,137*** Field of study (Ref: gender-mixed) male-dominated 0,630*** 0,498*** 0,552*** 0,538*** 0,318*** 0,244*** female-dominated 1,150 0,898 0,341*** 0,226*** 2,419*** 1,599*** Type of partnership (marriage) Unmarried cohabitation 0,734* 0,751* 1,158 1,168 0,973 0,985 Nationality (Ref: German) Non-German 2,641*** 2,729*** 0,490*** 0,493*** 0,265*** 0,265*** Size of place of residence (Ref: 20.000-500.000 inhabitants)

< 20.000 inhabitants 0,938 0,934 1,103 1,100 0,974 0,973 500.000 and more 0,861 0,886 1,024 1,027 0,653*** 0,658*** Place of residence (West-Germany) East-Germany 0,696*** 0,699*** 0,500*** 0,507*** 0,720*** 0,719*** Berlin 1,499** 1,476** 0,744** 0,747** 0,956 0,949 Age (Ref: 40-49 yrs old) 30-39 yrs old 0,803** 0,752*** 1,087 1,097 0,993 0,993 Interaction terms Female * Child < 3 yrs 2,636*** 1,015 1,032 Female*Child 3-9 yrs 1,897*** 0,759* 0,817 Female*Child 10-17 yrs 1,081 0,787 0,938 Female* Non-academic partner 2,052*** 0,760* 1,016 Female*Academic partner 3,135*** 0,973 1,307 Female*male-dominated 1,458 0,984 1,751*** Female*female-dominated 1,554 1,995*** 2,108*** Improvement of fit (df) 8583,4***

(48) 8458,7*** (69)

Coefficients are significant: * p<0.1, ** p<0.05, ***p<0.01, Source: Microcensus 2000, authors’ estimations

As shown by the first model, it is apparent that female academics have a higher risk of being

non-employed and a lower chance of being in private sector profession than men in

comparison to being employed in a non-profession. The interaction effects (model 2) clarify

that for women a family – both partner and children – increases their risk of being out of the

23

labor force; while for male academics the presence of a partner reduces this risk. Moreover

for men, a male-dominated field of study appears to protect from non-employment while the

same in not true for women.

With regard to private sector professions, the models reveal that women have a lower chance

than men of such an employment; and this regardless of their family commitments (as

indicated by odds ratio smaller than 1 in model 2). Differently, male academics with small

children have a higher chance of being in a private sector profession than childless male

academics. This might indicate that given the prevalent gender roles ideology in German

society, once men become fathers they try to maximize their monetary gains in the private

sector where generally higher wages are paid. Interestingly, both male and female academics

with a non-academic partner have a lower chance of a profession in the private sector than

single academics. This might indicate that couples in which partners have unequal human

capital (and thus unequal chances and rewards on the labor market) might be less prone to

take on the risks entailed in the private economy. Beside gender and family structures, also

the fields of study influence the chances of private sector employment. Yet, the effect appears

to differ for men and women: among the latter those who studied in female-dominated fields

of study have a higher chance of being professional in the private sector than those who

studied a gender-mixed subject. The opposite is true for men: here the chances are higher for

male academics with a degree in gender-mixed fields of studies. Finally, also the type of

degree obtained plays a role: university graduates have higher chances of being professional

in the private sector. Although is true, as argued in the previous section, that some of these

degrees (such as state examination) give privileged access to classic public sector professions

(such as medicine and law), it might be the case that after some years of experience in the

public sector some of these professional might decide to work in or establish own (private)

practices.

With regard to public sector professions, the first model shows that male and female

academics have equal chances of being in such an employment (model 1). Furthermore,

among academics who studied gender-mixed subject women have a lower chance than men of

such an employment (model 2).8 Male graduates from male-dominated subjects are much less

likely to enter this segment of the labor market, while for both men and women the chances of

being in public sector professions are considerably higher when they achieved a degree in a

8 The negative effect is confirmed by models with only one interaction term (gender*field of study); data is not shown here due to space reasons; but is available by authors upon request.

24

female-dominated field of study. This might be due to the fact that education is female-

dominated (see Appendix A) and thus the grand majority of graduates in this field works as a

teacher in the public sector. Also a university degree (casmin 3b) enhances the chances of a

public sector profession. Thus, for academics with a university degree the question appears to

be all or nothing: they are either employed as professionals in the public or private sector or

not employed at all; i.e. university graduates are in comparison more rarely employed in non-

professional positions.

As expected, professions in the public sector are, for women, not influenced by the presence

of children and partner. This signifies that those female academic who remain employed ‘in

spite’ of their family commitments have equal chances of working as non-professional or

pursuing a profession in the public sector. Differently, male academics with a less qualified

partner are less likely to have a public sector profession than being employed as non-

professionals.

In sum, during the family-intensive life phase the chances of professional employment are

indeed strongly shaped by gender and family commitments. Two features appear quite

striking. First of all, the risk of being out of the labor force is considerably higher for women

who have a partner and children; and this regardless of the type of higher education and fields

of study. Second, the chances of female academics in private sector professions are lower, in

turn, regardless of their actual family commitments. However, those women who stay in the

labor market despite their family commitments have equal chances to purse their professions

in the public sector.

Conclusions

The opening point made in this paper was the positive evaluation that increasing levels of

female participation in higher education will eventually lead to a decline of gender

inequalities in the labor market among the highly qualified. This development should be

supported by service sector expansion, which creates new job opportunities also at the upper

end of the status and pay hierarchy, as it is the case for professionals. By analyzing career

outcomes of female and male academics we asked whether this positive claim was justified.

Based on the theory of labor market segmentation our main hypothesis was that a public-

private order of professions exists, which offers specific career prospects. We assumed that

25

this public-private divide is likely to lead to horizontal sex segregation among professionals

due to the institutional and individual constraints men and women face in the course of their

careers.

Our empirical results indicate that the public-private order of professions indeed constitutes

gendered career outcomes. Even though men and women initially have equal chances to be

employed as professionals after graduation, women are less likely to find their first

professional employment in the private sector than men, regardless of their family status. This

indicates that the more external labor market structure of private sector professions offers

more male-typical career perspectives. The assumption that the more protected public sector

professions are more attractive for women is supported by the higher transition rates of young

mothers into this segment. The gender segregation between public and private sector

employment is already prepared by a particular gender-typing of fields of study. Especially

for men, a male-dominated field of study decreases the chances to be a professional in the

public sector, while their transition rates to the public sector increase if they studied a female-

dominated subject.

In the family-intensive life phase, the gender divide intensifies. While female chances in the

private sector remain lower when compared to men, the previously high employment chances

for women in the public sector become lower as well, irrespective of actual family

commitment. In addition, the risk of being not employed is considerably higher for women

who have a partner and children, regardless of the type of higher education and fields of

study. Only those women who stay in the labor market despite their family commitments have

equal chances to purse their professions in the public sector. Thus, institutional and individual

constraints related to family arrangements are becoming more important for gendered career

outcomes during the family-intensive life phase.

Taken together, we can conclude that the main reason for a horizontal sex segregation

occurring between public and private sector professionals immediately after graduation can be

mainly attributed to the gender-specific choice of fields of study rather than the family

situation. This is not surprising given the relative young age of graduates, where only a small

minority has already young dependent children. During the family-intensive phase the field of

study subject still has an influence on gendered career outcomes; however, the family

situation becomes much more important for sex segregation among professionals. Most

importantly, our findings show that the horizontal divide between public and private sector

26

professionals persists, but in addition vertical segregation between employment and non-

employment becomes aggravated in the family intensive phase.

Consequently, the public-private order of professions indeed constitutes a horizontal

segregation between female and male labor market outcomes in the professional sphere.

Obviously, the more sheltered career arrangement in the public sector is more attractive for

female graduates already immediately after graduation, and continues to be so in the family-

intensive phase. However, during this life phase, highly qualified women and men are not

only segregated horizontally, but also vertically as regards their labor market status. While

prime age men continue to have a high employment rate similar to the one after graduation, a

high proportion of women obviously opt out of the labor market. Thus, contrary to our

theoretical expectations the more sheltered internal labor market of public sector professions

does not provide enough protection to overcome the gendered division of labor within the

family.

Overall, our analyses demonstrated that even among highly qualified men and women

important patterns of segregation persist. Horizontal segregation is constituted by the public-

private order of professions, while vertical segregation follows from institutional and

individual constraints. Therefore, the optimistic beliefs that increasing higher education

participation of women in combination with service sector expansion will eventually lead to a

decline of gender inequalities can up to now not be confirmed.

27

References

Abbott, AD. 1988. The System of Professions. Chicago, London: The University of Chicago Press.

Allison, PD. 1984. Event History Analysis: Regression for Longitudinal Data. Beverly Hills: Sage.

Allmendinger, J, and A Podsiadlowski. 2001. "Segregation in Organisationen und Arbeitsgruppen." Kölner Zeitschrift für Soziologie und Sozialpsychologie Sonderheft 41:276-307.

Becker, PE, and P Moen. 1999. "Scaling Back: Dual-Earner Couples' Work-Family Strategies." Journal of Marriage and the Family 61:995-1007.

Becker, R. 1993. Staatsexpansion und Karrierechancen. Berufsverläufe im öffentlichen Dienst und in der Privatwirtschaft. Frankfurt/New York: Campus.

Behnke, C, and M Meuser. 2005. "Vereinbarkeitsmanagement. Zuständigkeiten und Karrierechancen bei Doppelkarrierepaaren." Pp. 123-139 in "Wenn zwei das Gleiche tun..." Ideal und Realität sozialer (Un)Gleichheit in Dual Career Couples, edited by Heike Solga and Christine Wimbauer. Opladen: Barbara Budrich Verlag.

Bien, W, T Rauschenbach, and B Riedel. 2007. Wer betreut Deutschlands Kinder? DJI-Kinderbetreuungsstudie. München: Deutsches Jugendinstitut.

Blossfeld, H-P, and G Rohwer. 1995. Techniques of Event-History Modeling. New Approaches to Causal Analysis. Mahwah, NJ: Erlbaum.

Blossfeld, H-P, and A Timm. 2003. "Who marries whom in West Germany." Pp. 19-35 in Who marries whom? Educational systems as marriage markets in modern societies, edited by Hans-Peter Blossfeld and Andreas Timm. Dordrecht: Kluwer.

Born, C. 2001. "Modernisierungsgap und Wandel. Angleichung geschlechtsspezifischer Lebensführungen?" Pp. 29-53 in Individualisierung und Verflechtung. Geschlecht und Generation im deutschen Lebenslaufregime, edited by Claudia Born and Helga Krüger. Weinheim/München: Juventa.

Boyle, P, TJ Cooke, K Halfacree, and D Smith. 2001. "A Cross-National Comparison of the Impact of Family Migration on Women's Employment Status." Demography 38:201-213.

Brater, M, and U Beck. 1981. "Berufe als Organisationsform menschlichen Arbeitsvermögens." Pp. 208-224 in Einführung in die Arbeits- und Industriesoziologie, edited by Wolfgang Littek, Werner Rammert, and Günther Wachtler. Frankfurt/Main, New York: Campus Verlag.

Brauns, H, and S Steinmann. 1997. "Educational Reform in France, West-Germany, the United Kingdom and Hungary: Updating the CASMIN Educational Classification." Working Papers Mannheimer Zentrum für Sozialforschung 21:1-55.

Charles, M. 2005. "National Skill Regimes, Postindustrialism, and Sex Segregation." Social Politics 12:289-316.

Costa, DL, and ME Kahn. 2000. "Power Couples: Changes in the Locational Choice of the College Educated, 1940-1990." The Quarterly Journal of Economics 115:1287-1315.

Deitch, CH, and SW Sanderson. 1987. "Geographic Constraints on Married Women's Careers." Work and Occupations 14:616-634.

Dettmer, S, and E-H Hoff. 2005. "Berufs- und Karrierekonstellationen in Paarbeziehungen: Segmentation, Integration, Entgrenzung." Pp. 53-75 in "Wenn zwei das Gleiche tun..." Ideal und Realität sozialer (Un)Gleichheit in Dual Career Couples, edited by Heike Solga and Christine Wimbauer. Opladen: Barbara Budrich Verlag.

28

Doeringer, PB. 1967. "Determinants of the Structure of Industrial Type Internal Labour Markets." Industrial Relations Review 20:206-220.

Doeringer, PB, and M Piore. 1971. Internal Labor Markets and Manpower Analysis. Lexington, Mass.: Heath.

Dressel, C. 2005. "Erwerbstätigkeit – Arbeitsmarktintegration von Frauen und Männern." Pp. 99-158 in Gender-Datenreport - Kommentierter Datenreport zur Gleichstellung von Frauen und Männern in der Bundesrepublik Deutschland, edited by Waltraud Cornelißen. München: im Auftrag des Bundesministeriums für Familie, Senioren, Frauen und Jugend.

England, P. 2005. "Gender Inequality in Labor Markets: The Role of Motherhood and Segregation." Social Politics 12:264-288.

England, P, and S Li. 2006. "Desegregation Stalled: The Changing Gender Composition of College Majors, 1971-2002." Gender and Society 20:657-677.

Estevez-Abe, M. 2005. "Gender Bias in Skills and Social Politics: The Varieties of Capitalism Perspective on Sex Segregation." Social Politics 12:180-215.

—. 2006. "Gendering the Varieties of Capitalism: A Study of Occupational Segregation by Sex in Advanced Industrial Societies." Pp. 1-27 in Government Department. Harvard University.

European Commission, D-GfE, Social Affairs, and Equal Opportunities. 2006. "Employment in Europe 2006." edited by Office for Official Publications of the European Communities. Luxembourg.

Fagan, C, J O'Reilly, and B Halpin. 2005. "Job Opportunities for Whom? Labour Market Dynamics and Service-Sector Employment Growth in Germany and Britain." edited by Anglo-German Foundation for the Study of Industrial Societies. London: Anglo-German Foundation for the Study of Industrial Societies.

Falk, S, R Sackmann, O Struck, A Weymann, M Windzio, and M Wingens. 2000. "Gemeinsame Startbedingungen in Ost und West? Risiken beim Berufseinstieg und deren Folgen im weiteren Erwerbsverlauf." Collaborative Research Center "Status Passages and Risks in the Life Course" (Sfb 186) Working Paper Series 65.

Fuchs, S, Jv Stebut, and J Allmendinger. 2001. "Gender, Science, and Scientific Organizations in Germany." Minerva:175-201.

Geenen, EM. 1993. "Brüche und Kontinuitäten in den Biographien von Wissenschaftlern." Pp. 41-65 in Frauenforschung in universitären Disziplinen, edited by Ursula Pasero and Friederike Braun. Opladen: Leske + Budrich.

Gottschall, K, and K Shire. 2006. "Comparing Employment Systems from a Gender Perspective." in Annual Meeting of the American Sociological Association. Montreal.

Haffner, Y, B Könekamp, and B Krais. 2006. Arbeitswelt in Bewegung. Chancengleichheit in technischen und naturwissenschaftlichen Berufen als Impuls für Unternehmen. Berlin: Bildungsministerium für Bildung und Forschung.

Haisken-DeNew, JP, and JR Frick. 2005. Desktop Companion to the Germany Socio-Economic Panel Study (SOEP). Berlin: DIW.

Hardill, I, AC Dudleston, AE Green, and DW Owen. 1999. "Decision Making in Dual-Career Households." Pp. 192-206 in Gender, Power and the Household, edited by Linda McKie, Sophia Bowlby, and Susan Gregory. London/New York: Macmillan Press/St. Martin's Press.

Heidenreich, M. 1999. "Berufskonstruktion und Professionalisierung. Erträge der soziologischen Forschung." Pp. 35-58 in Professionalisierung pädagogischer Berufe im historischen Prozeß, edited by Hans Jürgen Apel, Klaus-Peter Horn, Peter Lundgreen, and Uwe Sandfuchs. Bad Heilbrunn/Obb.: Julius Klinkhardt.

Hertz, R. 1986. More Equal than Others: Women and Men in Dual-career Marriages. Berkeley: University of California Press.

29

Hillmert, S. 2001. Ausbildungssysteme und Arbeitsmarkt: Lebensverläufe in Großbritannien und Deutschland im Kohortenvergleich. Wiesbaden: Westdeutscher Verlag.

Hoff, E-H, S Dettmer, S Grote, and H-U Hohner. 2002. "Formen der beruflichen und privaten Lebensgestaltung. Ergebnisse aus den Interviews im Projekt "PROFIL"." Berlin: FU Berlin, Bereich Arbeits-, Berufs- und Organisationspsychologie.

Huinink, J. 1995. Warum noch Familie? Zur Attraktivität von Partnerschaft und Elternschaft in unserer Gesellschaft. Frankfurt am Main: Campus Verlag.

Jacobs, JA. 1989. Revolving doors: sex segregation and women's careers. Stanford, California: Stanford University Press.

—. 1995. "Gender and Academic Specialties: Trends among Recipients of College Degrees in the 1980s." Sociology of Education . 68:81-98.

Janshen, D, and H Rudolph. 1987. Ingenieurinnen, Frauen für die Zukunft. Berlin: de Gruyter.

Jenkins, SP. 2004. "Survival Analysis." in Manuscript of the Essex Summer School in SSCA. University of Essex, Colchester.

Kalleberg, AL, and AB Sorensen. 1979. "The Sociology of Labor Markets." Annual Review of Sociology 5:351-379.

Kerr, C. 1954. "The Balkanization of Labor Markets." Pp. 92-110 in Labor Mobility and Economic Opportunity edited by E. Wight Bakke, P.M. Hauser, G.L. Palmer, C.A. Myers, D. Yoder, and C. Kerr. Cambridge, Mass.: Technology-Press of MIT.

Kilpatrick, AC. 1982. "Job Change in Dual-Career Families: Danger or Opportunity?" Family Relations 31:363-368.

Kirner, E, and E Schulz. 1992. "Das "Drei-Phasen-Modell" der Erwerbsbeteiligung von Frauen - Begründung, Norm und empirische Relevanz." Pp. 17-55 in Familie und Erwerbstätigkeit im Umbruch, edited by Notburga Ott and Gert Wagner. Berlin: Duncker & Humblot.

König, W, P Lüttinger, and W Müller. 1988. "A Comparative Analysis of the Development and Structure of Educational Systems. Methodological Foundations and the Construction of a Comparative Educational Scale." in CASMIN-Working Paper. Mannheim.

Konrad, AM, and K Cannings. 1997. "The Effects of Gender Role Congruence and Statistical Discrimination on Managerial Advancement." Human Realtions 50:1305-1328.

Krimmer, H, F Stallmann, M Behr, and A Zimmer. 2003. "Karrierewege von ProfessorInnen an Hochschulen in Deutschland." Münster: Institut für Politikwissenschaft.

Leuze, K. 2007. "Smooth Path or Long and Winding Road? Comparing the Institutional Embeddedness of Graduate Career in Germany and Britain." in Graduate School of Social Sciences (GSSS). Bremen: University of Bremen.

Levy, R, and M Ernst. 2002. "Lebenslauf und Regulation in Paarbeziehungen: Bestimmungsgründe der Ungleichheit familialer Arbeitsteilung." Zeitschrift für Familienforschung 14:103-131.

Lueptow, LB, L Garovich-Szabo, and MB Lueptow. 2001. "Social Change and the Persistence of Sex Typing: 1974-1997." Social Forces 80:1-36.

Marwell, G, R Rosenfeld, and S Spilerman. 1979. "Geographic Constraints on Women's Careers in Academia." Science 205:1225-1231.

Minks, K-H, and B Filaretow. 1996. Frauen aus technischen und naturwissenschaftlichen Studiengängen. Ein Vergleich der Berufsübergänge von Absolventinnen und Absolventen. Hannover: HIS.

OECD. 2004. OECD Handbook for Internationally Comparative Education Statistics: Concepts, Standards, Definitions and Classifications. Paris: OECD Publishing.

Reskin, BF, and I Padavic. 1994. Women and Men at Work. Thousand Oaks: Pine Forge Press.

30

Rosenfeld, R. 1992. "Job Mobility and Career Processes." Annual Review of Sociology 18:39-61.

Rothenbacher, F. 1999. "Der öffentliche Dienst in Europa - ein schrumpender Sektor? Sozialstruktur, Einkommen und soziale Sicherheit." Informationsdienst Soziale Indikatoren 21:1-4.

—. 2004. "The Welfare State of the Civil (or Public) Servants in Europe: A Comparison of the Pension Systems for Civil (or Public) Servants in France, Great Britain, and Germany." Working Papers Mannheimer Zentrum für Sozialforschung 74:1-63.

Rubery, J, and D Grimshaw. 2003. The Organization of Employment. An International Perspective. Basingstoke, New York: Palgrave Macmillan.

Rubery, J, M Smith, and C Fagan. 1999. Women's Employment in Europe: Trends and Prospects. London: Routledge.

Rusconi, A, and H Solga. forthcoming. "Determinants of and Obstacles to Dual Careers in Germany." Zeitschrift für Familienforschung/Journal of Family Research.

Schreyer, F. 2000. "Ein bewegtes Jahrzehnt. Maschinenbau, Elektrotechnik, Informatik." IAB-Materialien 4:6-7.

Schulz, F, and H-P Blossfeld. 2006. "Wie verändert sich die häusliche Arbeitsteilung im Eheverlauf. Eine Längsschnittstudie der ersten 14 Ehejahre in Westdeutschland." Kölner Zeitschrift für Soziologie und Sozialpsychologie 58:23-49.

Shavit, Y, and W Müller. 1998. From School to Work: A Comparative Study of Educational Qualifications and Occupational Destinations. Oxford: Clarenden Press.

Smyth, E. 2002. "Gender Differences and Early Labour Market Integration across Europe." Working Papers Mannheimer Zentrum für Sozialforschung 46:1-28.

Statistisches Bundesamt (Ed.). 2002. Datenreport 2002 - Zahlen und Fakten über die Bundesrepublik Deutschland. Bonn: Bundeszentrale für politische Bildung.

—. 2007. Hochschulen auf einen Blick. Wiesbaden: Statistisches Bundesamt. Stephens, J. 1999. "A Fight for Her Time: Challenges Facing Mothers who Work in Hospital

Medicine." Pp. 99-115 in Gender, Power and the Household, edited by Linda McKie, Sophia Bowlby, and Susan Gregory. London/ New York: Macmillan Press/ St. Martin's Press.

Stroh, LK, and AH Reilly. 1999. "Gender and Careers. Present Experiences and Emerging Trends." Pp. 307-324 in Handbook of Gender & Work, edited by Gary N. Powell. Thousand Oaks/London/New Delhi: Sage.

Swiss, DJ, and JP Walker. 1993. Women and the Work/Family Dilemma. How Today's Professional Women Are Finding Solutions. New York: John Wiley & Sons.

UN, UN. 2001. "Human Development Report 2001." Oxford, New York: Oxford University Press, for the United Nations Development Programme.

Vogel, U, and C Hinz. 2003. "Biographische Erfahrungen und Karriere-Entscheidungen bei Frauen auf dem Weg in Führungspositionen der Wissenschaft." Braunschweig: Institut für Sozialwissenschaften.

Ware, NC, and VE Lee. 1988. "Sex Differences in Choice of College Science Majors." American Educational Research Journal 25:593-614.

31

Appendix

Appendix A: Sample Description

GSOEP Microcensus 2000

Number of graduates 878 9168 Years of graduation (GSOEP) / Birth year (MC 2000) 1984 – 2001 1950-1970 Females 40.7 % 40,9 % Females with children under 6 yrs (GSOEP)/ Females with children under 6 yrs & under 18 yrs (MC 2000)

2 % 21.1 % 53,7 %

Non-German 8.2 % 5.4 % Father with higher education 26.7 % --- Vocational Training 20 % --- Mean age of graduation (Std. Dev.) 28 (4.102) --- CASMIN 3a 34.5 % 40.8 % CASMIN 3b 65.5 % 59,2 % Humanities/Arts (gender-mixed) 9.6 % 10 % Health, Welfare (gender-mixed) 7.7 % 7,5 % Engineering (male-dominated) 26.9% 28,5 % Science (male-dominated) 14.5 % 11,7 %

Soc. Sc., Business, Law (male-dominated) 34 % 24,9 % Education (female-dominated) 7.2 % 17,3 %

32

Appendix B: Professionals according to ISCO88 com

ISCO88 MAJOR GROUP 2: PROFESSIONALS

21 Physical, mathematical and engineering science professionals 211 Physicists, chemists and related professionals 212 Mathematicians, statisticians and related professionals 213 Computing professionals 214 Architects, engineers and related professionals 22 Life science and health professionals 221 Life science professionals 222 Health professionals (except nursing) 223 Nursing and midwifery professionals 23 Teaching professionals 231 College, university and higher education teaching professionals 232 Secondary education teaching professionals 233 Primary and pre-primary education teaching professionals 234 Special education teaching professionals 235 Other teaching professionals 24 Other professionals 241 Business professionals 242 Legal professionals 243 Archivists, librarians and related information professionals 244 Social science and related professionals 245 Writers and creative or performing artists 246 Religious professionals 247 Public service administrative professionals