Research and analysis

Main report

Published 22 October 2019

Women’s Progression in the Workplace

Laura Jones, Global Institute for Women’s Leadership, Kings College London

October 2019

Executive summary

The central aims of this research were to:

• provide an overview of the gender divide in progression in the workplace
• understand what is known about the barriers and facilitators of women’s progression
• critically review and synthesise the evidence on which organisational policy interventions have been demonstrated to be successful in improving women’s progression in the workplace
• identify possible untested organisational interventions which the evidence suggests could be successful

Progression was interpreted broadly as meaning not only moving up the occupational hierarchy, but also any job change that resulted in better pay, working conditions, responsibility or security.

The research took the form of a rapid evidence assessment of academic literature. Evidence on the extent of the gender divide in progression and the barriers and facilitators was limited to the UK, while evidence on organisational interventions was not geographically limited.

The gender divide in progression

The report looked at 2 key ways to measure progression – wage growth, and movement up a vertical occupational scale. In both cases the gender gap is minimal on labour market entry and widens significantly from the late 20s and early 30s as women’s progression plateaus. Gender differences in part-time work are important explanations for these differences, but a substantial amount remains unexplained.

Another key difference is that women who enter the labour market in low-paid jobs experience ‘sticky floors’ – rarely progressing upwards. By contrast, such jobs are ‘springboards’ for men into higher paid positions. This springboard vs sticky floor dichotomy has worsened over time.

Barriers to progression

This research condensed over 100 studies carried out in the UK between 2000 and 2018 to understand the organisational barriers to women’s progression, as well as any facilitators. Four key groupings of barrier were identified:

• Processes for progression that open up space for bias. In the absence of clear systems and transparent systems, decisions about pay and promotion are more likely to be made through processes that disadvantage women, including via networks and the process of social cloning, where those in positions of power champion those who are like themselves.

• Hostile or isolating organisational cultures including issues of sexual harassment and stereotyping.

• The conflict between external responsibilities and current models of working. In many workplaces persistent norms of overwork, expectations of constant availability and excess workloads conflict with unpaid caring responsibilities – the majority of which still fall on women. Unpredictable work demands linked to casualised forms of labour also pose challenges, as do the requirement in some fields of the necessity for geographic mobility to progress.

• Alternative ways of working do not currently offer parity. There continues to be a shortage of quality part-time work and part-time experience offers very little return on experience in terms of wage growth. While there is evidence of an increase in the proportion of female senior part-timers, the evidence suggests this is largely as a result of already senior women negotiating a reduction in hours. Meanwhile, part-timers in lower occupational jobs (most of whom are female) receive low wages with limited opportunities for progression. Evidence from the UK also suggests there is the potential for flexible workers to suffer negative career consequences.

Organisational interventions to support women’s progression

The review uncovered evidence of a range of different types of intervention designed to support women’s progression in the workplace. Some of the key conclusions from this section:

• Transparency and formalisation are key to reduce gender bias. Interventions to make the processes for promotion and progression standardised and transparent – such as formal career planning, clear salary standards and job ladders - are linked with improved career outcomes for women. This is especially the case when combined with senior oversight and accountability of these processes.

• Alternative working time policies need to go alongside efforts to reform organisational culture to include them. Flexible and part-time work are important factors helping women to maintain their labour market position following the transition to parenthood, but need to be combined with efforts to reform organisational cultures to truly accommodate them. Alternative working time policies without culture change are not enough and risk embedding gender inequality due to the negative effects these practices have on career progression in contexts where part-time and flexible working are seen as signalling a lack of commitment. Organisations which show ongoing top-down commitment to supporting part-time and flexible workers mitigate some of these barriers. Senior role models who talk openly about balancing work and family life, or who work part-time, are important for reducing flexibility stigma, as are supportive line managers and organisations which train line managers to deal with part-time and flexible workers.

Conclusions and implications for policy

The largest barriers to women’s progression in the workplace continue to arise from a conflict between current ways of organising work and caring responsibilities. Long-hours cultures, expectations of constant availability and a lack of part-time progression are enduring features of modern workplaces.

This suggests that the policy focus should be on reforming organisational cultures away from norms of overwork and supporting the construction of ‘non-extreme’ jobs, which do not require long hours and constant availability as a proxy for commitment, and which support part-time progression at all levels. Alongside policies to support work-life balance should come efforts to reform the deeper structures and workplace processes which encourage long hours.

At the same time, the stigma associated with part-time and flexible work is likely to persist so long as these ways of working continue to be associated with women, while men work long hours.

Policies that encourage and enable men to take on greater childcare responsibilities are thus essential if women are to be able to maximise their potential.

The other main barrier to women’s progression comes from organisational norms and processes that allow gender bias to creep into decision making. When there is a lack of clarity around the standards for recruitment, promotion or pay negotiation decisions are more likely to be made in ways that disadvantage women, whether because people in power seek those who are like them or because who you know is more important than what you know. Organisations should ensure that they have clear standards for promotion and advancement and create mechanism for organisational oversight of these processes.

1. Introduction

This chapter provides a brief overview of the context behind the commissioning of this study as well as of key relevant policy directives.

1.1 Policy context – the gender pay gap

The UK’s overall gender pay gap (GPG), the difference between men and women’s median hourly earnings, is 17.9% (ONS, 2018). Progress towards closing the gap in recent years has been slow, and has occurred against a backdrop of declining real wages, meaning that a narrowing in the GPG has come at the expense of wage growth for both men and women (Olsen and others, 2018).

The government is committed to taking action to close the GPG. Their flagship policy, gender pay reporting, came into force in April 2018 and requires all employers with over 250 employees to publish their GPG. This policy was designed to increase transparency and accountability among employers. The government also recognises that it is important to support employers to understand and tackle the drivers of their GPG. To this end it is working to build the evidence base to understand what employers can do to help reduce their own GPGs.

The GPG is measured as the difference in the average hourly wage of all men and women across the workforce. This means that if there are more women than men in less well-paid jobs within an organisation, and more men in the better paid jobs, the GPG will be bigger. Of the 10,000 employers who had submitted their GPG data to the government by August 2018, over 80% had more women in their lowest paid positions than in their highest paid positions (Government Equalities Office, 2018). While, overall, more women than ever before are employed, they continue to be overrepresented in low paid, part-time and insecure employment (De Henau and others, 2018). Understanding why women continue to outnumber men in less well-paid roles is therefore key to closing the GPG.

Government policy has sought to address barriers to women’s progression opportunities through policies designed to facilitate the reconciliation of work and caring responsibilities. These include the right to request flexible working. Introduced in 2002, this right entitles qualified employees to apply to their employers for a change to their terms and conditions of employment relating to the hours, times and location of work. When introduced, this right applied only to a limited category of employees with parental or caring responsibilities. The Children and Families Act 2014 extended the right to all employees with 26 weeks’ continuous employment (Pyper, 2018). Other initiatives have more directly targeted employers, including a focus on increasing the number of women on boards following the Hampton-Alexander Review and the compulsory GPG reporting.

Enabling women to fulfil their full potential in the workplace is not simply a consideration of equality. Research carried out by the Women and Work Commission estimates that the under- utilisation of women’s skills costs the UK economy between 1.3 and 2% of GDP every year (Women and Equalities Committee, 2016).

1.2 The GPG: drivers and protective factors

Analysing data from the United Kingdom Household Longitudinal Survey (UKHLS) relating to 2014/15, Olsen and others (2018) model the predictors of the GPG, referring to those predictors that increase the GPG as ‘drivers’, and those that decrease it as ‘protective factors’.

Figure 1 presents the results of this analysis along with an explanation underlying each of the drivers and protective factors.

Figure 1: Main drivers and protective factors of the pay gap in the UK

Drivers:

• Occupational segregation (19%, £0.31) – The types of job that women tend to do are less well paid than the types of jobs that men do. 
• Industrial sector (29%, £0.47) – The sectors of the economy that women tend to work in are less well paid than the sectors that men work in. 
• Unobserved factors (35%, £0.57) – Part of the GPG cannot be explained by the data we have — but factors could include discrimination, harassment, preferences, and choices (constrained or otherwise). 
• Labour market history (56%, £0.91) – Differences in the ways men and women participate in the labour market: 12.7% is accounted for by the fact that women tend to have more years spent out of the labour market and undertaking unpaid care work than men, and 43.6% is accounted for by the fact that women tend to have fewer years of full-time work experience than men. 

Protective factors:

• Part-time employment history (-20%, -£0.33) – The different types of part-time jobs that men and women do – women who work part-time tend to have higher quality jobs than men who work part-time. 
• Institutional factors (-16%, -£0.26) – Women tend to benefit from workplace factors such as public sector employment, union membership, firm size and job tenure 
• Education (-4%, -£0.06) – On average, women have slightly more years of education than men, and more highly educated women have 

Source: Olsen et al (2018)

1.3 About the research

In order to understand more about the barriers preventing women’s progression in the workplace, and how best to support employers to remove these barriers, the government Equalities Office (GEO) commissioned the Global Institute for Women’s Leadership to undertake a rapid evidence assessment (REA) on this topic. The review was commissioned in June 2018 and considered evidence up until August 2018.

1.3.1 Research aims and questions

The aim of this research was to provide an overview of the gender divide in progression in the workplace in the UK, an understanding of the barriers and facilitators of women’s progression, and a critical overview of organisational policies that employers can implement to successfully remove these barriers.

There are various different ways of measuring progression. In this report progression is interpreted broadly as meaning not only moving up the occupational hierarchy, but also any job change that results in better pay, working conditions, responsibility or security. The focus of the research is, therefore, loosely on what is termed vertical segregation – that is the ‘concentration of women and men in different grades, levels of responsibility or positions’ (European Institute for Gender Equality, 2019).

The main objectives of this research were to:

• establish the extent of the gender divide in progression in the workplace as identified in the UK literature
• understand what is known about the barriers and facilitators to women’s progression in the workplace
• critically review and synthesise the evidence on which organisational policy interventions have been demonstrated to be successful in improving women’s progression in the workplace
• identify possible untested organisational interventions which the evidence suggests could be successful

1.3.2 Research approach

To provide a comprehensive overview of the literature on this topic in a short time-frame, a rapid evidence assessment (REA) approach was adopted (for further details please see the appendix). The search was limited to peer-reviewed papers readily accessible online in English. After a series of scoping searches produced a large number of hits, a decision was made to limit the evidence for objectives 1 and 2 to UK based studies, while leaving the evidence for objectives 3 and 4 geographically unlimited. The search was time limited to return only literature published from 2000 until August 2018. For objectives 1 and 2 this still produced a large number of relevant papers and so all located literature published between 2010 and 2018 was read and then literature focusing on under-researched groups or occupations published prior to 2010 was prioritised.

Search terms were developed iteratively through a series of scoping exercises and agreed with the GEO prior to running the final searches. Searches were supplemented by ‘pearl-growing techniques’, including following up on the references of key texts, and papers subsequently referencing them. A set of agreed inclusion and exclusion criteria was drawn up, including an assessment of the quality of the methodology.

Through this approach, 175 articles were identified for inclusion.

1.4 Report structure

This report contains findings synthesised from the REA on the gender divide in progression in the workplace.

Chapter 2 sets out findings on the extent of gender divide in progression in the workplace. It considers 2 different ways of capturing progression. Section 2.1 looks at progression to higher pay and the role of parenthood and part-time work in explaining gender differences. Section 2.2 looks at progression up the occupational hierarchy, the explanatory role of part-time work in explaining gender differences and the gender divide in progression out of initial ‘bad’ jobs.

Chapter 3 contextualises the national data presented in chapter 3 by synthesising findings from mainly qualitative firm or occupation level studies on the barriers preventing women’s progression, as well as any identified facilitators. The evidence for this chapter was limited to studies conducted in the UK. These barriers and facilitators are grouped into 4 main categories. Section 3.1 looks at organisational norms and processes, Section 3.2 looks at hostile or isolating organisational cultures, Section 3.3 looks at the conflict between caring responsibilities and current models of working. Section 3.4 synthesises evidence on the current difficulties that alternative working patterns offer in terms of career progression. Finally, Section 3.5 links the findings of this chapter to broader labour market trends.

Chapter 4 provides a critical synthesis of evidence from studies which tested interventions to improve women’s progression in the workplace. The evidence for this chapter was broadened to include studies conducted in any country which addressed barriers identified in the UK context.

Chapter 5 concludes the report with a discussion of key findings and their implications.

2. The gender divide in workplace progression

For the purposes of this report we interpreted progression broadly, as meaning not only moving up the occupational hierarchy, but also any job change that resulted in better pay, working conditions, responsibility or security.

Because inter-occupational hierarchies are not comparable across occupations, progression can be difficult to capture quantitatively in a way that enables broad comparisons. In this section we look at 2 different ways of capturing progression: progression to higher pay (2.1) and progression up a ranked occupational hierarchy (2.2). This section is based on studies which analyse nationally representative survey data.

2.1 The gender divide in wage progression

2.1.1 A steady upward trajectory for men, a plateau in their late 20s for women

Costa Dias et al (2018) use data from the Labour Force Survey, British Household Panel Survey and Understanding Society from 1991 to 2015 to study the gender differences in wages over the life- course (figure 2). There is a very low gender wage gap on labour market entry. The gap widens gradually but significantly from the late 20s and early 30s, driven by the fact that men’s wages grow rapidly at this point, whereas women’s stagnate.

Figure 2. Mean hourly wages across the life cycle by gender and education

Source: LFS data presented in (Costa Dias and others, 2018)

2.1.2 Parenthood leads to a rapid increase in the gender wage gap

Using the same data, Costa Dias and others (2018) model gender differences in wage progression in years since the birth of the first child, including controls for industry and occupation and differences in full and part-time labour market experience (Figure 3). They find that a substantial wage gap exists before childbirth (about 10% overall and slightly less when industry and occupation are taken into account), but that the gap increases rapidly after the birth of a woman’s first child.

Figure 3. Gender wage gap by time to/since birth of first child, controlling for association between wages and other characteristics

Source: BHPS 1991 to x2008 and Understanding Society 2009 to 2015 presented in (Costa Dias and others, 2018)

2.1.3 Parenthood leads to part-time work for women and part-time work shuts down wage progression

As demonstrated by Figure 3, gender differences in full and part-time experience account for a large amount of the gender differences in wage progression– contributing far more than industry and occupation.

Figure 4 shows how the percentage of men and women working in part-time jobs (less than 25 hours) changes in the period around the birth of the first child. Prior to parenthood, there are only small differences in the rates of part-time work among men and women. When the first child arrives, a large proportion of women transition to part-time work, while there is barely any effect on men’s working patterns. Women with less education are more likely to persist in part-time work as their child grows up.

Figure 4. Percentage of all men and women in jobs of no more than 25 hours per week

Source: BHPS 1991 to 2008 and Understanding Society 2009 to 2015 presented in (Costa Dias and others, 2018)

Costa Dias and others’ analysis suggests that part-time workers are significantly disadvantaged with regards to wage progression compared to full-time workers. This disadvantage goes over and above what one might expect if part-time work offered pro rata the same return on experience as full-time work. While the average full-time worker would expect to see wage growth year on year due to her increased experience, they find that part-time workers experience virtually no growth.

Gender differences in rates of part-time and full-time work account for about half of the widening of the gender wage gap in the first 20 years of a family’s first child’s life.

Greater opportunity costs for highly educated women

Part-time working shuts down wage progression for women, regardless of education level but, because highly educated women would have seen the most wage progression if they remained in full-time work, this effect has a larger impact on highly educated women’s wage growth. A graduate who had worked for 7 years before childbirth would, on average, have seen her hourly wage rise by 6% from an additional year of full-time experience, compared to 3% for a woman with a GCSE level education. However, switching to part-time work would, on average, lead to negligible progression for both (Costa Dias and others, 2018).

The UK has one of the highest rates of part-time work in the EU

The UK labour market is characterised by a high proportion of women working part-time. As shown by Figure 5, among European countries the UK has one of the highest proportions of female part-time workers as a share of its total working female population.

Figure 5. Employment rate of the population aged 20 to 64 and percentage of part-time workers by gender in 2016

Source: Eurostat, LFS presented in (European Commission, 2017)

2.1.4 Women report higher work intensities while receiving lower wage growth

If men experience higher wage growth than women in the same occupation, then perhaps women are compensated by having lower work intensities. Lindley’s (2016) analysis of national data from the Skills Employment Survey finds that the opposite is, in fact, the case - women report higher work intensities – working at high speed – than men in the same occupation, while simultaneously seeing lower pay growth. These differences are insignificant on labour market entry but emerge subsequently.

This suggests that the relatively slow growth in female wages compared to male wages within the same occupation is not compensated by women having lower work intensities, and conversely that working very hard and working at high speed are not associated with compensation by higher pay.

2.1.5 A significant portion of the gender gap in wage progression remains unexplained

While gender differences in the accumulation of experience and other forms of human capital are able to account for some of the gender gap in wage progression, a substantial amount remains unexplained.

Manning and Swaffield (2008) consider not only human capital explanations of the gender gap in progression – including gender differences in occupation, training, labour market participation, and part-time work - but also explanations based on gender differences in job mobility or psychological attitudes – for example, towards risk taking, competition and self-esteem. They find that, of the overall wage gap observed 10 years after labour market entry, human capital explanations can explain about half, and differences in job mobility and psychological explanations only a small amount. A large gap remains meaning that ‘although men and women have similar earnings when entering the labour market, the women will be something like 8% behind the men 10 years later even if they have been in continuous full-time employment, have had no children, do not want any and have the same personality as a man.’ (p.1018)

Using more up to date data Costa Dias and others (2018) similarly report that ‘by the time the first-born child is 20, the difference in the hourly wages between men and women is about 30%. Of that gap, around one-quarter already existed when the first child arrived. Of the remaining three-quarters around half is due to factors other than the differences in rates of part-time and full-time paid employment after childbirth.’ (p.3)

As demonstrated in Figure 3 differences in occupation and industry are only able to account for a small amount of this gap.

2.1.6 The differences appear to be within firms rather than between them

One candidate factor to explain the remaining difference in wage growth between men and women is that women and men are likely to be sorted into more or less productive firms. If men tend to work in more productive firms then this might account for some of their higher wage growth (Costa Dias and others, 2018). Using linked employer-employee data from the Annual Survey of Hours and Earnings 2002 to 2016, Jewell and others (2019) ask whether the gender wage gap is driven by the sorting of men and women into different firms (and therefore the relative wage premiums that certain firms pay), after controlling for occupations, tenure and other influencing factors. They find that only a tiny fraction of the gender wage gap can be attributed to the allocation of men and women between different firms and occupations, concluding that ‘the clear majority of what explains the pay gap shows up within firms.’ (p.17)

2.2 The gender divide in occupational attainment

An alternative method of measuring for career progression is provided by Bukodi and others (2012). They construct a vertical ranking of occupations by first coding all occupations in an individual’s history against the 1990 Standard Occupational Coding frame (SOC90). This provides 77 occupational codes which they rank according to the average hourly pay within that occupation. Finally, they convert the ranking into scores between 1 and 100 which represent the relative positions of that occupation within the occupational distribution.

2.2.1 Men reach higher levels of occupational attainment than women, but the gap has narrowed over time

Using the ranking system described above, Bukodi and others (2012) compare the occupational mobility of men and women across 3 different birth cohorts - 1946, 1958 and 1970. Figure 6 shows the career progression over the life course for all men and Figure 7 shows the same for all women, and for those women who never worked part-time. They find that men reach higher levels of attainment than women in each cohort but that the divide has narrowed over time.

Figure 6. Men’s mean occupational earnings scores across age by cohort

Source: Bukodi and others (2012)

Figure 7. Women’s mean occupational earnings scores across age by cohort

Source: Bukodi and others (2012)

2.2.2 Part-time work is an important explanation of the differences between men and women in occupational attainment

Figure 7 demonstrates that women’s part-time work is an important explanation of the remaining gender differences in occupational attainment. Among the cohort born in the 1970s, women who had never worked part-time reach career levels very similar to men.

2.2.3 Badly paid jobs are sticky floors for women, springboards for men

Using the same vertical occupational scale, Bukodi and others (2012) analyse how first job on labour market entry affects later occupational attainment. They ask whether workers who enter the labour market in the lowest occupational level are able to use their initial job as a springboard into a better paid job, or whether they remain stuck in sticky floors. Their findings suggest that bad jobs are springboards for men, but sticky floors for women. While, initially, men and women move out of the lowest occupational category at similar rates, thereafter their trajectories differ. Men are more likely to experience a steady upward trajectory, whereas women’s trajectories stall and they are much more likely than men to experience subsequent downward mobility back to their original level.

Stewart (2009) similarly finds that single mothers who enter low-skilled employment experience very slow wage progression. While many leave employment again, those who remain rarely progress beyond low pay, suggesting that more needs to be done to foster wage progression among low paid workers.

The sticky floor/springboard gender divide has worsened over time

While Bukodi and others (2012) find that, overall, women’s occupational attainment has become more similar to men’s over time, this gender divide between springboards for men and sticky floors for women has deepened. They attribute this to an increasing polarisation of employment and occupational structures coupled with uncertainties about labour market conditions. Qualitative work explored further in Sections 3.3 to 3.4 suggests that some of this stickiness can be attributed to structural barriers to progression for women with caring responsibilities, including the demand for geographic mobility or longer working hours at higher occupational levels (Rainbird, 2007).

3. Barriers and facilitators

This section is based on evidence from 100 studies which were located in response to objective 2 (see appendix for a full outline of the search strategy). The search was limited to studies carried out in the UK and all relevant study designs were considered. Initial searches were limited to studies published after 2000. This produced a large number of relevant results and so those published post-2010, or those focusing on under-studied groups were prioritised.

The evidence that this section is based on is largely qualitative, drawn from interviews or ethnographies. Collectively, it provides the perspectives of hundreds of women and their colleagues from a diverse range of occupations and sectors. It thus helps to contextualise the national data set out in Section 2, providing an insight into the processes which create and sustain the observed differences in progression.

While it is difficult for qualitative studies to definitively prove cause and effect, or to generalise from findings observed in one firm or sector to another it provides an overview of the sorts of barriers that employers might encounter. Where pertinent, we have indicated the context from which the study was drawn. Where the source evidence permits, we also note where any barriers interact with other dimensions of identify such as ethnicity and class.

This section presents the findings from those studies. Section 3.1 explores organisational processes, Section 3.2 looks at organisational cultures, Section 3.3 explores the challenges of reconciling caring responsibilities with current working practices and 3.4 sets out the current difficulties of combining alternative working practices with career progression.

3.1 Organisational processes for progression that open up space for bias

This section provides an overview of how the organisational processes which govern workplace progression either generate gender differences or increase the likelihood that such differences will arise.

Women are disadvantaged by organisations which lack formality and transparency with regards to the criteria for recruitment, promotion or pay negotiation (3.1.1). In the absence of such criteria decisions on pay and promotion are more likely to be made via a process of social cloning, whereby those in a position of power champion those like them (3.1.2). In both informal and formal environments progression often happens via male-dominated social networks which women find difficult to access (3.1.3). The presence of an organisational sponsor was identified as a key facilitator to overcome some of these barriers (3.1.4).

3.1.1 Informality and lack of transparency leaves the way open for progression via processes that disadvantage women

Informality and an associated lack of transparency with regards to the criteria for recruitment, promotion or pay negotiation were identified as key organisational barriers to gender equality. As will be explored in the following sections, in the absence of clear systems advancement to more senior roles and higher pay grades can occur via processes that disadvantage women.

This barrier was located both in research on specific professions and sectors, including television (Leung and others, 2015), law (Tomlinson and others, 2013) and academia (Burkinshaw and others, 2018), as well as in more wide-ranging studies. In Corby and Stanworth’s (2009) study of 80 working women employed in 40 organisations, they find that nearly every private sector organisation lacked a structured pay system and, consequently, pay was determined opaquely, often by reference to what an individual had earned previously. This opacity, combined with an institutional culture that mandated against discussing pay with colleagues, left women with no way of knowing whether they were being paid the same as comparatively employed men.

Knowledge workers – those who (like lawyers, academics or television professionals) are required to ‘think for a living’, may be particularly likely to be exposed to this sort of informality, as employers seek ‘someone they can trust’ and where that trust is established in advance via contacts and reputation (Eikhof and Warhurst, 2013, Leung and others, 2015). Eikhof and Warhurst (2013) argue that, in certain industries, such as the television and wider creative industries, informality is not accidental, but a central part of a flexible, project-based model of production in which the majority of staff are freelancers, and recruitment and advancement happen primarily via personal networks. This increases the likelihood of advancement via social cloning, which is explored further in the next section.

3.1.2 Women are disadvantaged when those in positions of power seek those who look like themselves

Studies in the financial services sector (Atkinson, 2011), academia (White and others, 2012; Bagilhole, 2016; Shepherd, 2017) and among consultants (Kumra, 2010) identify occasions when promotion happens via a process of social cloning, whereby those in positions of authority seek to champion and promote those who are similar to themselves. This similarity can take many forms, including in terms of career path (White and others, 2012) or styles of leadership (Kumra, 2010; White and others, 2012). Given the male dominated nature of the senior ranks of many organisations, this puts women at a disadvantage, since women are less likely to possess, or be seen to possess, the relevant qualifications of the archetypal candidate.

This process of social cloning is made easier in informal systems where hiring and promotion are done by networks, or via trust relationships based on “affect and tacit judgements such as ‘he’s a good bloke’”(Leung and others, 2015).

Women are expected to fulfil a male-centred definition of merit, but may be penalised when they do

Perhaps unsurprisingly in these contexts, women report perceiving that they need to ‘behave like men to get on’ – to meet the qualifications of the archetypal candidate, which is defined by relationship to a male-centred definition of merit (Tomlinson and others, 2013; Charles, 2014; Woodfield, 2016). This is especially the case in environments (such as politics) where success, or occupational identity are closely tied to an ‘an aggressive, confrontational masculinity’ (Charles, 2014), but interestingly was also found in studies of feminized professions such as teaching, where headship was seen as requiring ‘tough and coercive’ qualities, something that was felt to be anathema to the female teachers (Smith, 2011a).

Peters and others (2017) investigate how this (lack of) fit between women’s perceptions of their own attributes and those of a successful leader affect their ambition and intention to persist in a career. They asked 62 female police officers to rate themselves on a series of character traits, while also asking them to rate how senior officers fared on those same traits. They found that, on average, female officers rated a mismatch between their own character traits (for example, collaborative and sociable) and those of senior officers (decisive, assertive, arrogant). This lack of fit was a significant predictor of a reduced level of ambition and higher propensity to exit the workforce.

There was some evidence from British organisations, too, of a well-documented concept from social psychology – the double bind. Women are seen as lacking the agentic qualities needed to be good leaders, but are penalised by others when they do, since these contradict the expected feminine qualities of warmth and helpfulness (Carli and Eagly, 2016). A study of female IT workers found that women perceived that when they demonstrated agentic traits modelled by their managers, such as being ‘forthright’, this was unfavourably received (Kirton and Robertson, 2018).

As will be explored further in Section 3.3.1, in some cases behaving like men means minimizing the visibility of family life (Tomlinson and others, 2013). For the black and ethnic minority lawyers interviewed by Tomlinson and others (2013) it also meant playing down other differences, such as by avoiding involvement in BME networks.

Work typically done by women can be devalued in terms of career capital

When progression is linked to social cloning the work that is typically associated with men is related to higher career capital. Conversely, work that is typically associated with women is devalued in terms of future returns on career progression. So that, for example, female academics reported higher levels of teaching and pastoral care for their students, activities that contributed little towards status and future promotion (Bagilhole, 2016). Similarly, the women working in trade unions studied by Guillaume and Pochic (2011) were over-represented in certain types of role where they were responsible for social activities, equalities or health and safety, while the roles typically held by men, such as grievance handling or contract negotiation were seen as being the most prestigious.

In some studies women reported being pushed towards or held in devalued roles by colleagues who perceived them as fitting with their gender-stereotypical skillsets (Moreau and others, 2007; Scholarios and Taylor, 2010; Bagilhole, 2016; Kirton and Robertson, 2018). In one study of call- centre workers the call-centre managers described female workers as being better at dealing with difficult customers. The researchers observed that this meant they were less likely than their male peers to be moved from customer-facing roles to more challenging or specialised positions (Scholarios and Taylor, 2010). Similarly female IT workers described being assigned to roles that involved ‘nagging the men’ to meet project deadlines, while a male manager in the same organisation described valuing his female staff’s ability to stop men from ‘taking lumps out of each other’ (Kirton and Robertson, 2018). In these situations ‘feminine skills end up being a catch-22 for women’ (Kirton and Robertson, 2018), in so far as they lead women to be herded into internal roles that offer lower career capital.

3.1.3 Women are disadvantaged when progression happens via social networks, which they find difficult to access

A large number of studies identified the key power that networks play in determining career advancement, along with women’s relative lack of access to these networks, as barrier to women’s career advancement (Miller and Clark, 2008; Tonge, 2008; Kumra, 2010; Pritchard, 2010; Atkinson, 2011; White and others, 2012; Eikhof and Warhurst, 2013; Tomlinson and others, 2013; Cahusac and Kanji, 2014; Wilson, 2014; Savigny, 2014; Leung and others, 2015; Bagilhole, 2016; Wright, 2016; Burkinshaw and others, 2018; Kirton and Robertson, 2018). Networks play a role not just in deciding pay and promotion but in allocating work assignments (Fearfull and Kamenou, 2006) which may play an important role in the build-up of subsequent career capital and, therefore, later progression.

Promotion via ‘grace and favour appointments’ (Burkinshaw and others, 2018) or based on ‘who you know’ (Wilson, 2014) is problematic, since women find it harder to gain access to the relevant male-dominated networks (Wilson, 2014), with ethnic minority women facing an additional hurdle of gaining access to primarily white networks (Fearfull and Kamenou, 2006).

Women can also be excluded due to the nature or location of the networking activity – for example, networking organised around sporting events (Tomlinson and others, 2013; Wilson, 2014) – or being held in a club ‘which still has rooms that women aren’t even allowed in’ (Tomlinson and others, 2013, p. 255). Timing is also a key issue, with the requirement for socialising and networking in the evening after already long working hours a barrier to those with caring responsibilities (Tonge, 2008; Pritchard, 2010; Atkinson, 2011; Cahusac and Kanji, 2014; Leung and others, 2015).

In some industries career networking is tied up with geographic mobility – for example, academic conferences – which similarly poses difficulties for those with caring responsibilities (Pritchard, 2010). Networking events centred around drinking were cited as challenging for some of the Muslim lawyers interviewed by (Tomlinson and others, 2013).

Networks may play an especially important role in informal environments which lack clear standards for career or pay progression and where the lack of transparency considerably obscures the proverbial ‘old boys networks’ (Eikhof and Warhurst, 2013). However, networks may also be used to make decisions about pay and promotion in contradiction of formal procedures (Kirton and Robertson, 2018). In certain occupations, such as law, extensive networking is a necessary pre-cursor to ‘rainmaking’ – bringing in and effectively managing the client relationships that are a key criterion for promotion (Tomlinson and others, 2013).

While the majority of studies that identified this barrier focused on professional women, a study of women working in the construction industry found similarly that in manual trades networks are sites of information exchange, important for the gaining of new skills and progression (Wright, 2016).

3.1.4 A champion or mentor can counteract some of this disadvantage

In some of the studies we located, women described how they had benefited from informal mentorship by senior figures within their own organisation, both female (Guillaume and Pochic, 2011; Norman, 2014) and male (White and others, 2012). Within these studies the description of the mentoring relationship went beyond providing advice to actively sponsoring their careers and was, in some cases, specifically designed to counter some of the effects of social cloning described in the preceding sections.

3.2 Hostile or isolating organisational cultures

This section looks at barriers that can arise for women working certain organisational cultures including isolation and lack of role models (3.2.1) and sexism and sexual harassment (3.2.2). Support networks are helpful for easing isolation (3.2.3).

3.2.1 Isolation and lack of role models can lead to a lack of confidence

Within male-dominated organisational cultures such as construction and sports coaching, social isolation was a problem for female study participants (Norman, 2014; Worrall and others, 2010). In some cases this was cited as leading to a lack of confidence that caused women to self-select out of working towards more senior roles (Norman, 2014).

3.2.2 Sexism, sexual harassment and bullying

A more severe issue associated with male-dominated organisational cultures was that of sexism and, in many cases, sexual harassment and bullying. Women working in construction (Worrall and others, 2010), finance (Wilson, 2014), retail managers (Broadbridge, 2007) and as pilots (McCarthy and others, 2015) described cultures in which sexist assessments of their competence from co-workers or clients affected self-esteem and their ability to progress. Fearfull and Kamenou (2006) and Umolu (2014) both note that, for ethnic minority women, racist or religious stereotypes sit alongside gender stereotypes so that, for example, a manager of Nigerian descent described being mistaken for the caterer (Umolu, 2014).

At the more extreme ends were cultures in which sexual harassment was common, with respondents feeling they had to turn a blind eye (Tomlinson and others, 2013) or even where the viewing of pornography at work and corporate entertaining involving the sex industry was part of organisational culture (Atkinson, 2011). Corby and Stanworth (2009) found that while some women had exited organisations in situations where bullying and sexual harassment had become intolerable, much more common was a resigned acceptance – ‘In short, they considered that a woman-hostile culture was a price worth paying for working in a predominantly male environment, especially where they were able to progress into managerial roles.’ (ibid, p.175)

3.2.3 Support networks can help in easing isolation

Some studies described the importance of wider support networks of colleagues for either easing the isolation felt within male-dominated working cultures (Wright, 2016; Papafilippou and Bentley, 2017) or providing emotional support in professions (such as medicine) where burn-out is common (Walsh, 2012).

3.3 Conflicts between external responsibilities and current models of working

As is suggested by the findings in Section 2, the majority of our studies cited the conflict between current models of working and caring responsibilities as being a significant barrier to women’s progression. This section explores the conflict in more detail.

Women continue to take on disproportionate responsibility for unpaid care – three-quarters of mothers say they have primary responsibility for childcare in the home, and women make up 58% of carers for the old or disabled (Women and Equalities Committee, 2016). Time use data from 2000 and 2015 analysed by the ONS finds that, while fathers with pre-school children are increasing the time they spend on care, mothers provided 74% of total childcare time in 2015 (ONS, 2017). Women also continue to perform the majority of housework (Wheatley and Wu, 2014). Ethnic minority women may face additional hurdles as work-life difficulties can be accentuated by cultural and religious expectations (Kamenou, 2008).

Working hours in the UK are long. On average, full-time employees in the UK work 42.3 hours per week, the highest number in the EU, against an Europe wide average of 40.3 (Eurostat, 2018).

Women’s progression and labour market participation is jeopardised not only by their inability to work long hours, but also by their partner’s long hours. Research with longitudinal data from the USA finds that having a partner who works long hours significantly increases a woman’s likelihood of leaving the labour market, although the reverse is not true (Cha, 2010). Analysis of BHPS data from the UK finds that women report greater levels of dissatisfaction with performing unpaid care and housework when the uneven division is linked to long working hours in their partners (Wheatley and Wu, 2014).

There is still evidence of direct discrimination against women with caring responsibilities (Section 3.3.1). However, more commonly we found evidence of prevailing organisational norms of overwork and expectations of constant availability as a proxy for commitment and merit. In many cases this expectation of constant availability stems not from the nature of work but from specific working practices and ideas of what makes a ‘committed worker’ (3.3.2). Unpredictable work demands linked to casualised forms of labour also pose a challenge for those with caring responsibilities (3.3.3) as do demands for geographic mobility in order to progress (3.3.4). Women find it hard to access training opportunities if they conflict (or are perceived to conflict) with caring responsibilities (3.3.5). Women’s working hours are affected by levels of pay and the UK’s high childcare costs (3.3.6) Access to childcare, alternative ways of working, flexible partners and changes to working time norms were found to be facilitators (3.3.7).

3.3.1 There is still evidence of direct discrimination against women with children regarding promotions

In a number of studies women described having to play down their caring responsibilities in order to avoid stigmatisation (Callan, 2007; Kumra, 2010; Tomlinson and others, 2013; Cahusac and Kanji, 2014). As one consultant put it ‘I think it’s definitely a requirement to show great commitment to the firm, which means not having children or, if you do have children, keeping extremely quiet about that.’ (Kumra, 2010, p. 235).

Examples of direct discrimination against women with caring responsibility were also cited, including claims that women were classed as less of an asset if they had children, and subsequently passed over for promotion (Walsh, 2012), or women being told they would not be offered a promotion as their children meant they were unable to be likely to leave and seek a better offer elsewhere (Savigny, 2014).

3.3.2 Women are disadvantaged where progression is linked to overwork and presenteeism

Much more common than direct discrimination, and the dominant theme in the studies we located, was of a conflict between organisational norms of overwork and boundless availability, and caring responsibilities (Dean, 2007; Moreau and others, 2007; Lyonette and Crompton, 2008; Miller and Clark, 2008; Corby and Stanworth, 2009; Broadbridge, 2010; Kumra, 2010; Scholarios and Taylor, 2010; Tomlinson and Durbin, 2010a; Vinnicombe and others, 2010; Worrall and others, 2010; Ford and Collinson, 2011; Ozbilgin and others, 2011; Smith and Elliott, 2012; Walsh, 2012; Tomlinson and others, 2013; Wattis and James, 2013; Cahusac and Kanji, 2014; Michielsens and others, 2014; Charlesworth and others, 2015; Jefferson and others, 2015; Munro and others, 2015; Papafilippou and Bentley, 2017; Budjanovcanin, 2018; Kirton and Robertson, 2018). Although the majority of these studies focused on professional and managerial occupations, similar themes were also found in studies of low-income parents (Dean, 2007), call-centre workers (Scholarios and Taylor, 2010), retail shop managers (Smith and Elliott, 2012) and service industry workers (Tomlinson, 2007).

In many cases ‘all hours working cultures’ (Ozbilgin and others, 2011) were bound up with occupational identities, so that succeeding in that occupation was seen as fundamentally at odds with outside caring responsibilities. This was demonstrated through the widespread acceptance of (un)spoken professional codes of conduct such as ‘the firm expected all or nothing’ (Vinnicombe and others, 2010, p. 180), ‘patients first, family second’ (Ozbilgin and others, 2011, p. 1591) or ‘paid care work trumps unpaid care work’ (Charlesworth and others, 2015, p. 608).

Conversely, anything less than ‘maximum flexibility’ - always being available for the organisation and the client, customer, patient, or the service user - is seen as demonstrating a lack of commitment either to the organisation or to one’s career (Kumra, 2010; Ozbilgin and others, 2011; Walsh, 2012; Tomlinson and others, 2013; Michielsens and others, 2014; Charlesworth and others, 2015; Jefferson and others, 2015; Munro and others, 2015; Kirton and Robertson, 2018).

Long working hours were viewed critically by the authors and subjects of the studies as being less down to the demands of the work than a proxy for commitment and merit. As such, in many workplaces opportunities for promotion are equated with the ability to demonstrate this unrestricted availability, and a culture of presenteeism (Tomlinson, 2007; Lyonette and Crompton, 2008; Corby and Stanworth, 2009; Atkinson, 2011; Crompton and Lyonette, 2011; Guillaume and Pochic, 2011; Tomlinson and others, 2013; Papafilippou and Bentley, 2017).

In other occupations, such as teaching or public sector management, excessively long hours were linked less with organisational pressures of presenteeism and more with an excess workload linked to the re-organisation of work around the principles of managerialism and modernization (Miller, 2009; Conley and Jenkins, 2011; Burkinshaw and others, 2018). In these workplaces top-down targets created an unbearable workload.

Although, in these examples, both male and female employees were subject to the same pressures, the fact that women continue to carry out the majority of unpaid care reduces their ability to work the long hours in evenings and weekends necessary to complete the work.

The pressure to work excessively long hours is linked with unsupportive line managers, not only in professional jobs in the private sector (Kirton and Robertson, 2018), but also in organisations in the public and voluntary sectors who ostensibly subscribe to work-life balance ideals and offer lower remuneration (Dean, 2007). Managers who failed to understand caring responsibilities expected workers to be able to cope with unexpected and spontaneous demands on their time (Wattis and James, 2013). In some cases superficially supportive line managers expressed sympathy with the difficulty that the conflict between caring responsibilities and long hours/maximum flexibility organisational culture created for women, but did not perceive a personal responsibility to find a solution (Kirton and Robertson, 2018). In this view the conflict between caring and work was a wider societal one, and not something produced by a certain and specific way of organising work.

3.3.3 Women are disadvantaged by unpredictable work demands

For some workers, particularly the low-paid, the unpredictability of work demands is built into their contracts of employment. Indeed, lower level workers are the group with the least autonomy over work time (Warren and Lyonette, 2018). In the last quarter of 2017 901,000 people were estimated to be on a zero-hour contract – 55% of these people are women. In total 2.7% of women in employment are on zero hours contracts (De Henau and others, 2018).

TUC research with zero-hours workers found that 51% have had shifts cancelled at less than 24 hours’ notice, and 73% had been offered work at less than 24 hours’ notice (Trades Union Congress, 2017), making it extremely difficult to make childcare arrangements. This precarity was borne out in the qualitative studies of low-paid retail workers and of retail store managers who had demands made on their time at short notice (Backett-Milburn and others, 2008; Smith and Elliott, 2012), while the complex and changing shift patterns of call centre workers lead to interference with family life (Scholarios and Taylor, 2010). Many respondents in Backett-Milburn and others’ (2008) study of low-paid women working in food retail commented that, were it not for the support of family members providing childcare, they would be unable to afford to continue working in their jobs.

Precarity of work and the unpredictability of hours also extend to freelance and contract workers. This was particularly notable in the television industry where women with children described the impossibility of obtaining childcare due to the unpredictable flow of work. This may go some way towards explaining the low retention rate of women in the film and television industries (Leung and others, 2015). In academia, where the use of short-term contracts is increasingly common, some women felt it necessary to delay pregnancy until they had secured a permanent contract (Pritchard, 2010).

3.3.4 Women are disadvantaged by the need for geographic mobility to progress

In a number of careers, constant geographic mobility is either a requirement of the job – for example, pilots, (McCarthy and others, 2015), politicians (Charles, 2014) – or key in building up the required career capital for progression – for example, academics (Shepherd, 2017; Miller-Friedmann and others, 2018) – and is difficult for those with domestic and caring responsibilities. This is not just an issue for professional women.

Harris et al (2007) found that the requirement for managers to transfer to another store meant that career progression would have caused problems for part-time working retail shop floor staff, for whom proximity to home and childcare are major considerations.

3.3.5 Women are disadvantaged when training opportunities are inaccessible to those with caring responsibilities

Some women reported difficulty in accessing training and development opportunities. Sometimes this took the form of direct bias. Women reported that comparable men were encouraged to attend training courses with a view to promotion, whilst they were unable to access them (Atkinson, 2011). In other cases, managers perceived development programmes as being too demanding for women with families (Kirton and Robertson, 2018). Despite the 2000 EU Part-Time work directive giving part-time workers the right to equal treatment, there was some evidence of part-time workers being denied access to training opportunities (Tomlinson, 2007).

In other cases, female workers were unable to take up training and development opportunities due the necessity to participate outside of (already long) working hours, creating a conflict with caring responsibilities (Kirton and Robertson, 2018). Part-time workers described being expected to attend training events held at times when they were not working, or in locations that presented additional family commitment issues (Harris and others, 2007).

3.3.6 High childcare costs mean that the ability to outsource care work is strongly linked to pay

In contrast to many EU and OECD countries the majority of childcare in the UK is provided by the private sector, and costs are high (Lyonette, 2015; ONS, 2018). In 2019, the average cost of childcare in a nursery for a child under 2 was £127 for 25 hours a week (£6,600 a year) and £242 for 50 hours a week (£12,600 a year) (Coleman and Cottell, 2019).

Due to the high cost of childcare in the UK the ability to reconcile work and caring responsibilities by outsourcing care work is dependent on pay. In a study by Charlesworth and others (2015) among relatively low paid social service workers, both workers and managers cited the low pay as a reason why those with caring responsibilities were unable to ‘afford’ to work in social services.

Backett-Milburn and others’ (2008) study of low paid women working in food retail produced similar findings, with many women commenting that, were it not for the support of family members providing childcare, they would be unable to afford to continue working in their jobs.

As a result of this, lower paid women with pre-school children are more likely than other women to reduce their hours of work, whilst simultaneously relying on informal provision of care from relatives (Lyonette, 2015). The 2018 Childcare and Early Years Survey of Parents in England reported that among mothers working part-time, almost a third (30%) said they would increase their hours if there were no barriers to doing so, while just over half of non-working mothers said that if they could access good quality childcare which was convenient, reliable and affordable, they would prefer to go out to work (Department for Education, 2018a).

As previously noted, the UK has one of the highest proportions of female part-time workers as a share of its total working female population (Eurostat, 2018). Recent evaluations of the extension of the early years entitlement in England, from 15 hours to 30 hours a week for 38 weeks of the year for 3 and 4 year olds with working parents, indicate that reforms to increase the affordability of childcare have a positive relationship to the number of hours that mothers work: over a quarter (26%) of mothers reported that they had increased their work hours since receiving the extended hours, with the proportion reporting impacts on parental work greater for families in the lower income group than for those in the higher income group (Department for Education, 2018b).

3.3.7 Facilitators

Changes in working time norms can reduce the conflict between work and caring responsibilities

Although rare we did find one example of a top-down commitment to gender equality leading to a change in working time norms. Charles (2014)’s study of the National Assembly for Wales described how a support for positive measures meant that the Assembly now sits only during school term time, its plenary sessions finish by 17.30, and Assembly members are only expected to be present 3 days a week.

Access to alternative ways of working are key in reconciling work and caring responsibilities

Studies frequently cited the availability and acceptability of part-time work and flexible work as key in enabling women to reconcile work and caring responsibilities (Broadbridge and Parsons, 2005; Backett-Milburn and others, 2008; Carmichael and others, 2008; Miller and Clark, 2008; Walsh, 2012; Wattis and James, 2013; Woolnough and Redshaw, 2016). Studies among low-paid workers in particular mentioned the flexibility to fit work around childcare responsibilities as a key factor in their employment choices (Backett-Milburn and others, 2008; Broadbridge and Parsons, 2005), although it is essential to make a distinction between employee-driven flexibility, where workers are able to organise work around caring responsibilities, and precarious employer-driven flexibility which leads to interference with family life (Backett-Milburn and others, 2008; Scholarios and Taylor, 2010; Smith and Elliott, 2012).

For professional women access to part-time or flexible work was cited as a factor in enabling their retention in the workplace (Wattis and James, 2013; Woolnough and Redshaw, 2016). As is explored further in Sections 3.4.4 to 3.4.5 the evidence suggests that alternative ways of working currently play a bigger role in enabling women’s retention in the workplace than their progression.

Flexible partners help in reconciling work and caring responsibilities

Both men and women mentioned a flexible partner as important in helping them to reconcile work and family life. Professional men described the support they received from ‘understanding wives’ (Jefferson and others, 2015; Kirton and Robertson, 2018). Equally female respondents in the study cited supportive partners, and especially partners who worked part-time or flexibly as key in enabling their progression (Lyonette and Crompton, 2008; Woolnough and Redshaw, 2016).

Access to childcare

Access to childcare is an essential factor enabling women to combine work and caring responsibilities. Due to the high cost of childcare among low paid workers support from family members is important in enabling them to ‘afford’ to work (Backett-Milburn and others, 2008). Among professional mothers, satisfaction with either family support or private nurseries is important in enabling women to combined motherhood with family life (Woolnough and Redshaw, 2016).

3.4 Alternative ways of working do not currently offer parity

Alternative ways of working, including part-time and flexible work, are currently proposed as 2 key solutions to overcome the conflict between current ways of working and caring responsibilities described in the previous section. In 2017, 41% of women in employment were working part-time, compared to 13% of men (ONS, 2019). An increasing number of workers are also making use of flexible working arrangements – a term which encompasses a wide range of arrangements allowing workers to work more flexibly, including flexitime – a worker’s ability to control their schedules, and teleworking – a worker’s ability to choose where to work freely – for example, being able to work from home (Chung, 2017).

This section presents evidence from the literature on access to these alternative ways of working and their relationship to women’s career progression. Access to alternative arrangements is still limited. There continues to be a shortage of quality part-time work and, while there are increasing numbers of part-timers in senior positions, lower-level part-time workers remain crowded into jobs that offer poor wages and restricted opportunities for progression (3.4.1). Similarly, there is some evidence that flexible working policies are predominately offered to high-skilled workers. Under the current law the success of flexible working requests are dependent on an employee’s bargaining positions, meaning they are not available to less ‘valuable’ employees (3.4.2). Sections 3.4 to 3.5 consider the evidence on the effect that taking advantage of alternative ways of working has on career progression. Without changes to underlying organisational cultures, and while such working practices are taken up only by women, they risk embedding gender inequality in contexts where full-time workers are the norm, and part-time and flexible working is seen as signalling a lack of commitment (3.4.3). Clear evidence demonstrates the negative effect of part-time working on career progression (3.4.4) and emerging qualitative evidence suggests that flexible working may similarly have negative consequences (3.4.5). A final consequence of implementing alternative working time policies without changing underlying organizational cultures is that they can lead to an intensification and extensification of work (3.4.6). Alternative working roles models and supportive managers, as well as workplaces that help managers to deal with part-time and flexible workers were found to be facilitators (3.4.7).

3.4.1 Access to quality part-time work is still restricted

There continues to be an issue with the availability and quality of part-time jobs in the UK (Warren and Lyonette, 2018). This can lead to women reducing their hours of employment and, whilst they almost always return to a job with the same occupational status as the one they left, over time this leads to a reduced likelihood of progression at work (Harkness and others, 2019). Harkness et al (2019) found that fewer than 1 in 5 of all new mothers follow a full-time career after maternity leave and, among those who were working full-time prior to childbirth, a majority either stop working or move to part-time work. Five years after childbirth the chance of having occupationally upgraded is far lower for women than men: while 26% of men have moved to a job with a higher occupational status, for women this figure is just 13% (ibid). Using a different measure of occupational downgrading, Dex and Bukodi (2012) find that women at the top-level of the occupational hierarchy and ‘high-flyers’ in male dominated organisations are most at risk of downgrading during the move from full time to part-time work.

This risk of occupational downgrading is greatly reduced if women are able to access part-time work with the same employer (Connolly and Gregory, 2008; Dex and Bukodi, 2012) but their chance of progressing is also lower compared to those who move to a new employer (Harkness and others 2019). Recent cohorts of women having children since the passage of the EU part-time workers directive appear to have benefited from a legislative environment providing greater protection for part-timers and increased flexibility (Bukodi and others, 2012). However, it remains the case that there is a scarcity of senior or ‘quality’ roles available on a part-time basis (Lyonette, 2015), greatly reducing opportunities for part-timers progression.

In some professions, most notably medicine, women segregate into parts of the profession where part-time working is more accessible. The over-representation of women in general practice, and under-representation in hospital medicine is attributable to the high percentage of part-time women in general practice (Taylor and others, 2009). Similarly, Tomlinson et al (2013) find that a significant number of female lawyers migrate towards specialisms such as employment, estates and trusts and family law which offer better work-life balance.

There are widening differences in the quality of part-time work by occupational class

National data examined by Warren and Lyonette (2018) found evidence of widening inequalities in the quality of part-time work by occupational class. There is an increase of part-timers in senior level jobs, and those women are moving closer to their peers in terms of job quality. However, post-recession lower-level part-time workers face enduring disadvantage, crowded into jobs that offer poor wages and restricted opportunities for progression.

This confirms a divide observed by Tomlinson (2006a) between ‘optimal’ and ‘restrictive’ part-time jobs. In her study of mothers working in the hospitality industry she observed that ‘restrictive’ part- time jobs, while providing women some control over their shifts, offered little prospects of promotion. By contrast ‘optimal’ part-time jobs tended to be held by those who had already developed their careers, and thus were in a stronger position to negotiate part-time work and preserve occupational status. This suggests that part-time work can currently contribute more towards maintaining women’s labour market position than towards progression, something explored more in Sections 3.4.4 to 3.4.5.

Part-time workers remain concentrated in certain occupations

Research carried out (Tomlinson and others, 2009, 2005) for the Women and Work Commission found that when women transition to part-time work, they crowd into 5 low skill, low pay occupations, ‘the 5 Cs’ - caring, cleaning, clerical, catering and customer service work. Data from the 2018 Labour Force Survey shown in figure 8 confirms that the 5 Cs continue to account for 70% of part- time jobs. While there has been an increase in the number of part-time professional jobs, only 3.7% of managerial workers are part-time.

Figure 8: Occupational distribution of part-time workers

Source: Analysis of LFS by (Tomlinson, 2018)

3.4.2 There is still limited access to flexible working

All employees with 26 weeks’ continuous service with their employer have the right to request Flexible Working; however, there is a disparity in how that is adopted in practice. In 2015, approximately a quarter of workers had access to flexitime – the ability to control their schedules - while just under 10% had complete autonomy over their working hours and schedules, slightly above the EU average, but well behind Sweden and Denmark, as shown in figure 9 (Chung, 2017).

Figure 9: The proportion of workers across 28 European countries with access to flexible schedules in 2015

Source: (Chung, 2017)

Flexible working may be less available to lower-skilled workers:

There is some evidence from Europe-wide data to suggest that flexible working is predominantly available to high-skilled workers. Managers, professionals and those in supervisory roles are more likely to have access to flexible schedules, with parental status making very little difference (Chung, 2017). This is supported by data from the Skills and Employment Survey Series, which finds that lower-level workers are the group with the least autonomy over the start and finish times of their work (Warren and Lyonette, 2018).

These quantitative findings are complemented by data we located from qualitative studies. Charlesworth and others’ (2015) study of social care workers found a lack of flexibility around 12-hour shift patterns, while a retail store manager (Smith and Elliott, 2012) observed that flexible working and job share policies were available to the human resources staff within their organisation but not to them. In general, most low-paid women work in small and medium sized enterprises where there are fewer formal family friendly policies and flexibility, where available, is on an informal basis (Backett-Milburn and others, 2008) However, there was also evidence of flexibility being less available at senior levels, something that went hand in hand with long-hours organisational cultures (Broadbridge, 2007; Vinnicombe and others, 2010).

Access to flexible working is often more available in policy than in practice

The literature points to a well-documented implementation gap, with organisations ostensibly committed to flexible working but not providing it in practice (Budjanovcanin, 2018). One key element of this is the gatekeeping of access by managers, with successful policy use highly dependent on their discretion (Callan, 2007; Backett-Milburn and others, 2008; Atkinson, 2011; Budjanovcanin, 2018; Michielsens and others, 2014). A recent study found that, while half of employers in the private sector had written policies on flexible working, 74% of those did not have any written procedure to help managers deal with the request, indicating that most requests are being dealt with on a case by case basis, leaving room for a large amount of managerial discretion (Department for Business, Innovation and Skills, 2014).

Managers, in turn, felt constrained by policies that, if fully implemented, they feared would hinder their ability to meet targets (Callan, 2007; Michielsens and others, 2014), indicating the importance of top down support from senior management, as well as line managers, if flexible working policies are to be successfully implemented.

While some requests for flexible working are successful, a weakness of the current law cited in the located studies is that the success of requests depends on the employee’s bargaining position, meaning that flexible working is not available to less ‘valuable’ employees (Atkinson, 2016).

3.4.3 Employees making use of alternative ways of working are marginalized

We found evidence in the studies we located of the marginalisation of part-time and flexible workers – a phenomenon produced by a mismatch between these ways of working and organisational cultures which equate commitment with the ability to work long hours; and which assume that those who make use of these schemes do not want to develop their careers.

The traditional full-time career path is seen as the norm

In many workplaces, career paths and, by extension, access to senior grades are constructed with full-time workers as the norm, automatically excluding part-time workers (Tomlinson, 2007). One consequence of this is that part-time workers have difficulty in accessing training opportunities, something that is particularly important for women in occupations such as the medical profession, with strict career paths dependent on completion of certain training (Miller and Clark, 2008).

Part-time and flexible workers seen as difficult and time-consuming to manage

A second consequence is that part-time and flexible workers are seen as difficult and time- consuming to manage, making decision makers reluctant to recruit for, or grant request for part- time hours (Conley and Jenkins, 2011). Since, as is discussed further in Section 3.4.6, the negotiation of part-time work often leads to reduced hours without a consequent reduction in work-load, part-time working can lead to resentment among co-workers who may feel they have to pick up the slack (Teasdale, 2013; Jefferson and others, 2015). Analysis of national data found that, in 2011, more than a third of workers believe that working flexibly creates more work for others, with men almost twice as likely to believe this than women (Chung, 2017).

Alternative working is seen as signalling a lack of commitment

Part-time and flexible working conflict with organisational cultures that demand long hours as evidence of commitment (Corby and Stanworth, 2009) or that understand professionalism as ‘patients first, family second’ (Munro and others, 2015; Ozbilgin and others, 2011). For this reason, flexible and part-time ways of working can be associated with a lack of commitment, professionalism, or seriousness about one’s career (Callan, 2007; Kornberger and others, 2010; Tomlinson and Durbin, 2010a; Ozbilgin and others, 2011; Tomlinson and others, 2013; Munro and others, 2015; Gascoigne and Kelliher, 2018), something which has serious consequences for future progression.

3.4.3 Part-time working has a detrimental effect on career progression

As demonstrated in Section 2.1.3, there is clear evidence that part-time work shuts down wage progression, offering extremely poor return on experience, and restricted possibilities for promotion (Connolly and Gregory, 2008; Dex and Bukodi, 2012; Costa Dias and others, 2018; Warren and Lyonette, 2018). Analysing data from the BHPS and Understanding Society (Costa Dias and others, 2018) show that the effect of part-time experience on women’s wage progression is negligible. This effect is particularly large for more highly educated women, who would otherwise have seen the most progression had they continued in full-time work.

Using national data, Warren and Lyonette (2018) find that, despite increases of part-timers in senior level jobs, there is a part-time penalty in perceived possibilities for promotion. In 2012, almost a quarter of part-timers reported no chance of promotion (as against 17% of full-time workers). While lower part-time workers were the group most likely to have no chance of promotion at all, the part-time/ full-time promotion chances gap was widest among higher-level workers. This suggests that even if there are increasing levels of ‘retention’ of part-time workers in highly skilled jobs – valued employees who previously worked full-time before managing to negotiate a reduction in hours - their opportunities for further progression are severely constrained.

These quantitative findings were born out by qualitative, occupational level studies which show that there are few opportunities to progress while working part-time. There was evidence of women being trapped in lower-level jobs due to a lack of part-time jobs at relatively more senior levels, or negative perceptions of part-time workers. This included retail sales assistants who felt opportunities to progress were limited to full timers (Harris and others, 2007) and part-time catering assistants who could not progress to supervisor due to unwritten rules that supervisors had to work late and early shifts (Tomlinson, 2007). Similarly, Rainbird’s (2007) study of low-paid local authority workers found that the necessity for longer working hours to progress reinforced the trap of low-paid work.

These findings also extended to professional and managerial women (Callan, 2007; Lyonette and Crompton, 2008; Corby and Stanworth, 2009; McQuaid and others, 2012), male-dominated professions such as IT and construction (Worrall and others, 2010; Kirton and Robertson, 2018) and female- dominated professions such as nursing and teaching (Moreau and others, 2007; McQuaid and others, 2012). Women who successfully negotiated part-time hours sometimes found their duties changed and downgraded (Conley and Jenkins, 2011).

Two qualitative studies in accountancy (Lyonette and Crompton, 2008) and the hospitality industry (Tomlinson, 2006b) suggested that senior women who worked part-time tended to have worked with the company for a long time, developing their careers and reaching senior position before negotiating part-time hours. Among the retail workers studied by Harris and others, (2007) there were similar findings - the few reported instances of women in managerial part-time work were those who had worked full-time in those roles before negotiating a reduction in hours.

The finding of the studies cited in this section suggest that part-time work contributes far more to the retention of (some) women’s occupational standing than their progression. Female employees who are perceived as valuable, with well-developed careers and long full-time tenures, are well placed to negotiate part-time hours, and thus avoid occupational downgrading. However, there remains very little prospect of progression, and this is a major factor contributing to the gender gap in wage progression (Costa Dias and others, 2018).

Woolnough and Redshaw (2016), whose qualitative study involved women from 2 cohorts (those whose first child was born between 1990 and 1995 and those whose first child was born between 2011 and 2016), found continuity rather than change with regard to the low status of part-time workers and the career penalties associated with working part-time.

There is a lack of senior part-time working role models

A dearth of senior part-time working roles models means that women (or their managers) can often not envisage progressing on a part-time basis, especially in organisational cultures where overwork is the norm among senior staff (Harris and others, 2007; Moreau and others, 2007; Tomlinson and Durbin, 2010a; Gascoigne and Kelliher, 2018). For instance Harris and others, (2007) in their study of shop retail staff found that while part-time shop employees felt that career opportunities ‘aren’t really aimed at people like me who work part-time’ shop-management’s view was that part-time staff self-imposed limits by not applying for opportunities. This also extended to the senior women with more career capital studied by Gascoigne and Kelliher (2018): ‘At the time, there were no partners working anything less than 5 days. I was quite a new partner and it [part-time] just seemed impossible, and it was certainly not spoken about. There were no role models.’

3.4.5 There is mixed evidence that flexible working can help or hinder career progression

There is good evidence to suggest that access to flexible working – both flexitime and teleworking – results in fewer women dropping out of the labour market or substantially reducing their working hours post-childbirth (Chung and van der Horst, 2018). However, the evidence on whether flexible working helps or hinders career progression is more mixed.

Chung (2018) finds that 32% of UK workers believe that working flexibly decreases the chances of a promotion and this ‘flexibility stigma’ substantially decreases the likelihood of workers taking up flexitime or teleworking. Workers in long-hours cultures are also less likely to take up flexitime even when it is available to them.

However, Lyonette and Baldauf (2019) does highlight that implementing a broad range of family- friendly working policies can lead to positive outcomes for employers including cost savings, better productivity, improved recruitment and retention and reduced absenteeism. Organisations need to adopt family-friendly working policies that work for both an employee and employer – flexibility designed only to benefit the employer (for example, in stretched economic times) may not necessarily result in positive outcomes for employees (Lyonette and Baldauf, 2019).

No large-scale quantitative analysis of data from the UK currently exists (although data from other countries will be explored in Section 4.4.1) but the qualitative evidence we located suggests that women who work flexibly may suffer a career penalty, especially in environments with long hour working cultures (Kornberger and others, 2010; Tomlinson and others, 2013; Michielsens and others, 2014; Kirton and Robertson, 2018). Kornberger and others (2010) find that a much celebrated flexibility program in a top accounting firm actually reinforced gender inequality within the firm, as it provided a mechanism for differentiating those who were ‘committed’ to their careers from those who were not, with a lack of commitment to one’s career synonymous with a lack of commitment to the client. Since flexible working violated a code of professional identity which emphasises all-hours availability, it thus provided a rationalisation for the lack of progress of women who chose to make use of the program.

Similarly, Budjanovcanin (2018) found that female lawyers who successfully pushed for flexible working policies to be implemented suffered career consequences, including subsequently receiving less interesting work and having to deal with the frustration of line managers. Kirton and Robertson’s (2018) study of workers and managers in an IT firm found that managers believed that flexible policy use was a legitimate reasons for excluding women from career development and advancement. Michielsens and others (2014) research with multi-national service corporations publicly recognized as being ‘good practice examples’ in terms of diversity and flexibility confirms that flexible working arrangements are seen as acceptable only as long they do not impact on the client. More generally they found a prevailing culture of presenteeism and ‘client comes first’ constrained take up of flexible working arrangements.

3.4.6 Alternative ways of working can lead to the intensification and extensification of work

The literature suggests that in some circumstances the potential for part-time and flexible work to address issues of work-life balance is currently unmet. When policies are enacted without changes to underlying long hours organizational cultures workers making use of such arrangements can experience an intensification of work – an expectation that they complete the same workload in a reduced number of hours – and an extensification of work – the intrusion of work into non-working hours.

Flexible working

Some evidence links flexible working to an increase in work-life tension Using data from the 2011 WERS study Chung (2017) finds that telework increases the feeling that is difficult to fulfil commitments outside of work because of the demands of your job. This remains true even controlling for a range of other factors, including working hours. She also finds that those making use of schedule control experienced decreased work-life tensions, but that the effect size was very small.

Findings from the quantitative and qualitative data point to some reasons why in some cases flexible working might be linked to an increase in work-life conflict. Using national representative data Chung (2017) finds that flexible working can increase overtime hours in some circumstances (when it is being used for performance enhancing purposes) but doesn’t reduce overtime hours even when used for family-friendly purposes. European data suggests that flexible working of all kinds increases workers’ likelihood of worrying about work when not at work, and of working during free time to meet work demands. Kelliher and Anderson (2010)’s study of knowledge workers in 3 industries found that those who worked flexibly reported working more intensively when they were at work and working long hours.

Part-time working

The increasing number of part-time workers in senior positions suggests that women are increasingly able to negotiate part-time hours within their existing posts, thus avoiding loss of status (Bukodi and others, 2012; Warren and Lyonette, 2018). However, qualitative evidence suggests that, in some cases, this involves simply a reduction in hours (and pay) rather than any meaningful reduction in workload (Lewis, 2001; Lyonette and Crompton, 2008; Kelliher and Anderson, 2010; Tomlinson and Durbin, 2010a; Crompton and Lyonette, 2011; Lyonette, 2015; Jefferson and others, 2015; Gascoigne and Kelliher, 2018). Thus, in some workplaces part-time work leads to an intensification of work, a reduction of pay and, therefore, a limited contribution towards work-life balance.

Similarly, part-time workers often reported an expectation that they would be available at all times, even on their non-working days (Tomlinson and Durbin, 2010a; Crompton and Lyonette, 2011; Lyonette, 2015; Kirton and Robertson, 2018). In some cases this was framed in terms of reciprocity in return for the granting of part-time hours (Tomlinson and Durbin, 2010a).

3.4.7 Senior role models who model work-life balance and workplaces that support managers to support employees are facilitators

The evidence bases suggest that role models making use of alternative working time arrangements, supportive managers, and workplaces which work to support those managers are facilitators which ease some of the barriers associated with the conflict between work and external responsibilities.

Alternative working role models

Senior part-time role models, or figures who openly talk about balancing work and family life are cited as important facilitators for reducing the stigma associated with alternative ways of working (Durbin and Tomlinson, 2014; Lyonette, 2015; Mavin and others, 2015).

Supportive line managers

Supportive line managers are also key in ensuring that alternative ways of work genuinely reduce work-life conflict (Dean, 2007; Lyonette, 2015; Atkinson, 2016; Woolnough and Redshaw, 2016). Examples of good practice ‘progressive’ workplaces identified by Lyonette (2015) were those that specifically trained line managers to deal with part-time workers to ensure that part-time work led to a genuine rebalancing of responsibilities. This includes ensuring that adequate cover is provided so that workers are not always expected to be available. One private sector organisation held one-to-one coaching sessions for future female leaders with young children working part- time. The facilitator helped them to delegate responsibilities to rebalance their work-load. The coaches then trained their managers in turn, and acted as positive roles models for colleagues within their department.

3.4.8 Do women choose to opt out?

Despite the poor quality and reduced career prospects of part-time (and potentially flexible) work, women in the UK continue to work part-time in large numbers. Many women say they like working part-time and national data suggests that working part-time brings about significant improvement in well-being for women (Gash and others, 2012). This has led to a vexed conversation about the role that ‘choice’ plays in the lack of progression for women in part-time jobs, or more generally. The literature located for this study seemed to support a more nuanced conclusion recognizing the way that structural factors both shape and constrain women’s choices.

While there were examples in the literature of women placing immense value on certain aspects of their job, such as social usefulness or day-to-day interactions with pupils, which they felt would be sacrificed were they to progress to a higher level (Rainbird, 2007; Smith, 2011b), we found little evidence that women (or part-time workers) differ from men in terms of their organisational commitment, or that this differs by parental status. Dick and Metcalfe (2007), using whole population surveys of 2 county police forces in the UK, find that female officers are just as committed as male officers. Walsh’s (2012) survey of lawyers finds that women with the strongest aspirations to reach partnership were just as likely to have young children and aspire to a balanced lifestyle as those with weaker aspirations.

Studies of professional (Lyonette and Crompton, 2008) and managerial (Tomlinson and Durbin, 2010b) women suggested a frustration with the lack of opportunities for progression but, so long as the choice to pursue top jobs meant committing to a certain model of working, that choice was severely constrained. Similar views were found in studies of lower paid women. Many of the low- paid part-time female workers in the service and retail industries interviewed by Tomlinson (2007) and Harris et al (2007) expressed frustration with their promotion prospects and would have welcomed the opportunity to progress if they could combine progression with other demands, but the reality of their workplaces meant that they discounted the possibility in advance.

Since the full-time working model remains dominant, the marginalization of part-time work is largely accepted, and the marginalization of part-time workers, or those unable to display commitment via boundless availability is widely accepted and justified by reference to personal choice (Broadbridge, 2010; Ozbilgin and others, 2011). However, in this instance, ‘choice does not always equate to preference’ when the choices are constrained by external and immovable individual and organisational factors (Broadbridge, 2010; Gascoigne and others, 2015).

‘Either the mother commits to the working practices of dominant masculinity, that is a boundless time schedule, a suppressed personal life and a reduced investment in care, reinforcing what some mothers feel is a destructive work paradigm, or they must accept lower-status work. Most mothers who stay in work in the UK choose the latter option’ (Cahusac and Kanji, 2014, p. 67).

3.5 Key conclusions and relationship to broader labour market trends

Section 3.1 set out the way that, in the absence of (or despite) formal structures, career progression and rewards will be shaped by informal processes that disadvantage women, due to their reliance on processes of social cloning and male-dominated social networks. Section 3.2 explored how social isolation, sexism and sexual harassment and bullying negatively impact women’s career progression. Section 3.3 examined the way that those with caring responsibilities (still disproportionately carried out by women) are disadvantaged in cultures characterised by overwork and the necessity for geographic mobility or unpredictability. Section 3.4 found that, while alternative ways of working offer valuable ways to tackle gender inequalities by allowing women to maintain their labour market position after childbirth, their relationship to women’s career progression is more mixed. Part-time work is linked to truncated career progression, and women who take up flexible working policies may be seen as less committed, receiving fewer rewards even as they intensify their work.

Given the synthesis of evidence in this chapter, it seems difficult to conclude that a move towards gender equality in career progression is simply a matter of time. Eikhof (2012) argues that 3 current developments in the world of work actually pull in the opposite direction, entrenching the barriers set out in the previous section. The first is the increase in knowledge work where ‘…the crucial resource of production lies with the individual worker and actual work practices are (much) less dependent on the physical infrastructure of an organisation’. Consequently, progression is more likely to be dependent on informal processes, networks and self-promotion and there is an increase in temporary and insecure employment. This exacerbates the barriers discussed in Sections 3.1 and 3.3. We might also add that the same trend is likely to deepen the issues of sexism and sexual harassment discussed in Section 3.2 by raising the costs associated with the reporting of sexual harassment. Workers with precarious employment contracts and freelancers are already more likely to experience sexual harassment (Women and Equalities Committee, 2018) and employees have more to lose by making an accusation when work is precarious and job opportunities rely on maintaining valuable contacts. The second trend is the spread of ICT which has served to decouple work from a particular location and time. While this has the potential to facilitate a reconciliation between work and caring responsibilities, it can also lead to and enable a culture of boundless flexibility, exacerbating the barriers set out in Sections 3.3 and 3.4.6. When this is combined with the final trend, the spread of work-life balance policies and their uptake by (mainly) women, there is clear potential for a deepening of gender inequalities.

Gascoigne and others (2015) similarly argue that a variety of forces at work in the labour market, including technology and globalization, high-performance work practices and work intensification resulting from global competition, and the encouragement of identification with organisational values which stress competitive promotion and rewards system based on the number of hours worked, all contribute towards and sustain the phenomenon of ‘extreme jobs’ – those that require long hours and boundless availability. When work-life balance policies are bolted onto deeper structures which encourage long hours, those who make use of them will inevitably be marginalised so that ‘organizations thus invest in training female professionals, but retain taken- for-granted working practices, job designs and career frameworks which are incompatible with working less or with fulfilling caring responsibilities. Women’s talents are then lost to the workforce.’ (Gascoigne and others, 2015, p. 465)

4. Organisational policy interventions to promote women’s career progression

This section is based on evidence from 75 studies which were located in response to objectives 3 and 4 (see appendix for a full outline of the search strategy). There were no geographical limits set, but we limited the search to studies published after 2000. For studies investigating the effectiveness of an intervention we limited the search to systematic reviews, meta-analyses and quantitative studies which made use of an adequate counterfactual. Mixed methods and qualitative studies that explored the implementation of an intervention were also included.

What constitutes success?

The aim of this section is to critically review and synthesise the evidence on organisational policy interventions to improve women’s progression in the workplace, identifying which were proven to be, or likely to be successful. As already discussed, this report has taken a broad approach to progression and, therefore, to what constitutes success, and we found this reflected in the literature. Among the studies we located there was a wide variation in the operationalisation of practice effectiveness. Figure 10, adapted from Nishii and others (2017), sets out a process model showing the various stages at which outcome variables were measured, and providing a framework for showing how interventions might link to outcomes for individuals and organisations.

Figure 10: A process model for understanding the effectiveness of diversity practices, adapted from Nishii and others (2017)

Some studies simply investigated employees’ reactions to organisational polices, others linked espoused policies (collected by asking HR representatives whether or not certain policies were in place) directly to outcomes for the organisation (such as the percentage of women in management), without investigating or unpacking the mechanisms linking policy to outcomes.

Wherever possible we have relied on studies which make use of objective outcomes in the latter stages of this process model. We also indicate (where appropriate) whether studies are examining espoused or enacted practices. This distinction is important, given the gap between policies offered on paper and those implemented in reality.

Section 4.1 presents the results of interventions which attempt to increase the career capital and reduce the social isolation of women. Section 4.2 looks at attempts to change organisational norms and processes. Section 4.3 examines interventions to address bias in individuals and section 4.4 looks at interventions to address the conflict between current ways of working and caring responsibilities. Finally, Section 4.5 presents evidence on how organisational policies should be communicated.

4.1 Interventions to increase the career capital of underrepresented groups

This section examines interventions to increase the career capital of underrepresented groups. The evidence suggests mentoring, networking and support groups have at best modest effects on progression, and studies fail to consider barriers that might prevent women from taking part in these schemes.

4.1.1 Mentoring and networking can have positive effects, but these tend to be small

As outlined in Section 3.1, women tend to have less access to career enhancing networks and capital than men. Mentoring and networking programmes offer an opportunity for women to boost their career capital and form relationships that can help them in their career advancement (Kalev and others, 2006).

Many studies which assess the outcomes of taking part in mentoring are cross-sectional in nature, looking only at outcomes at a single point in time (Eby and others, 2008; Kashiwagi and others, 2013). This makes it impossible to identify whether taking part in the mentoring scheme caused any change in outcomes for individuals (Eby and others, 2008). There is also a risk of selection effects in this type of research, due to the lack of experimental methods – people generally choose whether to take part in mentoring, rather than being randomly assigned. Those who choose to be mentored are likely be more self-motivated than those who do not, which is also likely to be associated with better outcomes.

Overall, the evidence suggests that mentoring has a small positive effect on objective career outcomes, such as earnings and advancement within organisations. A meta-analysis by Allen and others (2004) found evidence that mentoring is related to higher compensation, faster salary growth and receiving more promotions, but that the size of the effect was small. Eby and others (2008) reach similar conclusions in their multi-disciplinary meta-analysis; however, they note that the majority of studies included in the meta-analysis are cross-sectional and non-experimental.

Neither of the above meta-analyses broke down their results by gender, so we cannot be sure if these effects hold true for both men and women. This is important in light of qualitative research which suggests a difference between men and women in terms of their experiences with mentorship. Based on interviews with 40 high-potential men and women taking part in a mentoring programme in large multi-national corporation, Ibarra and others (2010) make the distinction between mentorship and sponsorship. Sponsorship involves a mentor going beyond providing advice and feedback and advocating for the mentee. The authors find that the men in their sample were more likely to describe examples of sponsorship from their mentors and that men’s mentors had more organisational clout.

Other studies that we located confirmed a positive relationship between mentoring and productivity among academic medics (Stamm and Buddeberg-Fischer, 2011; Holliday and others, 2014). However, neither of the designs was able to rule out the possibility of selection bias, that is that more highly motivated individuals were both more likely to seek out mentorship and be more productive.

Kalev and others (2006) use a quasi-experimental methodology to analyse data from 708 large U.S. firms and find that a firm’s adoption of a formal networking of mentoring schemes is moderately related to subsequent increases in the percentage of women in senior management, with white women benefiting the most, followed by black women. However, some caution should be urged in interpreting these results for at least 2 reasons. Although the study design was able to control for time-invariant differences between firms that adopt mentoring schemes and those that do not, it was unable to fully control for time-varying differences between such firms. These time-varying differences (such as managers and executive’s preferences) are likely to be highly relevant to women’s careers (Ferguson, 2015). Secondly, this study relied on asking HR managers to state whether a mentoring scheme had been in place during a specified time - that is it focused on the ‘what’ of the practice, not it’s specific implementation. This is problematic in view of much evidence to suggest a gap between actual and espoused diversity practices (Nishii and others, 2017).

Using a randomised control design, Fielden and Woolnough (2014) find positive results for mental health nurses assigned to a 12 month multi-faceted career development and mentoring programme. The intervention included mentor matching with a senior figure within their own organisation. Nurses in the treatment group reported considerably more promotions over the course of the study than the control group. However, this was a small scale study (with a mentored group of just 27 and a control group of 27) and the findings may be specific to this profession. It is also not possible to disentangle the effects of the mentoring relationship from those of the training and development events.

Studies typically report a larger impact of mentoring on subjective career outcomes. Eby and others’s (2008) meta-analysis concludes that attitude variables are much more amenable to being improved by mentoring than are hard career outcomes. Kashiwagi and others’s (2013) systematic review of mentoring programs among physicians in academic medicine in the USA found that mentees typically report that mentoring contributes to their career development. Using a randomised control design, Fielden and Woolnough (2014) find mentored female mental health nurse participants report an increase in personal development.

4.1.2 Employee resource groups are likely more useful for improving subjective rather than objective outcomes

Employee Resources groups are employee led networks organised around a shared identity, such as a women’s network. Such groups might be thought to improve women’s career progression by tackling social isolation.

We were unable to locate strong evidence that Employee Resources Groups had a positive impact on women’s progression. Using a quasi-experimental methodology, Kalev and others (2006) find that employee resource groups are linked to a modest rise in the share of white women in the management but, for the reasons set out in the previous section, this finding should be approached with caution. Nishii and others (2017), in their recent literature review, conclude that Employee Resource groups may be less valuable in promoting career progression per se, and more important for developing a perception of an inclusive culture and subjective career outcomes. This fits with the findings of the UK based qualitative studies in which support networks of colleagues were cited as important for tackling social isolation in male-dominated cultures (Wright, 2016; Papafilippou and Bentley, 2017).

4.1.3 Interventions to increase career capital rely on women having the time to participate

Interventions that seek to develop the human capital of women are ‘bottom up’, in that they rely on women opting in and devoting time to participation (Laver and others, 2018). Inevitably, therefore, they will be harder to access for some women. As discussed in Section 3.1.3, time constraints for those with caring responsibilities are already a major barrier preventing women from taking part in career enhancing network and training opportunities. Studies that we located did not discuss the factors that might prevent program participation in the first place, but it seems reasonable to suggest that, in the absence of specific measures to make them accessible, such opportunities will inevitably only be taken up by a select group of women.

4.2 Interventions to change organisational norms and processes

There is good evidence that interventions to induce transparency and formalisation into organisational norms and processes such as clear salary standards and formal career planning (4.2.1) can be effective, so long as they are combined with mechanisms to ensure oversight and accountability. In the absence of accountability mechanisms, symbolically meritocratic procedures have a negative effect (4.2.2).

4.2.1 Interventions which increase transparency and formalisation into promotion processes are effective in some situations

The evidence presented in Section 3.1 suggests that informality and lack of transparency in appointment and promotions disadvantage women. Informality opens up the possibility of decisions on pay and promotion being made less by reference to performance than by a process of social cloning and male-dominated networks. Studies located for this report tested whether introducing formal human resource practices can, therefore, counter some of these effects. Such practices define and make clear objective standards and formal decision rules for progression.

There is good evidence that these formalising processes are effective and have positive impacts on objective career outcomes for women. However, the evidence also suggests that the same practices can be ineffective – in fact damaging – if they are not combined with processes that hold decision makers accountable for upholding these standards.

Clear salary standards

There is strong meta-analytic (Mazei and others, 2015) and real-world experimental evidence (Bowles and others, 2005; Leibbrandt and List, 2014) that providing clear information about salary ranges reduces gender differences in negotiation. Bowles and others (2005) found that men tend to achieve better economic outcomes than women when negotiating, but that these differences disappear when negotiators are provided with information about the bargaining range. They also found that female executives secured higher salaries than male executives when they were negotiating on behalf of a more junior colleague. Mazei and others’s (2015) meta-analysis of 51 studies reached a similar conclusion. Men achieve slightly better economic outcomes than women, but these differences are reduced when women have information about the bargaining range and negotiate on behalf of someone else. Leibbrandt and List’s (2014) randomized control trial with real job seekers altered whether job adverts explicitly stated that wages were negotiable. They found that, when there was no explicit statement, men were more likely to negotiate wages than women.

When the job description explicitly stated that wages were negotiable, there was no gender gap in propensity to negotiate.

Formal career planning, job ladders, and job postings

Formal career planning, job ladders and job postings make knowledge of progression opportunities and eligibility for those opportunities open to all. It is suggested that this will reduce some of the barriers outlined in Section 3.1 and produce a system of transparency that will enable others to hold managers’ accountable for their decisions (Dobbin and others, 2015).

Dobbin and others (2015) analyse longitudinal data from 816 large American establishments to estimate the effects of various formalizing personnel reforms on subsequent managerial diversity, over an average period of 10 years. They find that formal job ladders, which clearly set out eligibility and onward paths from entry level jobs, have positive effects on the percentage of ethnic minority men, and some ethnic minority women in management positions, while reducing the share of white men in management. Job postings, which require hiring managers to post all promotion and transfer opportunities to current employees, increase the share of white and Hispanic women in management. According to their data, which covers U.S. establishments from 1971 to 2002, about 80% of firms have job postings, but only about 20% have job ladders. This study makes uses of the same methodology as Kalev and others (2006) and so similar caution in interpreting the results is advised for the reasons set out in Section 4.1.1.

Similar findings are reported by Abendroth and others (2017). They use matched employee/employer data from over 5,000 workers in 94 large German organisations to investigate how workplace contexts affect the operation of formal personnel policies. They find that formal career planning procedures are related to a reduction of gender differences in earnings between men and women in the same workplaces and similar occupations, and with similar accumulated human capital and family situations, but only in certain contexts. For highly qualified female workers working in organisations with relatively high shares of women in management, formal career planning reduces their earning disadvantage compared to similar men. They suggest this is because‘formal career planning gives women in management the tools they need to promote highly qualified women within their organisation’. However, when female workers are paired with male supervisors, formalized career planning increases the earnings gaps for women compared to similarly situated men. They suggest this may be because formal career planning produces resistance in male supervisors who resent the control of their discretion.

Written performance evaluations

Performance evaluations are theorised to have positive effects on gender equality by providing transparent (and perhaps objective) criteria for ranking candidates for promotion. Like other formal personnel procedures, it is argued that providing such transparent standards will reduce the influence of non-performance related factors on promotion decisions, thereby eliminating one source of gender bias. Contrary to this expectation, the evidence located for this study suggested a negative relationship between written performance evaluations and objective career outcome for women, but only under certain circumstances.

Dobbin and others (2015) analyse longitudinal data from 816 large American establishments and estimate the effects of performance evaluations on subsequent managerial diversity over an average period of 10 years. They find that performance evaluations have negative effects for the subsequent share of white women in management (with no effects for any other under- represented group), although the caveats noted in Section 4.1.1 apply. Similarly, Abendroth and others (2017) use matched employee/employer data from over 5,000 workers in 94 large German organisations and find that written performance evaluations are associated with lower earnings for female workers with male supervisors than for comparable male workers with male supervisors.

Neither of the above studies was able to establish through what mechanism written performance evaluations were linked to worse career outcomes for women. However, the researchers suggest that personnel practices which attempt to control managerial discretion – such as when they are required to allocate rewards based on performance evaluations – create resistance. They suggest this resistance is linked to the subsequent negative career outcomes for women.

Two studies by Castilla (2015) and Joshi and others (2015) are able to shed some further light on these hypotheses. Joshi and others (2015) carry out a meta-analysis of studies examining performance evaluations and organisational rewards conducted in a variety of work settings. Examining 190 effect sizes and 474,732 individuals, they find no average sex difference in performance evaluations but large sex differences in reward allocations (including salary, bonuses and promotions). In fact, sex differences in rewards (advantaged towards men) were 14 times larger than sex differences in performance evaluations. This performance-reward gap was larger in highly prestigious and male-dominated occupations, but was reversed in industries with higher proportions of female executives. Using personnel data from 9,000 employees from a large U.S. based service organisation, Castilla (2008) found a similar gap in supposedly merit-based pay rewards. Women, ethnic minorities and non-U.S. born employees were rewarded with lower pay increases compared to comparable U.S. born white men with the same performance evaluation scores.

Taken together, the evidence presented in this section suggests that performance evaluations can lead to negative career outcomes for women, not because women receive worse performance evaluations, but because performance evaluations are not subsequently linked to career outcomes. This emphasises the importance of combining reforms which increase transparency and formalisation with those that introduce oversight accountability.

4.2.2 Interventions which increase organisational accountability are effective and make interventions to increase transparency and formalisation more effective

The evidence presented in the previous section suggests clearly that a reform’s effects on gender equality lies not in the ‘what’ of the practice, but in its specific implementation (Nishii and others, 2017). The results of Joshi and Castilla’s studies imply that the negative relationship identified in the previous section between performance evaluations and career outcomes for women may arise from weaknesses in implementation and accountability which lead to the decoupling of performance and rewards. This suggests that the same interventions may be successful in promoting women’s careers if they are combined with mechanisms to increase organisational oversight and accountability.

While Castilla (2008) reports on an observed gender gap between performance evaluations and merit based pay, Castilla (2015) reports on the subsequent effect of new organisational measures to introduce accountability and transparency into the performance-reward system. These included measures such as a performance-reward committee made up of senior personnel with the power to challenge and modify pay decisions, training for managers on the criteria for pay-reward decisions, and annual reporting of aggregated pay reward data to senior staff. After instigating these measures, the gender, race and nationality gap in pay rewards disappeared. Although Castilla’s study makes use of a simple before-and-after design supplementary analysis, including assurances that no other HR changes had been made in the performance-reward system during that period, providing additional evidence that the observed effect can be attributed to the intervention, rather than down to other factors (that may have improved the conditions for underrepresented groups during that time frame) would have been beneficial.

Dobbin and others (2015) similarly find an interaction effect between performance evaluations and organisational oversight – when organisations have oversight mechanisms the negative effects of performance evaluations on women’s representation in management disappear, and they are in fact linked to a subsequent increase in the proportion of black and Hispanic women in management. Similarly, they find that organisational oversight improves the effectiveness of job postings and job ladders. They also find a main positive effect of diversity managers on the prevalence of all under-represented groups in management. Kalev and others (2006) report similar findings for structures that embed accountability (diversity managers, committees and taskforces).

In the absence of strong accountability measures, symbolically meritocratic personnel procedures can have a negative effect, as evinced by the decoupling of rewards from performance found by Castilla (2015) and Joshi and others (2015). Experimental evidence, using participants with managerial experience, supports the idea of a ‘paradox of meritocracy’. Using a randomized control trial methodology, Castilla and Benard (2010) asked their experimental subjects – all of whom had previous managerial experience - to make bonus, promotion and termination decisions for several employee profiles. The researchers manipulated the gender of the employee, and whether the firm’s core values emphasized meritocracy. They find that when an organisation proclaims itself to be meritocratic, managers are more likely to award larger monetary rewards to a male employee over an equivalent female employee. They suggest this is because the managers have internalised the idea that they are unbiased, leaving them free to act on their biases without fear of scrutiny.

The lessons for organisations wanting to increase gender equality with regard to pay and progression is to ensure that measures designed to increase meritocracy are coupled with mechanisms to oversee their implementation and link them to relevant outcomes. In the absence of these mechanisms the evidence base suggests that the symbolic adoption of policies may do more harm than good.

The importance of senior buy in

Lessons on the specific forms of the oversight and accountability can be learnt both from Castilla (2015) and from early qualitative evaluations of the Athena SWAN initiative – a program designed to increase gender equality in higher education which highlights some of the issues it has encountered. Castilla (2015) successfully reduced gender and racial disparities in performance related pay by introducing accountability and transparency into the company’s performance reward system. In Castilla’s intervention this accountability was achieved by creating a performance-reward committee made up of senior figures, supported by top executives and endowed with the power to modify decisions.

This contrasts with early qualitative evaluations of the Athena SWAN Charter, a system of 3 levels of awards, gold, silver and bronze, to which academic institutions and departments can apply. Receipt of a bronze award is dependent on a self-assessment of gender equality, a corresponding action plan and an organisational structure to implement the proposed action.

Further silver and gold awards rely on successful implementation of these plans. The evaluations found that, while staff were generally positive about the Athena SWAN program, they suggested that its impact was undermined by wider institutional practices which were beyond the programme’s remit, and by the fact that much of the burden of the Athena SWAN work fell on women (Caffrey and others, 2016; Ovseiko and others, 2017).

Thus, the evidence suggests that any attempt to induce oversight and transparency can only be achieved with senior buy-in, or alongside a remit to modify existing institutional practices, and that interventions which rely on the additional labour of the groups they seek to help may inadvertently reinforce gender inequality.

Joint candidate evaluation

One additional mechanism for bridging the gap between performance evaluations and pay or promotion decisions identified in the previous section is joint candidate evaluation. Using experimental methods with undergraduate subjects, Bohnet and others (2016) test whether evaluating candidates jointly rather than separately can increase the role that performance scores play in candidate evaluation (and decrease the role of gender stereotypes). They find that when evaluating candidates singly, decision makers were much more likely to recommend a male candidate for a mathematics task over a female candidate with the same past performance. The opposite effect was observed for candidates for a verbal task, with female candidates advantaged over male. When evaluators were tasked with evaluating candidates jointly, decisions were much more strongly linked to past performance than gender. They suggest that joint evaluation, currently common for hiring decisions but less for promotion or job assignment decisions, would help eliminate the role that non-performance related criteria play in decision making.

4.3 Interventions to address biased organisational cultures

The evidence suggests that these are not effective. There is no evidence linking diversity training to long-term changes in attitudes (4.3.1) and the limited evidence we have on gender diversity on selection committees suggest they do not lead to better outcomes for women (4.3.2).

4.3.1 There is no evidence that diversity training has a long-term effect on attitudes

Diversity training is designed to reduce levels of bias and discrimination among employees and, thereby, limit their influence on women’s careers. The evidence located from this study suggests that there is very limited reason to think that diversity training is effective. In their meta-analysis of 260 studies with almost 30,000 adult participants, Bezrukova and others (2016) examine the effects of diversity training on cognitive learning – the extent to which a trainee acquires knowledge about diversity issues – and attitudinal learning – changes in the trainee’s attitudes on diversity. They found that diversity training can have an effect on cognitive learning, and one which persists over time. However, they found no evidence that diversity training had any long-term effect on attitudes. Notably, they also found that training’s effects on cognitive learning were higher in educational settings, which they suggest is because, in these settings, training is not seen as something that ‘takes time away from work’. This finding, therefore, cautions pessimism in replicating the positive effects of diversity training on cognitive attitudes when the training is conducted in organisational environments.

The findings of this research are supported by evidence from the Kalev and others (2006) study of data from over 700 large US firms. They found that diversity training for managers was one of the most common diversity practices in the firms they studied, and was mandatory in 70 percent of firms that offered it and yet, after 5 years, it was associated with no statistically significant change in the percentage of white women in management, and a drop in the percentage of black and Asian women (Kalev and others, 2006; Dobbin and Kalev, 2016). They suggest this is because, far from removing bias, mandatory diversity training can have the opposite effect, creating anger and resistance among those forced to attend.

4.3.2 We have limited evidence on gender diversity on selection committees

We uncovered some limited evidence related to the effectives on gender diversity on selection committees on promoting female progression. Manuel Mendoza-Nunez and others, (2013) exploit a natural experiment based on the random assignment of candidates for the Spanish judiciary to evaluation committees with varying gender compositions. Based on a sample of 150,000 candidates between 1987 and 2007, they found that female candidates were less likely to be selected when the selection committee had a greater share of female members. By contrast, male candidates were more likely to be selected when evaluated by a committee with at least one female evaluator.

4.4 Interventions to address conflict between work and external responsibilities

International evidence suggests that a flexibility stigma is likely to affect workers making use of flexible working arrangements, with women facing harsher penalties in terms of future career progression. This may be because men and women use flexible work for different reasons, or because employers attribute different reasons to them. There is evidence linking a partner’s use of flexible working policies to increased wages for women (4.4.1). There is a dearth of evidence looking at how to reduce the negative career consequences associated with part-time or flexible working time arrangements, or how to tackle long-hours organisational cultures. However, there is some evidence to suggest that providing training for employees and supervisors to support the transition to flexible working can improve subjective career outcomes and reduce work-life tensions (4.4.2). Finally, some structural characteristics of occupations are associated with a reduced pay gap between mothers and childless women, including greater autonomy, and less teamwork and competitive pressure. This provides a potential basis for future research into ways to construct occupations so that they facilitate greater reconciliation of work and external responsibilities (4.4.3).

4.4.1 The evidence on flexible working suggests it has the potential to deepen gender differences in career progression

There is good evidence to suggest that access to flexible working – both flexitime and teleworking – results in fewer women dropping out of the labour market or substantially reducing their working hours post-childbirth (Chung and van der Horst, 2018). The evidence on whether flexible working helps or hinders career progression is more mixed.

There are 2 different theories on the effect of flexible working on career progression which pull in opposite directions (Langner, 2018). The evidence presented in Section 3.3.2 suggests that presenteeism and boundless availability are important signs of commitment. Thus, according to this ‘commitment theory’, requests for flexible hours, or to work from home, may be seen as signalling lower levels of commitment and be associated with reduced career progression (Glass, 2004).

Alternatively, if flexible working allows workers to meet their family commitments, this could reduce work-life conflict and increase productivity, resulting in improved career progression. This wage growth should be particularly noticeable for workers who would otherwise experience greatest work life stress (for example, those with caring responsibilities) (ibid).

A small group of studies have examined the effect of working flexibly on individual wage growth. Three cross-sectional studies (Gariety and Shaffer, 2001; McCrate, 2002; Winder, 2009) all find a positive relationship between the use of schedule flexibility and wages. However, selection biases hinder causal conclusions, since flexible working policies are more likely to be available to high- performing individuals within high-performing organisations, and these analyses cannot control for differences between individuals who make use of flexible working policies and those that do not.

We located several studies making use of longitudinal data which were able to avoid some issues of selection bias by using fixed effect models to control for time invariant differences between individuals.

Telecommuting during overtime hours is linked to a negative effect on wage growth

Using fixed effects regression of U.S. national data from the 1989 to 2008 National Longitudinal Study of Youth, Glass and Noonan (2016) find that telecommuting is linked to longer overtime working hours. They found no effect of telecommuting during normal hours on wages, but a negative effect of telecommuting during overtime on wages. Overtime hours worked on-site had almost twice as much impact on earnings as overtime hours worked off site. They suggest that this is because visible overtime worked on site, and its power as an indicator of commitment, is a more important factor in wage increases than the work completed in those hours in and of itself. While they found that hours worked at home show similar earnings return for women and men, their analysis also finds that mothers face more severe pay penalties for reducing their hours on- site below the traditional 40 hours per week. They suggest this is because managers hold mothers to a higher standard than other workers, and they pay a harsher penalty when their behaviour deviates from the ‘ideal worker’ norm.

Schedule flexibility leads to increased work hours for men and women, but wage increases only for men

Lott and Chung (2016) and Langner (2018) examine the effect of schedule flexibility on wage growth using national data from West Germany. Lott and Chung (2016) find that use of schedule control is associated with increases in overtime for both men and women, but only for men does this increased overtime lead to income gains. This suggests that flexible working may have a negative effect on organisational gender equality. Using similar data, Langner (2018) finds that, upon entering flexible work, men experience an immediate positive effect on their wages. For women, this effect was limited to those who worked flexibly for at least 4 years. Her analysis also showed that while fathers’ wages increased upon entering flexible work, mothers did not benefit significantly. Glass (2004), whose small sample consists only of Midwestern American mothers, also finds mild negative or no significant effects of schedule flexibility on wage growth.

Any stigma attached to flexible working may be down to reasons for its use, or reasons that others attribute for its use

Taken together, the evidence presented in the previous section suggests that both the commitment and signalling hypotheses are correct. Flexible working is linked to increases in worker intensity, perhaps because flexible workers perceive a debt to their employers, and seek to overcome stigma associated with its use (Chung, 2017). Yet all the studies located here seem to suggest that women do not benefit from increased intensity in this to the same extent that men do.

There are at least 2 potential explanations for this discrepancy. Firstly, men and women may have different motivations when accessing flexible work. Lott and Chung (2016) find that, when gaining schedule control, West German mothers increase their overtime hours to an even larger extent than women in general without receiving the same income gains as other workers. They suggest that mothers are, therefore, willing to forsake extra income for the chance to work flexibly. Men, by contrast, may make use of schedule control for career related reasons, rather than as a means of reconciling work and home. On the other hand, the observed gender discrepancy may be related less to the reasons that workers have for working flexibly, and more for the reasons that others attribute to them.

Experimental evidence from Almer et al (2004) indicates the existence of a flexibility stigma. The researchers found that respondents working for 3 US accounting firms viewed flexible workers as less committed and less likely to advance than childless individuals and individuals with children who were not working flexibly.

Other research suggests that whether flexible working is seen as signalling low commitment depends not only on whether it is, in fact, used for productivity purposes or to gain work-life balance, but also on what motivation managers attribute to workers. Using a randomized control trial design, Leslie and others(2012) find experimental participants saw use of flexible work practices as signalling low organisational commitment when they attribute use to a desire to balance personal life responsibilities, and as signalling high organisational commitment when they attribute use to a desire to increase productivity. The same paper reports similar findings using a field study with survey data from employees and managers from a Fortune 500 organisation.

These attributions are likely to be gendered, since employers are more likely to perceive that women are making use of flexible working policies for family reasons. Brescoll and others (2013) report on a randomized control trial in which respondents with previous managerial experience were randomly assigned to read one of 8 vignettes about an employee requesting a compressed work schedule. The vignettes differed across the status of the respondent (high vs low), the sex, and the reason for request (professional development vs family related). They found that the managers were more likely to grant requests from high status men seeking flexibility for professional development reasons than for equivalent women, while both men and women seeking flexibility for childcare responsibilities were equally as likely to have their request granted. For female employees, neither their status nor reasons for requesting flexibility had a significant effect on the likelihood of their request being granted. The researchers suggest that this is because the association of motherhood with women is so strong that, even when women are seeking flexibility for professional development reasons, it is not perceived as such by employers.

In summary, this evidence suggests that flexible working’s effect on career progression may have less to do with the reasons why we use it, than the attributions that others make towards us.

Flexible working is more likely to be seen as signalling low commitment when managers perceive workers are using it to balance work and family life, as opposed to for career development reasons. These attributions are likely to be gendered since employers are more likely to perceive that women are making use of flexible working policies for family reasons.

Partner’s flexibility is linked to increased wages for mothers

Langner (2018), in her study of West German workers, also examines the relationship between an individual’s wage growth and their (employed heterosexual) partner’s move into schedule flexibility. She found that partner’s flexibility showed a positive effect on both men and women’s wages, with the effect most pronounced for mothers. This suggests that ‘flexible working may be a man’s alternative to part-time work in order to support his spouse’s career’ and demonstrates the potential that flexible working has to decrease gender inequalities for women by increasing household level flexibility.

4.4.2 Interventions to address stigma associated with alternative models of working

There is little evidence looking at how to tackle long-hours organisational cultures.

Despite evidence presented in Sections 3.3 and 3.4 of a conflict between long-hours cultures and caring responsibilities and a stigma attached to alternative models of working, the review located few studies analysing conditions under which such stigma might be addressed, or long-hours cultures can be tackled.

One exception was a 3-year cluster-randomized control trial carried out by Grisso and others (2017). The trial evaluated the effect of an intervention to improve academic success among women assistant professors, while driving broader cultural change within a research-intensive medical school. 27 departments were randomly assigned to intervention or control groups. The 3-tiered intervention consisted of professional development skills offered to female faculty, faculty led task-forces to change departments and local environment, and close engagement with institutional leaders. The researchers found that while, overall, there was no difference in measurable career outcomes between the treatment and control departments, average hours worked per week declined significantly more for faculty in the intervention group compared to the control group, suggesting that the intervention may have facilitated a reduction in work intensity.

Another promising approach comes from a change initiative described by Perlow and Kelly (2014) and trialled within several U.S. organisations. The Results Only Work Environment (ROWE) was designed to change the structure and culture of organisations, moving them away from a culture of presenteeism and towards a focus on results. It was designed by the implementing line managers, and was sold by senior managers as being a ‘smart business move’. ROWE was designed to allow employees to ‘do whatever they want, whenever they want, as long as the work gets done’. It was implemented at an American Retail company, Best Buy Co. Inc. via a series of 3 participatory sessions between facilitators and employees including brainstorms around how to work effectively at different times, how to cut back on meetings and ‘refocus interactions on results and working effectively together, rather than on time.’ After 3 sessions ROWE principles and practices are implemented by teams. Perlow and Kelly (2014) report a number of positive effects of ROWE departments in Best Buy compared to traditional departments on employee and organisational outcomes, including reduced turnover and decreased work-family conflict.

Some evidence suggests that training for flexible employees and their managers can reduce work-life conflict among flexible workers

While the literature offers little guidance on how best to formulate alternative working time policies in order to improve career outcomes, it does provide some guidance on how such policies can best be made to reduce work-life conflict. A systematic review of studies to promote work-life reconciliation of employees (where the outcome of interest was work-life conflict) recommended that interventions focus not just altering working time, but also on providing personal resources such as training to supervisors and employees to support any changes (Ropponen and others, 2016).

Similarly a study by Moen and others (2016) offers some further guidance on how to support work-life balance among flexible workers. The study organized 867 employees from 2 locations of a US company into 56 study groups. Those groups were then randomized to usual workplace practice or to the STAR program – developed to promote workers’ subjective well-being by increasing control over their schedule, supervisor support for family and personal life and an orientation from long hours away to results. It consisted of 8 hours’ training for the worker, the flexibility to work from home and 4 hours’ training for the supervisors to support the worker’s family and personal life and professional development. The results of the study suggested that, over one year, the intervention increased well-being among IT workers and managers, compared to the control group, but these effects were more pronounced for female employees.

This evidence supports the findings from the qualitative evidence presented in Section 3.4.7 which suggests that supportive supervisors, and workplaces which provide training to those supervisors to deal with flexible workers, are key in ensuring that alternative working time policies support work-life balance.

4.4.3 Structural characteristics of occupations

While not strictly an intervention, a study by Yua and Kuo (2017) provides useful insight into the relationship between characteristics of occupations and the wage gap between mothers and non- mothers in the USA. Using fixed effects models to control for time-invariant personal characteristics, they examine data from the nationally representative National Longitudinal Survey of Youth from 1997 – 2013/4 and match it with information on the structural characteristics of occupations from the O*NET database (which compiles information on occupational attributes and requirements from surveys of job incumbents and experts.) They find that the wage reduction women experience for each child is reduced in occupations with greater autonomy, lower teamwork requirements and less competitive pressures, likely because work-life conflict is reduced in these occupations. This suggests that there is a limitation to purely employee-based policies, but also suggests that greater attention will need to be paid to the structural characteristics of occupations which facilitate the reconciliation of work and caring responsibilities, as well as the potential for those characteristics and working models to be incorporated into other areas.

4.5 Framing of Interventions

Finally, the literature offers some guidance on how efforts to promote women’s progression should be communicated. A systematic review of studies to promote work-life reconciliation of employees (where the outcome of interest was work-life conflict) recommended explicitly stating that work-life reconciliation (as opposed to productivity) was the aim and targeting the intervention towards both genders to avoid work-life reconciliation as being something that is seen simply as a problem for women. For this reason ‘work-life reconciliation’ is preferred as a term instead of ‘family-friendly’ to describe such policies (Ropponen and others, 2016).

This finding is echoed by a randomized control trial carried out with American undergraduates by (Cundiff and others, 2018). Across 2 studies they tested how students’ perceived sense of whether they would fit in at a company, and anticipation of fair treatment were affected by different diversity messages within a fictional recruitment brochure. Women and men both reported a reduced sense of fit and were more concerned with being treated unfairly by the company when the initiatives were described as targeting women employees, rather than being framed as all inclusive.

5. Key points and implications

The evidence base on women’s progression in the workplace is drawn from a heterogeneous range of studies. While there are some under-studied groups the barriers to women’s progression appear relatively well understood. By contrast, less is known about interventions to overcome these barriers.

5.1 Evidence gaps

There was little comparable information on how career progression varied for women across different sectors and what factors (such as lack of retention or slow progression) contributed to this. More evidence of this sort would help to identify structural factors of work organisation associated with women’s progression, as well as understanding in which sectors and occupations efforts should be most focused.

Much of the literature on barriers to women’s progression focused on women in more highly skilled occupations and on permanent contracts. More information on the challenges facing lower-income women and those in more casualised forms of employment would have been helpful.

Intersectional effects are under-studied, so there is little information on how (the barriers to) career progression differ across women from different social class or ethnic backgrounds.

In general, there was a mismatch between the focus of the intervention literature, the bulk of which focused on interventions to improve the human capital of women and change organisational norms and processes, and the barriers literature, in which long-hours cultures and stigma attached to alternative ways of working were the most commonly cited barriers.

There is little evidence on how the right to request flexible working is being implemented and, especially, the experience of lower-income women in weaker ‘bargaining’ positions.

There is very little evidence on how to reduce the negative career consequences associated with part-time or flexible working, or on how to change long-hours cultures, and in general a lack of ‘top-down’ interventions’ which start with changes in the practices of higher management.

We located no research on how the provision of childcare in a workplace setting might be related to women’s progression.

Many of the studies investigating interventions to overcome barriers to women’s progression do not have designs which allow firm conclusions about causality. Studies which are better placed to investigate causality are often lab-based RCTs leading to questions about their transferability into the workplace context. Some firm-based RCTs or well-designed before and after field studies provide good quality evidence but there is a need for more.

5.2 The gender divide in workplace progression

Because inter-occupational hierarchies are not comparable across occupations, career progression is difficult to capture quantitatively in a way that enables broad comparison. We made use of 2 different ways of capturing progression: progression to higher pay and progression up a ranked occupational hierarchy.

The GPG is minimal on labour market entry and widens significantly from the late 20s and early 30s. Men’s wages grow rapidly at this point, while women’s plateau. Controlling for industry and occupation only explains a small amount of the difference in wage growth, while a much larger proportion is explained by the move into part-time work that often occurs around childbirth. Part- time work offers very little return on experience in terms of wage progression. While gender differences in the accumulation of experience and other forms of human capital are able to account for some of the gender gap in wage progression, a substantial amount remains unexplained.

Using a different measure of career progression – one which makes use of a vertical occupational scale based on the average hourly earnings in each occupation - gives similar results. Men progress steadily upwards throughout their 20s and 30s, whereas women’s trajectory is flatter, meaning they do not reach the same levels of occupational attainment. Part-time work is an important explanation of this difference.

Another key difference is that women who enter the labour market in low-quality jobs experience ‘sticky floors’ – rarely progressing out of these jobs. By contrast, such jobs are ‘springboards’ for men into higher occupational positions. This springboard vs sticky floor dichotomy has worsened over time.

5.3 What are the barriers to women’s progression?

The evidence on barriers to women’s progression is largely qualitative and helps to unpack some of the drivers of the gender differences in progression established by the quantitative work set out in the previous section, providing an insight into the processes which create and sustain the observed differences in progression. While it is difficult for qualitative studies to definitively prove cause and effect, or to generalise from findings observed in one firm or sector to another, it provides an overview of the sorts of barriers that employers might encounter.

Sections 3.1 and 3.2 provide some insight into some of the processes underpinning the residual divide in progression once male-female differences in labour market history are taken into account. Women are disadvantaged by organisations which lack formality and transparency with regards to the criteria for recruitment, promotion or pay negotiation. In the absence of such criteria, decisions on pay and promotion are more likely to be made via a process of social cloning, whereby those in a position of power champion those like them. In both informal and formal environments, progression often happens via male dominated social networks which women find difficult to access. The presence of an organisational sponsor was identified as a key facilitator to overcome some of these barriers. Additional barriers for women working in male- dominated organisational cultures include isolation and lack of role models and sexism and sexual harassment. Support networks are helpful for easing isolation.

As presaged by findings looking at the impact of parenthood on career progression, many of our studies cited the conflict between current models of working and caring responsibilities as the most significant barrier to women’s progression. The evidence we located suggested that there was continuity (rather than change) with regard to studies on organisational culture around working time practices and their relationship to perceived commitment and progression. This is not just an issue for professional women – the expectation of constant availability and limitless flexibility was also a barrier to progression for low income women. The high cost of childcare in the UK limits the ability to outsource caring responsibilities.

Access to alternative working time policies is still limited. There continues to be a shortage of quality part-time work and, while there are increasing number of part-timers in senior positions, lower level part-time workers remain crowded into jobs that offer poor wages and restricted opportunities for progression. Similarly, there is some evidence that flexible working policies are predominately offered to high-skilled workers. Under the current law, the success of flexible working requests are dependent on an employee’s bargaining positions, meaning they are not available to less ‘valuable’ employees.

Alternative working time policies without culture change are not enough to support women’s progression – and can reinforce gender differences. When they are able to access them alternative ways of working, such as part-time and flexible work, have clear benefits in terms of retaining women within the workforce. However, without changes to the underlying organisational cultures, and while such working practices are taken up only by women, they risk embedding gender inequality due to the negative effects these practices have on career progression in contexts where full-time workers are the norm, and part-time and flexible working is seen as signalling a lack of commitment.

Part-time experience offers very little return on wage growth. While there is evidence of an increase in the proportion of female senior part-timers, the evidence suggests this is largely as a result of already senior women negotiating a reduction in hours. Meanwhile, female part-timers in lower occupational jobs receive low wages with limited opportunities for progression. Evidence from the UK also suggests the flexible workers may suffer negative career consequences and a flexibility stigma.

A failure to properly adapt organisational cultures and workplace responsibilities to alternative working time policies means that part-time and flexible workers are often still expected to take on the same responsibilities in reduced hours (and with lower pay), or to be available outside of their contracted hours. Implementation of these policies without reorganisation of responsibilities or training for line managers can mean their management is left to co-workers at the everyday level, which can lead to resentment. Organisations which show ongoing top-down commitment to supporting part-time and flexible workers mitigate some of these barriers.

5.4 What organisational policy interventions can support women’s progression?

Interventions, such as mentoring and employee support groups which attempt to increase the career capital and reduce the social isolation of women, can have positive effects on progression, but the evidence suggests they have greater impact on subjective outcomes, such as satisfaction.

Studies also fail to consider barriers, such as time constraints, that might prevent women from taking part in these schemes.

There is good evidence that interventions to induce transparency and formalisation into organisational norms and processes such as clear salary standards and formal career planning can be effective, so long as they are combined with mechanisms to ensure oversight and accountability. In the absence of accountability mechanisms, symbolically meritocratic procedures can have a negative effect.

Less effective are interventions to try to directly reduce bias. There is no evidence linking diversity training to long-term changes in attitudes and the limited evidence we have on gender diversity on selection committees suggests they do not lead to better outcomes for women.

While the provision of alternative working time arrangements – such as flexible working – can help women to maintain their labour market position following the transition to parenthood, the evidence on whether these arrangements help career progression is lacking. There are clear career penalties associated with part-time work and emerging evidence suggests that flexible workers may face a similar penalty. Some research documents the existence of a flexibility stigma, which is likely to affect workers making use of flexible working arrangements, with women facing harsher penalties in terms of future career progression. This may be because men and women use flexible work for different reasons, or because employers attribute different reasons to them. There is a dearth of evidence looking at how to reduce the negative career consequences associated with part-time or flexible working time arrangements, or how to tackle long-hours organisational cultures.

By contrast, there is some international evidence linking a partner’s use of flexible working policies to increased wages for women. This suggests that focusing on couple level flexibility is key to overcoming gender differences in progression.

There is some evidence to suggest that providing training for employees and supervisors to support the transition to flexible working can improve subjective career outcomes and reduce work-life tensions.

Finally, some evidence from the USA suggests that the wage gap between mothers and non- mothers is reduced in occupations with less teamwork, less competitive pressure and greater autonomy.

5.5 Conclusions and Implications for policy

The most commonly cited barriers to women’s progression in the workplace continue to arise from a conflict between current ways of organising work and caring responsibilities. Long-hours cultures and expectations of constant availability are an enduring feature of modern workplaces.

Part-time work offers a currently imperfect solution, but one that comes at the cost of greatly reduced career progression. Access to flexible working is vital for maintaining women’s labour market position, but is no panacea insofar as it may also lead to stigmatisation and reduced career progression, while also contributing towards the intensification and extensification of work.

Practices such as these that are ‘bolted on’ are likely to be less effective in promoting women’s progression than attempts to reform underlying organisational cultures and working structures. Couple level flexibility has positive effects on women’s career progression, suggesting that men’s long working hours are as much of a challenge to gender equality as women’s.

While current developments in the world of work suggest the potential for increased flexibility, they also contain hidden gender barriers if this flexibility increases the expectation of constant availability and the prevalence of casualised and precarious forms of employment, while work-life balance policies divert women onto the ‘mommy track’.

This suggests that the policy focus should be on reforming organisational cultures away from norms of overwork and supporting the construction of ‘non-extreme’ jobs, which do not require long hours and constant availability as a proxy for commitment. The aim should be for (employee driven) flexibility to be seen as the norm, rather than the exception, and as a way to support men to engage in care work, as much as it is for women to increase their labour market participation. Alongside policies to support work-life balance should come efforts to reform the deeper structures and workplace processes which encourage excessively long hours. Ensuring that full- time employees are not required to work long hours will be just as important as reducing the stigma associated with part-time or flexible employees.

At the same time, the stigma associated with part-time and flexible work is likely to persist so long as these ways of working continue to be associated with women and men who work long hours. Policies that encourage and enable men to take on greater childcare responsibilities are thus essential if women are to be able to maximise their potential.

There is scepticism within the literature about the potential for the right to request flexible working, as currently constructed, to challenge these underlying cultures. Instead, there is potential for these policies to perpetuate gendered divisions of care so long as flexible working is perceived as a ‘perk’ which exists primarily for mothers, and which is only available to more valuable employees.

The evidence that occupational class disparities among women continue to widen is worrying and it is, therefore, essential that policy focusing on the creation of ‘good jobs’ be informed by an understanding of the intersections of gender and occupational class (Warren and Lyonette, 2018).

The other main barrier to women’s progression comes from organisational norms and processes that allow gender bias to creep into decision making. When there is a lack of clarity around the standards for recruitment, promotion or pay negotiation decisions are more likely to be made in ways that disadvantage women, whether because people in power seek those who are like them or because who you know is more important than what you know. While the processes producing these gender differences are relatively well understood, less is known about how to alter them, and there is some difficulty in drawing causal inferences from existing studies due to variations in the operationalisation of practices, studies which look at espoused as opposed to actual practices, and in designs which fail to account for the fact that firms which implement these policies are likely to be systematically different from those that do not, and these differences are likely to correlate with other factors affecting women’s progression.

Facilitating work and caring responsibilities - recommendations for organisations

The long-term goals for organisations should be away from a culture that sees full-time long-hours work as the norm, and employees making use of alternative working time practices as lacking commitment. Organisational cultural change away from this expectation will be hard to achieve but it seems clear that, in the absence of this cultural shift, women will continue to be disadvantaged so long as they take disproportionate responsibility for unpaid caring responsibilities.

In the short term, the following steps seem promising intermediate steps, given the evidence summarised in this report:

1. Advertise all jobs as being available flexibly and part-time unless there is good reason not to.

2. Work-life balance should be modelled from the top of the organisation. Promote and champion senior figures who work part-time and flexibly.

3. Communication campaigns that promote the idea that work-life policies are for everyone, not just for parents, and which emphasise the positive effects of flexibility on productivity might also help to challenge flexibility stigma.

4. Managers are key in ensuring access to flexible working policies and quality part-time work. When managers are inexperienced in dealing with these workers, alternative working time policies can lead to more work being done in less time, or the intrusion of work into other areas of life. Employers should provide training to part-time and flexible employees and to their line managers to ensure a proper rebalancing of responsibility.

5. Employers should be pro-active in reducing or reshaping workload to match a part-time or flexible working commitment, exploring different approaches to work and reviewing cultural processes and values that pull in the opposite direction. Similarly, employers should pro- actively review whether current models of career progression or access to development opportunities exclude part-time or flexible workers.

6. Monitor uptake of flexible working policies to ensure that employees making use of them are not overlooked for promotion.

7. Create clear criteria for progression and promotion focused on (realistically achievable) outputs rather than hours.

Reforming organisational norms and processes - recommendations for employees

1. Review progression and promotion criteria. Carry out audits of previous promotion decisions to understand which criteria, qualification or experience predicted performance in the subsequent job, and which are just there because they always have been. Consider whether criteria that are more typically associated with women are (unduly) overlooked.

2. Ensure that there are clear standards for promotion and advancement, that these are clearly communicated and that they do not depend on long hours as evidence of commitment.

3. Ensure that networking and training opportunities are held in normal working hours.

4. Create mechanisms for organisational oversight of formalised processes such as performance evaluations via diversity committees or joint evaluation. Ensure these are made up of senior figures and have power to change decisions.

Recommendations for government

The long-term aim for government should be to understand how to support the construction of jobs where flexible and part-time work are the norm and away from a culture of overwork. It is concerning to find that overwork was cited as an issue by women in public sector occupations such as nursing, teaching and medicine. Overwork, or long working hours, is not just an issue for gender equality but is also linked to reduced productivity, well-being and job satisfaction and can affect burnout and retention (Gascoigne and others, 2015). There should also be a complementary focus on increasing the quality of part-time jobs, as well as improving wage progression within part-time work.

While the focus of this report was on organisational interventions to promote women’s career progression, there is clearly also much scope for increased public support for childcare to increase maternal employment and reduce the relatively high rates of part-time work among mothers in the UK. Analysis of international evidence by Thompson and Ben-Galim (2014) suggests that systems which provide 30 or more hours a week of subsidised childcare support high levels of maternal employment.

In the short term, the following steps seem promising, given the evidence in this report:

• the government is considering requiring employers to say in each job advert whether it can be done flexibly. The evidence presented in this report lends support to this proposal.
• Interventions similar to those implemented by Moen and others (2016) that focus on training managers to support flexible workers should be piloted.

Alongside this, future research should focus on:

• Working practices that mitigate the need for long and unpredictable hours. More comparative cross-sectoral or cross-occupational evidence on career progression and part-time/flexible work might help in this by identifying occupations where long and unpredictable hours are more/less common.
• More comparative cross-sectoral or cross-occupational evidence on women’s career progression would help in understanding which structural factors of work organisation are associated with women’s progression, as well as understanding in which sectors and occupations efforts should be most focused.
• Further comparative quantitative research on career progression across sectors and its relationship to overwork would be useful in an understanding of how overwork contributes to gender segregation via reduced retention of women in occupations were overwork is common (Cha, 2013).
• Similarly, further research on the relationship between overwork, the GPG and employee productivity/ burn-out in different sectors would help to both build a business case against overwork and link overwork more clearly to the GPG.
• While the barriers to women’s progression in professional and managerial occupations seem relatively well understood, more evidence on the barriers to progression for low paid women would be useful.
• There is a need for stronger evidence on which processes support women’s career progression, since much evidence comes from the USA and is based on studies from which it is difficult to draw causal conclusions. The government should support further field-based intervention studies.

Appendix

Research method

To provide a comprehensive overview of the literature on this topic in a short time-frame, a rapid evidence assessment (REA) approach was adopted. The search was limited to peer-reviewed papers readily accessible online in English. After a series of scoping searches produced a large number of hits, a decision was made to limit the evidence for objectives 1 and 2 to UK-based studies, while leaving the evidence for objectives 3 and 4 geographically unlimited. The search was time limited to return only literature published from 2000 until August 2018. For objectives 1 and 2 this still produced a large number of relevant papers and so all located literature published between 2010 and 2018 was read and then literature focusing on under-researched groups or occupations published prior to 2010 was prioritised.

Search

Search terms were developed iteratively through a series of scoping exercises and agreed with the GEO prior to running the final searches. Searches were supplemented by ‘pearl-growing techniques’, including following up on the references of key texts, and papers subsequently referencing them. A set of agreed inclusion and exclusion criteria was drawn up, including an assessment of the quality of the methodology.

Objectives 1 and 2

In order to make sure that the terms we used to search the literature reflected the research question, we investigated various potential frameworks. Given the nature of the data (with very few control groups or interventions), we opted for a SPIDER framework. The framework and search terms used are detailed below.

Table 1: SPIDER framework for objectives 1 and 2

Sample: UK Female Workers

Phenomena of Interest: Gender differences in workplace progression

Design: All relevant

Evaluation: Barriers/Facilitators of Progression

Research: Quantitative, qualitative, mixed

Context: Workplace

Search terms

Sample: Gender OR women OR woman OR female OR sex

Phenomena of Interest: Barrier* OR facilitator* OR inhibitor* OR bias OR inequality OR sexism OR “Gender imbalance” OR “sex imbalance” OR “sex inequal” OR “gender inequal” OR “sex equal” “gender equal” OR stereotype

Design: n/a

Evaluation: Progression or development OR trajecto* OR leader* OR advancement OR promotion OR pay OR success OR “gender gap” OR “gender divide” OR “vertical segregation” OR “glass ceiling” OR “glass labyrinth” OR “glass cliff” OR “mummy track” OR “parent track” OR “sticky floor” OR “occupational mobility” OR “gender differentiation” OR “occupational achievement” OR “sex differentiation” OR “career opportunities” OR hierarch*

Research: n/a

Context: Career OR workplace OR job OR employment

Geography: UK OR England OR Britain OR “united kingdom” OR Wales OR Scotland

OR “Northern Ireland” OR British OR English OR Welsh OR Irish

Boolean Operators: Sample AND (Phenomena OR Evaluation) AND Context AND Geography

Restrictions

Geographical restrictions: Studies carried out in the United Kingdom

Language restrictions: Studies in English

Date restrictions: 2000 to 2018. We will be pragmatic in terms of selecting studies to include within this period, given the time scale. We will work through the studies and use a sampling strategy on the basis of conceptual and sectoral saturation. So that if we have included 5 studies on a particular sector and the findings are broadly similar, we may then omit previous studies of the same sector, but will continue to read studies of previously uncovered sectors.

Objectives 3 and 4

For objectives 3 and 4 a PICO framework was adopted. There were no geographical limits set, but we limited the search to studies published after 2000. For studies investigating the effectiveness of an intervention, we limited the search to systematic reviews, meta-analyses and quantitative studies which made use of an adequate counterfactual. Mixed-methods and qualitative studies that explored the implementation of an intervention were also included.

Table 2: PICO Framework for objectives 3 and 4

Population: Female workers

Impact/Intervention: Organisational interventions to improve female progression

Control: Context with no intervention

Outcome: Differences in workplace progression/ intermediate outcomes that might affect progression

Context: The workplace

Search terms

Population: Gender OR women OR woman OR female OR diversity OR sex

Impact/Intervention: Intervention OR “organisation* polic” OR mentor OR flex* OR “work life” or “family friendly” or “work-life” or “family- friendly” or training or target* OR initiative OR practice OR diversity

Control (Context with no intervention): Barrier* OR inhibit* OR bias OR inequality OR sexism OR “Gender imbalance” OR “sex imbalance” OR “sex inequal” OR “gender inequal” OR “sex equal” “gender equal” OR stereotype

Outcome: Progression or development OR trajecto* OR leader* OR advancement OR promotion OR pay OR success OR “gender gap” OR “gender divide” OR “vertical segregation” OR “glass ceiling” OR “glass labyrinth” OR “glass cliff” OR “mummy track” OR “parent track” OR “sticky floor” OR “occupational mobility” OR “gender differentiation” OR “occupational achievement” OR “sex differentiation” OR “career opportunities” OR hierarch*

Context: Career OR workplace OR job OR employment

Boolean Operators: (All in Title or Abstract) Population AND Impact AND Control AND Outcome AND context

Scope of the Work – RQ 3

Geographical restrictions: None

Language restrictions: English

Date restrictions: 2000 to 2018

Databases

The above terms were used to search 4 academic databases – Proquest’s Social Science Premium Collection, EBSCOhost Business Source, Psycinfo and the Web of Science core collection.

Evidence quality criteria

Each study was critically appraised based on the appropriate appraisal template modified from Collins and others (2015). Studies which scored lower than one star overall were removed from the analysis.

Quantitative study evidence quality criteria – adapted from Collins et al 2015

	Criteria	Score (//)	Comments
Methodology	Is the sample population used in the study representative of the overall population that is the subject of the study and it is relevant in the context of the evidence statement (for example, relevant to UK)
	Was an adequate control group used? Was this similar to the population receiving the intervention?
Analysis	Were the analytical methods appropriate
Findings	Were the conclusions based on the information gained? Are the conclusions supported by the data, are they appropriate to the sample/approach?
Summary	Overall how well are the above criteria met?

Qualitative study evidence quality criteria – adapted from Collins et al 2015

	Criteria	Score (//)	Comments
Methodology	Are the interviewees clearly identified
	Are the interviewees the most suitable and representative – was the size of the group suitable for diversity of opinions?
Synthesis	How well has diversity of perspective been explored? Were minority opinions stated?
Findings	Were the conclusions based on the information gained from the interviewees (that is, are the conclusions supported by the data, are they appropriate to the sample/approach?)
Summary	Overall how well are the above criteria met?

Bibliography

Abendroth, A.K., Melzer, S., Kalev, A., Tomaskovic-Devey, D., 2017. Women at Work: Women’s Access to Power and the Gender Earnings Gap. Ilr Rev. 70, 190–222. https://doi.org/10.1177/0019793916668530 

Allen, T.D., Eby, L.T., Poteet, M.L., Lentz, E., Lima, L., 2004. Career Benefits Associated With Mentoring for Proteges: A Meta-Analysis. J. Appl. Psychol. 89, 127–136. https://doi.org/10.1037/0021-9010.89.1.127

Almer, E.D., Cohen, J.R., Single, L.E., 2004. Is it the Kids or the Schedule?: The Incremental Effect of Families and Flexible Scheduling on Perceived Career Success. J. Bus. Ethics 54, 51–65.

Atkinson, J., 2016. ‘Letters aren’t good’: the operation of the right to request flexible working post-maternity leave in UK small and medium-sized companies. J. Soc. Welf. Fam. Law 38, 380–393. https://doi.org/10.1080/09649069.2016.1239346 

Atkinson, J., 2011. Gendered organizations and women’s career progression in the UK financial services sector. J. Soc. Welf. Fam. Law 33, 243.

Backett-Milburn, K., Airey, L., McKie, L., Hogg, G., 2008. Family Comes First or Open All Hours?: How Low Paid Women Working in Food Retailing Manage Webs of Obligation at Home and Work. Sociol. Rev. 56, 474–496. https://doi.org/10.1111/j.1467-954X.2008.00800.x 

Bagilhole, B., 2016. An analysis of male cultural hegemony in senior management in UK Academia. Investig. Fem. 7, 9–23. https://doi.org/10.5209/INFE.53778 

Bezrukova, K., Spell, C.S., Perry, J.L., Jehn, K.A., 2016. A meta-analytical integration of over 40 years of research on diversity training evaluation. Psychol. Bull. 142, 1227–1274. https://doi.org/10.1037/bul0000067 

Bohnet, I., van Geen, A., Bazerman, M., 2016. When performance trumps gender bias: Joint vs. separate evaluation. Manag. Sci. 62, 1225–1234. https://doi.org/10.1287/mnsc.2015.2186

Bowles, H.R., Babcock, L., McGinn, K.L., 2005. Constraints and triggers: Situational mechanics of gender in negotiation. J. Pers. Soc. Psychol. 89, 951–965. https://doi.org/10.1037/0022-3514.89.6.951

Brescoll, V.L., Glass, J., Sedlovskaya, A., 2013. Ask and Ye Shall Receive? The Dynamics of Employer- Provided Flexible Work Options and the Need for Public Policy. J. Soc. Issues 69, 367.

Broadbridge, A., 2010. Choice or constraint? Tensions in female retail executives’ career narratives. Gend. Manag. Int. J. 25, 244–260. https://doi.org/10.1108/17542411011036437

Broadbridge, A., 2007. Dominated by women: managed by men? The career development process of retail managers. Int. J. Retail Distrib. Manag. 35, 956–974. https://doi.org/10.1108/09590550710835193

Broadbridge, A., Parsons, E., 2005. Gender and career choice: Experiences of UK charity retail managers. Career Dev. Int. 10, 80–97.

Budjanovcanin, A., 2018. Actions speak louder than words: how employees mind the implementation gap. Int. J. Hum. Resour. Manag. 0, 1–20. https://doi.org/10.1080/09585192.2018.1443959

Bukodi, E., Dex, S., Joshi, H., 2012. Changing Career Trajectories of Women and Men Across Time. Edward Elgar Publishing.

Burkinshaw, P., Cahill, J., Ford, J., 2018. Empirical Evidence Illuminating Gendered Regimes in UK Higher Education: Developing a New Conceptual Framework. Educ. Sci. 8. https://doi.org/10.3390/educsci8020081

Caffrey, L., Wyatt, D., Fudge, N., Mattingley, H., Williamson, C., McKevitt, C., 2016. Gender equity programmes in academic medicine: a realist evaluation approach to Athena SWAN processes. Bmj Open 6. https://doi.org/10.1136/bmjopen-2016-012090

Cahusac, E., Kanji, S., 2014. Giving Up: How Gendered Organizational Cultures Push Mothers Out. Gend. Work Organ. 21, 57–70. https://doi.org/10.1111/gwao.12011

Callan, S., 2007. Implications of family-friendly policies for organizational culture: findings from two case studies. Work Employ. Soc. 21, 673–691. https://doi.org/10.1177/0950017007082876

Carli, L.L., Eagly, A.H., 2016. Women face a labyrinth: an examination of metaphors for women leaders. Gend. Manag. 31, 514–527. https://doi.org/10.1108/GM-02-2015-0007

Carmichael, F., Hulme, C., Sheppard, S., Connell, G., 2008. Work-life imbalance: Informal care and paid employment in the UK. Fem. Econ. 14, 3–35. https://doi.org/10.1080/13545700701881005

Castilla, E.J., 2015. Accounting for the Gap: A Firm Study Manipulating Organizational Accountability and Transparency in Pay Decisions. Organ. Sci. 26, 311–333. https://doi.org/10.1287/orsc.2014.0950

Castilla, E.J., 2008. Gender, Race, and Meritocracy in Organizational Careers. Am. J. Sociol. 113, 1479– 1526. https://doi.org/10.1086/588738

Castilla, E.J., Benard, S., 2010. The Paradox of Meritocracy in Organizations. Adm. Sci. Q. 55, 543–676. https://doi.org/10.2189/asqu.2010.55.4.543

Cha, Y., 2013. Overwork and the Persistence of Gender Segregation in Occupations. Gend. Soc. 27, 158–184. https://doi.org/10.1177/0891243212470510

Cha, Y., 2010. Reinforcing Separate Spheres: The Effect of Spousal Overwork on Men’s and Women’s Employment in Dual-Earner Households. Am. Sociol. Rev. 75, 303–329. https://doi.org/10.1177/0003122410365307

Charles, N., 2014. Doing Gender, Practising Politics: Workplace Cultures in Local and Devolved Government. Gend. Work Organ. 21, 368–380. https://doi.org/10.1111/gwao.12042

Charlesworth, S., Baines, D., Cunningham, I., 2015. “If I Had a Family, There Is No Way That I Could Afford to Work Here”: Juggling Paid and Unpaid Care Work in Social Services. Gend. Work Organ. 22, 596–613. https://doi.org/10.1111/gwao.12111

Chung, H., 2018. Gender, Flexibility Stigma and the Perceived Negative Consequences of Flexible Working in the UK. Soc. Indic. Res. 1–25. https://doi.org/10.1007/s11205-018-2036-7

Chung, H., 2017. Work Autonomy, Flexibilty and Work-Life Balance: Final Report. 2017.

Chung, H., van der Horst, M., 2018. Women’s employment patterns after childbirth and the perceived access to and use of flexitime and teleworking. Hum. Relat. 71, 47–72. https://doi.org/10.1177/0018726717713828

Coleman, L., Cottell, J., 2019. Childcare Survey 2019. Coram Family and Childcare.

Collins, A., Coughlin, D., Miller, J., Kirk, S., 2015. The Production of Quick Scoping Reviews and Rapid Evidence Assessments A How to Guide.

Conley, H., Jenkins, S., 2011. Still ‘a Good Job for a Woman’? Women Teachers’ Experiences of Modernization in England and Wales. Gend. Work Organ. 18, 488–507. https://doi.org/10.1111/j.1468-0432.2011.00573.x

Connolly, S., Gregory, M., 2008. Moving Down: Women’s Part-Time Work and Occupational Change in Britain 1991–2001*. Econ. J. 118, F52–F76. https://doi.org/10.1111/j.1468-0297.2007.02116.x

Corby, S., Stanworth, C., 2009. A price worth paying? Equal Oppor. Int. 28, 162–178. https://doi.org/10.1108/02610150910937907

Costa Dias, M., Joyce, R., Parodi, F., 2018. Wage progression and the gender wage gap: the causal impact of hours of work. The Institute of Fiscal Studies.

Crompton, R., Lyonette, C., 2011. Women’s Career Success and Work-life Adaptations in the Accountancy and Medical Professions in Britain. Gend. Work Organ. 18, 231–254.  https://doi.org/10.1111/j.1468-0432.2009.00511.x

Cundiff, J.L., Ryuk, S., Cech, K., 2018. Identity-safe or threatening? Perceptions of women-targeted diversity initiatives. Group Process. Intergroup Relat. 21, 745–766. https://doi.org/10.1177/1368430217740434

De Henau, J., Harris, S., Stephenson, M.-A., 2018. 2018 WBG Briefing: Women, Employment and Earnings.

Dean, H., 2007. Tipping the Balance: The Problematic Nature of Work–Life Balance in a Low-Income Neighbourhood. J. Soc. Policy 36, 519–537. https://doi.org/10.1017/S004727940700116X

Department for Business, Innovation and Skills, 2014. The Fourth Work-Life Balance Employer Survey 2013.

Department for Education, 2018a. Childcare and Early Years Survey of Parents in England, 2018. Department for Education, 2018b. 30 hours free childcare: evaluation of the national rollout.

Dex, S., Bukodi, E., 2012. The Effects of Part-Time Work on Women’s Occupational Mobility in Britain: Evidence from the 1958 Birth Cohort Study. Natl. Inst. Econ. Rev. 222, R20–R37. https://doi.org/10.1177/002795011222200103

Dias, M.C., Joyce, R., Parodi, F., n.d. The gender pay gap in the UK: children and experience in work 28. Dick, G., Metcalfe, B., 2007. The progress of female police officers?: An Empirical Analysis of Organizational Commitment and Tenure Explanations in Two UK Police Forces. Int. J. Public Sect. Manag. 20, 81–100. https://doi.org/10.1108/09513550710731463

Dobbin, F., Kalev, A., 2016. Why Diversity Programs Fail. Harv. Bus. Rev.

Dobbin, F., Schrage, D., Kalev, A., 2015. Rage against the Iron Cage: The Varied Effects of Bureaucratic Personnel Reforms on Diversity. Am. Sociol. Rev. 80, 1014.

Durbin, S., Tomlinson, J., 2014. Female Part-Time Managers: Careers, Mentors and Role Models. Gend. Work Organ. 21, 308–320. https://doi.org/10.1111/gwao.12038

Eby, L.T., Allen, T.D., Evans, S.C., Ng, T., DuBois, D.L., 2008. Does mentoring matter? A multidisciplinary meta-analysis comparing mentored and non-mentored individuals. J. Vocat. Behav., Mentoring 72, 254–267. https://doi.org/10.1016/j.jvb.2007.04.005

Eikhof, D.R., 2012. A double‐edged sword: twenty‐first century workplace trends and gender equality. Gend. Manag. Int. J. 27, 7–22. https://doi.org/10.1108/17542411211199246

Eikhof, D.R., Warhurst, C., 2013. The promised land? Why social inequalities are systemic in the creative industries. Empl. Relat. 35, 495–508. https://doi.org/10.1108/ER-08-2012-0061

European Commission, 2017. European Semester Thematic Factsheet: Women in the Labour Market. European Institute for Gender Equality, 2019. Vertical Segregation. Gend. Equal. Gloss. Thesaurus.

Eurostat, 2018. How many hours do Europeans work per week? [WWW Document]. URL https://ec.europa.eu/eurostat/web/products-eurostat-news/-/DDN-20180125-1 (accessed 4.3.19).

Fearfull, A., Kamenou, N., 2006. How do you account for it?: A critical exploration of career opportunities for and experiences of ethnic minority women. Crit. Perspect. Account. 17, 883–901. https://doi.org/10.1016/j.cpa.2005.08.006

Ferguson, J.-P., 2015. The Control of Managerial Discretion: Evidence from Unionization’s Impact on Employment Segregation. Am. J. Sociol. 121, 675–721. https://doi.org/10.1086/683357

Fielden, S., Woolnough, H., 2014. The impact of a career development and mentoring programme on female mental health nurses: A longitudinal, qualitative study. Gend. Manag. Int. J. 29, 108–122. https://doi.org/10.1108/GM-05-2013-0049

Ford, J., Collinson, D., 2011. In search of the perfect manager? Work-life balance and managerial work. Work Employ. Soc. 25, 257–273. https://doi.org/10.1177/0950017011398895

Gariety, B.S., Shaffer, S., 2001. Wage Differentials Associated with Flextime Work Arrangements in the New Economy. Mon. Labor Rev. 124, 68–75.

Gascoigne, C., Kelliher, C., 2018. The transition to part-time: How professionals negotiate “reduced time and workload’’ i-deals and craft their jobs.” Hum. Relat. 71, 103–125. https://doi.org/10.1177/0018726717722394

Gascoigne, C., Parry, E., Buchanan, D., 2015. Extreme work, gendered work? How extreme jobs and the discourse of ‘personal choice’ perpetuate gender inequality. Organization 22, 457–475. https://doi.org/10.1177/1350508415572511

Gash, V., Mertens, A., Gordo, L.R., 2012. The Influence of Changing Hours of Work on Women’s Life Satisfaction*. Manch. Sch. 80, 51–74. https://doi.org/10.1111/j.1467-9957.2011.02255.x

Glass, J., 2004. Blessing or Curse?: Work-Family Policies and Mother’s Wage Growth Over Time. Work Occup. 31, 367–394. https://doi.org/10.1177/0730888404266364

Glass, J.L., Noonan, M.C., 2016. Telecommuting and Earnings Trajectories Among American Women and Men 1989–2008. Soc. Forces 95, 217–250. https://doi.org/10.1093/sf/sow034

Government Equalities Office, 2018. 100% of UK employers publish gender pay gap data [WWW Document]. GOV.UK. URL https://www.gov.uk/government/news/100-of-uk-employers-publish- gender-pay-gap-data (accessed 10.23.18).

Grisso, J.A., Sammel, M.D., Rubenstein, A.H., Speck, R.M., Conant, E.F., Scott, P., Tuton, L.W., Westring, A.F., Friedman, S., Abbuhl, S.B., 2017. A Randomized Controlled Trial to Improve the Success of Women Assistant Professors. J. Womens Health 26, 571–579. https://doi.org/10.1089/jwh.2016.6025

Guillaume, C., Pochic, S., 2011. The Organisational Nature of Union Careers. Eur. Soc. 13, 607–631. https://doi.org/10.1080/14616696.2011.580855

Harkness, S., Pelikh, A., Borkowska, M., 2019. Employment pathways and occupational change after childbirth, 50 - 57 https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/840062/Bristol_Final_Report_1610.pdf 

Harris, L., Foster, C., Whysall, P., 2007. Maximising women’s potential in the UK’s retail sector. Empl. Relat. 29, 492–505.

Holliday, E.B., Jagsi, R., Thomas, C.R., Wilson, L.D., Fuller, C.D., 2014. Standing on the Shoulders of Giants: Results From the Radiation Oncology Academic Development and Mentorship Assessment Project (ROADMAP). Int. J. Radiat. Oncol. Biol. Phys. 88, 18–24. https://doi.org/10.1016/j.ijrobp.2013.09.035

Ibarra, H., Carter, N.M., Silva, C., 2010. Why Men Still Get More Promotions Than Women [WWW Document]. Harv. Bus. Rev. URL https://hbr.org/2010/09/why-men-still-get-more-promotions-than-women (accessed 8.9.18).

Jefferson, L., Bloor, K., Spilsbury, K., 2015. Exploring gender differences in the working lives of UK hospital consultants. J. R. Soc. Med. 108, 184–191. https://doi.org/10.1177/0141076814558523

Jewell, S., Razzu, G., Singleton, C., 2019. Who Works for Whom and the UK Gender Pay Gap? Br. J. Ind. Relat. Forthcom.

Joshi, A., Jooyeon Son, Hyuntak Roh, 2015. When Can Women Close the Gap? A Meta-Analytic Test of Sex Differences in Performance and Rewards. Acad. Manage. J. 58, 1516–1545. https://doi.org/10.5465/amj.2013.0721

Kalev, A., Kelly, E., Dobbin, F., 2006. Best Practices or Best Guesses? Assessing the Efficacy of Corporate Affirmative Action and Diversity Policies. Am. Sociol. Rev. 71, 589–617.

Kamenou, N., 2008. Reconsidering Work–Life Balance Debates: Challenging Limited Understandings of the ‘Life’ Component in the Context of Ethnic Minority Women’s Experiences. Br. J. Manag. 19, S99–S109. https://doi.org/10.1111/j.1467-8551.2008.00575.x

Kashiwagi, D.T., Varkey, P., Cook, D.A., 2013. Mentoring Programs for Physicians in Academic Medicine: A Systematic Review. Acad. Med. 88, 1029–1037. https://doi.org/10.1097/ACM.0b013e318294f368

Kelliher, C., Anderson, D., 2010. Doing more with less? Flexible working practices and the intensification of work. Hum. Relat. 63, 83–106. https://doi.org/10.1177/0018726709349199

Kirton, G., Robertson, M., 2018. Sustaining and advancing IT careers: Women’s experiences in a UK- based IT company. J. Strateg. Inf. Syst. 27, 157–169. https://doi.org/10.1016/j.jsis.2018.01.001

Kornberger, M., Carter, C., Ross-Smith, A., 2010. Changing gender domination in a Big Four accounting firm: Flexibility, performance and client service in practice. Account. Organ. Soc. 35, 775–791. https://doi.org/10.1016/j.aos.2010.09.005

Kumra, S., 2010. Exploring career “choices” of work‐centred women in a professional service firm. Gend. Manag. Int. J. 25, 227–243. https://doi.org/10.1108/17542411011036428

Langner, L.A., 2018. Flexible men and Successful Women: The Effects of Flexible Working Hours on German Couples’ Wages. Work Employ. Soc. 32, 687–706. https://doi.org/10.1177/0950017017708161

Laver, K.E., Prichard, I.J., Cations, M., Osenk, I., Govin, K., Coveney, J.D., 2018. A systematic review of interventions to support the careers of women in academic medicine and other disciplines. Bmj Open 8. https://doi.org/10.1136/bmjopen-2017-020380

Leibbrandt, A., List, J.A., 2014. Do Women Avoid Salary Negotiations? Evidence from a Large-Scale Natural Field Experiment. Manag. Sci. 61, 2016–2024. https://doi.org/10.1287/mnsc.2014.1994

Leslie, L.M., Manchester, C.F., Park, T.-Y., Mehng, S.A., 2012. Flexible Work Practices: A Source of Career Premiums or Penalties? Acad. Manage. J. 55, 1407–1428. https://doi.org/10.5465/amj.2010.0651

Leung, W.-F., Gill, R., Randle, K., 2015. Getting in, getting on, getting out? Women as career scramblers in the UK film and television industries. Sociol. Rev. 63, 50.

Lewis, S., 2001. Restructuring workplace cultures: the ultimate work‐family challenge? Women Manag. Rev. 16, 21–29. https://doi.org/10.1108/09649420110380256

Lindley, J.K., 2016. Lousy pay with lousy conditions: the role of occupational desegregation in explaining the UK gender pay and work intensity gaps. Oxf. Econ. Pap. 68, 152.

Lott, Y., Chung, H., 2016. Gender Discrepancies in the Outcomes of Schedule Control on Overtime Hours and Income in Germany. Eur. Sociol. Rev. 32, 752–765. https://doi.org/10.1093/esr/jcw032

Lyonette, C., 2015. Part-time work, work-life balance and gender equality. J. Soc. Welf. Fam. Law 37, 321.

Lyonette, C., Baldauf, B., 2019. Family friendly working policies and practices: Motivations, influences and impacts for employers, 31 - 48 https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/840061/Warwick_Final_Report_1610.pdf 

Lyonette, C., Crompton, R., 2008. The only way is up?: An examination of women’s “under‐achievement” in the accountancy profession in the UK. Gend. Manag. Int. J. 23, 506–521. https://doi.org/10.1108/17542410810908857

Manning, A., Swaffield, J., 2008. The gender gap in early-career wage growth. Econ. J. 118, 983–1024.

Manuel Mendoza-Nunez, V., Mecalco-Herrera, C., Ortega-Avila, C., Mecalco-Herrera, L., Luis Soto- Espinosa, J., Alfredo Rodriguez-Leon, M., 2013. A randomized control trial: training program of university students as health promoters. Bmc Public Health 13, 162. https://doi.org/10.1186/1471-2458-13-162

Mavin, S., Williams, J., Bryans, P., Patterson, N., 2015. “Woman as a project”: Key issues for women who want to get on. Handb. Gendered Careers Manag. Get. Get. Get. Out 305–321.

Mazei, J., Hüffmeier, J., Freund, P.A., Stuhlmacher, A.F., Bilke, L., Hertel, G., 2015. A meta-analysis on gender differences in negotiation outcomes and their moderators. Psychol. Bull. 141, 85–104. https://doi.org/10.1037/a0038184

McCarthy, F., Budd, L., Ison, S., 2015. Gender on the flightdeck: Experiences of women commercial airline pilots in the UK. J. Air Transp. Manag. 47, 32–38. https://doi.org/10.1016/j.jairtraman.2015.04.001

McCrate, E., 2002. Working mothers in a double bind: Working moms, minorities have the most rigid schedules, and are paid less for the sacrifice. Economic Policy Institute.

McQuaid, R., McIntosh, B., Munro, A., Dabir‐Alai, P., 2012. Motherhood and its impact on career progression. Gend. Manag. Int. J. 27, 346–364. https://doi.org/10.1108/17542411211252651

Michielsens, E., Bingham, C., Clarke, L., 2014. Managing diversity through flexible work arrangements: management perspectives. Empl. Relat. 36, 49–69. https://doi.org/10.1108/ER-06-2012-0048

Miller, K., 2009. Gendered nature of managerialism? Case of the National Health Service. Int. J. Public Sect. Manag. 22, 104–113. https://doi.org/10.1108/09513550910934510

Miller, K., Clark, D., 2008. “Knife before wife”: an exploratory study of gender and the UK medical profession. J. Health Organ. Manag. 22, 238–53. https://doi.org/10.1108/14777260810883521

Miller-Friedmann, J., Childs, A., Hillier, J., 2018. Approaching Gender Equity in Academic Chemistry: Lessons Learned from Successful Female Chemists in the UK. Chem. Educ. Res. Pract. 19, 24–41.

Moen, P., Kelly, E.L., Fan, W., Lee, S.-R., Almeida, D., Kossek, E.E., Buxton, O.M., 2016. Does a Flexibility/Support Organizational Initiative Improve High-Tech Employees’ Well-Being? Evidence from the Work, Family, and Health Network. Am. Sociol. Rev. 81, 134–164. https://doi.org/10.1177/0003122415622391

Moreau, M.-P., Osgood, J., Halsall, A., 2007. Making Sense of the Glass Ceiling in Schools: An Exploration of Women Teachers’ Discourses. Gend. Educ. 19, 237–253.

Munro, A., McQuaid, R., McIntosh, B., 2015. The impact of gender perceptions and professional values on women’s careers in nursing. Gend. Manag. Int. J. 30, 26–43. https://doi.org/10.1108/GM-12-2013-0135 

Nishii, L.H., Khattab, J., Shemla, M., Paluch, R.M., 2017. A Multi-Level Process Model for Understanding Diversity Practice Effectiveness. Acad. Manag. Ann. 12, 37–82. https://doi.org/10.5465/annals.2016.0044

Norman, L., 2014. A crisis of confidence: women coaches’ responses to their engagement in resistance. Sport Educ. Soc. 19, 532–551. https://doi.org/10.1080/13573322.2012.689975

Olsen, W., Gash, V., Sook Kim, M.Z., 2018. The gender pay gap in the UK: evidence from the UKHLS.

ONS, 2019. Labour market overview, UK: September 2019 [WWW Document]. URL https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/bulletins/uklabourmarket/september2019 [ (accessed 9.16.19).

ONS, 2018. Gender pay gap in the UK - Office for National Statistics [WWW Document]. URL https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/earningsandworkinghours/bulletins/genderpaygapintheuk/2018 (accessed 10.29.18).

ONS, 2017. Changes in the value and division of unpaid care work in the UK - Office for National Statistics. Ovseiko, P.V., Chapple, A., Edmunds, L.D., Ziebland, S., 2017. Advancing gender equality through the Athena SWAN Charter for Women in Science: an exploratory study of women’s and men’s perceptions. Health Res. Policy Syst. 15. https://doi.org/10.1186/s12961-017-0177-9

Ozbilgin, M.F., Tsouroufli, M., Smith, M., 2011. Understanding the interplay of time, gender and professionalism in hospital medicine in the UK. Soc. Sci. Med. 72, 1588–1594. https://doi.org/10.1016/j.socscimed.2011.03.030

Papafilippou, V., Bentley, L., 2017. Gendered transitions, career identities and possible selves: the case of engineering graduates. J. Educ. Work 30, 827–839. https://doi.org/10.1080/13639080.2017.1375088

Perlow, L.A., Kelly, E.L., 2014. Toward a Model of Work Redesign for Better Work and Better Life. Work Occup. 41, 111–134. https://doi.org/10.1177/0730888413516473

Peters, K., Ryan, M., Haslam, A., 2017. In Our Leader’s Footsteps: How Perceived Similarity to Leading Members of an Occupation Shapes Career Motivation.

Pritchard, R., 2010. Gender inequalities among staff in British and German universities: a qualitative study. Compare 40, 515–532. https://doi.org/10.1080/03057925.2010.490375

Pyper, D., 2018. Briefing Paper: Flexible Working. House of Commons Library.

Rainbird, H., 2007. Can training remove the glue from the “sticky floor” of low-paid work for women? Equal Oppor. Int. 26, 555–572. https://doi.org/10.1108/02610150710777042

Ropponen, A., Känsälä, M., Rantanen, J., Toppinen-Tanner, S., 2016. Organizational Initiatives for Promoting Employee Work-Life Reconciliation Over the Life Course. A Systematic Review of Intervention Studies. Nord. J. Work. Life Stud. 6. https://doi.org/10.19154/njwls.v6i3.5529

Savigny, H., 2014. Women, know your limits: cultural sexism in academia. Gend. Educ. 26, 794–809. https://doi.org/10.1080/09540253.2014.970977

Scholarios, D., Taylor, P., 2010. Gender, choice and constraint in call centre employment. New Technol.Work Employ. 25, 101–116. https://doi.org/10.1111/j.1468-005X.2010.00242.x

Shepherd, S., 2017. Why Are There so Few Female Leaders in Higher Education: A Case of Structure or Agency? Manag. Educ. 31, 82–87.

Smith, A., Elliott, F., 2012. The demands and challenges of being a retail store manager: ‘Handcuffed to the front doors.’ Work Employ. Soc. 26, 676–684. https://doi.org/10.1177/0950017012445091

Smith, J., 2011a. Aspirations to and Perceptions of Secondary Headship: Contrasting Female Teachers’ and Headteachers’ Perspectives. Educ. Manag. Adm. Leadersh. 39, 516–535. https://doi.org/10.1177/1741143211408450

Smith, J., 2011b. Agency and Female Teachers’ Career Decisions: A Life History Study of 40 Women. Educ. Manag. Adm. Leadersh. 39, 7–24. https://doi.org/10.1177/1741143210383900

Stamm, M., Buddeberg-Fischer, B., 2011. The impact of mentoring during postgraduate training on doctors’ career success. Med. Educ. 45, 488–496. https://doi.org/10.1111/j.1365-2923.2010.03857.x

Stewart, K., 2009. Employment and Wage Trajectories for Mothers Entering Low-skilled Work: Evidence from the British Lone Parent Cohort. Soc. Policy Adm. 43, 483–507. https://doi.org/10.1111/j.1467-9515.2009.00675.x

Taylor, K.S., Lambert, T.W., Goldacre, M.J., 2009. Career progression and destinations, comparing men and women in the NHS: postal questionnaire surveys. Bmj-Br. Med. J. 338, b1735. https://doi.org/10.1136/bmj.b1735

Teasdale, N., 2013. Fragmented Sisters? The Implications of Flexible Working Policies for Professional Women’s Workplace Relationships. Gend. Work Organ. 20, 397–412. https://doi.org/10.1111/j.1468-0432.2012.00590.x

Tomlinson, J., 2018. Part-time work and career progression: prospects and opportunities.

Tomlinson, J., 2007. Female part‐time workers’ experiences of occupational mobility in the UK service industry. Women Manag. Rev. 22, 305–318. https://doi.org/10.1108/09649420710754255

Tomlinson, J., 2006a. Routes to Part-Time Management in UK Service Sector Organizations: Implications for Women’s Skills, Flexibility and Progression. Gend. Work Organ. 13, 585–605. https://doi.org/10.1111/j.1468-0432.2006.00326.x

Tomlinson, J., 2006b. Routes to part-time management in UK service sector organizations: Implications for women’s skills, flexibility and progression. Gend. Work Organ. 13, 585–605. https://doi.org/10.1111/j.1468-0432.2006.00326.x

Tomlinson, J., Durbin, S., 2010a. Female part-time managersWork-life balance, aspirations and career mobility. Equal. Divers. Incl. 29, 255–270. https://doi.org/10.1108/02610151011028859

Tomlinson, J., Durbin, S., 2010b. Female part-time managersWork-life balance, aspirations and career mobility. Equal. Divers. Incl. 29, 255–270. https://doi.org/10.1108/02610151011028859

Tomlinson, J., Muzio, D., Sommerlad, H., Webley, L., Duff, L., 2013. Structure, agency and career strategies of white women and black and minority ethnic individuals in the legal profession. Hum. Relat. 66, 245–269. https://doi.org/10.1177/0018726712460556

Tomlinson, J., Olsen, W., Neff, D., Purdam, K., Mehta, S., 2005. Examining the Potential for Women Returners to Work in Areas of High Occupational Gender Segregation 106.

Tomlinson, J., Olsen, W., Purdam, K., 2009. Women Returners and Potential Returners: Employment Profiles and Labour Market Opportunities—A Case Study of the United Kingdom. Eur. Sociol. Rev. 25, 349–363. https://doi.org/10.1093/esr/jcn053

Tonge, J., 2008. Barriers to networking for women in a UK professional service. Gend. Manag. Int. J. 23, 484–505. https://doi.org/10.1108/17542410810908848

Trades Union Congress, 2017. Great Jobs with Guaranteed Hours.

Umolu, A.O., 2014. Gender and Race in UK Organisations: Case Study of Nigerian and Indian Women. Gend. Behav. 12, 6045–6058.

Vinnicombe, S., Singh, V., Anderson, D., 2010. Women partners leaving the firm: choice, what choice? Gend. Manag. Int. J. 25, 170–183. https://doi.org/10.1108/17542411011036383

Walsh, J., 2012. Not Worth the Sacrifice? Women’s Aspirations and Career Progression in Law Firms. Gend. Work Organ. 19, 508–531. https://doi.org/10.1111/j.1468-0432.2012.00607.x

Warren, T., Lyonette, C., 2018. Good, Bad and Very Bad Part-time Jobs for Women? Re-examining the Importance of Occupational Class for Job Quality since the ‘Great Recession’ in Britain. Work Employ. Soc. 32, 747–767. https://doi.org/10.1177/0950017018762289

Wattis, L., James, L., 2013. Exploring order and disorder: Women’s experiences balancing work and care. Eur. J. Womens Stud. 20, 264–278. https://doi.org/10.1177/1350506813484241

Wheatley, D., Wu, Z., 2014. Dual careers, time-use and satisfaction levels: evidence from the British Household Panel Survey. Ind. Relat. J. 45, 443–464. https://doi.org/10.1111/irj.12071

White, K., Bagilhole, B., Riordan, S., 2012. The Gendered Shaping of University Leadership in Australia, South Africa and the United Kingdom. High. Educ. Q. 66, 293–307. https://doi.org/10.1111/j.1468-2273.2012.00523.x

Wilson, F., 2014. May the best man win: Understanding the lack of meritocracy of women in UK banking. Equal. Divers. Incl. 33, 361–371. https://doi.org/10.1108/EDI-11-2013-0095

Winder, K.L., 2009. Flexible Scheduling and the Gender Wage Gap. BE J. Econ. Anal. Policy 9. https://doi.org/10.2202/1935-1682.2197

Women and Equalities Committee, 2018. Sexual harassment in the workplace. Women and Equalities Committee, 2016. Gender Pay Gap. House of Commons.

Woodfield, R., 2016. Gender and the achievement of skilled status in the workplace: the case of women leaders in the UK Fire and Rescue Service. Work Employ. Soc. 30, 237.

Woolnough, H., Redshaw, J., 2016. The career decisions of professional women with dependent children What’s changed? Gend. Manag. 31, 297–311. https://doi.org/10.1108/GM-03-2016-0038

Worrall, L., Harris, K., Stewart, R., Thomas, A., McDermott, P., 2010. Barriers to women in the UK construction industry. Eng. Constr. Archit. Manag. 09699988 17, 268–281. https://doi.org/10.1108/09699981011038060

Wright, T., 2016. Women’s Experience of Workplace Interactions in Male-Dominated Work: The Intersections of Gender, Sexuality and Occupational Group. Gend. Work Organ. 23, 348–362. https://doi.org/10.1111/gwao.12074

Yua, W.H., Kuo, J.C.L., 2017. The Motherhood Wage Penalty by Work Conditions: How Do Occupational Characteristics Hinder or Empower Mothers? Am. Sociol. Rev. 82, 744–769. https://doi.org/10.1177/0003122417712729