Non-standard employment contracts & promotion - McMaster ...

NON-STANDARD EMPLOYMENT AND PROMOTIONS: A WITHIN GENDERS ANALYSIS* by Isik U. Zeytinoglu, PhD and Gordon B. Cooke, PhD

* This paper has been accepted for publication in the Journal of Industrial Relations and the final (edited, revised and typeset) version of this paper will be published in the Journal of Industrial Relations, 2008, 50(2), 319-337 by SAGE Publications Ltd, All rights reserved. © Industrial Relations Society of Australia. For more information please visit: www.sagepublications.com . This study is supported by a grant from the Social Sciences and Humanities Research Council of Canada. Access to data for this paper was received through Statistics Canada Research Data Centre at McMaster University. The research and analysis are based on data produced by Statistics Canada and the opinions expressed do not represent the views of Statistics Canada. For correspondence, contact either [email protected] or [email protected]

Addresses: Dr. Isik U. Zeytinoglu Professor of Management & Industrial Relations DeGroote School of Business McMaster University Hamilton, On, Canada L8S 4M4 E-mail: [email protected] Fax: (905) 521 8995 Tel: (905) 525 9140, ext.23957

Dr. Gordon B. Cooke Assistant Professor of Industrial Relations Faculty of Business Administration Memorial University of Newfoundland St. John’s, NL, Canada A1C 5S7 E-mail: [email protected] Fax: (709) 737 7680 Tel: (709) 737 6204

Non-standard employment & promotions, Page 2

NON-STANDARD EMPLOYMENT AND PROMOTIONS: A WITHIN GENDERS ANALYSIS

ABSTRACT This paper examines promotion experiences of workers in non-standard employment as compared to those in regular full-time employment. Since females dominate non-standard employment, we analyse the female and male labour forces separately. Non-standard employment refers to regular part-time, temporary full-time, and temporary part-time employment. Data comes from the Workplace and Employee Survey (1999) of Statistics Canada. Results are generalized to Canadian workers. Results show that within the female labour force, workers in all three types of non-standard employment are less likely to be promoted than workers in regular full-time employment. Within the male labour force only those in temporary part-time employment are less likely to be promoted. Working in regular part-time or temporary full-time contracts has no impact on promotion for male workers. Overall results suggest that all three types of non-standard work adversely affect females’ promotion experiences but for males only for those in temporary part-time jobs are adversely affected.


INTRODUCTION Non-standard employment has been increasing globally over the past several decades (Burgess & Connell 2004; ILO 2003; Zeytinoglu 2002). Initially global restructuring and competition led employers in industrialized countries to create non-standard jobs. More recently however, influential intergovernmental organizations such as the Organisation for Economic Cooperation and Development (OECD 2003 & 2006a & b) and the European Union (Goetschy 2006) have begun promoting non-standard employment to increase flexibility for employers and workers, and to create jobs for working-age populations who are under-represented in the workforce. A significant employment strategy of these intergovernmental organizations is part-time jobs for women with young children. While non-standard employment may be seen by policy makers as a win-win strategy for employers and women, there can be longer-term negative consequences for workers in these jobs (Fudge & Owens 2006; Gonäs 2004). We focus on one of the consequences and ask whether workers in non-standard employment ever get promoted. Our purpose is to examine promotion experiences of workers in non-standard employment contracts as compared to those in regular full-time employment contracts. Since females dominate the non-standard labour force, we also conduct a within genders analysis for the female and male labour forces. Promotion is a career opportunity to which most workers aspire. It is a change in job duties with increased responsibilities and can lead to an increase in earnings. Promotion also shows the high value the employer places on the worker. In this paper we focus on a specific question in the 1999 Workplace and Employee Survey (WES) of Statistics Canada that asks workers ‘Have you ever been promoted since working for this employer [giving the name of the employer]?’ By promotion, the survey says, ‘we mean a change in job duties and responsibilities that leads not only to an increase in earnings but also an increase in the


complexity and responsibility of the job.’ The WES data allows us to examine promotion experiences of workers controlling for a large number of variables not usually available in other data sets. As early as 1983, the Canadian government’s Commission of Inquiry into Part-Time Work in Canada (Wallace Report 1983) identified a lack of advancement opportunities as one of the long-term employment problems of part-time and temporary workers. About a decade later Zeytinoglu (1994) raised similar concerns showing that Canadian employers treated workers differently: full-time workers given priority in promotions, followed by those in regular part-time jobs, and casual part-time workers were considered last. In the mid1990s, another Canadian federal government commission urged a closer study of employment futures of workers in non-standard employment (Sims Commission 1997). Researchers continue to raise concerns about whether these workers are becoming permanently marginalized in the labour market with few career opportunities (see, for Canada, Bolle 1997; Hall 2004; ILO 2003; Maxwell & Saunders 2003; McKeown 2005; Peetz 2005; Watson 2005). With the long tenure of non-standard workers in the labour market (Dickens 2003; Fudge & Owens 2006; Zeytinoglu 2002), it is important and timely to analyse promotion experiences of these workers.

DEFINITIONS, THE CONTEXT AND THE OECD POLICY INITIATIVES PROMOTING NON-STANDARD WORK We refer to non-standard employment as regular part-time, temporary full-time and temporary part-time. This type of categorization of non-standard workers is commonly used by researchers (see, for example, Cranford et al. 2003; Kapsalis & Tourigny 2004; Watson 2005; Wooden & Warren 2004). In Canada, the share of non-standard employment in the labour force increased until the late 1990s and stabilized in the early 2000s (Cranford et al.


2003). Currently, the Labour Force Survey shows that 11% of the labour force in Canada is in regular part-time employment, 6% in temporary full-time employment, and 4% in temporary part-time employment (Vosko et al. 2003). An earlier study using the WES data shows that the proportions for regular part-time, temporary full-time and temporary part-time are 12%, 3% and 3%, respectively (Zeytinoglu & Cooke 2005). In Australia recent research shows that 18% of workers are in permanent part-time employment, 3% in casual full-time employment, 2% in fixed-term contracts, and 9% in casual part-time employment (Louie et al. 2006). In terms of gender, female workers dominate non-standard employment in Canada and elsewhere (see, for example, Broadbent 2002; Cranford et al. 2003; Fudge & Owens 2006; Gonäs 2004; Markey et al. 2003). In the mid-1990s both the Organisation for Economic Co-operation and Development (OECD 2003, 2006a & b) and the European Commission (Gonäs 2004; Rubery et al. 2004) formulated a set of policy recommendations (or guidelines in the European Union case) with the goal of reducing high and persistent unemployment and facilitating job creation for difficult to employ segments of the population; i.e. women with young children, older workers, younger workers and immigrants. Since Canada is a member of the OECD but the European Employment Strategy guidelines do not apply to Canada, we focus on the OECD policy recommendations. The 1994 OECD Jobs Strategy recommended that member countries increase flexibility of working time, make wages and labour costs more flexible, and reform employment security provisions that inhibit the expansion of employment (OECD 2006a). These recommendations presumed that in some member countries, i.e. Continental European countries, labour markets were too rigidly structured and flexibility in employment was needed to achieve the ‘job creation success’ of the U.S. Flexibility in working time, pay, and benefits along with relaxed job security were promoted.


In reassessing its Jobs Strategy policies, the OECD noted that in 2006 the challenges faced by policy makers widened (OECD 2006a). While females’ labour force participation rates continue to increase, older workers, younger workers, immigrants, less-educated workers, and women with children still face barriers. Labour markets remain segmented in some countries with a significant share of low-paid workers finding it difficult to climb the job ladder and/or maintain continuous employment. While acknowledging that there is no single successful approach, the Restated OECD Jobs Strategy recommends that ‘familyfriendly policies, including childcare support, as well as working-time arrangements which help reconcile work and family life, should be implemented so as to remove barriers to employment for those with family commitments’ (OECD 2006a: 21). The Restated OECD Jobs Strategy refers to women in the context of family-friendly policies and promotes flexible working-time arrangements and part-time work, among other policies, to enable their labour force participation (OECD 2006b).

THEORETICAL APROACHES TO PROMOTION EXPERIENCES OF NONSTANDARD WORKERS AND EMPIRICAL FINDINGS Non-standard employment contracts were initially explained by the dual internal labour markets theory (Osterman 1992) and the core-periphery conceptualization (Atkinson 1987; Beechy & Perkins 1987). Existence of job ladders, i.e. promotion opportunities, with pay increases and job security were defining characteristics of the internal labour markets. Workers in regular full-time jobs were placed in the core and workers in non-standard employment contracts were placed in the periphery of their firms and as a result were less likely to enjoy promotion prospects (Osterman 1992). Building on the dual internal labour markets theory and focusing on part-time contracts – the most common non-standard employment contract in the 1980s – earlier


studies showed that part-time workers tend to have fewer promotion opportunities than regular full-time workers (Warme et al. 1992). Recent research on earnings, benefits, and a variety of other employment conditions of non-standard workers provides further evidence that the treatment of workers differs according to their employment status (see, for example, Lautsch 2002; Markey et al. 2003; McKeown 2005; Peetz 2005; Zeytinoglu & Cooke 2005). The European Foundation reports that in European Union member countries, on average, part-time workers are less likely to be promoted (European Foundation 2004). McGovern and colleagues (2004), using the Working in Britain 2000 survey, showed that non-standard employment arrangements are generally more common in jobs with no career ladders and thus with no promotion prospects. The restructuring of the labour markets in the 1980s and the dualism of flexible labour markets created both exclusion and advantages for non-standard workers based on gender, race and class dimensions (Zeytinoglu & Muteshi 2000). As Forrest argued (1996), the ‘heterosexist’ assumptions, gendered social practices and racial understandings about the family and men’s and women’s roles within it have been extremely influential in how and where women can be employed. Hierarchies and gendered organizations (Acker 1990; Amott & Matthaei 1991) place women on the periphery regardless of women’s location in the labour market, firm or the workplace. Feminist researchers have long argued that the theories of industrial relations and dual internal labour markets use male patterns of employment as the norm (see, for example, Forrest 1993; Dickens 2003; Wajcman 2000). They contend that women’s work and the gendered nature of organizations has been ignored (Dickens 2003; Forrest 1996; Mills 1989). Changes in the labour market are increasingly leading to a disconnect between the prevailing institutional theoretical framework for work and employment and the sets of assumptions applied to female workers (Coutu & Murray 2005; Fudge & Owens 2006).


There is ample research evidence to show that gender enhances the labour market capacity of men but undermines that of women (McGovern et al. 2004). Women are less likely than men to be promoted when employed in the same or similar occupations (Catalyst 2005; Economist 2005a & b). Research from the U.K. shows that women in part-time jobs are more likely to stay in the same position or be demoted (Fagan 2006). Most theoretical explanations for this experience focus on sex discrimination (Cohn 2000; Deaux & Kite 1987; Padavic & Reskin 2002). The literature also shows that employment and careers for women are still discussed within the context of family-friendly flexible work arrangements despite mounting evidence that most women, particularly those in male-dominated fields, have ambitions and attitudes towards work similar to men’s. Part-time work is often an ‘option’ for women with children largely due to a lack of affordable, good quality child care services (for earlier studies on these themes, see, for example, Wallace Report 1983; Warme et al. 1992; Zeytinoglu 1994; and for recent discussions, see, Edwards & Wajcman 2005; Gonäs 2004; Kirton & Greene 2005; Owens 2006; Yerkes & Visser 2006). Organizational career paths are changing and there is much debate that perhaps in the new or transformed organizations, gender inequality at work will be eliminated (Hebson & Grugulis 2005). However, Edwards and Wajcman’s study (2005) shows that the predictions of change are overstated and the traditional organizational career model still continues to be the experience of most workers. In all organizations, old and new, the traditional career model, however, is eroded rather than transformed (Edwards and Wajcman 2005). Gendered career paths (including part-time employment and the ‘mommy track’) still apply to women’s experiences, while marriage and fatherhood continue to not present serious career dilemmas for men. Even in new organizations, research does not show an erosion of gender divisions. Instead, the risks of precariousness such as job insecurity are concentrated on lower-paid jobs


that are dominated by women (Hebson & Grugulis 2005). We build our analysis on these theoretical approaches and empirical studies of the promotion experiences of non-standard workers and women and men in the labour force. However, we go beyond the debates of differences between genders in promotion experiences and the gendered nature of non-standard work. We focus within the female and male labour forces and examine the differential treatment of non-standard workers as compared to regular full-time workers. Often it is difficult to untangle the effects of gender and non-standard work on employment experiences, including opportunities for promotion (McGovern et al. 2004). In our study we first control for the effects of gender and confirm that non-standard workers have lower promotion experiences than regular full-time workers. Then we focus on each gender’s labour market separately and argue that regardless of gender those in non-standard employment contracts are less likely to be promoted. Thus we hypothesize that within the female and male labour forces, those in non-standard employment will have lower promotion experiences in comparison to workers in regular full-time employment. In addition to gender many factors can influence non-standard workers’ promotion experiences. In some unionized jobs, promotion possibilities might be limited. In others, the collective agreement can define the characteristics to be considered for promotion, standardizing the process. There is ample research showing the benefits of unionization for workers (Fang & Verma 2002), though many non-standard workers are not covered by collective agreements (Zeytinoglu & Cooke 2005). We include in our study the collective agreement coverage variable to control for the effects of unionization. Workplace characteristics such as the size and percentage of part-time and temporary workers in the workplace can also affect promotion experiences. Larger workplaces can provide more promotion opportunities than smaller workplaces. A larger percentage of part-


time or temporary workers can be an indication of an overall ‘bad quality’ workplace where promotion possibilities might be limited. A few studies in Canada using the workplace parttime rate and the temporary worker rate as control variables showed that the higher the percentage of part-time and temporary workers, the lower the wages (Drolet 2002), benefits (Zeytinoglu & Cooke 2005), and training provided (Turcotte et al. 2003). Workplace quality is not the focus of this study, and therefore we control for these workplace characteristics. We also control for the industry characteristic, since the industry that the worker is employed in can also affect promotion experiences. Having a high level of education, longer full-time experience, long-tenure at the workplace, and being in a professional or managerial position can positively affect the promotion experience. In our study we control for these human capital factors. Training received at the workplace can increase the probability of promotion though recent research incorporating part-time and temporary worker variables in their analysis shows that these workers are less likely to receive training (Böheim & Booth 2004; van Velzen 2002). In Canada, employer investment in training is lower than most other industrialized countries (Goldenberg 2006). Emerging research from Canada that includes non-standard workers in its analysis shows that non-standard workers are less likely to receive training (Turcotte et al. 2003). In terms of personal characteristics, research shows that due to gendered perceptions, being married with children is positively associated with promotion for males but the opposite tends to be true for females (Hersch & Stratton 2000). Research shows that recent immigrants earn lower wages (Thomson 2002) and benefits (Zeytinoglu & Cooke 2005) than Canadian-born workers and earlier cohorts of immigrants. Possible effects of these human capital and personal characteristics are included as control factors in our study so that we can isolate the association between non-standard employment and promotion in our genderfocused labour market analysis.


METHODOLOGY Data We use Statistics Canada’s 1999 Workplace and Employee Survey (WES), with employee microdata linked to workplace (i.e. employer) microdata. The WES is designed to explore a broad range of issues relating to employers and their employees. The survey is unique for Canada in that the employer responses and employee responses are linked at the microdata level; and employees are selected from within sampled workplaces. There are, however, similar linked data sets in the UK and Australia (see, for example, Böheim & Booth 2004; Gray & Tudball 2003). The WES covers all firms regardless of size, and examines a range of workplace issues. The 1999 WES has data on 24,597 employees from 6,351 workplaces, with a response rate of 83% and 94% respectively, and represents more than 10 million workers on a weighted basis. Using weighted micro-data and the bootstrapping technique allow us to generalize from the results to the Canadian labour market. (For more on sampling and sample design, see WES Compendium (2001)). The unit of analysis in this paper is the individual worker. The employee survey is the basis of the data and it is linked with the employer survey for workplace and industry variables. Variables The dependent variable is whether the worker has ever been promoted since working for the employer (1=promoted, 0=not). The questionnaire defines promotion as a change in job duties and responsibilities which led to both an increase in pay and to a complex or higher responsibility job. The independent variable is the employment contract grouped into four. Based on Statistics Canada’s definitions, we categorize those working 30 hours or more per week as


full-time, and those working less as part-time. We further group them based on whether their employment status is regular (continuous) or temporary. This yields a typology consisting of regular full-time (coded as 1=yes, 0=otherwise, and reference group) as standard employment. Regular part-time, temporary full-time and temporary part-time are nonstandard employment categories (and each coded as 1=yes, 0=otherwise). Temporary categories include workers in casual or on-call contracts or term employment. Although gender is technically categorized as a control variable, it also receives significant attention for the earlier discussed reasons. Female workers are coded as 1, and male workers as 0. Control variables consist of collective agreement coverage, workplace, industry, and individual (human capital and personal) characteristics. If the worker is a union member or covered by a collective agreement, they are coded as 1. Otherwise they are coded as 0. For workplace characteristics, we include size (total employment at the workplace in the last pay period of March 1999), log form of size (size transformed via log 10), and percentage of parttime and temporary workers in the workplace calculated as total number of part-time or temporary (i.e. non-permanent) employees on the payroll at the workplace in the last pay period divided by total employment at the workplace in the last pay period). Industry is coded as dummy variables of primary sector (forestry and mining), manufacturing and related sector (as the reference group), and service sector. For human capital characteristics, highest education attained is coded into four dummy variables (1=yes, 0=otherwise) of less than high school, completed high school, some post secondary, or university degree or higher with the first category as the reference group. Years of full-time work experience (and squared) are included as the number of years. To control for low tenure status that can affect promotion experience, we include a variable of low tenure status (with 1=start date at current employer is within the past five years, 0=not). We control for whether at least one of two types of training has been received by the worker:


employer-sponsored classroom training related to the job and/or on-the-job training in the past 12 months (1=yes, received one or both, 0=no, did not receive either type of training in this time period). Occupation is coded as a set of dummy variables (1=yes, 0=otherwise) representing managers, professionals (reference group), lower-white collar, and blue collar. In terms of personal characteristics, we control for marital status (1=married/common law partner, 0=single/ separated/ divorced/ widowed) and having dependent children (1=yes, 0=no). Immigrant status is coded as 1=if immigrant, i.e. not born in Canada, 0=born in Canada. Recent immigrants coded as 1=immigrated in/after 1990, 0=otherwise. Analysis We begin the analysis with descriptive statistics and correlations between variables. Results presented use weighted microdata. We then proceed to logistic regression analysis. Results include odds ratios, bootstrap regression coefficients, bootstrap standard errors, and significance levels. Regressions are conducted using Stata 8.0 software. First, we examine the association between employment status and promotion holding all other factors constant. Next the regression model for promotions is regenerated for workers separated into the female and male labour force. We conducted all of the analysis using weighted microdata accessed at the Statistics Canada’s McMaster University Research Data Centre (RDC). Statistics Canada strongly recommends the use of bootstrapping in multivariate analysis using the WES dataset due to its complex survey design. Bootstrapping refers to a process of repeatedly drawing random samples, with replacement, from the data at hand (Hamilton 2003). Bootstrapping involves ‘resampling’ the data with replacement many, many times in order to generate an empirical estimate of the entire sampling distribution of a statistic. Any data that involves complex sampling methodology requires special analytic consideration of bootstrapping (Mooney & Duval 1993). In all regression results, we use Statistics Canada’s mean bootstrap weights


using the revised Stata “bswreg” ado file by Chowhan and Buckley (2005). Limitations of the Study There are a few limitations of this study. First, the WES data slightly under represents the proportion of non-standard workers in the labour force. Only workers on the company payroll are eligible for sampling under the WES methodology. Thus, agency temporaries that could represent up to 3% of Canadian workers (derived from Acsess 2006) are only captured if their temporary work agency is included as an establishment by the WES. Self-employed contractors, many of whom can be included in the temporary full-time work force, are not captured in the WES data since they are not employees. About 15% of employed consider themselves self-employed in Canada (Cranford et al. 2003). Second, our study focuses on promotion since working with the current employer. This is only one means of advancement for employees and moving between employers is also important for workers. This aspect of promotion is not captured in our study. Third, it should be noted that the ‘employer-provided training received’ variable refers to the incidence, not the extent, of training taken. Although adequate for this study, it would be ideally augmented by a measure of the extent or quality of training.

RESULTS Promotion Proportions, Employment Contracts and Worker Characteristics Among all respondents 39% have been promoted since working for their current employer. (It should be noted that since the questionnaire gives the name of the employer, it is not clear if this means the workplace or the establishment.) When considering employment contract status, however, a distinct gap highlights the challenges facing non-standard workers. While 44% of those in regular full-time contracts have ever been promoted, the comparable proportions among non-standard workers are 18% for regular part-time, 23% for


temporary full-time, and 11% for temporary part-time workers. The large majority of workers, at 83%, are in regular full-time employment. The proportions for those in regular part-time, temporary full-time, and temporary part-time employment are 12%, 2%, and 3%, respectively. About 28% of the workers in this survey are covered by a collective agreement. In terms of organizational characteristics, the mean number of employees per workplace exceeds 400, while the mean proportion of part-time and temporary worker rates per workplace are 28% and 12%, respectively. Almost two-thirds of workers are in the service sector, with one-third in manufacturing and related, leaving only a very small proportion in the primary sector. In terms of worker characteristics, more than two-thirds have some post-secondary education, with the majority as college graduates taken college courses. Slightly more than one-fifth are in lower level white collar occupations, and close to half are in blue collar jobs. The remaining third is split evenly between managers and professionals. The average amount of full-time work experience is 16 years, although almost half of the workers are considered to have low tenure. Of all respondents, 55% received employer-provided training in the year of the survey. Regular full-time workers are somewhat more likely to receive training than non-standard workers. When sorting by employment status, the proportion that received training is 57%, 48%, 45%, and 48% for regular full-time, regular part-time, temporary fulltime and temporary part-time workers respectively. Slightly more than half of the respondents are female, though we should note that data has not been adjusted on these variables to account for known population. More than two-thirds are married or in a common-law relationship, and almost half have dependent children. Less than one-fifth are immigrants, with 3.5% immigrating in or after 1990.


Correlations Bivariate correlations between promotion and employment contract status show that having regular full-time employment is positively correlated with having ever been promoted, while working in any one of the non-standard employment categories is negatively correlated. All three non-standard employment categories are positively correlated with being female, while the opposite holds for regular full-time employment. Moreover, gender is negatively correlated with having ever been promoted. All correlations were significant at p chi-sqr 0.000 Significance Levels: *** p