Lewchuk, de Wolff and King 23

FROM STRAIN TO STRAIN: HEALTH EFFECTS OF PRECARIOUS EMPLOYMENT

Wayne Lewchuk Labour Studies Program, McMaster University, Hamilton, Canada Alice de Wolff Community University Research Alliance on Contingent Employment, Toronto, Canada Andy King United Steelworkers of America, Toronto, Canada Michael Polanyi Faculty of Kinesiology and Health Studies, University of Regina, Regina, Canada

The growth of precarious is as important as exposure to employment relationships described dangerous substances and exposure in this issue of Just Labour raises a to biomechanical risks in range of bargaining and policy understanding work-related health issues (Vosko, Zukewich and outcomes in permanent, full-time Cranford 2003; Fudge, this issue). (Cooper 1998). Our challenge Many of the daily concerns of has been to design a research workers in precarious employment method that captures the unique relationships centre on the issues of characteristics and effects of the stress and health. This article organisation of precarious reports on our investigation into the employment. Precarious relationship between health and the employment is, in our unique organisation of precarious understanding, a cumulative employment. In 2002-2003, we combination of atypical employment designed and conducted an contracts, limited social benefits, Employment Strain and Health poor statutory entitlements, job Survey with over 400 workers in insecurity, short tenure and low precarious employment wages. We have developed the relationships in Ontario. The concept of “employment strain” and preliminary analysis and findings indicators of “employment that we discuss here have uncertainty” to capture the implications for contracts, characteristics of precarious bargaining, workplace health and employment, and these are central safety and broader social policy. to the design of the survey and to Since the 1970s research has our analysis. shown that the organisation of work 24 JUST LABOUR vol. 3 (Fall 2003) Forum on Precarious Employment

HEALTH RISKS AND CONTROL employment than for those in OVER WORK ORGANISATION standard employment relationships.

Much of the research on work organisation and health traces its THE RETURN TO PRECARIOUS theoretical roots to the Job Demand- EMPLOYMENT RELATIONSHIPS Control model developed by AND LOSS OF CONTROL Karasek (1979) and Karasek and Theorell (1990). Job Demand- In 2002, approximately one-third of Control studies have shown that workers in Canada were part-time, employees’ control over how work temporary, working on contract, is done and their workload each holding multiple jobs or own- affect health directly, and that the account self-employed. This figure interaction between these two has increased from approximately elements has a further health one-quarter of employed workers in impact. Jobs characterised by low the early 1990s. While the majority levels of worker control and high of workers in precarious expenditures of psychosocial effort employment are young, female appear to expose employees to “job and/or recent immigrants, the new strain”. "Job strain" appears to lead growth in this kind of employment to lower , exhaustion has been, proportionally, among and depression, and in the long run mid-career men and women in to stress-related illness, including temporary jobs and self- cardiovascular disease. Jobs employment (see Vosko, Zukewich characterised by high levels of and Cranford 2003; Cranford and control and expenditure of Ladd this issue; Fudge this issue). psychosocial effort are considered Precarious employment “active jobs”. “Active jobs” include relationships are not new in the challenges, opportunities, and Canadian labour force. The labour learning on and off the job which history of much of the first half of can lead to positive health outcomes. the 20th century can be described as Control over how work is done is the successful struggle to reduce recognised as providing employees workers’ uncertainty and to gain with a buffer from the negative some control over the precarious health effects of workload-related conditions of their work. Early in stress. It provides “the opportunity the 20th century, many dock for individuals to adjust to demands workers, construction, agricultural, according to their needs and food processing, forestry, circumstances" (Wall, et al. 1996). and garment workers were hired at Control, however, is likely to be very plant gates, street corners or hiring different for workers in precarious halls for a day, a week or a month at a time. The subsequent movements Lewchuk, de Wolff and King 25 to establish “living wages” were through the intermediary of a concerned with raising wages and subcontractor. These casualised reducing uncertainties: employment contracts significantly limit labour relationships became less casual and relations protection for the worker were governed by legally and transfer additional enforceable contracts that defined responsibilities to the individual rights to continuing employment, worker (Cameron 2001). Together, the terms and conditions of work, business and governments have and in many cases, seniority-based restructured employment so that job rights. These were significant increasing numbers of workers are victories for the predominantly male “free agents” in a weakly regulated workforce that created control over labour market. access to, and the terms and conditions of full-time permanent CONTROL AND PRECARIOUS wage work. These victories, EMPLOYMENT RELATIONSHIPS however, left workers in other forms of employment unprotected. Many workers in precarious The current growth of precarious employment face constant employment is a predictable uncertainty about their future outcome of employment strategies employment prospects and the and policies of both private sector terms and conditions of their work. employers and governments of the Low pay and lack of benefits can past fifteen years, and reflects a loss create added uncertainties, of control over employment by including workers’ ability to provide larger groups of workers. Through for their basic household needs. the 1990s many institutions cut back Further, workers in precarious on core permanent workers and employment may need to search for increased the periphery of flexible work on a regular basis, manage pay workers. Employers and policy systems based on completed tasks makers have used fears about global rather than hours expended, balance competitiveness, and tools like the multiple jobs at multiple work sites, privatisation of services, just-in-time and provide their own equipment production and flexible human and training. These types of resource management to create a uncertain work organisation are climate where employers have associated with three categories of permission to break the implicit work-related health risks: weak agreements associated with labour market regulation; increased standard employment relationships. injury and illness risks that are Employers have casualised their similar to permanent workers; and relationship with employees in two over-all stress-related health risks. ways: either directly by creating temporary and part time jobs, or 26 JUST LABOUR vol. 3 (Fall 2003) Forum on Precarious Employment

Weak labour market regulation injury and illness. Researchers have Occupational health and safety consistently found evidence of research identifies three major increased injury and ill health from factors that contribute to injury and outsourcing, labour shedding illness prevention in “the (restructuring) and casualisation. workplace”: management (Quinlan et al 2001). In 1999 Quinlan commitment, worker participation, demonstrated a worsening of health and regulatory compliance (Walters outcomes associated with three and Frick 2000). Precarious broad causal factors: economic and employment relationships weaken reward systems (competition, long and undermine these factors in a hours, piecework, etc); number of ways. Laws regulating disorganisation (ambiguity of rules, the basic relationships between splintering occupational health and workers and employers, the right to safety management systems, etc); organise and bargain, minimum and increased likelihood of labour standards, and worker regulatory failure (laws do not apply compensation schemes, both assume to these employment relationships). and are designed to support the He revealed increased risks of injury standard employment relationship and illness in very different sectors, (Cranford and Ladd, this issue; and found that the consistent factor Fudge, this issue). Precarious was minimal regulatory protection. employment is largely defined by Workers in precarious employment the absence of laws, regulations and relationships are poorly protected in practices that support the standard hazardous work situations, and their employment relationship. The legal often unregulated hours of work, relationships relating to precarious intense workloads and limited employment are blurred as a result decision making latitude contribute of third party employment agencies, to high rates of workplace injury own-account self-employment, and illness. The point here, however, temporary and short-term contracts is that while these risks are higher that make labour market regulations for workers in precarious less effective. Every manner in employment, they are workplace which the legislation is blurred or specific and qualitatively similar in where workers are not covered has a nature to those of workers in negative impact on prevention in the standard employment relationships. workplace. From job strain to employment strain Increases in work-related injury and Our study suggests that workers illness in precarious employment also face The focus of most current qualitative health risks that are not research on health and precarious specific to one workplace. These employment is on work-related risks are a product of the insecurity Lewchuk, de Wolff and King 27

of employment itself, the with a trade-off of longer hours and uncertainty regarding the terms and more intensive employment. These conditions of work, and the need to analyses minimise the effect of the expend additional effort searching legal constraints that limit for work and balancing multiple employers in standard employment jobs at multiple work sites. relationships from hiring and firing Studies that have expanded the at will. They also permit employees, notion of workplace control beyond through collective action, some decision authority, use of skills, controls over which job they do capacity for collective action and (seniority), the right to participate in supportive colleagues have been of setting wages and benefits, and particular use to us. Brooker and sometimes the ability to affect how Eakin (2001) have found that social the work is organised. power and its uneven distribution Workers in precarious shapes the variance in health employment relationships lack outcomes, rather than decision control over these critical areas of latitude or control at work. Other work, relative to employees in studies further suggest that social standard employment relationships. support outside of the workplace Precarious employment can positively buffer “job strain” relationships involve radically and its health effects (Johnson 1991). different power relationships, since We build on these ideas to argue limited (or non-existent) contractual that the standard employment rights apply, and few stable relationship, and consequently accepted customs and practices Karasek’s notions of control and exist. Precarious employment workload, assumes a very specific relationships create new types of allocation of power and support at “control” uncertainty over access to work and in society. Some work future work, level of income and organisation researchers have benefits, location of work, work looked at precarious employment schedules, who one works with, but have not, we argue, adequately one’s supervisors and even the jobs understood precarious employment one must accept. Workers in relationships. Goudswaard and precarious situations may also face Andries (2002), for instance, suggest increased workload and effort that while precarious employment associated with searching for work, results in less control, it involves travel time between multiple jobs at fewer demands compared to multiple sites, and constantly permanent work. Cooper (2002) has adapting to new work locations, co- suggested that the rise of a “short- workers and supervisors. They are term contract culture” actually gives also likely to bear more employees more control and choice responsibility for their own training over their working lives, perhaps and work related equipment. 28 JUST LABOUR vol. 3 (Fall 2003) Forum on Precarious Employment

To incorporate these elements of care workers who did not know in the organisation of precarious advance how many hours, or even employment into the “job strain” whether they will work each week. model of workplace health, we have Auto parts workers have felt developed the idea of “employment insecure about their job futures since strain” which includes seven new the recession and layoffs of the early control and workload variables. 1990s. Employees in smaller They are illustrated in Table 1. We workplaces, many of which are propose that “employment strain” handling sub-contracted work, have captures a dimension of work fewer basic employment standards organisation that has a health effect protections. We see the that is independent from employees’ development of the notion of experience with any one workplace. “employment strain” as a tool that We expect that the uncertainty will allow us to assess the effects of associated with precarious the erosion of the standard employment increases stress and employment relationship and the work-life conflicts, increases overall continuum of precarious workloads, and that this may in turn employment that is growing in the lead to low social support and poor Canadian labour market. health, regardless of the level of “job strain”. We think that high levels of PRECARIOUS EMPLOYMENT employment strain can be correlated AND HEALTH – THE SURVEY with high levels of job strain, but this need not be the case in all Over the last year we have designed sectors. Where there is a strong and conducted a survey to examine connection between employment “employment strain” with over 400 strain and job strain, we would workers in precarious employment expect employment strain to relationships in Ontario. Survey compound the health effects respondents include contract and associated with job strain. part-time workers and workers While the notion of employment hired through employment agencies strain is based in the realities of in home care, construction, social precarious employment, we services, manufacturing and recognise that workers in so-called support. The results reported here permanent jobs also experience are based on a preliminary analysis these uncertainties. In reality, many of 137 surveys. They indicate that jobs that are called “permanent” no stress related health issues are longer have all of the characteristics significant for workers in precarious of a standard employment employment relationships and begin relationship. In the course of the to provide a description of the study we have, for instance, met components of employment strain. with “permanent”, unionized home Lewchuk, de Wolff and King 29

Table 1: Employment Strains

Control / Components Demand Measured By

1. Employment Control over · average length of contracts Uncertainty access to · perceived uncertainty regarding current work employers offering more work · presence / absence of a union to enforce workplace rights · perceived influence of day to day work performance and attitude evaluations on future offers of work · favouritism in getting new work. 2. Earnings Control over · presence/absence of written pay records Uncertainty future · EI/CPP deductions from earnings earnings · whether employee is paid when sick · whether employee is paid on time · the degree to which employee can plan on future earnings 3. Household Control/ · number of dependants in household Precarious-Ness demand · individual and household earnings providing · individual and household benefit coverage basic needs

4. Control over · length of advance notice of work schedule Scheduling work · number of hours to be worked Uncertainty schedule and hours 5. Location Control over · number of work locations Uncertainty work location · length of advance notice of work location

6. Task Uncertainty Control over · perceived influence of day to day use of skills evaluations of attitude over work tasks and job assigned assignment · the number of different supervisors and groups of co-employees · frequency of working in an unfamiliar location. 7. Employment Demand · time spent looking for work Uncertainty required to · time spent travelling between jobs Workload manage · conflicts from holding more than one job. employment uncertainty 30 JUST LABOUR vol. 3 (Fall 2003) Forum on Precarious Employment

Our respondents were a mid-aged precarious employment respondents group of workers, just over half reported poorer health than (55%) were female and over one- employed Canadians who were quarter (27%) were immigrants who asked the same question. had been in Canada for less than ten Figures 2, 3 and 4 compare levels years. They were highly educated: of pain, tension and exhaustion 45% had some university education, between our respondents and while only 10% had less than workers in standard employment secondary education. Most were in relationships at a large unionized short-term employment: 58% had manufacturing plant in Ontario. short-term jobs; 20% had part-time Both groups were asked the same jobs; 17% had full-time jobs that questions to allow for this were either temporary or where comparison. As Figure 2 shows, their weekly hours were not actually working with pain appears to be a assured; and 3% were self larger concern for workers in the employed. One in five (20%) standard employment relationship belonged to a union. Just under half than for precariously employed (46%) had relied on precarious respondents. And, while 20% employment for a year, while 16% reported working with severe pain had been working in precarious jobs in the previous month, this was less for over five years. On average they than the 30% of the workers in had worked for 1.7 employers in the standard employment relationships previous month. who worked with severe pain. We Figure 1 compares the self- can hypothesise older workers, reported health status of the employed in more standard respondents with the responses of employment relationships, may Canadians in the National work in pain more than younger Population Health Survey 1998. Our workers in more precarious

Figure 1: Self Reported Health, Canada, 1998, Age 18 to 64

45% Precarious respondents 40% Canada, NPHS 35% 30% 25% 20% 15% 10% 5% 0% Excellent Very Good Fair/Poor Good Lewchuk, de Wolff and King 31

Figure 2: Days Working in Pain Last Month

45% Precarious Respondents 40%

35% Workers in Standard Employment Relationship 30%

25% 20%

15% 10%

5%

0% Every Most Half Few Never day days days

Figure 3: Days Tense at Work Last Month

60% Precarious Respondents

50% Workers in Standard Employment Relationship 40%

30%

20%

10%

0% Every Most Half Few Never day days days

Figure 4: Days Exhausted After Work Last Month

45% Precarious Respondents 40% 35% Workers in Standard Employment Relationship 30% 25% 20% 15% 10% 5% 0% Every Most Half Few Never day days days 32 JUST LABOUR vol. 3 (Fall 2003) Forum on Precarious Employment employment. income is not predictable for more The health issues that appear to than half (58%) of the group. Many be more severe for workers in are reliant only on their own precarious employment are stress- resources for any “insurance” when related tension and exhaustion. they are sick: 64% lose earnings Figure 3 indicates that our when they are sick and 21% were respondents were tense at work not covered by Employment more often than those in Insurance or Canada Pension Plan. manufacturing jobs. Figure 4 shows Significantly, while 71% lived with that exhaustion was more frequently another adult, workers in this group a concern for more workers in are not relying on a partner’s secure precarious employment than for standard employment relationship manufacturing workers. at home. They belong to precarious households, where other earners EMPLOYMENT STRAIN raise the household income only UNCERTAINTIES slightly, and where two-thirds have no household drug plan, 78% have Tables 2 to 7 begin to describe the no vision plan and 81% have no characteristics of employment strain pension plan (Table 4). among our respondents. Table 2 Working in multiple locations shows that their level of uncertainty with different schedules results in about future employment was high, uncertainties (Table 5) that also even though 20% were unionized. create stress and disruption in part Many didn’t have a written contract, because of unfamiliarity with the a surprising number were working workplace, but also because they on weekly contracts (9%), and 41% create other forms of managing did not know if their current work, like rescheduling childcare employer would offer them more and finding the appropriate work. Ongoing formal and informal transportation. Our survey performance evaluations, respondents worked either favouritism and discrimination all sequentially or in multiple feature significantly in respondents’ workplaces (2.2 on average). Less understanding of whether their than half of precariously employed employer will offer them more respondents (48%) knew their hours, longer contracts or new schedules a week or more in contracts. advance half of the time; 40% didn’t Table 3 and Table 4 depict a get much more than a day’s notice group where 72% are earning at a half of the time. level that is close to poverty in Table 6 suggests that task Ontario, particularly for the 29% of assignments and co-workers respondents who were supporting changed consistently: on average children. And even that level of respondents had 2.2 supervisors Lewchuk, de Wolff and King 33

Table 2: Employment Uncertainty Table 3: Earnings Uncertainty Length of contract 40% no contract Can plan on same 58% disagree 9% < 1 week income in 6 months 40% < 6 months Bargain collectively 20% Lose pay when sick 64% Current employer 41% unlikely or don’t Receive written 35% not all offer more work? know record of pay workplaces Evaluation of 21% a lot Paid on time 9% less than half the performance affects time hours of work Evaluation of 17% a lot Pay EI/CPP 21% no attitude affects hours of work Favouritism affects 31% al lot hours of work

Table 4: Precarious Households Table 5: Scheduling and Location Uncertainty Income from work 72% < $25,000 One or more weeks 48% half the time or notice of work less schedule Household income 68% < $35,000 One or more days 40% half the time or notice of work hours less Benefits 48% no benefits Advance notice of 34% half the time or work location less Pension plan 84% no plan Average number of 2.2 work locations last month Household pension 81% no plan Work in unfamiliar 19% at least weekly plan location last month Household drug plan 66% no plan Household vision 78% no plan plan

Table 6: Task Uncertainty Table 7: Employment Uncertainty Workload Worked with 46% at least half the Days spent looking 32% at least half the different co-workers days for work last month days last month Average number of 2.2 Daily unpaid travel 38% two or more supervisors last to work hours per day month Conflicting demands 14% at least half the by multiple time employers Conflicting demands, 11% at least half the multiple work sites time Do extra things to get 10% at least half the more work time 34 JUST LABOUR vol. 3 (Fall 2003) Forum on Precarious Employment over the previous month, and almost characteristics of precarious half had worked half of the previous employment and health outcomes. month with different co-workers. Table 7 illustrates the workload of ACKNOWLEDGEMENTS managing the uncertainty of whether, where and when one was The authors would like to thank Niki Carlan, going to work. This workload is Simon Enoch, Brian Gibson, Erika Khandor specific to precarious employment. and Syed Naqvi for their contributions to the research upon which this paper is based. One-third of this group had spent at The research was funded by the Social least half their days in the previous Sciences and Humanities Research Council month looking for work, over one- and is part of the Community University third spent significant unpaid time Research Alliance on Contingent getting to the work they had, and Employment (ACE). some spent significant time managing conflicting demands by REFERENCES multiple employers or multiple worksites. Brooker, Ann-Sylvia and Eakin, Joan 2001. Gender, Class, Work Related Stress and Health: Toward a Power Centred CONCLUSIONS Approach. Journal of Community and Applied Social Psychology 11, 97-109. We have developed the concept of Cameron, Barbara 2001. The Occupational Health and Safety Implications of “employment strain” to attempt to Nonstandard Employment. A Report to the capture the unique characteristics of Ontario Workplace Safety and Insurance precarious employment. Preliminary Board, September. evidence suggests that workers in Cooper, C.L. 1998. Theories of Organizational precarious employment Stress. Oxford, Oxford University Press. Cooper, C.L. 2002. The Changing relationships report poorer overall Psychological Contract at Work. health than other workers in the Occupational Environment Medicine 59, 355. National Population Health Survey Goudswaard, A. and F. Andries 2002. and higher levels of stress than Employment Status And Working Conditions. Dublin: European Foundation for the workers in standard employment Improvement of Working and Living relationships. They face high levels Conditions. of uncertainty regarding access to Johnson, J.V. 1991 Collective Control: work, the terms and conditions of Strategies For Survival In The Workplace. that work, and future earnings. In J. Johnson and G. Johansson (eds), The Psychosocial Work Environment: Work They engage in additional effort , Democratization And Health. searching for work and balancing Amityville: Baywood Publishing the demands of multiple employers. Company, pp.121-132. They have low earnings, few Karasek, Robert 1979. Job Decision Latitude, Job Demands and Mental Strain: benefits, and reside in low-income Implications for Job Redesign. households. Future research will Administrative Science Quarterly, 24(June): explore the links between these 285-308. Lewchuk, de Wolff and King 35

Karasek, Robert and T. Theorell, 1990. Healthy Work, Stress, Productivity and the Reconstruction of Working Life. New York: Basic Books. Quinlan, Michael, Claire Mayhew and Philip Bohle 2001. The Global Expansion of Precarious Employment, Work Disorganization, and Consequences for Occupational Health: A Review of Recent Research. International Journal of Health Services, 31(2): 335-414. Vosko, L., N. Zukewich and C. Cranford 2003. Nonstandard Work and Precariousness in the Canadian Labour Market. Perspectives on Labour and Income, Fall (forthcoming). Wall, T. et al 1996. The Demands-Control Model Of “Job Strain”: A More Specific Test. Journal of Occupational and Organizational Psychology 69: 153-66. Walters, David and Frick, Kaj 2000. Worker Participation and the Management of Occupational Health and Safety: Reinforcing or Conflicting Strategies. In K. Frick et al (eds), Systematic Occupational Health and Safety Management. Oxford: Elsevier Science Ltd (Pergamon) and Oxford University Press, pp.42-65.