Labour / Le Travail
Issue 90 (2022)

Research Note / Note de recherche

Profits First, Safety Second: Canada’s Occupational Health and Safety System at 50

Jason Foster, Athabasca University
Susan Cake, Athabasca University
Bob Barnetson, Athabasca University

Abstract: This article argues that while workplaces are safer today than they were 50 years ago, the degree to which this change is due to Canada’s occupational health and safety (ohs) system is unclear. Examining the literature and reflecting upon the authors’ own experiences with workplace safety, the article suggests that fundamental flaws embedded in the principles of the system undermine its effectiveness at keeping workers safe. Specifically, the premise of joint responsibility – which is given life in the internal responsibility system (irs) – appears to ignore the conflicting interests and unequal power relations that exist in Canadian workplaces. The circumstances that contributed to the historical effectiveness of the irs no longer exist, undermining the ability of workers to realize safe and healthy workplaces.

Keywords: occupational health and safety, history of workplace safety, workplace injury, internal responsibility system, ohs reforms

Résumé : Cet article soutient que même si les lieux de travail sont plus sécuritaires aujourd’hui qu’ils ne l’étaient il y a 50 ans, la mesure dans laquelle ce changement est attribuable au système de santé et de sécurité au travail du Canada n’est pas claire. En examinant la documentation et en réfléchissant sur les propres expériences des auteurs en matière de sécurité au travail, l’article suggère que des defaults fondamentaux intégrés aux principes du système compromettent son efficacité à assurer la sécurité des travailleurs. Plus précisément, la prémisse de la responsabilité conjointe – qui prend vie dans le système de responsabilité interne – semble ignorer les conflits d’intérêts et les relations de pouvoir inégales qui existent dans les lieux de travail canadiens. Les circonstances qui ont contribué à l’efficacité historique du système de responsabilité interne n’existent plus, minant la capacité des travailleurs à créer des lieux de travails sûrs et sains.

Mots clefs : santé et sécurité du travail, histoire de la santé et sécurité du travail, lésions professionnelles, système de responsabilité interne, réformes de la santé et sécurité du travail

It has been 50 years since Canada’s first modern occupational health and safety (ohs) law was enacted in Saskatchewan. Since that time, Canadian accident rates have gone down. Yet, despite improvements in workplace safety, hundreds of thousands of workers in Canada continue to be injured at or because of work each year. And the centrepiece of Canadian legislation – the internal responsibility system (irs) – does not appear able to effectively address unsafe and unhealthy workplaces.¹ This growing failure reflects a profound erosion of the circumstances that once allowed the irs to act as an effective check on employers trading workers’ health for profits. Declining state regulation, weakening union power, hazards that are more complex and costly to control, and the growth of precarious employment all contribute to the diminished utility of the irs.

An all-too-tragic illustration of this failure can be seen in the covid-19 outbreak at the Cargill meat-packing plant in High River, Alberta, in the spring of 2020, which infected almost 1,000 and led to the deaths of two workers.² The outbreak and subsequent illness and death were a direct result of a breakdown in the irs and the inability of the ohs system to adequately protect the lives of these mainly racialized and unionized workers. Despite the workers’ efforts to utilize their rights under the irs, they were unable to get the employer or the government to act until the outbreak had reached a crisis point. Indeed, it appears the government colluded with the employer to hide the degree of hazard facing the workers and to encourage them to continue working.³

As the Cargill case highlights, and as will be discussed in this article, the irs affords employers a great deal of control over whether and how to control hazards. This power, combined with the profit motive, means safety is prioritized only when it is consistent with improving the employer’s bottom line. Profits take precedence over safety. As a consequence, hundreds of thousands of serious workplace injuries continue to occur each year.

This article examines the state of Canada’s ohs system 50 years after its inception through a political economy lens that focuses attention on interests, power, and the allocation of benefits. In this view, ohs is an arena of contestation between labour and capital, with the degree of safety being determined by the relative power of the actors. This approach differs from a view of ohs as a technical activity where the actors share an interest in a safe workplace. This political economy approach allows us to analyze how structural shortcomings in the system and contextual changes have limited its effectiveness in protecting workers’ health. This article also discusses some options for reforming the existing ohs system to better protect workers, including enhancing existing worker rights under the system, bolstering the power of workers to act in a concerted fashion, incentivizing citizens to report non-compliance, and increasing direct action by workers in defence of their own safety.

Canada’s OHS System

The central premise of Canada’s contemporary ohs system is that almost all injuries can be prevented if employers organize (or reorganize) work in ways that control the hazards to which workers are exposed. This principle reflects the fact that most injuries are the result of exposing workers to hazards in the workplace, rather than the result of worker incompetence or (mis)behaviour. In practice, however, the conflicting interests of labour and capital around the extraction and distribution of the surplus value of labour are a significant barrier to effective injury prevention. Specifically, improving safety often erodes employer profitability by, for example, increasing input costs and/or slowing production. Consequently, employers can be reluctant to control workplace hazards unless the financial return on the controls is positive.⁴ As Peter Dorman notes, the nature of cost accounting means the cost of controls is visible and accrues to the employer. By contrast, the value of a healthy and safe workplace is very difficult to quantify and record and mainly accrues to the workers.⁵ This dynamic, in conjunction with the profit imperative, encourages employers to adopt a cost-benefit approach to ohs. The contemporary use of incentives in ohs (e.g. experience-rated workers’ compensation premiums, financial penalties) seeks to alter employers’ cost-benefit analyses.

Not surprisingly, the safety of work has long been a zone of contestation between workers and employers. In the latter half of the 19th century, workers protested unsafe working conditions, sought compensation for injuries through the courts, and staged walkouts and rallies seeking government intervention.⁶ The state responded to risks created by new technologies and working arrangements that emerged from industrialization by regulating – often ineffectively – steam-powered machinery, railway construction and operations, mining, and factories (with particular emphasis on regulating the work of women and children).⁷ Growing unrest led to a Royal Commission in 1886 that recommended state intervention to address some of the most egregious safety issues. Recommendations included inspection and regulation of factories, railways, and ships to improve workplace safety and to compensate injured workers.⁸ Action on the recommendations was slow. In 1899, the federal government instituted some regulations for the federal public service. Workers had to wait until 1914 for the creation of a system for compensating injured workers.⁹ In the following five decades, few significant steps were taken by governments to regulate workplace safety.

The next major overhaul of occupational safety in Canada came in the 1970s with the emergence of our contemporary health and safety system. This change was also the direct result of worker protest. A series of strikes across Canada in the 1960s centred on lack of safety protections for workers. A growing movement of worker safety activists emerged in the late 1960s and early 1970s to make workplace safety a political issue.¹⁰ In 1971, the ndp government of Allan Blakeney in Saskatchewan established a task force to implement a new approach to regulating workplace safety. Following the approach set out in the United Kingdom’s Robens Report, the province’s subsequent Occupational Health and Safety Act became the first legislation in North America to regulate workplace safety across all industries with a system designed to codify specific rights and obligations for workers, employers, and the state.¹¹ All provinces eventually responded to rising worker concerns about safety and introduced similar legislative schemes. Commenting on these new legislative regimes, critical scholars would later assert that in addition to making workplaces safer, Canada’s ohs system channelled class conflict into a highly legalistic but low-cost system, wherein workers’ power to force safety improvements (by disrupting the capital accumulation process) was attenuated.¹² This new approach to injury prevention maintained the long-standing focus on controlling obvious hazards (e.g. those that could result in acute, physical injury) while failing to engage with less obvious hazards (e.g. those giving rise to occupational diseases or mental health conditions).

A key component of this new legislation was the creation of the internal responsibility system. The irs assumes workers and employers share an interest in safe and healthy workplaces and divides responsibility for achieving that between employers, workers, and the state. Employers, owing to their control over the workplace, are deemed to have the primary obligation for ensuring a safe workplace by identifying and controlling hazards. The degree to which employers must control hazards is, however, limited to controls that are reasonably practicable to implement. The reasonably practicable standard means employers can weigh the risk posed by a hazard against the cost of specific control strategies and decline controls where the cost is disproportionate to the risk. Employers are expected to engage with workers to address safety issues.

Workers’ role in the irs is mostly consultative and includes an obligation to comply with employer-mandated safety rules. To bolster workers’ ability to meaningfully participate in the irs, governments granted workers three safety rights: to know, to participate, and to refuse. The right to know requires employers to provide workers with necessary information about workplace hazards, such as providing material safety data sheets for workplace chemicals as well as copies of hazard assessments. Since this information comes from the employer, the employer has an opportunity to decide, to some degree, what information is shared and to shape the meaning of it before it reaches workers.

The right to participate is designed to ensure workers an avenue through which to express concerns about safety and, ideally, to assist in resolving issues. The primary mechanism for this right is the establishment of joint health and safety committees (jhscs) where workers comprise a minimum of 50 per cent of members and the committee is granted certain powers, such as the right to investigate incidents and inspect the workplace. Importantly, the committees can only recommend actions; the final decision about implementing controls remains with the employer.

The right to refuse dangerous work is the worker failsafe in the irs. If the other processes fail to properly control a hazard that endangers the worker, a worker can refuse unsafe work, without fear of retaliation, until the work is deemed safe. According to Robert Sass, the creator of Saskatchewan’s ohs system in the 1970s, “Such a refusal was seen as the most effective way for a worker to raise a problem of health and safety so that it had to be confronted and dealt with before the working conditions produced an injury or health hazard.”¹³ As designed, the three rights were seen as interdependent and mutually necessary. According to Sass, “All three rights act like different gears in a complex mechanism: remove one and the machinery breaks down, becoming immobile and useless.”¹⁴

The state’s role in the ohs system is to establish legal obligations, educate workers about their rights, investigate serious incidents, and police employer compliance, through inspections (whether random, targeted, or in response to complaints) and, less commonly, sanctions. In this way, state enforcement backstops the irs, which is designed to be the primary mechanism of injury prevention.¹⁵ Off-loading responsibility for hazard control and injury prevention to employers and workers was intended to address the limits of state-driven regulation. The continued presence of state enforcement recognizes that despite the assertion that workers and employers share an interest in safer workplaces, there will be instances where their actions and interests conflict. Over time and between jurisdictions, Canadian governments have exercised their enforcement role unevenly, and the number of inspections and prosecutions has fluctuated significantly.¹⁶

Changes in Work since 1971

Over time, the workplace circumstances for which the irs was designed have changed significantly and thereby shifted the underlying risk factors for workers. Generally, Canada has witnessed a shift in economic activity away from manufacturing and primary industries and toward a greater focus on services and knowledge. Deindustrialization has shifted the underlying risk factors as work has moved to workplaces with different hazard profiles, including fewer or lower severity hazards as well as hazards that give rise to injuries with longer latency periods and murkier causality. Deindustrialization also reflects, in part, the growth of multinational trade agreements alongside the economic development of the Global South.¹⁷ In this context, governments and employers have faced pressure to minimize production costs, including those associated with worker safety. This has resulted in a shift in state injury-prevention strategies, including a declining willingness to act as a backstop to the irs.¹⁸

Another significant change is in the labour relations landscape in Canada, which has seen a decline in union density since the 1980s, particularly for men and young workers.¹⁹ In part, these changes are related to deindustrialization, whereby jobs that traditionally had high unionization rates are replaced by jobs with lower unionization rates. Research has found that the irs generally works better in unionized work sites, likely because unions provide a countervailing source of worker power.²⁰ Although Canada’s unionization rate has never exceeded 38 per cent, many sectors did (and do) exceed this average. Some researchers suggest that unionization helps reduce the power imbalance in the workplace and is “an unstated premise of the system” that is required for the irs to function properly.²¹

Declining unionization has been paired with the growth in Canada of precarious employment, which provides workers with less job stability, lower pay and benefits, and less access to statutory protections.²² Not surprisingly, precarious workers report more injuries than do workers in more standard work relationships. They also report more stress and ill health resulting from their job insecurity, but their lack of access to sick leave masks their true level of ill health.²³ Similarly, while precarious workers face a greater level of risk in the workplace, their concerns are often not voiced (for fear of retaliation) or not addressed.²⁴ That said, research has found that the ohs concerns of precarious workers are more likely to be addressed when the concerns resemble those found in traditionally male-dominated and highly unionized sectors like mining and heavy industry, leaving women, racialized workers, and others more vulnerable. Concerns such as stress, harassment, and even insecure employment are rarely addressed or even recognized as legitimate ohs matters under the irs.²⁵

Are Workers Safer Today?

Research suggests that at a macro level, Canada’s workplaces appear to be safer today than they were in 1971. Equipment, procedures, and practices that were commonplace in the later 1960s and early 1970s are widely rejected today as irresponsible by most employers and workers. Injury data reflects these changes. In 1974, before most safety legislation had taken hold, the number of industrial incidents resulting in a recorded injury in Canada topped 1,000,000 for the first time, following a steady climb for 40 years.²⁶ Under the current system, the trend has reversed. The most complete data set we have for tracking occupational injuries begins in 1982. In that year, there were 479,558 accepted lost-time claims, which represent the most serious injuries.²⁷ Lost-time claims rose to 620,979 in 1989 before entering a long decline to reach 271,809 in 2019.²⁸ Total claims, which include less serious injuries and injuries “managed” without time loss, have also trended downward, with approximately 650,000 claims in 2018.

While this data is encouraging, it is worth remembering that workers’ compensation claims data undercounts injuries; some workers are excluded from the ambit of workers’ compensation, and only serious injuries resulting in a workers’ compensation claim must be reported.²⁹ Reportable injuries themselves are subject to underreporting in the range of 40 per cent to 69 per cent, in part due to employer suppression.³⁰ Further, occupational diseases and psychological injuries have historically been subjected to both significant underreporting and high rates of claims denial.³¹

With these concerns in mind, the data does suggest a significant (43 per cent) reduction in accepted lost-time claims between 1982 and 2019. This reduction may, in part, reflect the long-term shift in employment toward less physically hazardous occupations and changes in production and management techniques that have reduced the risk of injury (e.g. automation, safety design, training). Thus, this apparent improvement in safety needs to be viewed in context. Important contextual factors include a 49 per cent increase in the population and a gradual expansion in the kinds of injuries deemed compensable, both of which suggest a more significant improvement in safety than the raw numbers alone reveal.³² By contrast, the improvements in workers’ compensation data may, in part, reflect government and employer efforts to convert lost-time claims into modified work claims, which often do not have to be reported, and thus lower the apparent “injury” rate without appreciably improving safety.³³

Workplace fatalities are similarly complex. The number of work-related fatalities has grown slowly over the past few decades, stabilizing since 2010 at 900 to 1,000 per year.³⁴ Nevertheless, fatalities per 100,000 workers have dropped from approximately 11 in the 1970s to 7 in the 1990s to under 5 today.³⁵ This data is subject to most of the same caveats as lost-time claim data. Additionally, fatality statistics only include deaths officially recognized by workers’ compensation boards as work related, thereby excluding many categories of work-related fatalities. Steven Bittle, Ashley Chen, and Jasmine Hebert estimate that the actual number of work-related deaths in Canada “is at least ten to thirteen times higher than the approximately 900 to 1,000 annual average fatalities reported by the awcbc [Association of Workers’ Compensation Boards of Canada]. This makes work-related fatalities one of the leading causes of death in this country.”³⁶ The study attributes this discrepancy to unrecognized occupational illnesses, excluded workers and industries, fatalities during commuting, deaths of “non-workers” in workplace incidents, and numerous other factors.

Even considering the concerns about data collection and reporting (which are significant), the official data suggests that Canadian workplaces are safer today than they were in 1971 when the ohs system was enacted. It is unclear how evenly safer working conditions have been distributed among workers in different sectors. The ascribed characteristics of workers (e.g. gender, ethnicity) as well as their right of residency and degree of employment precarity may make some workers more likely to experience unsafe work or less able to access their ohs rights than others.³⁷ Still, most workers today are relatively safer when compared with their predecessors in the 1970s. That being the case, two immediate questions emerge. First, how much of the increased safety in workplaces can be attributed to the existing ohs system, and to the irs specifically? Second, would workers be safer if federal and provincial governments implemented a reformed or alternative system?

The IRS Impact on Safety

To answer the first question, we need to consider the role of economic change in creating safer workplaces. We also need to examine more closely the irs in practice to ascertain its specific impact on safety. The role of economic change cannot be understated in the transformation of Canadian workplaces. As discussed, deindustrialization, increased automation, and technological advances have had the effect of removing more dangerous types of work and more effectively insulating workers from risks of injury. The nature of work performed in mines and factories entails greater risk for injury than does work performed in offices or retail stores.³⁸ This is not to say the latter workplaces are free of safety hazards – new forms of employment create new types of hazards. But there is an intuitive logic that an economy that employs fewer workers on farms, in factories, and in mines will produce fewer work-related injuries and fatalities.

It should also be noted that the technological and macroeconomic changes witnessed in the last couple of decades that led to reduced injury took place primarily in the interests of profit rather than safety. Technology lowered the cost of production or increased productivity; increased safety was of secondary concern. Deindustrialization moved much of the “unsafe” work formerly performed by Canadian workers to cheaper jurisdictions of the Global South, exporting the safety hazards with it. Disentangling the relative impacts of economic change on workplace safety is an impossible task, as the forces driving innovation and change cannot be fruitfully isolated from behaviour incentivized by government rules and regulations. Consideration of these broader changes is a reminder to avoid reading too much into the multiple-decade reduction in work-related injury and fatality rates.

A more fruitful way to evaluate the effectiveness of the irs is to examine how it has worked in practice over time and, specifically, to analyze workers’ experience with the irs and exercising of their ohs rights. Researchers have identified several recurring problems. Despite improvements over time, a sizable minority of workers face barriers to ohs participation based on lack of awareness of their ohs rights (24 per cent) and/or lack of empowerment to exercise them (35 per cent).³⁹ Although the strength of the relationships varies, workers who are employed on a temporary basis, who were born outside of Canada, whose first language is not English (or French in Québec), who are younger, and who are employed in smaller workplaces typically exhibit greater ohs vulnerability.⁴⁰ A slight majority of workers (58 per cent) receive some form of ohs training each year; however, only a minority of new workers receive training.⁴¹ Still, workers often do not have access to basic health and safety information (such as hazard assessments) in the workplace.⁴² Absent basic awareness and empowerment, workers may have difficulty avoiding injury as well as meaningfully participating in injury prevention.

Joint health and safety committees are the main mechanism through which workers exercise the right to participate, although these committees typically exist only in larger workplaces (usually those with more than twenty workers). In many jurisdictions, smaller workplaces are required to have a worker health and safety representative. Where jhscs exist, their effectiveness appears mediated by union representation, involvement of workers, management attitudes, and the degree of external regulation.⁴³ The most salient criticism of jhscs is that they lack the authority to compel employers to act on safety issues.⁴⁴ This limit can be somewhat offset by the behaviour of worker representatives, but employers continue to have an oversized role in determining which hazards are identified and how they are controlled.⁴⁵ The existence of jhscs also channels worker energy into a system that manages discontent while delegitimizing the use of direct action to address safety issues.

Workers infrequently use their right to refuse when confronted with unsafe work, citing fear of retaliation.⁴⁶ While employers are prohibited from disciplining workers for exercising their right to refuse, retaliation does occur and workers may have difficulty seeking redress.⁴⁷ When workers do exercise their right to refuse, employers may apply pressure to the worker or simply assign the work to another worker who is more compliant.⁴⁸

Other features of the irs further undermine workers’ ability to exercise their ohs rights. In its design, the irs was intended to establish a direct relationship between workers and their employer regarding safety issues, where the two parties would collaborate to find workplace-specific solutions. The role of workers has diminished as ohs has become increasingly technical and professionalized. Technical and scientific information is privileged in discussions of safety, and highly specialized tools for measuring hazards (e.g. audiometers, chemical analysis devices) have been developed that require specialized training. An ohs profession has emerged in Canada over the past 50 years that sees its practitioners as objective and neutral experts on safety, separated from the employment relationship.⁴⁹ Unions have contributed to this development through the hiring/appointment of ohs representatives with specialized training and a focus only on ohs (rather than on broader collective representation matters).⁵⁰

While a reliance on technical information is, to a degree, inevitable given the nature of ohs issues, the emergence of a professional class of ohs experts has made the worker role more passive. In the 1970s and 1980s, ohs advocates were “knowledge activists,” emphasizing worker knowledge and worker organizing over technical information.⁵¹ Professionalization leads to the marginalization of workers’ knowledge as “scientific” knowledge is privileged and silences workers by discouraging organizing. Using the veneer of “objectivity,” professionals have entrenched a form of cost-benefit analysis in ohs, wherein the benefits of controlling a hazard are balanced against the control’s cost to the employer.

A further impediment to the effective operation of the irs has been uneven and declining state enforcement of ohs laws. State enforcement is designed to act as a backstop to the irs. Government willingness to resource enforcement and apply meaningful sanctions to non-compliant employers has varied over time and between jurisdictions.⁵² Dorman suggests that there is a trend away from state inspection because the hazards that were amenable to this type of regulation and remedy were largely addressed. The hazards that remain are typically more difficult to identify and control, and states are emphasizing various forms of self-regulation by enterprises and industries.⁵³

Some governments have introduced more market-oriented incentives (e.g. experience rating) designed to shape employer decision-making about ohs. These sorts of incentives are vulnerable to employers’ engaging in cost-shifting strategies, such as claims suppression, rather than the expected cost-reducing strategies (i.e. making workplaces safer). Other governments have emphasized collaborative self-regulation. This includes having employers establishing and interpreting regulatory requirements.⁵⁴ In this approach, industry safety associations – funded by employers, often through workers’ compensation board premiums – take an active role in education, workplace auditing, and acting as the “voice” of industry on safety matters. Increased influence of industry associations has occurred alongside (and facilitated) the professionalization of ohs in Canada, which has intensified the ohs profession’s alignment with employer interests.

Workplace violence offers an important opportunity to examine the impact of employer involvement in regulating hazard recognition and control.⁵⁵ Since 2000, media attention on worker fatalities in the convenience store industry has generated pressure on governments to address risks caused by “gas-and-dash” and workplace violence. Pay-before-you-pump legislation is an effective control of the hazards arising from gas-and-dash and has been adopted by several jurisdictions. A pay-before-you-pump system entails little to no extra direct cost for employers and reduces revenue loss from theft. In this way, gas-and-dash is an easy, even desirable hazard to control. At the same time, employers have evaded (costly) controls that would protect workers from violence on the job (such as installing barriers and restricting employee movement throughout the store) through subtle reframing. Specifically, employers have supported controls aimed at reducing robberies (such as cash control and good lighting) while misleadingly labelling them as violence-prevention initiatives. Only a small portion of worker injuries in convenience stores is related to robbery. Controlling only the robbery hazard leaves serious risk associated with violence uncontrolled. This is hard for most workers and the public to see because of the mislabelling of the controls. Governments have gone along with this sleight of hand because of employer pushback over the infrastructure and operational costs of controls that would prevent non-robbery-related violence.⁵⁶

The combination of worker underutilization of their rights, the distortion of irs principles, and the ineffectiveness of government enforcement has led to the system’s weakened ability to make workplaces safer. The theorized benefits of having workers and employers work co-operatively to improve workplace safety do not materialize when either workers do not use their rights under the system or employers sidestep the system by utilizing other mechanisms for addressing ohs issues.

Little research has investigated the link between the irs and reduced rates of injury. One recent study explicitly attempted to find a link between injury and the use of irs rights. The study, conducted for the Government of Alberta as part of an effort to construct performance measures for the irs, consisted of a survey of 2,000 Alberta workers.⁵⁷ The researchers constructed twelve measures of employers’ compliance with ohs rules and workers’ engagement with the irs, broken into the three safety rights. They found no significant correlation between engagement with the irs and rates of injury. In other words, at workplaces where the irs processes were functioning as intended (i.e. employers complied with regulatory requirements and workers actively used their rights), workers reported just as many injuries as those in workplaces where the irs was ignored. Injury rates were better explained through other variables, including industry and occupation, exposure to hazards, and workers’ demographic characteristics.⁵⁸

While this study has a number of limitations, including being restricted geographically and relying on participant self-reporting, it is useful in highlighting two things. First, it confirms a high level of employer non-compliance with basic irs requirements. Between 25 and 50 per cent of employers did not engage in the basic elements of irs, such as providing safety information and training, including workers in hazard assessments, and possessing a functioning joint committee.⁵⁹ This tells us that in a significant minority of these workplaces, the irs has not yet been fully implemented, 50 years after its inception. Second, the study reveals that a functioning irs system in a workplace does not lead to reduced injuries. This finding is particularly damning. If irs worked as intended, we should expect to see a difference between compliant and non-compliant employers.

David Walters and Theo Nichols helpfully summarize the international evidence that worker representation can reduce the risk of workplace injury, often through the intermediary step of improving ohs management practices. Preconditions for effective worker representation include strong legislative provisions and state enforcement, management commitment, competent hazard assessment and control, and effective worker representation, including union support.⁶⁰ Declining interest in state enforcement and lower unionization coupled with a shift toward more precarious employment (which is emblematic of employers prioritizing profit making) and limited employer compliance with even basic legislative safety requirements have all eroded the preconditions in Canada under which the irs was developed and is most likely to operate effectively. This, in turn, suggests that workers may have little ability to achieve positive outcomes, despite having safety rights and access to elements of the irs.

A Case Study in Internal Responsibility Failure

A single, compelling case underscores the shortcomings of the irs in improving safety. The Cargill meat-packing plant in High River (a small town south of Calgary) employs 2,000 workers, many of them recent immigrants and migrant workers. It is unionized by the United Food and Commercial Workers, Local 401. In spring 2020 this plant became the location of the largest covid-19 outbreak in Canada, with almost 1,000 workers becoming infected and two dying. The size and severity of the outbreak were due to a series of breakdowns of the irs and public health systems in the workplace and the province.

Meat-packing is physically demanding, fast-paced work that exposes workers to a wide range of serious safety hazards, including sharp objects, biological hazards, repetitive strain, and physical exertion.⁶¹ Workers work in close quarters and plants often have poor ventilation and climate control. These working conditions mean the emergence of the covid-19 virus became, for these workers, a serious and pressing biological hazard.

Early in the outbreak, Cargill unilaterally announced a series of measures to address the covid-19 risk, including screening and temperature checks, rearranged break schedules, increasing break-room space, increased frequency of cleaning, prohibition of visitors, and minor adjustments to the workspace.⁶² Some of these measures were implemented immediately while others, including temperature checks and screening, were put in place only after the outbreak began. Cargill also increased wages by $2 an hour and offered a one-time $500 bonus payment for workers who showed up for every shift.

On 6 April, the first positive case was detected at the plant. The union requested an emergency jhsc meeting, which the employer declined. The union also called for enhanced personal protective equipment for every worker and redesigned workspaces to increase distance between workers. The union and workers asked for more information on how many workers were at risk as “close contacts” but received no information. On 13 April, 38 cases were confirmed, and the union called for the plant to be shut down for two weeks to stop the spread.⁶³ This call was rejected and the company called the request “inflammatory.”⁶⁴ The next day, Cargill laid off half of its workers to prevent the spread of covid-19.

On 14 April, Government of Alberta ohs officers conducted a virtual inspection of the plant via FaceTime and declared the plant as safe as “reasonably practicable.”⁶⁵ The inspection was virtual owing to inspectors’ fear for their own safety. The union filed a complaint about the nature of this inspection, but the provincial government ignored the complaint. On 18 April, Cargill hosted a telephone town hall for plant workers. The Minister of Agriculture, the Chief Medical Officer of Health, and other government officials were in attendance, declaring the plant safe and urging workers to continue to go to work despite the provincial health authority reporting that 200 Cargill workers had tested positive to date.⁶⁶ Union representatives were barred from attending this meeting. The following day the union held its own virtual town hall with its members. At the meeting, the local president informed workers of their right to refuse under the ohs Act. The union reported that over the next 24 hours, hundreds of workers refused to show up for work, citing their right to refuse unsafe work.⁶⁷

On 19 April, the first Cargill worker died as a result of covid-19. The next day, Cargill announced a two-week shutdown of the plant. It also announced that it would provide pay during this time for workers who showed up for their shifts in the days leading up to the shutdown, in effect punishing workers who exercised their right to refuse. The plant reopened with two shifts on May 4 in spite of a complaint made by the union to the provincial labour relations board and to ohs officers demanding it remain closed. On this date, the Alberta government linked over 900 positive covid-19 cases to the Cargill plant. On 7 May, the second Cargill worker died from covid-19.

At the same time as the Cargill outbreak, a similar outbreak occurred at the jbs Foods (formerly Lakeside) meat-packing plant in Brooks, Alberta. At that plant, 650 workers were infected and one worker died. The jbs plant reduced production but never closed during the outbreak. In February 2021, a third Alberta meat-packing plant, Olymel in Red Deer, experienced an outbreak where over 500 workers were infected and three workers died. A second, smaller outbreak occurred at the High River Cargill plant in February 2021. As of writing, an rcmp investigation into the death of one of the Cargill workers is underway, as is a class-action suit by people who came into close contact with infected Cargill workers, citing the company’s inadequate covid response.

The Cargill case illustrates several problems with the ohs system. First, the employer was able to act unilaterally to impose safety measures despite the presence of a jhsc. The workers were unable even to attain a meeting of the joint committee during the crisis. This suggests that their right to participate was a hollow right, verging on being meaningless. Second, the workers did not receive needed information from the employer regarding the outbreak and did not participate in any processes for evaluating the hazards. This suggests that their right to know about hazards was equally devoid of meaning in practice. Third, while it appears that workers utilizing their right to refuse (encouraged by the union) may have contributed to the decision to shut down the plant, those workers faced reprisals in the form of deducted pay. Further, no formal refusal investigation is known to have taken place. In short, the workers’ exercise of their rights under the irs proved to be ineffective at triggering the employer to control an immediate and obvious hazard.

The government also failed in its role to enforce safety laws and ensure the safety of workers in the plant. The inspection (done virtually because the inspectors deemed the plant unsafe to enter) failed to order any additional measures to be taken, and officers made no attempt to investigate the worker illnesses. More importantly, government officials collaborated with the employer to downplay workers’ safety concerns at the town hall. In other words, the government allied with the employer to protect the employer’s economic interests at the expense of workers’ health and safety.⁶⁸ As of June 2022, no charges or penalties have been assessed against Cargill by the provincial government, a further indication of the government’s unwillingness to hold employers accountable for failure to ensure a safe work site.

The hazards of the covid-19 pandemic are unique and particularly dangerous, and everyone has struggled to meet the challenges of keeping people safe. The lack of effective action taken early in the pandemic could be partially explained by a lack of knowledge of the virus and its virulence. A third meat-packing plant outbreak almost a year later, however, suggests that ignorance of the virus is not the reason employers failed to take necessary steps to protect workers. Instead, they failed to take necessary steps because profits mattered more than safety. The Cargill case study is relevant not because the employer’s efforts were inadequate and an outbreak occurred but because even when workers (and their union) were actively advocating for their safety, the employer’s economic interests took precedence. And the ohs system, including the irs, failed to ensure even basic protections for these workers under the circumstances.

Discussion

The Canadian Centre for Occupational Health and Safety, a tripartite organization mandated to provide safety education, describes the irs in this way:

The internal responsibility system is the underlying philosophy of the occupational health and safety legislation in all Canadian jurisdictions. Its foundation is that everyone in the workplace – both employees and employers – is responsible for his or her own safety and for the safety of co-workers. Acts and regulations do not always impose or prescribe the specific steps to take for compliance. Instead, it holds employers responsible for determining such steps to ensure health and safety of all employees. Internal responsibility system does the following: Establishes responsibility sharing systems; Promotes safety culture and communication; Promotes best practice; Helps develop self reliance; Ensures compliance.⁶⁹

This description succinctly sets out the assumptions upon which Canada’s ohs system is based, emphasizing mutual responsibility, co-operation, internal autonomy, and a minimal role for government. These qualities are presumed to lead to safer workplaces. A close reading of the description reveals two other important characteristics of the irs. The mutual responsibility and co-operation in the irs tacitly assume the parties possess a common interest in making workplaces as safe as possible. Further, the description elides the power imbalance that exists in Canadian workplaces. These two hidden characteristics fundamentally undermine the effectiveness of Canada’s irs and the ohs system.

As the Cargill covid case demonstrates, employers do not possess the same level of interest in worker safety as workers. Employers’ economic interests often take priority over safety, especially when the price of keeping workers safe is perceived as high or simply can be avoided. Cargill also reminds us that employers wield both a great deal of power in the workplace and influence with governments. This structural power imbalance allows employers to pursue their interests at the expense of worker interests. Consequently, the irs frames workers as being partners in safety but does not provide them with any meaningful authority or tools with which to achieve safe workplaces when employers are reluctant to act. The three process-based rights are insufficient to overcome the employer’s substantive control over the workplace.

It is not that the legislation entrenching the current ohs system has done nothing to improve workplace safety. Most observers readily accept that workplaces are safer in 2021 than they were in 1971. Employers – sometimes voluntarily, sometimes compelled by government – have taken steps to create safer workplaces. But, under the irs, employers retain vast discretion over which hazards to control and how. That discretion, exercised with the profit imperative in mind, favours safety improvements that also improve the bottom line. When safety does not pay, improvements are generally not made. This dynamic helps explains why hundreds of thousands of serious workplace injuries still take place each year. It also suggests that the adage “safety pays,” which has emerged as a central rhetorical flourish in ohs circles, might be more accurately phrased as “safety only when it pays.”

Canada’s ohs regime should properly be seen as a later addition to the series of compromises crafted between capital, labour, and the state during the 20th century, wherein class conflict was managed through capital conceding to workers slightly improved working conditions, institutionalized unions, and modest socialized benefits in return for increased labour peace and a more compliant workforce. In the realm of ohs, workers received modest safety rights and sporadic state enforcement in exchange for not disrupting production over safety issues. The effect of Saskatchewan’s groundbreaking ohs legislation and all that followed was to shift the location of struggle for workplace safety from picket lines and factory floors to meeting rooms and professionals’ offices.

The current ohs system also parallels those historic compromises in its differential impact on groups of workers. The workers who have historically been best able to exercise their newly found safety rights were those who are already advantaged in the workplace: educated, unionized men. The Cargill case suggests that the ability of unionized workers to exercise their rights may be diminishing, although it is difficult to disentangle this effect from other worker characteristics, such as gender, ethnicity, right of residency, and employment precarity. Further, qualitative aspects of the work environment – such as pace, repetitiveness, and de-skilling – that flow from management decisions about when, where, and how to produce things are matters that are largely out of workers’ reach under the irs.⁷⁰

Like those other compromises, the creation of the formalized ohs system has been both a blessing and a curse for workers. For some groups of workers, there have been some modest but tangible gains in safety and some marginal increases in workplace participation. The downside is that the system entrenches the power imbalance in the employment relationship and partially removes safety from the frontier of control. As the preconditions for workers to effectively exercise their rights have deteriorated, workers have become less able to generate effective responses to workplace hazards.

The Future of OHS

The past 50 years have taught us that workplace safety is intrinsically a part of the struggle over the frontiers of control. Consequently, efforts to make workplaces safer must address the question of the power imbalance between workers and employers. The complex reality of unions demonstrates that the imbalance is a feature of capitalist relations, but workers can shrink the gap. There are three potential pathways toward safer workplaces within the existing capitalist framework: (1) incremental change to the ohs system and the irs, (2) bolstering workers’ power by giving them access to external sources of power, and (3) direct action by workers.

An incrementalist approach might focus on strengthening or expanding existing approaches to ohs such that they are more effective. jhscs could be provided with decision-making powers to compel employers to attend to hazards. Refusals could be strengthened by allowing for group refusals and lowering the bar as to when a refusal is legal. Employers could be required to provide safety information and training by outside sources and workers could be offered resources to seek out information independently. A slightly bolder approach would be enacting “concerted activity” rights that would render illegal any employer retaliation for group actions (e.g. safety strikes). This principle exists in US labour law and was used by workers there to protect themselves during the covid pandemic.⁷¹ Giving workers a way to put economic pressure on employers may compel employers to deal with workers directly, without the deflection of joint committees or government officers. In many respects, it reflects the actions taken by workers in the 1960s to force employers to act on safety.

These kinds of reforms might increase the effectiveness of the existing tools, but they are vulnerable in two main ways. First, employers may simply ignore these changes, knowing that both the risk of being penalized by the state and the cost of any penalty will likely be relatively small. Second, these changes are vulnerable to statutory erosion. For example, in 2017, Alberta’s New Democratic government made modest changes to the province’s ohs code for the first time since 1976, including strengthening the right to refuse and introducing jhsc and safety representatives; by 2021, the United Conservative government had rolled back these changes to the point that jhscs are largely meaningless.⁷²

Another incrementalist approach is to seek more active state enforcement, including more aggressive enforcement of existing rules, greater resources for enforcement, higher penalties for contraventions, and strengthened regulatory requirements. All these changes would benefit workers, in the same manner that increasing the minimum wage helps low-income workers. Such changes are likely to exact incremental improvements in safety, but the history of ohs enforcement in Canada suggests changes in law, policy, and practice are often short-lived. Pressure from employer interests on government to scale back such initiatives can be intense, much as was seen in the late 1980s and 1990s.

A second approach would be to bolster the power of workers to meaningfully exercise their current rights. This might be done by giving workers access to new external sources of power to supplement that provided by the waning union movement. For example, community-based non-governmental organizations may be more effective than government inspectors or unions in engaging with traditionally disadvantaged workers’ groups to identify and seek remedies for unsafe working conditions.⁷³ Such groups could also be empowered (via legislative change) to conduct workplace inspections, identify unsafe working conditions, and issue fines to employers for non-compliance. Open questions regarding this approach include how to fund such work and whether the state should be off-loading this work to communities. Legislative change requires government co-operation (and is thus vulnerable to being rolled back) but empowering third parties to enforce the laws may offer states a way to reduce the direct conflict with employer interests that aggressive state enforcement can generate.

A more market-based approach might be to incentivize citizens to report employer non-compliance. For example, citizens can report violations of New York City’s anti-idling law and receive a portion of the fine.⁷⁴ This approach dramatically expands the potential inspectorate at no cost to the state, can alter employer cost-benefit calculations, and may normalize unsafe work as both unacceptable and something about which to take action. It also creates citizen pressure on the state to impose fines for violations. Applied to ohs, it is important to recognize that some violations, especially those that are more visible to the public and more apt to cause immediate injuries (such as not wearing fall protection on residential roofing projects), are more amenable to this sort of crowdsourced enforcement than others. However, it is important to point out that the current irs system is also more functional for certain kinds of injuries than others, as previously mentioned. Additionally, there is also the risk the program could be structured in a way that punishes workers rather than employers.

Worker safety advocates might also shift the emphasis of their work away from training workers to participate in the irs to holding employers to public account for unsafe working conditions. Public shaming of employers over
safety issues may be more effective than sporadic government inspections at changing employer behaviour, by heightening the cost to employers of unsafe workplaces.⁷⁵ This approach would benefit from better, more complete, and more immediate data about workplace injuries and fatalities. These options for bolstering worker power emphasize that employers are responsible for ensuring safe workplaces while the role of workers is primarily to hold employers to account. This shift in emphasis (away from co-operative, shared responsibility) brings attention to the class-based conflict the irs often obscures, manages, and contains.

Finally, workers can engage in direct action around safety risks beyond their existing right to refuse, including tactics typical of grassroots union organizing (e.g. workplace slowdowns and unilateral job process changes, marching on the boss). These tactics are available even without the concerted activity protections discussed above. However, even with concerted activity protections any direct action by workers can result in employer retaliation. This raises the question, for whom is direct action most accessible? Worker activism in the United States during covid-19 (such as mass quits) has been most prevalent among workers with the most precarious and least desirable jobs. This makes intuitive sense (given the low quality of their jobs) but may also reflect that the costs of direct action for these workers are low because comparable replacement jobs are easy to find.

Ultimately, creating safer workplaces bumps up against the same hurdles as any effort to improve conditions for workers. Progress is only made if workers organize and agitate. The existing ohs system is a product of collective worker action in the 1960s and 1970s. Taking the next step in making safer workplaces likely requires a reprise of this organizing. Whether that is possible in the current context of globalization, neoliberalism, and increasing precarity is a big question.

Conclusion

Fifty years after the creation of Canada’s contemporary ohs system, workplaces are safer. Nevertheless, this relative improvement in workers’ fortunes at work cannot be fully credited to the system. As constructed, Canada’s ohs system is bedevilled by a series of incorrect assumptions that undermine its effectiveness at creating safe workplaces. Most significantly, the failure to recognize power imbalances inherent in the employment relationship, coupled with the declining power available to workers, has rendered the rights afforded to workers under the system weak and largely unused. Any effort to reconstruct the ohs system must start from a recognition of the existing power imbalance in the workplace and build possibilities for strengthening the power available to workers.

How to cite:

Jason Foster, Susan Cake, and Bob Barnetson, “Profits First, Safety Second: Canada’s Occupational Health and Safety System at 50,” Labour/Le Travail 90 (Fall 2022): 179–202. https://doi.org/10.52975/llt.2022v90.007

Copyright © 2022 by the Canadian Committee on Labour History. All rights reserved.
Tous droits réservés, © « le Comité canadien sur l’histoire du travail », 2022.