Research on the Current Status of Safety Systems in the Steel Industry and Alternative Solutions

Hyung-ryeol Kim, Korean Institute of Labor Safety and Health
Jin-woo Lee, Hanil Hospital
Hye-eun Lee, Hallym University
Ju-hee Jeon, Korean Institute of Labor Safety and Health
Min Choi, Korean Institute of Labor Safety and Health
Gyeong-geun Kim, Labor Research Institute of the Korean Metal Workers’ Union
Gyeong-won Na, Korean Metal Workers’ Union

Translated by Michelle Jang

Korean Institute of Labor Safety and Health
2025

Research Summary

1. Introduction

This study aimed to examine the workability of the safety and health system for workers in the steel industry. The degree to which companies accept the participation of workers and labor unions in revealing and resolving risks in the working environment is very important. It is easy to think that companies sufficiently guarantee the participation of workers in the legally guaranteed Occupational Safety and Health Committee (OSH). However, this alone does not mean that companies recognize workers as co-stakeholders in the conception, planning, implementation, and evaluation of the safety and health system.

The same goes for how trade unions view occupational safety and health activities. Most labor union activities are carried out within legally guaranteed participation and institutional categories. It is necessary to confirm whether debates are taking place within the system related to safety to fundamentally address risks at the root and whether collective rights and capabilities in the workplace are being fostered through labor safety. This includes selecting protective equipment,  measuring the work environment, investigating musculoskeletal hazards, assessing risks, and establishing an Occupational Safety and Health Committee.

Based on awareness of these issues, this study examined the following six points:

1. Through analysis of the types of accidents that occur repeatedly in the steel industry, the tendency of risks to repeat and lead to accidents and their causes were explored.
2. By analyzing collective agreements related to labor safety, we examined the differences between the standards for labor safety activities at individual workplaces and the legal standards.
3. We analyzed the actual practice of labor union participation and labor union authority by securing and analyzing the minutes of the Occupational Safety and Health Committee for each branch for the past five years.
4. We examined how accident investigations are conducted and the extent to which labor unions intervene and participate in the process. In addition, we analyzed how accident investigations affect the restructuring of the safety system.
5. In order to examine whether the enhanced safety after the enforcement of the Serious Accidents Punishment Act is controlling or democratic, we analyzed the contents of the safety core rule – 10 key safety rules (SCR) being implemented at Hyundai Steel.
6. In order to examine the impact of risk multiplication on the safety system or how the safety system resolves risk multiplication, we looked into the risks of Hyundai Steel’s subsidiaries and in-house subcontractors.

II-1. Review of Occupational Safety and Health Items in Collective Agreements of the Steel Division of the Metal Workers’ Union

The collective agreements subject to analysis were collective agreements of branches affiliated with the Steel Division of the Metal Workers’ Union, which were collected in bulk through the Metal Workers’ Union. The collective agreements of the Steel Division branches were compared with the relevant laws and regulations of the Occupational Safety and Health Act and the Model Collective Agreement of the Metal Workers’ Union.

Among the provisions related to outsourcing of risks, the Steel Division Collective Agreements did not include three articles of the Model Collective Agreement of the Metal Workers’ Union: Article 184 (Protection of the Right to Health of Non-Regular Workers), Article 185 (Prohibition of Subcontracting Hazardous and Dangerous Work), and Article 186 (Prohibition of Outsourcing Risks). The Model Collective Agreement includes provisions for a joint Occupational Safety and Health Committee in which the general contractor and subcontractor participate, and for the general contractor and subcontractor unions to jointly investigate and take measures to prevent recurrence in the event of a serious accident involving subcontracted workers. None of the collective agreements analyzed included such provisions.

The Model Collective Agreement  of the Metal Workers’ Union includes provisions that guarantee the participation of the labor union in accident investigations, state that the purpose of an accident investigation is not for punishment, prohibit actions and statements that may intimidate the victim, and prohibit disadvantages in personnel affairs. However, in the collective agreements of the steel branches, only the Poroltech branch includes joint labor-management investigations and prevention of intimidation of the victim, while other branches stipulate company-led accident investigations with insufficient specification of the participation of the labor union. It is necessary to stipulate the right of the labor union to participate in an accident investigation in the collective agreement, and to add a clause to the collective agreement that prohibits intimidation of injured workers. The Model Collective Agreement of the Metal Workers’ Union emphasizes the importance of serious accidents by  including separate provisions on accident investigations and work stoppages in the event of serious accidents. It includes provisions such as joint accident investigations with the labor union, participation of subcontracted worker representatives in the event of a subcontracted worker accident, reflection of the opinions of all workers in lifting the work stoppage, and consultation on payment of leave allowances to subcontracted workers during the work stoppage period. None of the collective agreements in the steel industry branches included a separate clause on serious accidents, and there was no mention of compensation for wage loss in the case of subcontractor accidents or work stoppages. Considering the characteristics of steel industry workplaces with a high incidence of serious accidents, it is necessary to add a clause related to serious accidents based on the Model Collective Agreement of the Metal Workers’ Union.

In the provisions related to safety and health management regulations, the Model Collective Agreement of the Metal Workers’ Union comprehensively stipulates the safety and health obligations of the company and does not separately mention the obligations of workers. In contrast, there were many collective agreements in the steel sector that stipulated the obligation of union members to comply with safety and health rules. It is necessary to ensure that the obligation to comply with safety and health rules does not lead to an excessively controlling company-led safety and health management system.

In the provisions related to the right to stop work, the Model Collective Agreement of the Metal Workers’ Union expanded the right to stop work to all workers and the union’s Occupational Safety and Health Committee members, and expanded the requirements for exercising the right to stop work from ‘imminent danger’ to ‘potential for risk of injury.’ In the case of steel sector collective agreements, only some branches of Poroltech and Hyundai BNG Steel expanded the subject of exercising the right to stop work to committees, etc. Most collective agreements set the subject of the right to stop work at the same level as compliance with the law or more narrowly and the requirements were stipulated as ‘imminent danger’ or ‘urgent and serious danger.’ It is necessary to strengthen the authority of the labor union regarding the right to stop work beyond the legal level.

In conclusion, the steel industry collective agreements were found to be somewhat lacking in strengthening the authority of the union’s safety and health activities and work stoppage rights compared to the Model Collective Agreement of the Metal Workers’ Union. It is necessary to add and supplement items related to serious accidents, such as clearly stipulating the union’s participation in accident investigations and conducting joint investigations including the general contractor and subcontractor unions in the event of serious accidents involving subcontracted workers.

II-2. Analysis of Occupational Safety and Health Committee Meeting Minutes

The minutes of Occupational Safety and Health Committee meetings for the last five years were collected from the Steel Division branches of the Metal Workers’ Union. The minutes were analzed against the legal provisions of the Occupational Safety and Health Act, as well as the labor union’s demands for outsourcing, worker participation, reward and punishment systems, and zero-accident campaigns discussed at the Occupational Safety and Health Committee meetings of each branch.

The analysis showed that there were many cases where the format of the minutes was not followed. Minutes are important records that serve as supporting materials for deliberation and resolution. The obligation to prepare and keep minutes is to ensure the fulfillment of the commitments made by labor and management to prevent industrial accidents. Some of the submitted minutes did not include required information such as date and time of the meeting and attendance of members. It was also confirmed that in some cases, matters that required deliberation and resolution were classified as miscellaneous or other matters, or only the title of the discussion contents was left in the meeting results, making it impossible to confirm the specific contents. In cases where there is an ad hoc or special Occupational Safety and Health Committee, the activities should be reported to the committee regularly, but most branches omitted doing so.

There were many cases where meeting reports were recorded in a superficial manner. In some cases, branches recorded the same list of 8 essential deliberation and resolution items of the Occupational Safety and Health Committee in the minutes and then just filled in superficial content without conducting any discussion. In other cases, irrelevant results were entered due to a lack of understanding of the agenda item. We also saw minutes that were filled in like a calendar in the same quarter every year without any discussion in the results section. There were also many essential items that were not addressed. In the minutes of the Occupational Safety and Health Committee that were reviewed, the main agenda item was improvements in equipment. While improving the work environment and equipment are important areas of safety and health, there are more comprehensive and diverse agenda items that can be addressed in an Occupational Safety and Health Committee meeting, but this is not being fully utilized.

There were many agenda items that did not have any discussion results in the minutes. There were also many cases where agenda items were consistently brought up but the results of the discussions were not confirmed. ‘Matters related to inspection and improvement of the work environment, such as work environment measurement’ and ‘Matters related to health management, such as health checkups for workers’ are mandatory items that must be implemented by the Occupational Safety and Health Committee each time. However, only the implementation date was reported, and there were almost no branches where specific discussions were confirmed.

Several problems were identified regarding the contents related to accident investigations. Accident investigations were conducted without the participation of the labor union, and accident prevention measures focused on blaming and disciplining injured workers. In addition to the atmosphere of shifting responsibility for safety and health to workers, a small-scale safety activity certification system was introduced company-wide. The company’s responsibility for safety and health was substituted with workers’ efforts for safety awareness which were rewarded with money under this system. This type of arrangement can lead to accusations that a worker did not receive money due to a co-worker’s accident, and the cause of the accident and the preparation of countermeasures can also be blamed on the co-worker’s lack of awareness. Recently, in even more extreme cases, disciplinary action against accident victims for Safety Core Rule violations have been discussed in practical meetings.

It is necessary to: 1) share best practice examples of accident investigation cases; 2) make efforts to secure maintenance time through joint labor-management safety inspections; and 3) operate integrated general and subcontracting Occupational Safety and Health Committees.

II-3. Ten-year (2013-2023) Analysis of Steel Mill and Subcontractor Occupational Accidents By Case Type

A total of 1,901 accident cases were confirmed by searching for “Hyundai Steel,” “POSCO,” and “Dangjin” in the entire occupational accident occurrence data from 2013 to 2023. There were 861 cases at Hyundai Steel’s Pohang, Incheon, and Dangjin plants; 257 cases at POSCO’s Pohang and Gwangyang plants; and 783 cases at other partner companies and subsidiaries.

We examined the incidence of occupational accidents that occurred among subcontractor workers in the steel industry over the past 10 years by categorizing them based on the cause of the accident. Unlike the cases of chronic diseases such as musculoskeletal disorders and noise-induced hearing loss, or relatively mild accidents reported by workers at the original contractors, accidents among subcontractors were relatively serious. In particular, two types of serious preventable accidents occured: 1) accidents that occured when maintenance was not properly carried out after a stoppage of operation; and 2) accidents that occurred when work was restarted before completion of maintenance. The analysis also showed that work was performed without securing a safe workspace, and falls and entrapment accidents frequently occured in these cases. Failure to complete maintenance before restarting work constantly exposes workers to such risks, and it is necessary to conduct a large-scale inspection and establish preventive measures to secure safety. In the steel industry, when doing dangerous work such as working at height and in hot and dusty conditions, serious accidents are still occurring due to the complex subcontracting employment structure and pressure to speed up work, despite preventable measures such as engineering and management measures that are not functioning properly.

II-4. Controlling Safety System: Hyundai Steel SCR System and Small-Scale Safety Activity Certification System

After the implementation of the Serious Accidents Punishment Act, there has been a trend of strengthening worker disciplinary action or punishment in the event of an accident, claiming to establish or improve the safety and health management system. These are safety measures that consider the cause of industrial accidents to be human error and focus on workers’ compliance with safety rules. The SCR system (safety core rule, hereinafter referred to as the SCR system) was introduced in 2015. It is literally a core safety rule. The purpose of the SCR system was stated in company documents as “establishing a safety culture at the steel mill through selection and implementation of core safety rules” in order to prevent serious accidents.

The SCR system causes several problems. First, accident investigations fail to get to the root of the accident. Most of the causes of accidents emphasized in the SCR, such as failure to wear protective gear and failure to follow work procedures, are based on the idea that accidents are due to the dangerous behavior and negligence of workers. This justifies disciplinary action by shifting responsibility for accidents to the worker. However, this approach misses the opportunity to identify the conditions that actually made unsafe behavior possible. In addition, it fails to address the employer’s duty to take measures to ensure safety even when unsafe behavior occurs. Second, it leads to concealment of occupational accidents. Since a lost-time occupational injury due to SCR violations can lead to disciplinary action, we have observed situations after 2023 where workers are pressured to obtain company-provided financial assistance for medical treatment instead of filing for workers’ compensation. Even after obtaining company assistance, workers are still required to report for work. In addition to the SCR system card, the small-scale safety certification system linked to it also discourages workers from filing for occupational accidents and regards injured workers as criminals. Under an oppressive safety and health system, occupational accidents are concealed and the right to receive medical treatment, proper rehabilitation, and return to work is violated. Third, the SCR system is a means of on-site control. In particular, cases such as non-compliance with work procedures can be applied to any accident. This is because work procedures cannot accurately reflect the on-site situation, and on-site work cannot be operated solely by work procedures or manuals. If the tacit knowledge of on-site workers is not operating, production cannot proceed smoothly. However, the fact that one quarter of all SCR cards issued are for non-compliance with work procedures suggests that the autonomy that on-site workers exercise during their work process may be excessively infringed upon.

Despite these problems, the labor union’s response has been inadequate. Except for the Dangjin plant, Hyundai Steel labor union branches have had almost no response to the SCR issue. All branches interviewed had critical perceptions of SCR or the small-scale safety certification system, but Dangjin was the only one to directly raise issues or respond to them. Critical perceptions also do not surface unless union members are disciplined, or are considered as subjects of interest or discussions by the Occupational Safety and Health Committee. There were also cases where the small-scale safety certification system was first introduced without any particular objections due to thinking that workers ‘could get extra money from the company.’ In some cases, the labor union took the initiative to package the small-scale certification systems as a no-accident movement. After the implementation of the Serious Accidents Punishment Act, the trend of strengthening worker surveillance under the pretext of safety, or disciplinary action or punishing workers when accidents occur, is not a problem of one or two individual workplaces, but a phenomenon that occurs in various industries and workplaces. In contrast, labor unions have not been able to sensitively detect and respond to these problems.

Companies emphasize safety as long as it does not affect production. The essence of this kind of safety policy is to force the workplace to operate in a ‘company-permitted manner’ for ‘company-permitted safety.’ This trend is not unique to Hyundai Steel or the steel industry. Since the enactment of the Serious Accidents Punishment Act, there have been cases of increased surveillance of workers through CCTV, etc., punishing workers for ‘not exercising the right to stop work’ while emphasizing the right to stop work under the law, and disciplinary actions against workers for not following safety rules. However, the labor union’s response is limited to individual workplaces. In most cases, individual workplaces fail to respond effectively to the company’s tactics that emphasize the value of ‘safety.’ Broader solidarity and joint response are needed so that a production order that guarantees worker safety and health is established in the workplace beyond ‘company-permitted safety,’ and so that workers become the main players in safety and health.

II-5. Accident Investigation Practices and Problems: The Case of Hyundai Steel’s Serious Accidents

Hyundai Steel is a workplace with a high incidence of serious accidents, and a workplace where similar accidents occur repeatedly. From 2010 to 2025, 53 workers  died in serious accidents at Hyundai Steel (including deaths from overwork), and 41 of them (77%) were non-regular workers.

Hyundai Steel has not conducted an in-depth investigation into the causes of these accidents, despite the fact that accident investigation reports have been prepared and made public through the accident investigation committee. Analyzing and sharing not only the direct cause of an accident but also the management and structural causes will have a profound and fundamental effect in changing work practices and safety awareness, even if it does not lead to immediate institutional improvement measures.

The importance of accident investigation lies in the fundamental prevention of industrial accidents. Risks are structural and general, while accidents are accidental and localized. In other words, before a risk is revealed as an accident, we can only predict the risk, but we cannot accurately identify the nature of the risk. However, it is revealed in the form of a specific accident in a space and time where specific coincidences overlap. Therefore, since eliminating only the direct cause of an accident does not fundamentally resolve the risk, a comprehensive management direction for a wider range of risks that can cause an accident must be established. For this, a structural cause investigation of an accident is necessary. Therefore, listing measures to prevent recurrence without a proper accident investigation not only cannot verify its effectiveness and operability, but also increases safety fatigue within the organization.

The Hyundai Steel accident investigation reports deal with the causes of accidents very narrowly, rather than broadly addressing management and structural risks and establishing measures to prevent recurrence. This is linked to the SCR, which reinforces ‘negligence of the victim,’ and thus a vicious cycle of poor accident investigation and negligence of the victim is repeated. As a result, there are no records of discussions and agreements between labor and management regarding the cause of an accident in the special Occupational Safety and Health committee that deals with serious accidents, and recurrence prevention measures are listed without any basis in discussions. It is unclear why the recurrence prevention measures are submitted in relation to the accident. In addition, the recurrence prevention measures are not structured in terms of ‘comprehensive diagnosis and measures,’ including their importance or priority and the relationships between them. From the union members’ perspective, there is a lack of understandable explanation as to why safety systems that are strengthened or newly established are necessary whenever a serious accident occurs. As a result, safety is perceived as synonymous with regulation or control, and as additional safety systems are excessively introduced on top of systems that do not function properly, overall distrust and fatigue about safety increases. Accordingly, the current organizational culture of Hyundai Steel can be seen as a paradoxical situation in which, instead of systematically increasing awareness of risks and interest in safety after a serious accident, they accept risks insensitively and become indifferent to safety.

In addition, Hyundai Steel is a workplace where outsourcing of risk is serious, and as the number of fatal accidents involving subcontracted workers is increasing, the safety control of subcontracted workers is also being strengthened. Regulations and controls for safety are necessary to a certain extent, but regulations that do not guarantee the participation, intervention, discussion, and acceptance that workers demand and want can be controlling regardless of their purpose. In particular, a system that classifies and selects subcontracted workers based on ‘safety’ and ‘health’ is actually intended to select workers that are willing to accept dangerous workplaces and to exclude others, instead of strengthening ‘workplace safety.’ How could measures be prepared to select and exclude subcontracted workers in cases of fatal accidents involving subcontracted workers? Such measures could be established because even the Occupational Safety and Health Committee related to fatal accidents involving subcontracted workers could be decided by the general contractor’s labor and management without the participation of the subcontractor’s labor union. At least in relation to serious accidents, it is necessary to have subcontractor labor unions participate in the special Occupational Safety and Health Committee or to establish a separate consultative structure similar to the general contractor and subcontractor Occupational Safety and Health Committee.

II-6. Selective Inclusion of Safety Systems and Labor Union Responses: In-House Subcontracting, Subsidiaries

Hyundai Steel established a subsidiary to avoid direct employment after subcontractor workers won a lawsuit against the company for illegal dispatch of subcontracted workers. However, since the establishment of the subsidiary, the risks of the subsidiary and subcontracted workers have not improved, and instead, the structure of indirect employment has become more complex, further multiplying risks.

Like POSCO’s in-house subcontracting, most subcontracting labor unions are not actively carrying out safety and health activities due to limitations in the structure of the subcontracting company. This is not only due to the employment structure, but is also related to the fact that the main demands of the subcontracting labor unions are focused on direct employment or improvement of working conditions, and safety issues are pushed to a lower priority. However, the cases of POSCO and Hyundai Steel’s irregular workers’ branches are important cases to consider for how in-house subcontracting labor unions should carry out labor safety activities. It is important to recognize that no one but the in-house subcontracting labor union can solve not only direct employment but also subcontracted processes and risks. In addition, the more efforts that are made to solve risks, the more structural problems between the general contractor and the subcontractor are revealed, which clearly shows why in-house subcontracting labor unions should strategically address labor safety issues.

In the case of subsidiaries, we found that the form of ‘quasi-manpower outsourcing’ that appeared in the conversion to subsidiaries of the public sector still existed as well as the problems of division, discrimination, and hierarchy in terms of safety issues. However, as the size of the business grows and individualized workers in small businesses are converted to one company, labor unions are established and the collective voice and capacity of workers regarding safety have increased compared to those of in-house subcontractors. However, it is difficult to conclude that subsidiaries are safer than in-house subcontracting.

As can be seen in the case of Hyundai Steel, the establishment of a subsidiary under the subcontracting structure and the controlling and hierarchical safety system did not contribute to improving or democratizing the existing safety system at all. Rather, as internal conflicts deepened, safety deteriorated further, and it was also confirmed that even after the establishment of an integrated safety and health system, irregular workers were only partially included. The case of occupational safety and health activities carried out by labor unions of in-house subcontractors and subsidiaries in the steel industry somewhat indicates the direction of labor safety activities for ‘indirectly employed unstable workers.’ In fact, there has been an emphasis on the ‘structural problems’ of those activities – for example, under the condition of ‘indirect employment’, the joint occupational safety and health committees with employers that do not own facilities are ineffective, or it is difficult to exercise the right to stop work due to the insecurity of re-contracting. However, the case of the steel industry clearly shows why labor safety activities by indirectly employed labor unions are necessary and how they should be carried out despite the limitations of low organization rate.

III. Conclusion

In the steel industry, where serious incidents have occurred repeatedly, safety has been emphasized and reinforced in some way since the Serious Accidents Punishment Act entered into force. The problem is the meaning of safety and the way it is emphasized. The larger the company, the more formal and procedural is the established safety system. The smaller the company, the more likely that the safety system is absent or operated in a very weak state. There are various systems in place for regular employees of general contractors to participate in the safety system, but there is no independent safety system for subsidiaries or subcontracted workers. Instead, subcontracted workers are partially included in the head office’s safety system or are mostly excluded and selectively managed. Under the same safety system, a gap and inequality in safety are occurring between general contractor and subcontractor, and between the parent company and the subsidiary.

In the steel industry, the controlling nature of safety efforts has been reinforced under superficial and procedural safety systems. This seems to have become even more severe since the enactment of the Serious Accidents Punishment Act. The emphasis on safety and worker participation in safety were emphasized to the point of obsession, but labor unions and workers perceived it as mobilization and control rather than participation. In order to change punitive regulations and a controlling safety system into a more democratic safety system, it is important to secure a collective decision-making structure for workers. To this end, joint practice among the member unions of the steel industry sector of the Korean Metal Workers’ Union and the collective decision-making structure for workers should be strengthened.

Strengthening joint response of labor unions in the steel industry: It is necessary to build joint safety capabilities such as joint responses to SCR and other controlling safety systems, and joint accident investigations into serious accidents.

It is necessary to compile cases of controlling safety systems that have been further strengthened since the introduction of the Serious Accidents Punishment Act, and strengthen joint responses to them. In addition to Hyundai Steel’s SCR and small-scale safety certification system, it is necessary to compile cases of controlling safety systems that blame worker negligence when accidents occur, including POSCO’s zero-accident movement. Based on analysis of these cases, it is necessary to form public opinion and seek joint practices to abolish controlling safety systems.

Establishment of an integrated Occupational Safety and Health Committee at Hyundai Steel: Joint responses to serious accidents, SCR, and other company-wide safety systems are necessary.

The existing Occupational Safety and Health Committee system is established at each business site, making it difficult to intervene and discuss safety and health policies planned and implemented at the level of the headquarters safety management office. This is a limitation of existing laws, but it is possible to demand an integrated Occupational Safety and Health Committee through collective bargaining agreements, etc. Through this, it should be possible to raise a collective voice about the headquarters’ responsibility and authority related to safety management. It is important to create a joint action at the labor union level by raising issues such as the abolition of SCR, which are urgent issues, as a joint demand.

A joint response to the demands of the subcontracting and subsidiary unions for the general and subcontracting Occupational Safety and Health Committee is necessary.

The response to the gas leak accident in December 2025 was the first case in which the Hyundai Steel general and subcontracting unions jointly conducted the entire process from accident investigation to periodic supervision through the activities of the Metal Workers’ Union Countermeasures Committee. In this response, the case of the Hyundai Steel Non-Regular Workers’ Union (Dangjin) responding to the gas leak accident in 2020 was an important reference. Although there is conflict between regular workers, in-house subcontracted workers, and subsidiary workers, it is important to form the awareness and practice that a joint response is important in matters of safety.

It is necessary to establish a safety system support plan for small- and medium-sized businesses in the steel industry.

It is necessary to establish a separate plan to strengthen the occupational safety and health activities of small- and medium-sized businesses such as Hyundai Special Steel and Hyundai BNG Steel at the labor union level. It is necessary to establish a separate labor safety manager meeting within the steel division of the Metal Workers’ Union to ensure exchange and communication between workplaces, or to provide separate support for strengthening safety and health activities. At the very least, it is necessary to create an action of joint response related to serious accidents within the steel industry sector of the Metal Workers’ Union. The safety system of the steel industry, which is divided into large companies, small- and medium-sized companies, and general contractors and subcontractors, should ultimately emphasize corporate participation based on information sharing and cooperation at the steel industry level, including POSCO and Hyundai Steel, beyond the size of the company and employment type.