Abstract
This study was conducted to investigate the causes and consequences of the vulnerability of occupational health and safety (OHS) regulations to deregulation during a period of economic crisis in the Republic of Korea. Analysis of Korea’s national regulation database revealed that the vulnerability of OHS regulations to deregulation was related to the fact that OHS policy included many regulations without direct deregulatory impacts on workers. The most affected victim of this characteristic was information regulation that provided a legal basis for government’s monitoring and inspection of OHS activities. The massive relaxation of information regulation has the potential to weaken government oversight and to tempt businesses to hide industrial accidents. Since changes in regulations without direct deregulatory impacts are not easily identifiable by workers, careful monitoring of deregulation is necessary to prevent policy impacts harmful to workers’ health and safety.
Introduction
In the event of an economic crisis, governments can implement policies stimulating their national economy. Deregulation, that is, the abrogation or relaxation of regulation, is a powerful tool to support businesses by reducing certain costs that may arise from various regulations. But in the field of labor policy, deregulation can be a disaster and harm the lives, health, and welfare of workers. The study presented here attempted to understand the impact of deregulation on workers’ health by examining the experience of the Republic of Korea during an economic crisis. It examined why protective occupational health and safety (OHS) regulations were targeted for deregulation and explored the consequences of this deregulation. A research goal was to provide policymakers with evidence to prevent and mitigate deregulatory impacts detrimental to the health and safety of workers.
In principle, regulations are set to protect the interests of the public.1–6 There are, of course, various interpretations of government regulations that focus on interest groups, policy ideas, and institutional effects as well as public interests. 7 Regardless of those diverse perspectives, the topics addressed by labor regulations, such as basic rights, employment relationships, industrial relations, and social security, certainly aim to protect and promote the interests of workers. 8 Thus, the deregulation of labor policies needs to be thoroughly monitored, particularly in the context of an economic crisis in which the government has focused on supporting business activity.
Changes in Labor Regulations: 1998–2002.
Note. Values denote number of regulations (percent). Pearson chi-square (OHS and the others) = 5.561 (p = .018). OHS = occupational health and safety.
At that time in Korea, the workplace fatality rate was one of the highest among Organisation for Economic Co-operation and Development countries. 15 Although social welfare policies such as national pensions and national unemployment insurance had been implemented, labor policy deregulation posed a major threat to workers’ lives and well-being, and the disproportionate reduction of OHS regulations worsened the situation.14,16
Methods
The research consisted of four steps. First, the contents and structures of all labor regulations registered in the national regulations database were characterized on the basis of policy termination literature (explained below). In the second step, multiple logistic regression analyses were conducted to find the factors that had affected the probability of deregulation (i.e., deregulation factors). In the third step, the effects of the deregulation factors found in the second step on the deregulation of OHS regulations were examined to understand why OHS regulations had been more vulnerable to deregulation than other labor regulations: the factor causing the vulnerability was labeled the vulnerability factor. Finally, to identify the actual impact of the vulnerability factor on the deregulation in OHS, an investigation was conducted on deregulation experiences in various OHS policy areas, such as the OHS information system and the quality of OHS management (Figure 1).
Summary of research framework and results.
For the analysis of labor regulations, this study used a database compiled by the Regulatory Reform Committee (RRC) of the Korean government. The RRC database is a national database of government regulations, including laws, orders, and notices. This database contained almost all “administrative regulations” defined by the Administrative Regulation Act as “activities of the central or local government that impose obligations on people or restrict people’s rights to achieve specific administrative purposes.” 17 That is, “administrative regulation” means a coercive way for governments to control the private sector. This definition of administrative regulation is close to the usual use of “government regulation:” government’s coercive intervention, setting limits on private activities to protect the public interest.1–6 In this study, the term “government regulation” was used to refer to the “administrative regulation” of the database.
Although Korea experienced two recent economic crises, this study used only deregulation data following a crisis in 1998 because there was not much deregulation data from the later crisis in 2008. Since the deregulation data at that time were not available due to the alteration of the RRC database, this study used the data obtained on 1 January 2003, when the term of President Dae-Jung Kim (February 1998–February 2003), who pushed ahead with deregulation and downsizing the government throughout his term, was nearly over. Thus, the dataset for the analysis included all deregulation cases registered in the RRC database for about the five years after August 1998. The deregulation cases from the beginning of the foreign exchange crisis in October 1997 to the establishment of the RRC database in August 1998 were not analyzed due to lack of data availability, but it was not likely that many cases existed during this period.
The dataset used in the analysis consisted of 561 regulations covering the entire range of responsibilities accorded to the Ministry of Labor. When cleaning data, officially acknowledged duplicate records were removed. Classification errors found in the original official classification of regulatory changes were corrected, and the regulations in ambiguous categories were reclassified according to the changes that had actually occurred. Regulations that had been changed more than once were reclassified based on the last change: for example, a regulation that had been strengthened and then relaxed was classified as deregulated.
After the dataset for analysis was prepared, some variables were extracted from the contents and structures of the remaining 548 labor regulations. Deregulation, literally the removal of regulation, means the rollback of state intervention in numerous areas of social life. 18 In this study, regulations that were relaxed (loosening of restrictions or obligations) or abrogated (abolition of code-defining restrictions or obligations) were classified as the “deregulated” group and regulations that were unchanged, strengthened (tightening of restrictions or obligations), or newly established during the observation period (from 31 August 1998 to 1 January 2003) were classified as a “continued” group. This binary variable “changes in regulation” was used as the dependent variable in multiple logistic regression analyses.
In addition, various characteristics related to the contents and structures of regulations were explored and used as independent variables for analysis. The exploration of characteristics that could affect the probability of deregulation was based on the literature on “policy termination” rather than “policy change.” In fact, policy termination has generally been regarded as a stage in the policy process and has therefore not attracted much research attention. 19 However, while studies on policy processes are concerned with the general aspects of policy change, the ending phase of policy may be different from the other stages. For example, the structure of the advocacy coalition in the policy termination stage may be different from that in the policy establishment stage because stakeholders related to the policy might have changed over time. 20 For this reason, this study relied on the literature on policy termination, not policy change.
Unfortunately, however, very few studies have suggested research frameworks for policy termination. Unlike Kaufman, who denied the existence of patterns in the termination process and emphasized the role of chance, some authors argued that there are several factors facilitative of or antagonistic to policy termination: Bardach proposed facilitative factors such as a change in administration, a delegitimation of the ideological matrix, a period of turbulence, cushioning the blow of deregulation, and designing policies for termination, while deLeon proposed antagonistic factors, such as intellectual reluctance, institutional permanence, dynamic conservatism, antitermination coalitions, legal obstacles, and high start-up costs.21,22
This study was able to extract several variables related to these facilitative or antagonistic factors from the contents and structures of regulations and incorporate them into the analysis model. First, regulatory issues and deregulatory impacts were considered as content characteristics. The issue characteristic of each regulation was determined by whether the regulation addressed social or economic issues, based on the definition by Organisation for Economic Co-operation and Development: economic regulations “intervene directly in market decisions such as pricing, competition, market entry, or exit,” whereas social regulations “protect public interests such as health, safety, the environment, and social cohesion.” 23 This classification of regulatory issues was expected to reflect the policy ideas of the day strongly supporting the deregulation that would reduce the cost burden on businesses and ameliorate the impact of the economic crisis.
The impact characteristic of deregulation was determined by whether each instance of deregulation could have any direct impact on the beneficiaries of the regulation, primarily workers. If a change in a regulation could have affected the beneficiaries of the regulation directly, it was classified as a “direct impact.” In fact, while most deregulation cases had direct impacts, the remainder affected beneficiaries only indirectly through their effect on other groups, such as government or agencies. In the classification process, indirect deregulatory impacts were found to occur in three different ways: (1) limiting information available to government administrators, (2) causing a financial loss to the government (fund) that could result in a loss to the beneficiaries (workers), and (3) reducing the qualification requirements for conducting labor-related businesses (market entry of and job training for agents and agencies). This variable could be related to one of deLeon’s “start-up costs”: the protests of those damaged by the deregulation that deregulators must endure. 24 If the beneficiaries of a regulation were unaware of the deregulation because there were no direct impacts on them, they would not oppose the deregulation.
Second, three variables indicating structural characteristics related to the enactment and enforcement of regulation were identified: duration, delegation of execution, and level of legal change. The “duration” of a regulation (i.e., the period during which the regulation existed) was determined by calculating the time elapsed from the enforcement of the regulation to the point of the change (for deregulated cases) or when the data were collected for analysis (for continued regulations).
The variable “delegation of execution” was determined by whether the regulation was implemented by the government or by another delegated organization. Regulations that were not implemented directly by the central or local government but executed by other organizations or associations (either public or private) were classified as delegated regulations. The effect of delegation on the likelihood of deregulation was expected to be negative since delegated agencies would have sought “institutional permanence,” the tendency of institutions to maintain themselves, at the time of deregulation. 24
The “level of legal change” of a regulation referred to the highest legal level that should be involved in the abrogation or relaxation of a regulation. For this purpose, three categories were used corresponding to the legal hierarchy: statute, order (presidential or ministerial), and administrative notice. This variable was designed to examine deLeon’s “legal obstacles” that were related to the constraints of due process. 24
When all these variables were prepared, multiple logistic regression analyses were conducted to find factors that could affect the deregulation of labor regulations. The SPSS version 19 package was used for analysis.
After finding deregulation factors, statistics regarding deregulation in OHS regulations were compared with those in the other labor regulations to find the OHS deregulation factor that made OHS regulations more vulnerable to deregulation than the other labor regulations (vulnerability factor). In addition, in order to identify the actual impact of the vulnerability factor on deregulation in OHS, various OHS policy domains and their deregulation experiences were investigated using cross-tabulations and chi-square statistics.
Results
In this study, policy domains in the area of labor regulations were categorized into OHS, employment stabilization, working conditions, labor relations, and so forth, according to the laws to which the regulations belonged (Table 1). The domain of OHS contained regulations concerning OHS insurance (run by the government), OHS management (OHS programs or activities in businesses), work systems (people, equipment, materials, work methods, and environment put into the work processes of businesses), and disease (the law relating to pneumoconiosis was enacted independently). 25
Table 1 presents the proportion of labor regulations that were deregulated during the period of massive deregulation from 1998 to 2002; approximately half of all labor regulations were deregulated. Among the labor regulations, the proportion of deregulated regulations was significantly higher in the domain of OHS (54.5 percent) than in the other domains (43.6 percent). Among OHS regulations, those in the OHS management area presented the highest proportion of deregulation (69.3 percent).
The Characteristics of Labor Regulations.
Note. Values denote number of regulations (percent). OHS = occupational health and safety.
p < .1, **p < .05.
Factors Influencing Deregulation of Labor Regulations: Multiple Logistic Regression Analyses.
Note. Values denote odds ratio (95 percent confidence interval). OHS = occupational health and safety.
p < .05.
The three factors affecting deregulation can be applied to labor regulations in general, but must be filtered to be applied to OHS regulations, that is, to be the vulnerability factors of OHS regulations: factors that made OHS regulations more vulnerable to deregulation than the other labor regulations. In Table 2, the mean durations of both groups of regulations were not significantly different. In addition, in Table 3, the odds ratio of the duration was less than 1 only in OHS regulations, indicating a negative effect of the variable on OHS deregulation, contrary to what happened in the other labor regulations. Thus, although the factor affected the deregulation of labor regulations, it could not explain why OHS regulations were more vulnerable to deregulation.
The delegation of execution is also not adequate for explaining the vulnerability. If the delegation of execution was responsible for the vulnerability of OHS regulations, the proportion of delegated regulations in the OHS domain should be lower than that in the other labor domains, since delegation reduced the likelihood of deregulation. However, in fact, the delegated proportion was higher in the OHS domain (34.7 percent) than in the other labor domains (18.9 percent; Table 2).
Regulations without direct impacts were vulnerable to deregulation in the OHS domain as well as in the others (Table 3). At the same time, they were more prevalent in the OHS domain (31.7 percent) than in the others (24.1 percent; Table 2), although the statistical significance was somewhat low.
Therefore, among the three deregulation factors, the nondirectness of deregulatory impact was the most likely factor causing OHS vulnerability. In particular, the vulnerability factor would expand its effect in such an area containing many regulations without direct deregulatory impacts. The OHS management area was the prominent example of such an area: the proportion of regulations without direct deregulatory impacts in the area was 41.2 percent. This fact may explain why OHS management regulations were deregulated on a large scale as shown in Table 1. To confirm the effect of the vulnerability factor on OHS management regulations, an additional analysis was carried out.
Effects of the Nondirectness of Deregulatory Impact in OHS Management Area.
Note. Values denote number of regulations (percent). OHS = occupational health and safety.
Direct: regulations expected to have direct deregulatory impacts on the beneficiaries of the regulations (workers).
Not direct: Regulations expected to have no direct deregulatory impacts.
p < .1, **p < .05.
The relaxed or abrogated information regulations included various duties of businesses to inform the government: information on their OHS status and activities such as the outbreaks of major accidents, the results of working environment measurement and workers’ health check, the preparation of process safety reports and accident prevention plans, the appointment of OHS management personnel, and the use of OHS budgets. The deregulation also involved information regulations concerning the duty of businesses to cooperate with government investigations, such as inspection of the implementation of accident prevention plans, penalties for the refusal of inspections, and summoning employers to government meetings for OHS issues.
Discussion
Although the details of OHS activities may vary from country to country, the OHS management system in Korea is similar to those in other countries in that employers have duties to assess and control risks in workplaces and the government oversees the activities. Medium- and large-sized enterprises must appoint qualified full-time health and safety managers for risk management in workplaces or delegate the tasks to external OHS agencies, while small-sized enterprises with fewer than fifty employees are subject to a government program supporting OHS activities.26–28 By establishing various standards and monitoring, the government regulates the OHS activities of businesses and OHS agencies.
To find the factors causing the vulnerability of OHS regulations to deregulation, this study examined various characteristics of labor regulations and found three factors affecting deregulation: duration of regulation, delegation of execution, and (non) directness of deregulatory impacts. Among them, only the nondirectness of deregulatory impacts could explain the vulnerability in OHS. The actual effect of the factor on OHS regulations was the large-scale deregulation of information regulations that allowed the collection of information for government oversight of corporate OHS activities (Figure 1).
In order for the government to adequately supervise the OHS activities of businesses, it is necessary to stipulate OHS standards and observe the compliance of businesses with those standards. If the government has failed to properly monitor the situation of businesses due to the lack of information, it cannot establish OHS policies appropriate for the situation.
The large-scale deregulation in the reporting and inspection of OHS activities might have resulted in a lack of information not only on OHS activities but also on OHS status, such as illness and injury. Figure 2 shows the changes in occupational accident and disease outbreaks reported to the Industrial Accident Compensation Insurance, a national insurance program in Korea. For ten years from 2004 to 2014, the number of companies and workers covered by the insurance increased enormously: companies doubled and workers increased by 70 percent. (In 2004, 75 percent of all companies and 93 percent of all workers were covered by the insurance, and by 2014 almost all companies and workers were covered.) However, the reported numbers of occupational accident and disease outbreaks remained virtually unchanged. How can we interpret the drastic reduction in incidence and mortality rate?
Changes in occupational accident and disease outbreaks in Korea (among those covered by the Industrial Accident Compensation Insurance). Source: Statistics Korea.
29
First, changes in the industrial structure of Korea would not have contributed significantly to the reduction. Most of the increase in workers (81 percent) came from service industries such as retail, transport, and finance. It might have lowered the incidence of occupational accidents and diseases because the number of outbreaks in these industries was about half of those in manufacturing industries in Korea. However, 57 percent of the increase in workers came from small-sized enterprises (fewer than fifty employees), where the incidence rate is more than three times higher than that in other enterprises in Korea. Thus, even if mortality might be reduced due to the increase in small-sized service industries, the incidence could not.
Second, the effect of changes in the health status of the population can be considered. For example, during the period, the general population of working age experienced a dramatic reduction in deaths caused by cardio-cerebrovascular diseases (CVDs), along with an increase in life expectancy. This change in the health status of the general population has been reflected in the statistics of CVD deaths among workers, as shown in Figure 2; that is, without CVD, the reported reduction in mortality could not be so large. However, it is very difficult to explore this kind of effect of changes in the health of total population on OHS statistics.
Third, many government policies could have contributed to the reduction. However, considering that the number of businesses doubled and the number of workers increased by 70 percent in just ten years, it is difficult to explain such a large reduction in the incidence of accidents and diseases solely by the effect of government policies. Moreover, since the newly established businesses were mainly small in size, it was difficult to expect them to be prepared for the prevention of occupational accidents and diseases.
As the above interpretations could not give a plausible explanation, the possibility of information failure leading to an underestimation of the scale can be considered. In particular, the incidence rate can be underestimated since fatal injuries are difficult to be concealed. This would have contributed to the fact that the ratio of nonfatal to fatal injuries is lower in Korea (84) than in developed countries such as the United Kingdom (540) and United States (337), and even in countries such as Hong Kong (127) and Taiwan (96) where economic conditions are comparable. 30
In Korea, injuries requiring treatment for more than three days have to be reported to the government and treated through the Industrial Accident Compensation Insurance program. Injuries milder than these can be regarded as “Gong-sang,” which literally means injuries due to business affairs, and can be treated at the employer’s own expense without reporting to the government. However, the “Gong-sang” has often been used to conceal serious accidents at work in order to reduce accident rates. In fact, an investigation of subcontract workers in the shipbuilding industry revealed that only 7 percent of the injured workers had been compensated by the Industrial Accident Compensation Insurance, 60 percent had been treated via the “Gong-sang” system to conceal occupational accidents, and the rest had received healthcare services personally or had not been treated at all. 31
For businesses, maintaining the status quo is a virtue, because they benefit from it; that is, a high incidence of accidents and diseases above the usual level would result in a government investigation and an increase in OHS insurance contributions. Government officials also favor the status quo, as they are accountable for an unusual upsurge in accidents in their jurisdictions. In addition, OHS agencies in contractual relationships with the businesses tend to protect the interest of the businesses, insofar as they hope to continue with the contracts.
The finding that most OHS information regulations were the victims of deregulation supports the possibility of information failure. Although the information deregulation may not have been the fundamental cause of underreporting of accidents and diseases, it may have enticed businesses to hide accidents and diseases. In other words, the weakening of government surveillance could have facilitated the concealment of OHS accidents and diseases. Thus, to make OHS statistics and monitoring systems more reliable, it is necessary to restore the government functions that were weakened or replace them with new protective institutional arrangements.
In recent years, the importance of OHS surveillance has increased. In the public policy sector, regulatory restructuring has become a continuous event in contemporary society, as the flexibility and responsiveness of government regulation has been emphasized. That is, governments increasingly rely on corporate self-regulation through negotiations with industry to reflect the changing business environment.32–34 In this atmosphere, self-regulation in the OHS area is also emphasized. The conversion of OHS guidelines from prescriptive standards to process- or performance-based standards becomes a key issue in this area and calls for participatory mechanisms, which can be characterized by stakeholder self-regulation and government supervision: in brief, regulatory governance.35,36 This kind of change in regulation and governance made the situation more vulnerable to information failure; that is, inadequate and insufficient information can lead to more serious consequences than before. Self-regulation without proper monitoring can result in catastrophic consequences because it may not even expose OHS problems, as suspected in OHS statistics in Korea (Figure 2).
In methodological aspects, this research consulted several concepts related to policy termination and identified three regulatory features that facilitated deregulation: longer duration, nondelegation of execution, and nondirectness of deregulatory impacts. It is reasonable that being in force for a longer period of time renders a regulation more vulnerable to deregulation. Indeed, circumstances change over time, and such changes can reduce the usefulness of the old regulation, possibly leading to its termination. Conversely, however, when confined to the OHS area, the duration of regulation lowered the likelihood of deregulation (Table 3). Therefore, the effect of duration on deregulation needs to be investigated further.
Interest group theories of regulation argue that regulatory agencies are vulnerable to “capture” by interest groups and then protect related regulations. 37 The effect of delegation of execution on deregulation observed in this study is consistent with this argument, as it reflects whether the regulation has been executed by government departments or other organizations (i.e., regulatory agencies). The delegated organizations were governmental or private agencies that were partially or fully independent of the government. In the process of deregulation, these agencies may have protected the regulations related to them.
Nondirectness of deregulatory impact can also be a plausible explanation for facilitating deregulation. As described in the Methods section, regulations with this characteristic were about information, government funds, and the qualification of agencies. Because these issues rarely have direct influences on workers, regulators may relax or abolish related regulations without worrying about workers’ opposition.
Conclusions
Political pressures in the context of the national economic crisis resulted in Korean OHS deregulation that exceeded that of other labor regulations. This study revealed that the vulnerability of OHS regulations to deregulation was related to the nondirectness of deregulatory impacts on the beneficiaries of regulations, that is, workers.
In particular, this vulnerability factor was found in most regulations that allowed the government to collect information on the OHS activities of businesses. It is illogical to remove low-cost OHS regulations such as information regulations to mitigate business costs in order to overcome an economic crisis. Such deregulation is convenient for businesses but likely will trigger the concealment of industrial accidents. In fact, in Korea, many accidents that must be reported to the government and covered by industrial accident insurance are handled in such a way that the company or the workers themselves pay for the treatment. The large-scale deregulation of OHS information collection that took place in Korea seems to have facilitated the persistent and systematized concealment of industrial accidents. Government measures to overcome an economic crisis that include easing regulation of industry must not compromise protection of workers’ health and safety. This cannot resolve an economic crisis, and it merely shifts more costs to the victims of hazardous workplace exposures.
Limited OHS data collection could also lead to distrust of national OSH statistics. The fact that there has been little change in Korean OHS statistics despite the radical doubling of the number of companies and workers in a decade from 2004 to 2014 amplifies this suspicion. The practice allows the Korean government to establish OHS policies based upon unreliable statistics. Underreporting can lead to failures in the evaluation of OHS policies and also to the establishment of new policies which would be destined to fail. Even if the government’s ignorance of hidden injuries and illnesses is not intentional, the responsibility for failure is entirely with the government. Until the government strives to actively grasp the situation in the workplace, industrial accidents will continue to be concealed and the government’s efforts to prevent industrial accidents will be meaningless.
Thus, urgent restoration or strengthening of the government’s OHS monitoring function needs to be accomplished through reregulation to ensure accurate information collection and to establish policies that require businesses to accurately report their OHS data. New policies must give particular attention to the protection of workers in small business and precarious jobs. Currently, these sectors are in the blind spots of government surveillance.
Recently, the many national governments have established OHS regulations and programs that rely on employer self-regulation. Such programs require transparency to ensure proper monitoring from outside the industry. Workplace inspection and surveillance are needed in order to protect workers’ health and safety. The Korean example described in this paper indicates that strengthening government data collection for OHS likely is an essential task for building proper governance in all countries.
In addition to strengthening government OHS monitoring capabilities, regulatory changes in OHS must be monitored actively by all stakeholders, including workers, industry, and civic groups. However, workers themselves are probably the best watchers and activists because maintaining vigilance in this regard is in their own interest. Thus, the labor organizations, such as the trade unions, must play a critical role advocating for and providing oversight of appropriate worker protection regulations and compliance measures. In particular, the deregulation of information should be observed with caution, because it has great potential for being invisible to workers due to the lack of direct deregulatory impacts.
In the process of engaging in government policies, the labor community must obtain a strong alliance with other social stakeholders. It is necessary to cooperate with various groups such as politicians, researchers, and journalists, and especially other social movements such as public health and environmental protection to form a social consensus on the prevention of industrial accidents and to influence policy decisions. Further research regarding OHS policy initiatives is needed. In particular, as interest in regulatory governance in OHS continues to grow, research on the relationship between the structure and function of OHS governance and effective OHS management, including reliable data collection is urgently required.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Occupational Health Scholarship Program of the Korea Occupational Safety & Health Agency. This study was also supported by Soonchunhyang University, the Republic of Korea.