key: cord-0072009-cods81aa
authors: Jae Moon, M.; Suzuki, Kohei; Park, Tae In; Sakuwa, Kentaro
title: A comparative study of COVID-19 responses in South Korea and Japan: political nexus triad and policy responses
date: 2021-03-18
journal: nan
DOI: 10.1177/0020852321997552
sha: 976c9726d17d2f8bf691b514e939f0da6a0fab20
doc_id: 72009
cord_uid: cods81aa

Korea and Japan, neighboring democratic countries in Northeast Asia, announced their first COVID-19 cases in January 2020 and witnessed similar patterns of disease spread but adopted different policy approaches to address the pandemic (agile and proactive approach versus cautious and restraint-based approach). Applying the political nexus triad model, this study analyzes and compares institutional contexts and governance structures of Korea and Japan, then examines the differences in policy responses of the two Asian countries. This study first reviews the state of COVID-19 and examines changes in the conventional president-led political nexus triad in Korea and the bureaucracy-led political nexus triad in Japan. Then, this study examines how the differences in institutional contexts and governance structures shaped policy responses and policy outcomes of the two countries in managing the COVID-19 crisis. POINTS FOR PRACTITIONERS: • Institutional and governance structure in a society are likely to affect policymaking processes as well as selection of policies among various policy alternatives. • Government officials often need to refer to government capacity as well as citizens’ voluntary participation in resolving wicked policy problems like COVID-19. • Policy decisions made by government officials affect policy outcomes while political environment and political leadership are equally important to policy effectiveness.

COVID-19 has been overwhelming the world since it broke out in Wuhan, China, in December 2019. The number of confirmed cases continues to increase rapidly and reached 10 million in the six months since the first case was identified. This is in stark contrast to other coronaviruses, such as severely acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS), which reached only 8439 and 2519 confirmed cases, respectively. Countries, irrespective of economic level, region, or political system, have not been able to mitigate COVID-19 effectively. Speed of transmission and asymptomatic spread has made it difficult to mitigate the COVID-19 pandemic, forcing governments to test, trace, and isolate confirmed patients from the rest of the population.

Japan and Korea identified the first confirmed cases on January 16 and January 20, respectively. Following similar infection patterns, the number of confirmed cases in the two countries had increased to 70,000 in Korea and to 300,000 in Japan as of mid-January 2021. Having found the first confirmed case related to Wuhan, China, around the same time in mid-January, both countries contained the spread of COVID-19 effectively without implementing a hard lockdown policy, which has been the case in several nations, including China, France, Italy, and the US. Based on its experience with MERS, the Korean government has taken a proactive and aggressive testing, tracing, and treatment approach, while the Japanese government has relied on a cautious and self-restraint-based approach that basically focuses on testing only certain people who experience several symptoms, such as high fever, and treating high-risk patients while heavily relying on citizens' voluntary restraint (Jishuku) to prevent the spread of This study aims to examine how South Korea (hereafter, Korea) and Japan, two neighboring countries in Northeast Asia, have been responding to and mitigating the spread of COVID-19. We find that despite having the same target, which is to prevent the spread of COVID-19, the two countries have taken remarkably different approaches. The Korean government has taken an agile and proactive approach, while Japanese government policy is characterized by a cautious and restraint-based approach. Compared to many other Organisation for Economic Co-operation and Development (OECD) countries, however, both countries seem to have relatively low numbers of confirmed cases and deaths per million.

With the institutional and organizational perspectives on policy responses to COVID-19, we adopt the framework of the political nexus triad (PNT) (Moon and Ingraham, 1998) in order to examine major analytical dimensions of this special issue, including institutional context, governance arrangements, and actor strategies for policy actions of Korea and Japan. Based on the comparative studies of two selected countries, we argue that the difference in policy actions and approaches is partly due to the differences in institutional contexts and governance structures that are largely incorporated into the PNTs of the two countries. To analyze Korea's and Japan's policy responses to COVID-19, this study first reviews the background and spread of the disease. Then, it examines the policy responses of the two countries based on the PNT framework, focusing on the institutional and political contexts, as well as the governance structures and coordination mechanisms through which policy responses to COVID-19 are formulated. This study also discusses the results and impacts of the policy responses in the two countries, and then offers conclusions and policy implications.

The number of confirmed cases in Korea began increasing rapidly in the middle of February due to contagion spread by Shincheonji (a religious group) members. In Japan, the sharp increase in cases began at the end of March and in early April 2020. The number of confirmed cases in Japan began to exceed that in Korea on April 20, 2020, when the numbers of confirmed cases were 10,674 and 10,751 in Korea and Japan, respectively. Korea reached a plateau in the middle of April and Japan at the end of May 2020 but both continued to experience three waves of the COVID-19 pandemic.

The Johns Hopkins University Center for Systems Science and Engineering (CSSE) COVID-19 database shows how the number of daily confirmed COVID-19 cases and deaths per million changed. In the first and the second waves, the highest numbers of daily new confirmed cases per million in Korea and Japan were 11.88 in early March and 12.5 in early August, respectively. While the highest number of daily new confirmed cases per million was somewhat similar in both countries in the first and the second waves, it seems different in the third wave, which began at the end of 2020. The number of daily new confirmed cases per million in Japan increased to 50.96 on January 11, 2021, while that in Korea was 20.43 on December 25, 2020. As of January 15, 2021, the number of deaths (23.7) and number of confirmed cases per million (1390) in Korea are much smaller than those in Japan (32.6 and 2457), which are still much lower than those of other OECD countries. 2 Although Korea and Japan seem to have done well in their efforts to prevent the spread of COVID-19, citizens' assessments of the governments' approaches largely differ between the two countries. Korean citizens appeared to be largely satisfied with the government's actions against the virus. A national survey indicated that a majority (74.4%) 3 of citizens were satisfied with the transparent communication and agile response to the problem, which helped boost President Moon's approval rating and led to a great victory in the general parliamentary election in the middle of April 2020 (Moon, 2020) .

Japanese citizens, however, seem to be less satisfied with their government's policies despite the surprisingly low numbers of infections and deaths. According to a Gallup International poll conducted at the end of March 2020, only 23% of respondents were satisfied with the Japanese government's responses to the pandemic, leaving it ranked second from the bottom of 17 countries. This number is significantly different from the high satisfaction rates reported for Korea (74%) and other countries 4 (Gallup International Association, 2020). Public opinion polls conducted by the Japanese media also show low rates of citizen satisfaction (38% in Nihon Keizai Shimbun (2020a) and 37.4% in Jiji Tsushin (2020)). 5 Japanese citizens' dissatisfaction does not seem to reflect the relatively low number of virus infections in Japan compared to other major advanced countries, such as the US and European countries. However, it seems to reflect their frustration with the government's cautious approach in relying on citizens' self-restraint and personal hygiene. In fact, this gap has led to Japan being seen as a mysterious case by Western media. Japan was considered a "most likely case" for disastrous results due to several factors, including high population density, lack of virus testing, a soft approach without financial penalties, and the widespread perception that Japan mishandled the Diamond Princess cruise ship situation in February (McCurry, 2020; Sturmer and Asada, 2020) .

Institutional contexts in Korea and Japan for emerging COVID-19: the PNT models of the two countries Both Japan and Korea are advanced democratic countries, similar to most other countries examined in this special issue. Both are largely characterized by strong government intervention in the economy, elite bureaucracy, and bureaucratic discretion (Ginsburg, 2001) . However, they differ in many respects in terms of institutional and political contexts, as well as governance structures, which may have affected how these two countries responded to the pandemic. To compare institutional and political contexts of the two countries, we adopt the PNT model proposed by Moon and Ingraham (1998) . The PNT model basically depicts "an interactive power structure that is formed by the processes of politicization in which politicians, bureaucrats, and citizens communicate with each other and attempt to protect and increase their political and administrative power" (Moon and Ingraham, 1998 : 78) . For example, an ideally balanced PNT model might be characterized as an equilateral triangle indicating equal power relationships among politicians (i.e. president, party, legislative body), bureaucrats, and civil society (i.e. citizens, civil organizations, interest groups). However, each country has its own uniquely skewed model based on the power of the major political actors.

As shown in Figure 1 , 6 Moon and Ingraham (1998) suggest that Korea, traditionally defined as a president-led PNT, has been transformed into a balanced PNT through increases in bureaucratic autonomy, as well as the growing influence of citizen participation and the legislative body, whereas Japan, traditionally a bureaucracy-led PNT, has become a balanced system where the power of the Diet and prime minister continues to increase and "challenge the conventional model of bureaucratic supremacy" (Moon and Ingraham, 1998 : 94) . It should be noted that the unique PNT model of each country is not static, but dynamic and evolutionary, because interactions and power relationships among major political actors continue to change and often vary depending on the nature of particular policy issues.

As indicated in the PNT models, one major difference between the two countries is that Japan's parliamentary system places more constraints on leadership than Korea's presidential system (Est evez-Abe and Kim, 2014) . While Korea is unicameral and a president is directly elected by its citizens, Japan has a bicameral system, with a prime minister elected by members of the legislature. Although the second Abe cabinet recorded the longest consecutive tenure, Japanese prime ministers typically have very short terms, while Korea's presidents have only a single fixed term of five years, without the possibility of re-election. 7 Due to the electoral system, Korea's presidents tend to be more sensitive to the needs and voices of the majority of voters than Japanese prime ministers, who are repeatedly re-elected by the party. This creates "a strong incentive to appeal to the majority of the fellow Diet members from the same party" (Est evez-Abe and Kim, 2014: 672) . Bureaucrats in both countries are recruited mainly through competitive exams (Moon and Hwang, 2013) , and political influence on the personnel management of civil servants is limited. 8 However, the two countries differ in terms of the influence of bureaucrats relative to politicians, as depicted in the PNT models. While bureaucratic decisions are often bound by political dynamics and primarily by the president in the conventional Korean PNT model, Japan's conventional PNT model suggests that the bureaucracy enjoys more autonomy from politics. As noted earlier, the PNT models are not static, but continue to evolve, reflecting dynamic changes in the interactions of politicians, bureaucrats, and civil society (citizens) (depicted as solid arrow lines in Figure 1 ).

For example, while the image of a strong bureaucracy in Japan remains (Aoki, 2018) , as depicted in the original PNT model, recently, the Japanese bureaucracy has experienced gradual declines in its size, capacity, and influence. According to Maeda (2014 Maeda ( , 2018 , Japan records a remarkably low number of civil servants per capita compared to that in other advanced democratic countries. In fact, according to the OECD (2020), the OECD average for employment in the general government as a percentage of total employment was 18.06% in 2015, while Japan's rate was only 5.94% and Korea's was 7.61%. The overall number of personnel in the Japanese central government significantly declined after several government reforms (Nakamura, 2012) . In fact, Japanese researchers and practitioners have discussed the Japanese government's low capacity, especially the shortage of labor in the Ministry of Health, Labor and Welfare, in relation to its measures against COVID-19 (Kitamura, 2020; Sensho, 2020) . Furthermore, although it was not articulated in the original PNT model, Japanese bureaucracies adopt a careerbased closed system of recruitment for civil servants. Recruitment is strictly limited to the entry level, and civil servants enjoy lifetime employment. Mobility between public and private organizations is largely limited (Dahlstr€ om and Lapuente, 2017; Maeda, 2018) . Korea's president-led PNT model has also evolved into a more balanced model, where professionalism in bureaucracy has been enhanced by the introduction of open competition-based recruitment of government officials in addition to conventional closed exam-based recruitment.

Having introduced the institutional contexts of the two countries based on the PNT models, we examine the governance structures and coordination mechanisms through which main political and administrative actors have responded to the COVID-19 crisis. In Korea, while President Moon showed his strong political will in fighting COVID-19 from the beginning and the presidential office often works closely with the related agencies in deciding major policy positions and policy instruments (restrictions on schools and businesses, economic subsidies, etc.), the Korean professional bureaucracy has played a critical and autonomous role in handling and implementing the various policies. In contrast to its approach to international and inter-Korean affairs, in making COVID-19-related policy decisions, the presidential office has largely relied on the policy positions and scientific judgment of the responsible agencies and related professional communities, such as the medical community. This is partially due to the establishment and upgrading of the Korean Centers for Disease Control and Prevention (KCDC) and the policy lessons learned from the unnecessary politicization of infectious disease issues in the course of handling SARS in 2003 and MERS in 2015. The KCDC was strengthened and became an autonomous and professional vice-ministerial-level agency after the MERS crisis in 2015. During the COVID-19 crisis, the KCDC has been expanded and upgraded to an autonomous and independent agency, whose employees increased in number from 907 to 1476 on September 12, 2020 (Kwon, 2020) . In fact, the Korean government learned important policy lessons after the painful MERS experience, which became a strong foundation for strengthening its institutional capacities, building collaborative partnerships with local governments and private organizations, and establishing effective procedures for testing, tracing, and treating potential and confirmed cases (Lee et al., 2020; Moon, 2020) . This immediately became a key approach as soon as the first confirmed case was identified in early 2020.

Although the presidential office works with the executive branch, the highest official authority in the Korean government for the COVID-19 response is the Central Disaster and Safety Countermeasure Headquarters (CDSCH), headed by the Prime Minister. The Minister of Health and Welfare and the Minister of Interior and Safety assume the Vice Head 1 and Vice Head 2 positions, respectively, in the CDSCH. The CDSCH involves 23 ministries, as well as the 17 provinces and cities of Korea. One of the most significant purposes of the CDSCH meeting is the enhancement of effective communication between the central and local governments at the highest level to address crucial problems and find collaborative solutions. In practice, the CDSCH meeting has convened every day 9 since late February, and the Minister of Health and Welfare has begun to chair committee meetings for in-depth discussions twice a week, once with the ministries and once with the provinces and cities. The Korean media has frequently used the English word "control tower" to describe the important function of the ad hoc authority in controlling and managing the national crisis. Since raising the country's infectious disease alert to the highest level on February 23, 2020, the Korean government has assembled the CDSCH to bolster a whole-government approach. The response system in Korea is essentially government-driven; however, the close partnership with expert groups and ordinary citizens has made a significant contribution to the effective countermeasures against COVID-19.

Since the launch of the CDSCH, it has played a central role in preparing for potential massive disease spread by strengthening response capacities in collaboration with local governments and medical communities. The Korean government made efforts to encourage citizens to participate in nonpharmaceutical interventions (NPIs), such as social distancing, hand washing, and mask wearing, while it took an aggressive and proactive approach by testing and tracing potentially infected individuals, followed by identifying and separating confirmed patients for medical treatment. This helped to flatten the COVID-19 curve, despite the sudden peak caused by Shincheonji members, and the number of daily new confirmed cases dropped from over 900 in March to under 10 in the middle of April, allowing Korea to effectively hold its general parliamentary election (Lee et al., 2020) .

It is also worth noting how Korea's political environment is faring in the face of the crisis caused by COVID-19. Political issues, such as the general parliamentary election and presidential approval ratings, were affected by the pandemic. Above all, COVID-19 emerged as the most important topic of conversation for Korea's 21st general election, held on April 15. The ruling party appealed to citizens to empower the government and the president during this national crisis, emphasizing that the Korean government and President Moon Jae-in have been praised for effectively handling the COVID-19 pandemic. In contrast, the opposition party strongly criticized the failure of government policies, which, according to them, resulted in a decline in economic growth and an increase in unemployment, pointing out that the situation worsened because the government did not initially block entry from China in the early stages of the pandemic. Overall, however, there were more Korean citizens who trusted the government and expressed satisfaction with the government's COVID-19 response, which led to an overwhelming victory for the ruling party and a voter turnout of 66.2%, the highest in the 28 years since the 14th general parliamentary election, which was held on April 15, 2020. Thanks to citizens' support for the COVID-19 response of the Korean government, President Moon's approval rating exceeded 60% after the general parliamentary election, though it later declined and stayed at approximately 50%.

While the president's political power is still dominant in the Korean system, the ongoing shift from the dominant president-led PNT to a more balanced distribution of power among the president, professional bureaucrats, and civil society (particularly the policy community) has helped the Korean government effectively cope with this serious problem. Many point out that the government's past failure in its initial response to the MERS outbreak in 2015 partly, if not completely, contributed to the effective policy responses designed to mitigate the surge of the virus (Comfort et al., 2020; Kim, 2020; Moon, 2020; Normile, 2020) . Comfort et al. (2020) contend that these experiences increased the level of "cognition" of COVID-19 in Korea as a serious risk among professionalized public agencies such as the KCDC and individual citizens.

The balanced PNT model also enabled the Korean government to take an agile and proactive approach, which contributed to the effective mitigation of COVID-19. The Korean government primarily relied on three Ts (aggressive testing, tracing, and treatment) in mitigating COVID-19, based on the initial policy positions shaped jointly by the president and professional bureaucrats. With strong support from the president, the Korean government used several policy instruments, such as free testing for those potentially exposed to infected individuals and for highrisk populations, such as nursing house residents and international entrants. It also aggressively used public information campaigns to promote NPIs and encourage citizens to voluntarily participate in social distancing and individual sanitization. Various open data-based apps, such as Coronamap or Maskfinder, have also been developed as outcomes of coproduction (Lee et al., 2020; Moon, 2020) .

The Korean government also adopted highly adaptive and flexible methods as the number of confirmed patients increased (Moon, 2020) . One of the globally acknowledged examples is the drive-through and walk-through screening stations that have now been introduced in many other countries. The Korean government sought to put "science" over "policy" or "politics" by making science-based policy decisions mainly through the KCDC and medical and scientific communities. On September 12, 2020, the status of the KCDC was further strengthened when it was promoted to a state-level, independent, and autonomous agency overseeing the country's infectious disease response. This occurred when the cabinet approved the elevation of the KCDC and separated it from the Ministry of Health and Welfare (MOHW) to become the Korea Disease Control and Prevention Agency (KDCA), granting it autonomy in infectious disease policymaking and enforcement. Additionally, to enhance specialization, the position of the second vice minister in charge of health and medical care was also established at the MOHW.

Input from scientists and frontline health workers, rather than political calculation, has influenced the crisis management efforts of the Korean government. This helped to prevent overpoliticization of the COVID-19 issue, which could have occurred under the conventional president-led PNT. In fact, policy decisions and responses to COVID-19 outbreaks in Korea actually reflect the shift from the president-led PNT to a more interactive and balanced PNT in which political interactions among the three actors-the president, bureaucrats, and citizensare becoming increasingly balanced through the president's sense of accountability and leadership, bureaucrats' administrative power based on professional expertise (including the medical policy community's professional contributions), and citizens' voluntary engagement.

From the perspective of opportunity management, it should also be noted that the Korean government has taken several key actions to accelerate digital transformation of the economy and society in preparing the post-COVID-19 era under the name of the Korean new deal policies (Digital New Deal and Green New Deal). 10 The new deal policy initiative is basically the Korean government's strategic action to turn the COVID-19 crisis into future opportunities by transforming its economy and society. In response to COVID-19, the Korean government also continues to stress inclusive growth and social values in order to resolve increasing inequalities, while aiming to upgrade its digital government services and to enhance the quality of public healthcare services and administrative capacity of the public healthcare agency.

Different from those of Korea, the governance structure and coordination mechanisms in Japan's policy responses to the COVID-19 response are characterized by rather looser policy coordination and stronger restrictions upon the administration than in Korea. First, unlike Korea, Japan does not have an independent government agency that controls responses to infectious diseases-the ad hoc Novel Coronavirus Response Headquarters (NCRH) is the highest authority for the COVID-19 response. Second, the Japanese legal tradition does not allow for granting the government emergency powers. The ruling Liberal Democratic Party's (LDP's) proposals to expand the government's emergency powers faced strong opposition from bar associations, law professors, and news organizations, despite the Abe administration's electoral popularity (Repeta, 2020) . Third, economic policy has been central to the LDP, especially under Abenomics, so the mitigation of COVID-19 was not initially considered a top policy priority, whereas the Korean government put an emphasis on the mitigation of COVID-19. During the 2011-2016 period, donations grew for five straight years, with nearly 90% going to the LDP. The steep rise in LDP donations came amid annual calls since 2014 by the Japan Business Federation, also known as Keidanren, for member companies to make contributions (Nikkei, 2017) . The centrality of economic policy and connections with the business community was an underlying political circumstance of the Abe administration.

The cabinet decided to set up the NCRH, led by Prime Minister Abe, on January 30, 2020, when Japan had 11 cumulative confirmed cases. The Ministry of Health, Labor and Welfare (MHLW) was the main bureaucratic actor involved in decision-making. On February 16, 2020, the Japanese government established the Novel Coronavirus Expert Meeting, which is chaired by the head of the National Institute of Infectious Disease (NIID). The Expert Meeting, consisting of scientists with relevant expertise, mainly reports scientific findings and offers policy advice to the Prime Minister and the cabinet on matters related to COVID-19. Later, the NCRH was officially established when a related law was passed on March 26, 2020. The NCRH, led by the Prime Minister, is supported by two major ministers, the Chief Cabinet Secretary and the Minister of the MHLW.

Coordination between the parties involved has not gone smoothly from the beginning. It has been repeatedly reported that these important actors have often encountered coordination problems, especially between the MHLW and other actors. For example, due to the shortage of personnel, it was reported that the MHLW strongly opposed the Prime Minister's idea in early March to require all travelers from overseas to be quarantined in designated facilities and let returnees with no COVID-19 symptoms go home by public transport (Yomiuri Shimbun, 2020a).

The Abe administration was also concerned about the potential impact of COVID-19 on the Tokyo 2020 Olympics, which was eventually postponed to the following year after negotiations between the Japanese government and the International Olympic Committee (IOC) on March 25, 2020. Before the postponement decision was made, the Abe administration had advocated strongly for the Summer Olympic Games to be held as scheduled. In his statement to the Parliament on February 6, 2020, Prime Minister Abe made clear that his administration would not cancel or delay the Tokyo 2020 Olympics, despite international concerns about COVID-19 (Reynolds and Hirokawa, 2020) . The political context of the Tokyo 2020 Olympic Games appeared to influence the somewhat passive initial responses of the Abe administration to COVID-19. In fact, the number of confirmed cases began to rise sharply after the postponement decision was made.

Unlike the KCDC in Korea, Japan does not have an independent agency for controlling and preventing infectious diseases; it only has the NIID, which is responsible for research and administrative support. Without having previously experienced severe outbreaks, such as MERS, the Japanese government was not well prepared for COVID-19, and it was less proactive than the Korean government in coping with the virus. As Figure 2 suggests, Korea and Japan have similar institutional arrangements for fighting against COVID-19. In both countries, the prime minister heads the central COVID-19 response headquarters (i.e. the CDSCH in Korea and the NCRH in Japan), and under them, two major ministers play critical roles as vice heads. Despite these similarities, the KCDC, as the central disease control headquarters, plays a more essential role in making critical decisions on the response to COVID-19 than the NIID in Japan. Although the KCDC is part of the CDSCH headed by the Prime Minister, it serves as the command center for infectious disease control and prevention based on its professional expertise in infectious disease (MHLW, 2020) .

The NCRH lacks coordination without a central agency like the KCDC, and it is fair to say that the Japanese government has not been successful in terms of proactive responses and communication management with regard to COVID-19. The ineffective quarantine measures enacted on the British-flagged Diamond Princess cruise ship in February 2020 drew much criticism from the national and international press. The sudden official announcement on February 27 requesting all the country's schools to close beginning on March 2 also created confusion and shock among teachers and parents (Rich et al., 2020) , since it was a request, not a mandate. The decision to close schools was left to local authorities and schools; however, almost all schools closed. In contrast, the Korean government's school-related decisions were made by the Ministry of Education in consultation with the KCDC and expert groups. Parents were not informed about the school closure set to begin the following week until Friday afternoon. Moreover, approximately 60% of households with children under the age of 18 are dual-income households (Kuga, 2020) , and schools were not ready for a shift to digital forms of teaching, sowing further confusion among teachers and families (O'Donoghue, 2020).

Loose policy coordination was again observed when the central government lifted the state of emergency in all prefectures on May 25. For example, local governments, such as Tokyo's, implemented a gradual reopening of economic activities, and all the requests (with a few exceptions) for business suspension were lifted on June 19, accompanied by requests that stores and businesses implement measures to prevent COVID-19 infections and actively conduct health checkups of employees (Jiji Press, 2020). When the government declared a state of emergency in seven prefectures, including Tokyo and Osaka, on April 7, the Abe administration called on citizens to exercise "self-restraint" (jishuku in Japanese) in refraining from activities that involve direct interhuman contact, and contacts subsequently decreased by 80%. Following the declaration, local governments "requested" targeted industries, ranging from museums and schools, to bars and nightclubs, to suspend business. These rules were requests, and there was no penalty for noncompliance-the administration has been reluctant to enforce measures that restrict private rights. On June 15, Prime Minister Abe insisted that the introduction of penalties should be considered with caution (Yomiuri Shimbun, 2020b). Such requests have been quite effectively enforced, even without legal penalties. Although the reason for this has not been systematically explained, it is sometimes attributed to Japan's unique culture of social norm compliance. In extreme cases, stores that (legally) continued business were insulted and pressured to stop operation by local citizens, called jishuku keisatsu, or the "self-restraint police" (Fujii and Hata, 2020) . Those who defied social norms (including mask wearing), authorities' stay-at-home requests, and business suspension requests were often harassed by these local citizens (Osaki, 2020) . The reason for reliance on self-restraint is also attributed to Japan's unique constitutional regime. The contemporary Japanese constitution lacks explicit emergency provisions, and there has been strong resistance to legislation that would give the government emergency powers. The prewar Imperial Constitution had emergency provisions that often led to the abuse of power during "national crises," such as the Great Kato Earthquake of 1923 and the period of war spanning the 1930s and 1940s. Therefore, the legal basis for compulsory measures during a crisis has been severely limited (Oya, 2020) .

The role of experts in crisis management is relatively limited, in that they "advise" the policymaking of the administration and bureaucracy; however, experts neither make decisions nor implement policies. The closed employment system for civil servants and limited use of external experts may be one reason for the passive use of experts in the policymaking process. Japan also lacks an expert-led independent decision-making body similar to the Centers for Disease Control and Prevention (CDC) in the US or KCDC in Korea. The NIID is an outpost agency of the MHLW and is not granted decision-making autonomy. The Novel Coronavirus Expert Meeting, led by NIID Director Takaji Wakita, is a collective body consisting of academics and infectious disease experts; however, its role is limited to advisement-the nationwide emergency closure of schools, "requested" by Prime Minister Abe, was not even advised by the Expert Meeting (Nihon Keizai Shimbun, 2020b).

As Korea launched the new deal initiative, the Japanese government has also taken the crisis as a window of opportunity to promote remote working and digital transformation. Upon the inauguration of the Suga administration, for example, the Japanese government announced its plan to strengthen digital government by establishing a digital agency, which helps to promote digital reform in the public sector (The Economist, 2021).

In summary, policy responses to COVID-19 in Japan reflect to some degree the ongoing shift from a bureaucracy-led PNT model to a new model in which the policy influence of the bureaucracy is gradually decreasing while the prime minister and the Diet are gaining more policy influence over the bureaucracy than before. Under the new PNT model, policy coordination among government agencies tends to be much looser than before, while policy responses to major social issues such as COVID-19 are easily politicized by the prime minister's policy priorities and ideological orientation, as seen in the initial impacts of Abenomics and the Tokyo Olympic Games on the policy response to COVID-19. Despite cautious and rather loose policy coordination among various policy actors, Japan relied on the strong self-restraint of citizens, which allowed Japan to mitigate COVID-19 more effectively than many other European countries.

Since their first confirmed cases were detected in the middle of January 2020, both the Korean and Japanese governments have paid attention to the potential risks of COVID-19 and have made efforts to mitigate the unprecedented disease amid a number of political, institutional, and socioeconomic challenges. As Figure 3 shows, the ratio of confirmed cases to the number of tests is 0.014 in Korea and 0.047 in Japan. This means that there is only one confirmed case per one hundred COVID-19 tests in Korea while the ratio is much higher in Japan than Korea, which is primarily thanks to cautious and passive testing policy in Japan. The fatality rates of Japan, however, are 0.015 which is similar to that of Korea, one of the lowest rates among OECD member countries. This suggests that the different approaches of two countries are closely associated with spreads of COVID-19 but not necessarily with fatality rates though the number of deaths per million of population is much higher in Japan than Korea.

Due to its painful, though valuable, past experiences with SARS and, particularly, MERS, the Korean government appeared to be better prepared to respond to COVID-19 in an agile and adaptive way (Lee et al., 2020; Moon, 2020) . Agile and proactive COVID-19 testing was a primary approach that helped mitigate and prevent the virus effectively, particularly in the initial stage. Korea's agile and proactive approach based on massive and aggressive testing, tracing, and treatment (the three Ts) appears to be effective in mitigating COVID-19 and has kept the fatality rate lower than that of Japan. The Korean government was able to introduce the agile and proactive approach in the early stage due to the President's strong leadership and his explicit reliance on various policy recommendations made by the KCDC, which played a central role in key policy decisions based on professional and science-based evidence. The Korean government also quickly raised the alert to the highest level and launched the CDSCH, through which interagency as well as intergovernmental communication and collaboration were effectively monitored and promoted in responding to daily developments, such as changes in the number of new confirmed cases and deaths. The CDSCH, headed by the Prime Minister, also played a critical role in coordinating policies for mitigating and preventing massive infection by collecting policy information and coordinating policy instruments among various policy actors, including central government agencies, local governments, various interest groups (businesses, retailers and restaurants, schools), medical communities and expert groups, and civil organizations. There was also a mature sense of citizenship and voluntary participation in personal hygiene, social distancing, and other constraints, such as voluntary quarantines, facial mask requirements in public facilities, and entry check-in for restaurants. This suggests how the conventional president-led PNT model has been shifting toward a more balanced PNT model, in which the president, bureaucrats, and civil society are fairly well balanced.

Compared to Korea, Japan took a relatively cautious and self-restraint-based approach in the initial phase. The Japanese government primarily relied on the self-restraint of individual citizens and was less proactive in identifying those who were eligible for COVID-19 testing and tracing potentially infected individuals, partially due to its initial concern about the potential impact on the 2020 Tokyo Olympics. The Prime Minister's policy influence appeared to be more influential than that of the bureaucracy in making key decisions regarding COVID-19. Japan adopted rather loose policy coordination among central government agencies, as well as between the central and local governments, partially because of the lack of legal basis for compulsory crisis measures (Oya, 2020) and nonexistence of an independent administrative agency for disease control and prevention, such as Korea's KCDC. Without clear policy prioritization of a proactive response to COVID-19 by the Prime Minister, Japan's homogeneous and closed careerbased bureaucratic system (Aoki, 2018; Dahlstr€ om and Lapuente, 2017; Mishima, 2017) appears to be ill equipped to deal with this unprecedented crisis. Despite the rather loose and cautious policy coordination, it should also be noted that citizens' self-restraint and their adherence to social norms and strong awareness of public hygiene are key factors contributing to the effectiveness of the Japanese government's policies. This shows that these institutional and social contexts partially contributed to the Japanese government's relatively cautious and self-restraint-based approach, rather than the agile and proactive response adopted in Korea.

Relatively speaking, the two selected neighboring Asian countries effectively managed the outbreak without completely closing their borders or instituting massive domestic lockdowns. The two different approaches of Korea and Japan demonstrate how different political and institutional factors affect policy choices, such as the decision to adopt an agile and proactive approach or a cautious and selfrestraint-based approach. The former requires strong central leadership in mobilizing the necessary financial, human, and technological resources, as well as in coordinating policy alternatives among key policy actors, while the latter entails the strong voluntary self-restraint of citizens, as well as loose policy coordination without strong top-down leadership. The ratio of confirmed cases out of tests as well as the number of deaths per million began to rise much steeper in the recent third wave in Japan than Korea, which seem to suggest that a loose policy coordination model primarily relying on self-restraint is much more vulnerable and unstable than strong policy coordination with active voluntary participation of citizens (Moon, 2020) . Of course, no single theory prevails regarding how institutional and governance factors are associated with crisis management and its success simply because there are many different explanatory factors, including government capacity, past experiences and learning, interaction between policy and science communities, intersectoral collaboration, intergovernmental cooperation, and citizen participation. Future studies need to conduct both theory-based and practice-oriented comparative studies by investigating different countries with diverse political and institutional arrangements, as well as social norms and citizenship, and then assessing the alternative policy responses and their policy outcomes in the fight against COVID-19.

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. 

M. Jae Moon https://orcid.org/0000-0002-6499-5700 Kohei Suzuki https://orcid.org/0000-0002-5403-4826 Tae In Park https://orcid.org/0000-0002-3485-814X Kentaro Sakuwa https://orcid.org/0000-0002-5921-851X Notes 1. The data were acquired from the following website: https://ourworldindata.org/ coronavirus 2. For example, as of January 15, 2021, the number of deaths per million was 1337 in Italy, 1269 in the UK, and 1174 in the US. 3. An international citizen satisfaction survey (Gallup International Association, 2020) also confirmed that the satisfaction rate of Korean citizens with the government's approach to COVID-19 is very high (74%), as noted later.

4. For example, Austria (88%), India (83%), Malaysia (77%), and the Netherlands (79%).

Participants were asked, "How strongly do you agree or disagree with the following statements?-I think the Government is handling the coronavirus well." Responses ranged from "strongly agree" to "strongly disagree," or respondents could choose "do not know." Respondents who selected "strongly agree" or "agree" were considered satisfied citizens. 5. Those respondents who selected "rate it highly" (hyouka suru) were considered satisfied. 6. The arrow line indicates the shift from a conventional model to a new model, while the solid triangle and dotted triangle are the conventional triad and new triad, respectively. 7. Prime Minister Abe, followed by Prime Minister Suga, announced his resignation on August 28, 2020. 8. For details on civil service systems in South Korea and Japan, see Moon and Hwang (2013) . Both countries are ranked very high in terms of administrative professionalism, as measured by the degree of political influence in the personnel management of civil servants. The two countries belong to a group of countries with high levels of administrative professionalism, such as New Zealand, Ireland, Norway, Hong Kong, Denmark, Sweden, Switzerland, the UK, the Netherlands, and Canada. For international comparisons, see Dahlstr€ om et al. (2012) and Suzuki and Demircioglu (2019) . 9. There have been very few exceptions, including the day of the nationwide elections. 10. Digital New Deal includes stronger integration of DNA throughout the economy, digitalization of education infrastructure, fostering the noncontact industry, and digitalization of social overhead capital. Green New Deal includes free transition of infrastructure, low-carbon and decentralized energy, and innovation in the green industry (UNDP Seoul Policy Center, 2020). 

Japan's bureaucracy in international perspective

Republic of Korea (no date) Korea's response to COVID-19 and future direction

Japan's crisis management responses to COVID-19 and prospects. Issue report, the Asan Institute for Policy Studies

Crisis decision making on a global scale: Transition from cognition to collective action under threat of COVID-19

Organizing the Leviathan: How the Relationship Between Politicians and Bureaucrats Shapes Good Government

(eds) Good Government: The Relevance of Political Science

Presidents, prime ministers and politics of care-Why Korea expanded childcare much more than Japan

Kyugyo semaru "jishuku keisatsu" dema kakusan de kyaku gekigen-sutoresu de kishimu shakai [The "restraint police" are pressing for closure. The number of customers

The coronavirus: A vast scared majority around the world

Dismantling the "developmental state

Health Policy Bureau (2020) ICU-tou no byosho ni kansuru kokusaihikaku ni tsuite

To no kyugyo yosei ga zenmen kaijo shinjukuku "yorunomachi" sekkyoku kensa-shingata korona [The Tokyo Metropolitan Government's request to close the office was fully lifted

Naikakushiji 38%, fushiji 61% Shingatakorona taiou 6 wari hyoukasezu jiji yoronchousa [Cabinet approval rate 38%, disapproval rate 61%. 60% of the respondents did not evaluate the government responses to the new coronavirus

2020) Weekly report on the COVID-19 situation in the Republic of Korea (as of

South Korea's fast response to coronavirus disease: Implications on public policy and public management theory

Nihon no gyousei wa surimu sugiru

Shingata korona, kyukoude kosodatekatei daikonranno mittsuno haikei [Three reasons why the novel corona virus and school closures are wreaking havoc on families raising children

KCDC, upgraded by to 1,476 government officials

Policy learning and crisis policy-making: Quadruple-loop learning and COVID-19 responses in South Korea

Costly lessons from the 2015 Middle East Respiratory Syndrome coronavirus outbreak in Korea

From near disaster to success story: How Japan has tackled coronavirus. The Guardian

Shimin wo yatowanai kokka [A State Not Actively Employing Its Citizens

The Japanese civil service

Iryo shisetsu (seitai, dotai) chosa, byoin hokoku no gaikyo [Summary of the medical facilities (static and dynamic) survey and hospital reports

MHLW (2020) Shingata koronawirusu nikanrenshita haienno kanjanohassei nitsuite (1reime) [Outbreak of pneumonia associated with the novel coronavirus

Keizai sangyoshou no shiensaku [METI's support measures

A big bang for Japanese mandarins? The civil service reform of 2014

Republic of Korea) (no date) Korea's evolving response to COVID-19 (as of 11

Fighting COVID-19 with agility, transparency, and participation: Wicked policy problems and new governance challenges

The state of civil service systems in the Asia-Pacific region: A comparative perspective

Shaping Administrative Reform and Governance: An Examination of the Political Nexus Triads in Three Asian Countries

Asian model of government re-examined in the aftermath of the global economic crunch: A Japanese perspective from the experience of the triple disasters in

NHK (2020) Naikaku Shijiritsu

Korona taiou hyoukasezu 55% yoronchousa naikakushijiritsu yokobai 49% [Disapproval rate for government's response to corona 55%, cabinet approval rate unchanged at 49%

Nihon wa senmonka no shireito-soshiki nashi kengen to sekinin no meikakuka fukaketsu

Japan's ruling party gets lion's share of business donations

NNN and the Yomiuri Shimbun National Polls (2020) Yoron Chousa

have-dropped-sharply-south-korea-whats-secret-its-success O'Donoghue JJ (2020) In era of COVID-19, a shift to digital forms of teaching in Japan. The Japan Times

Government at a Glance -2017 Edition: Public employment and pay

? QueryId=78408# (accessed October 9

Japan's "virus vigilantes" take on rule-breakers and invaders. The Japan Times

Nase Nihon wa Jishuku no Onegai Shika Dekinai no ka: Oya Takehiro san no Baai Zempen

The coronavirus and Japan's Constitution

Japan shocks parents by moving to close all schools over coronavirus. The New York Times

Hitodebusoku to gyoumu kata de kasumigaseki houkai no pinchi [Shortage of labor and overwork will cause the collapse of Kasumigaseki

Japan was feared to be the next US or Italy. Instead their coronavirus success is a puzzling "mystery

The association between administrative characteristics and national level innovative activity: Findings from a cross-national study

Japan's new prime minister drags government into the digital era

Korea new deal for the post-COVID-19 era

Teiryu Kantei Ni Hampatsu [The MHLW falls behind in border measures: Disagree with the cabinet's idea for stopping for quarantine

Kyugyoyosei ni bassoku hiteiteki shusho "shiken ni okina seiyaku