Differential Assessment for the Effect of Government Epidemic Prevention Policies on Controlling the COVID-19: The Experience of Taiwan

ABSTRACT

The effect of government epidemic prevention policies on controlling COVID-19 is of great importance to health, politics, and development economics. This paper thus investigates the impacts of government responses on confirmed cases related to COVID-19 in Taiwan for the period January 1, 2020 to May 13, 2021 by employing ordinary least squares (OLS) estimation. Overall, our empirical results indicate that there is a significantly impact of government responses on COVID-19 pandemic spread in Taiwan. In addition, the speed of government responses would significantly affect confirmed cases of COVID-19 in Taiwan. The earlier government epidemic prevention responses led to fewer confirmed cases. After conducting a series of robustness checks, the above conclusions are still robust.

KEYWORDS:

• Government responses
• COVID-19 pandemic
• Taiwan

JEL:

• H51
• H75
• I18

Acknowledgments

Jun Wen thanks financial support from the National Science Foundation of China (Grant No. 72074176). Chyi-Lu Jang and Chun-Ping Chang thank support from Ministry of Science and Technology ,110WFA0810556. We are solely responsible for any errors and omissions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/1540496X.2022.2073818

Notes

1. COVID-19 is a kind of epidemic that refers to infectious diseases that can be transmitted to a large number of people in a short period. Infectious diseases similar to COVID-19 include the Black Death in medieval Europe, Ebola in Africa, H1N1 influenza virus, Severe Acute Respiratory Syndromes (SARS), and so on.

2. It is worth noting that since the outbreak of the new crown epidemic and its widespread global spread, Taiwan’s epidemic prevention and control has performed well in the previous year and a half. However, since mid-May of this year, the epidemic situation in Taiwan has suddenly deteriorated. The epidemic situation is currently in a “high-altitude period” with high confirmed cases, an increase in deaths, and a mortality rate higher than the global average, which poses a huge challenge for Taiwan’s government to intervene.

3. For example, Hogan et al. (2020) demonstrate that the increasing cases cause substantial damage to the health system.

4. Serafini et al. (2020) denote that many people suffered from various psychological problems such as stress, anxiety, depression and frustration during the outbreak and continuous spread of the pandemic.

5. Resources of OxCGRT are obtained from the website: https://github.com/OxCGRT/covid-policy-tracker.

6. Among all the government response measures, S_closing, I_support, T_policy and F_coverings are ordinal variables which are used to measure the severity/intensity of these response measures. Other measures such as GI_support, E_investment and I_vaccines are numeric variables which measure a specific number.

7. The reasons why the speed of government response is calculated as the number of days between Taiwan recorded its first COVID-19 case and when it reached a stringency level of 30 out of 40 are as follows: the variable of the stringency level ranges from 0 to 100. The larger the value is, the stricter is the government’s corresponding measures. Hale et al. (2020c) calculated the speed of government response as the number of days between when a country records its first COVID-19 case and when it reaches a stringency level of 40 out of 100. However, during the sample period, the maximum value of stringency level in Taiwan is no more than 40. Therefore, the Speed variable in Taiwan is calculated as the number of days between when Taiwan recorded its first COVID-19 case and when it reached a stringency level of 30 out of 40 in this paper.

8. The results of E_investment, I_vaccines and Speed are not reported because in the limited sample, the number of these variables are the same without fluctuations, which makes the regression results omitted.

9. It is worth noting after expanding the sample period that there are no data about E_investment and GI_support in the website of https://github.com/OxCGRT/covid-policy-tracker. Therefore, there are no empirical results of these two variables.

10. It is worth noting that we also conduct other robustness tests by using other methods or alternative definitions and dealing with the possible endogeneity problems caused by various causes (Feng et al. 2021; Long et al. 2021; Wang et al. 2022; Yang et al. 2022; Zheng et al. 2021), which are listed in the supplementary materials. If the reader needs, you can download it yourself.