2019冠狀病毒疾病封城措施對總體經濟之影響：時間序列分析

Abstract

背景與目標：新型冠狀病毒(COVID-19)於2019年底爆發後，迅速擴散到世界各地，造成全球大流行，疫情仍在持續延燒，截至2022年5月，全球已超過五億多人確診，六百多萬人不幸死亡。為了控制疫情，各國紛紛採取封城及各項管制措施，以遏止疫情擴散，例如關閉學校、工作場域及非必要商店等。儘管封城措施能有效防止COVID-19疫情擴散，卻可能導致經濟活動放緩，對總體經濟產生不利影響。本研究旨在利用量化分析，探討英國封城措施是否對該國經濟帶來不利的影響。 方法：本研究使用時間序列模型與動態迴歸模型，針對五個不同的總體經濟面向：勞動力市場、製造面、消費面、國際貿易與股票市場進行量化分析，並納入防疫嚴格指標(封城效應)、COVID-19確診人數、死亡人數及基準利率作為輸入變量，進行英國封城措施對於經濟影響之實證研究。另外，本研究亦將不封城的日本、臺灣作為描述性對照，與封城的英國進行比較，探討不同防疫策略對於總體經濟指標的影響。 結果：研究結果顯示，COVID-19的封城措施將為勞動力市場、製造面與消費面市場帶來不利的影響。失業率的時間序列模型為ARIMA(4, 1, 2)，納入輸入變量後，防疫嚴格指標的參數估計為0.0015(p=0.0218)。製造業採購經理人指數(PMI)的時間序列模型為ARIMA(2, 1, 1)，納入輸入變量後，防疫嚴格指標的參數估計為-0.15(p<.0001)。服務業採購經理人指數(NMI)的時間序列模型為ARIMA(3, 0, 1)，納入輸入變量後，防疫嚴格指標的參數估計為-0.23(p=0.0087)。零售銷售指數(RSI)的時間序列模型為ARIMA(0, 1, 3)，納入輸入變量後，防疫嚴格指標的參數估計為-0.3(p<.0001)。國際貿易、英國股市與封城效應未存在顯著相關，但基準利率與英國股市存在顯著負相關。本研究亦發現，相較於不封城的臺灣與日本，英國的經濟活動放緩更為嚴重。 結論：COVID-19封城措施將導致經濟活動放緩，對於勞動力市場及供應鏈，甚至是實質GDP皆帶來不利的衝擊，總體經濟影響廣泛且充滿不確定性，因此，若要在防疫與經濟之間取得平衡，本研究並不建議以嚴格的封城措施作為防疫手段。

Background and objective: COVID-19 has spread rapidly across the world and caused a pandemic. As of May 2022, the number of confirmed cases surpassed 500 million and over 6 million deaths. Many countries enforced lockdown measures or restrictions to contain the spread of coronavirus. Shutting down public places, such as schools and non-essential shops has been the leading lockdown measure. However, lockdown measures have often been framed as a trade-off between protecting people’s health and protecting the economy. To discuss this issue, we aimed to analyze whether lockdown measures lead to an adverse impact on the economy in the United Kingdom. Methods: We used the time series model and dynamic regression to analyze five different economic aspects: labor market, manufacturing, consumption, international trade, and stock market. Using stringency index (lockdown effect), COVID-19 confirmed cases, deaths, and bank rate as input variables to prove that lockdown measures may impact the economy in the United Kingdom. We also took non-lockdown countries, such as Taiwan and Japan as a descriptive comparison to discuss the economic impact of two different measures. Results: Based on our results, the major adverse macroeconomic impact of COVID-19 lockdown is on the labor market, manufacturing, and consumption. The unemployment rate time series is ARIMA(4, 1, 2) model, and after adding the input variables, the coefficient of the stringency index is 0.0015 (p=0.0218). The manufacturing purchasing managers' index (PMI) time series is ARIMA(2, 1, 1) model, and after adding the input variables, the coefficient of the stringency index is -0.15 (p<.0001). The non-manufacturing purchasing managers' index (NMI) time series is ARIMA(3, 0, 1) model, and after adding the input variables, the coefficient of the stringency index is -0.23 (p=0.0087). The retail sales index (RSI) time series is ARIMA(0, 1, 3) model, and after adding the input variables, the coefficient of the stringency index is -0.3 (p<.0001). International trade and the UK stock market are not significantly associated with the stringency index, but the bank rate significantly decreased the UK stock price. We also found that the slowdown in economic activity of the UK is more severe than non-lockdown countries. Conclusion: This study found that the COVID-19 lockdown caused a slowdown in economic activities. The negative macroeconomics impact will be wide-ranging in different aspects on labor markets, supply chains, and GDP levels. To strike a balance between public health and the economy, we don’t recommend taking lockdown measures to contain the spread of this pandemic.