The Impact of the COVID-19 Pandemic on Consumer and Business Conﬁdence Indicators

: The COVID-19 pandemic and induced economic and social constraints have signiﬁcantly impacted the conﬁdence of both consumers and businesses. Despite that, comprehensive studies of the impact of the COVID-19 pandemic on the consumer and business sentiment are still lacking. Thus, in our research we aim to identify consumer and business conﬁdence indicators’ reaction to the spread of the COVID-19 pandemic in the Eurozone, the United States, and China. For this purpose, we used the method of correlation–regression analysis. We chose the consumer-conﬁdence index, manufacturing purchasing manager’s index, and services purchasing manager’s index as dependent variables; and the number of conﬁrmed cases of COVID-19, the number of deaths caused by COVID-19, and the mortality rate of COVID-19 infections as independent variables. The results showed a relatively rapid and robust effect of COVID-19 in the short period, but longer-term results depended on the region and were not so unambiguous: in the case of the Eurozone, the spread of COVID-19 pandemic did not affect the consumer-conﬁdence index (CCI) or, in the cases of the United States and China, affected this index negatively; the purchasing managers’ index (PMI) in the services sector was signiﬁcantly negatively affected by the mortality risk of COVID-19 infection; and the impact on the purchasing managers’ index (PMI) in the manufacturing industry appeared to be mixed.


Introduction
The COVID-19 pandemic has been a big shock for the global economy, but this crisis is different from other types of situations-especially a financial or banking crisis. The main difference is the level of impact and the causes of the problem. Consumer confidence and economic sentiment are the critical drivers for future economic growth. The health crisis of the COVID-19 pandemic and lockdowns had a significant impact on society. Those changes can be seen in consumer and business confidence indicators. Thus, it is essential to analyze how this demand and business sentiment changed in the COVID-19 environment to better analyze the possible consequences in the future.

of 23
The economic consequences of the COVID-19 pandemic, as well as its impact on the general economy and consumer and business sentiment, were analyzed by van der Wielen and Barrios (2020), Coibion et al. (2020), Andersen et al. (2020), Barro et al. (2020), and Chronopoulos et al. (2020).
As stated by van der Wielen and Barrios (2020), the COVID-19 crisis and crisis-induced constraints drastically affected households' economic sentiment in Europe. Negative trends in consumption and the labor market emerged. Andersen et al. (2020) analyzed customer spending changes and indicated a spending drop that was larger in the sectors of goods and services directly affected by COVID-19 pandemic-induced restrictions. Baker et al. (2020) analyzed the indicators of newspaper-based economic uncertainty and subjective uncertainty in business-expectation surveys, and indicated an unprecedented decrease of these measures in the face of the COVID-19 pandemic.
Although research on the impact of the COVID-19 pandemic on economic sentiment has become more widespread recently, most of the studies have focused on one area (consumer sentiment or business sentiment), or one region or country (the United States, the United Kingdom, Europe, etc.). Our research is not only focused on the analysis of the impact on different sentiment indicators (both consumer and business), but also includes the analysis and comparison of several regions.
Some authors analyzed the sentiment of economics or consumers on a country level, while others focused on a regional level. Savin and Winker (2011) analyzed Germany and Russia's cases, and pointed to the specifics of tendencies in different countries. Bhattacharyay et al. (2009) studied early-warning indicators impacting economic and financial risks in Kazakhstan. Kanapickiene et al. (2020) found that at the beginning of the COVID-19 crisis, there were different reactions in separate European countries. However, for more extended periods (i.e., covering both the onset of the pandemic in China and its global spread (the first and the second waves)), there is a lack of studies. With our research, we add value to the literature covering the COVID-19 pandemic impact. We use a longer time horizon and attempt to identify the differences in consumer and business sentiments in different regions. Armantier et al. (2020) analyzed the impact of the COVID-19 pandemic on consumer confidence in the United States. They tried to identify how the pandemic's impact had changed over time and among different demographic groups.
Considering that, this paper's primary purpose is to assess the impact of the spread of the COVID-19 pandemic on consumer and business sentiment in different regions.
We aim to identify consumer and business confidence indicators' reaction to the COVID-19 pandemic in the Eurozone, the United States, and China. For this purpose, we chose the consumer confidence index as well as the purchase manager's indices for manufacturing and services.
After analyzing the relevant scientific literature, we continue with the description of the research methodology (provided in Section 2) and the presentation of our results (provided in Section 3).
To assess the impact of the COVID-19 pandemic, we chose to analyze economic sentiment in three different regions: the Eurozone, the United States, and China. For each area, we selected three different economic sentiment indicators: (i) consumer confidence index, (ii) purchase manager's index in the manufacturing sector, and (iii) purchasing manager's index in the services sector.

Literature Review
As stated by Nowzohour and Stracca (2017), "Sentiment may be used to describe economic agents' views of future economic developments that may influence the economy because they influence agents' decisions today". The economic sentiment includes two opposing dimensions-confidence and uncertainty (van der Wielen and Barrios 2020). The COVID-19 pandemic and social and economic constraints induced by this pandemic have undoubtedly brought more tension to the economy (Baker et al. 2020), while at the same time affecting the confidence of households and businesses. As economic-sentiment indicators are one of the most critical indicators showing the economy's overall health, it is crucial to assess how the spread of the COVID-19 pandemic affects those indicators. Our analysis contributes to the theoretical issues analyzing how consumer and business sentiments can change in periods of external shocks. Consumer confidence is often described as a fundamental driving force of the economy, because when consumers are optimistic, consumption increases, and we have economic growth; while in pessimistic periods, consumers pull down the economy.
Sentiment issues are significant in analyzing economic growth and tendencies in financial markets, and especially when trying to identify future trends. The sentiment of financial markets, consumers, and business entities can be measured by various surveys, after which different sentiment indices are created based on the responses to specific questions about current and expected economic conditions. These indices are used for forecasting household expenditures, consumer spending habits, tendencies in the labor market, or even industrial production growth. However, some authors have an opinion that sentiment surveys have little power for predictions (Roberts and Simon 2001).
Not only are quantitative indices used in the forecasting process, but qualitative assessment of textual sentiment in news is becoming more and more popular. Ardia et al. (2019) found that news-based sentiment values help increase the accuracy of forecasting methods trying to identify the growth rates of industrial production in the United States. So, sentiment can be measured by using various survey quantitative indicators and including qualitative information such as news.
Understanding the sentiment itself is also very important. When trying to describe the concept of sentiment, scientists usually use such proxies such as fear and uncertainty. Barone-Adesi et al. (2018) revealed that sentiment and fear are complementary risk-aversion measures that are linked with uncertainty.
Sentiment indicators are a part of leading indicators that are used to predict future financial and economic trends, as those indicators change before factual changes in economy or business. Leading indicators are not new phenomena and include not only economic, business, or consumer sentiment indicators. Different indicators are essential and have been analyzed for various purposes. Leading inflation indicators are necessary for the monetary policy decision-making process. Ripatti (1995) investigated the mentioned inflation indicators by using Finland's case and applying a pairwise analysis of Granger causality and cointegration. We even found news-sentiment indicators created using digital technology such as machine learning (Nguyen and La Cava 2020;Lee et al. 2012). Some authors (Gurcihan et al. 2013) studied leading indicators for unemployment rate forecasting. Gründler and Potrafke (2020) pointed out an exciting moment when experts changed their policy decisions because of sentiment analysis. Burri and Kaufmann (2020) created a leading indicator using financial market and news data. Benhabib and Spiegel (2017) investigated how sentiment or consumer-confidence shocks can influence state outputs and consumptions, and they revealed a significant effect during a one-year horizon. The impact of consumer sentiment on consumption was analyzed by Gillitzer and Prasad (2016). Golinelli and Parigi (2004) investigated consumer sentiment and economic activity and, using a large set of observations, confirmed the consumer-confidence indices' forecasting ability in the sample and out-of-sample periods.
National sentiment and economic behavior can be crucial in the sports field of trying to bet on final match results (Braun and Kvasnicka 2013). Sentiments have strong power in any area. Charoenrook (2005) analyzed the University of Michigan Consumer Sentiment Index and found that consumer-sentiment changes were positively related to contemporaneous excess market returns, and that they were negatively related to future excess market returns at different horizons. The author pointed out that a shift in consumer sentiment can improve asset-return predictions. Consumer-confidence issues were also analyzed in the research of Daas and Puts (2014), who investigated country-level social-media messages and compared them with consumer confidence. Using the Granger causality test, the mentioned authors revealed that "changes in consumer confidence precede those in social media sentiment than vice versa" (Daas and Puts 2014). Fuhrer (1993) paid a lot of attention to consumer-sentiment analysis and tried to identify the role of consumer sentiment in the Unites States' macroeconomy. The author stressed that "consumer sentiment, or consumer confidence, is both an economic concept and a set of statistical measures" (Fuhrer 1993). Consumer confidence as a household-sentiment indicator can help to explain tendencies in the consumer-loans segment (Rakovská et al. 2020).
Asset-return predictions are modeled not only using consumer-confidence indicators, but also including investor-sentiment indices. Chen et al. (2013) used a principalcomponent approach and constructed an investor-sentiment index for the Chinese stock market and found that the created index had good out-of-sample predictability. Investorsentiment issues were analyzed in the studies of Chu et al. (2015). The authors revealed that economic variables could increase forecasting accuracy when investor sentiment was low, and lose their prediction power when investor sentiment was high. Another author (Dieckelmann 2021) used corporate-bond and stock-market data as proxies for investorsentiment measurement, and used those proxies not only for forecasting tendencies in the financial market, but also for predicting banking crises and economic cycles. The role of market sentiment is very important for forecasting tendencies in the financial market, and a study (Frydman et al. 2019) showed that market sentiment was not related to the state of the economy. Investor sentiment was also analyzed by Jiang et al. (2020) While consumer-confidence analysis is much broader, there is still a lack of literature analyzing business sentiment. There are business-sentiment indicators for different sectors explaining the mood of separate parts of the economy. These business-sentiment indices mostly are based on a survey about present and future tendencies. As those indicators include information about the future, they can be used for forecasting as well. Vanhaelen et al. (2000) investigated the Belgian industrial-confidence indicator and tried to answer whether this country-level sentiment indicator precedes euro-area business cycles. The authors concluded that the turning points in the Belgian industrial confidence indicator significantly impact turning points in the euro area. Using graphical examination, correlation analysis, and Granger causality tests, Santero and Westerlund (1996) revealed that business-sentiment measures gave valuable information when trying to assess the present economic situation and to forecast future economic trends. Kukuvec and Oberhofer (2018) analyzed EU business-sentiment indicators and found substantial spillover effects.
We agree that all sentiment indicators are more focused on the expectations about the future. A present economic situation has a very short impact, so such kind of indicators help in future trend predictions. It is essential to analyze how sentiment indicators interact with each other, especially in such critical moments as the COVID-19 pandemic. In our paper, we attempt to add value to the literature analyzing sentiment issues during critical moments and shocks, and at the same time, we want to stress regional aspects, as the COVID-19 pandemic is global.

Methodology
Seeking to evaluate the impact of the spread of the COVID-19 pandemic on economic sentiment, the economic-sentiment indicators of three different regions-the Eurozone, the United States, and China-were selected for further investigation. As we aim to analyze the reaction of both consumers and businesses to the COVID-19 pandemic, three different economic sentiment indicators-the consumer-confidence index (CCI), and the purchase manager's indices for manufacturing and services (manufacturing PMI and services PMI, respectively) were selected.
Our research consisted of two stages. In Stage 1, the trends of the selected economic sentiment indicators (indices) were analyzed (Section 4.1). Afterward, in Stage 2, the impact of the COVID-19 pandemic on the sentiment of both consumers and businesses was assessed (Section 4.2).
In Stage 1, we used the method of graphical and statistical analysis and analysis of the trends of consumer and business confidence indicators in the face of the COVID-19 pandemic. We also used the method of correlation analysis (Pearson correlation coefficient) in order to identify possible similarities or differences between the dynamics of the selected economic-sentiment indicators in different regions.
In Stage 2, the impact of the spread of the COVID-19 pandemic on selected consumer and business confidence indicators was assessed.
When assessing the economic impact of the COVID-19 pandemic, many researchers (Verma et al. 2021;Chen et al. 2020;Lee 2020;Pavlyshenko 2020;Vasiljeva et al. 2020;Fetzer et al. 2020;Kanapickiene et al. 2020; Albulescu 2020; Ashraf 2020, and others) used the regression approach. For example, Verma et al. (2021) used the method of correlationregression analysis in order to assess how COVID-19 was correlated with economic growth, and evaluated the impact on stock markets. Ashraf (2020) used panel-data regression models to assess the impact of COVID-19 on stock-market returns. Pavlyshenko (2020) discussed different regression approaches for modeling COVID-19's impact on the stock market. Albulescu (2020) used simple OLS regression to evaluate the impact of COVID-19-induced uncertainty on the volatility of financial markets. Fetzer et al. (2020) also used the regression technique in order to assess the relationship between the spread of COVID-19 and economic anxiety. Chen et al. (2020) provided the regression models of high-frequency indicators allowing the assessment of the economic impact of COVID-19. Lee (2020) conducted a correlation-regression analysis to explore the initial impact of COVID-19 sentiment.
Given the wide application of the regression techniques in COVID-19-induced economicimpact studies, in our research, we used the methods of correlation and regression to assess the impact of the spread of the COVID-19 pandemic on selected economic-sentiment indicators.
First of all, at the starting point of the assessment, seeking to identify a possible linear association between selected consumer and business sentiment indicators and COVID-19 related variables, the correlation was assessed (the Pearson correlation coefficient was calculated and interpreted). Second, the simple linear (bivariate) regression models for each pair of dependent and independent variables were constructed. Considering the statistical characteristics of these models (t-value, p-statistics, R squared), the conclusions regarding the impact of the spread of the COVID-19 on consumer and business economic sentiment were made. It is important to note that in our research, we were not able to construct multiple regression models due to: (i) a relatively small number of observations, and (ii) multicollinearity of regressors.
Based on previous studies (for example, Albulescu 2020, Ashraf 2020, Verma et al. 2021, and others), we selected five groups of COVID-19-related variables: (i) the cumulative number of cases of COVID-19 confirmed in each region selected and globally, (ii) new cases of COVID-19 confirmed in each region selected and globally per month, (iii) cumulative number of deaths from COVID-19 reported in each region selected and globally, (iv) new deaths from COVID-19 reported each region and globally per month, and (v) COVID-19 fatality rate in each region and globally. By choosing these groups of variables, we intended to check: (i) whether confidence was affected by the total prevalence of COVID-19 or its monthly growth, (ii) whether consumers and businesses tended to react to the increase of COVID-19 cases or its caused deaths, (iii) whether the changes of the fatality rate of COVID-19 affected confidence indicators, and (iv) whether the reaction to country-level and global-level changes differed.
Based on this logic, 20 independent COVID-19-related variables were chosen for our research. Research variables (dependent and independent), their abbreviations, and data sources are provided in Table 1. In our research, we analyzed the period of January 2020 to January 2021 and used monthly data (as the data of selected economic-sentiment indicators is provided on a monthly basis). The data of the selected economic-sentiment indicators were retrieved from Thompson Reuters database, and the data for COVID-19-related variables were collected from World Health Organization coronavirus disease 2019 (COVID-19) situation reports and the Our World in Data Coronavirus Pandemic (COVID-19) (Roser et al. 2020) database (see Table 1). Data were analyzed using Eviews 11 software.
It is important to emphasize that we did not seek to analyze a comprehensive set of economic-sentiment factors in our research. We only aimed to identify the presence or absence of the reaction to the COVID-19 pandemic (and its direction).
The descriptive statistics of selected dependent (economic sentiment (CCI, manufacturing PMI, services PMI)) and independent (COVID-19-related) variables are provided in Table 2. The dynamics of COVID-19 related variables are shown in Appendices A-D.

Results and Discussion
This section analyzes the dynamics of the selected economic-sentiment indicators in the Eurozone, United States, and China. The effect of the COVID-19 pandemic on different sectors' economic-sentiment indicators in three regions was assessed.

Analysis of the Trends of Economic Sentiment Indicators in the Face of the COVID-19 Pandemic
The Eurozone dynamics and the United States' and China's CCI, manufacturing PMI, and services PMI are provided in Figure 1. the Eurozone, United States, and China. The effect of the COVID-19 pandemic on differe sectors' economic-sentiment indicators in three regions was assessed.

Analysis of the Trends of Economic Sentiment Indicators in the Face of the COVID-19 Pandemic
The Eurozone dynamics and the United States' and China's CCI, manufacturing PM and services PMI are provided in Figure 1.  Table 1. pandemic and was on average the highest of all regions analyzed (mean-119.517) (Figure 1a).
In terms of business sentiment in the manufacturing and services sectors, the data in Figure 1 allowed us to assume that in the face of the COVID-19 pandemic: (i) Both manufacturing PMI and services PMI were on average the lowest in the Eurozone (the means were 49.046 and 42.762, respectively), and experienced the sharpest decline in comparison with the same indices in the other analyzed regions (Figure 1b,c); (ii) During the first peak of the pandemic, business sentiment in the services sector declined more than business sentiment in the Eurozone and China's manufacturing sector. However, this was not the case in the United States; (iii) In the cases of the Eurozone and the United States, the lowest values for the businesssentiment indicators were observed in April 2020; while in the case of China, the lowest values were observed in February 2020, which coincided with the peak of new cases and deaths in the country (Appendix C); (iv) The lowest volatility of business sentiment in the manufacturing and services sector (st. deviation-3.601 and 8.760, respectively) was observed in the United States.
To investigate possible differences in the response of consumer-and business-confidence indicators to the spread of the COVID-19 pandemic, we also assessed the interrelationship between these indicators in different regions. The results of a correlation analysis (Appendix E) of the Eurozone, the United States, and China economic-sentiment indicators inter alia showed a strong or robust correlation between PMI (both manufacturing and services sectors) in the Eurozone and the United States, a substantial correlation between CCI in the Eurozone and the United States, and moderate negative correlation between the China PMI in the services sector and the United States CCI.
Further, it was analyzed how the spread of the COVID-19 pandemic was related to recent changes in consumer-and business-confidence indicators.

Assessment of the Impact of the Spread of COVID-19 Pandemic on Consumer and Business Sentiment
First, the correlation between economic sentiment and COVID-19 related indicators was assessed. The results of the correlation analysis are provided in Table 3.
The results of the correlation analysis (Table 3) showed that: (i) The Eurozone consumer-confidence index was not correlated with any COVID-19 variables, while CCI in the United States was strongly negatively correlated with total cases of the COVID-19 pandemic confirmed globally, and CCI in China was strongly negatively correlated with COVID-19 fatality rate both in China and globally; (ii) Interestingly, in all analyzed regions, business sentiment in the manufacturing sector strongly positively correlated with the global spread of COVID-19 (cases and deaths); while at the same time, the United States manufacturing PMI positively correlated with, while the China manufacturing PMI negatively correlated with, country-level COVID-19 indicators; and the Eurozone and the United States manufacturing PMIs negatively correlated with COVID-19 fatality rate both in at the country and global levels; (iii) As with manufacturing PMI, the Eurozone and the United States services PMIs were negatively correlated with the COVID-19 fatality rate at the global level; while in China, the services PMI was positively correlated with the global spread of COVID-19 and negatively correlated with country-level COVID-19 indicators. Second, to reveal a clearer picture of the impact of the spread of the COVID-19 pandemic on consumer-and business-sentiment indicators, a regression analysis was conducted. The results are provided in Tables 4-6. These results allowed us to make assumptions regarding the reaction of different economic-sentiment indicators in different regions.
The linear regression models for the COVID-19 impact on consumer and business sentiment in the Eurozone are provided in Table 4.
The results in Table 4 (t-values, p-statistics, and R-squared) allows us to state that, when taking into account the period of January 2020 to January 2021, in the case of the Eurozone: (i) No statistically significant impact of COVID-19-related indicators on the consumerconfidence indicator was identified; i.e., in the long period, no statistically considerable reaction of CCI was observed (although the short-time effect as specified in Section 4.1 was); (ii) Interestingly, the Eurozone PMI for the manufacturing sector demonstrated a positive reaction to the increase of the cases and deaths of COVID-19 at both the in-country level and globally; i.e., a statistically significant positive impact of the COVID-19related variables was observed; (iii) The fatality rate of COVID-19 infection (at the global level) appeared to have a statistically significant negative impact on the business-sentiment indicators in the manufacturing and services sectors (the manufacturing sector PMI also demonstrated a reaction to the changes in the country-level fatality rate); i.e., as in the case of the short period (Section 4.1), in the long period, an adverse business-sentiment reaction was observed.  The regression models for COVID-19's impact on consumer and business indicators in the United States are provided in Table 5.
The results in Table 5 (t-values, p-statistics, and R-squared) allows us to state that, when taking into account the period of January 2020 to January 2021, in the case of the United States: (i) The consumer-confidence indicator in the United States demonstrated a negative reaction to the growth of deaths caused by COVID-19 infections (cumulative and monthly) both at the country and global levels; i.e., a statistically significant negative impact of the COVID-19 pandemic was confirmed (as it was also observed in Section 4.1); (ii) As in the case of the Eurozone, PMI in the manufacturing sector in the United States demonstrated a positive reaction to the increase in COVID-19 cases and deaths both at the country level and globally; i.e., a statistically significant positive impact of the COVID-19-related variables was observed; (iii) As in the Eurozone case, the fatality rate of COVID-19 infections (at the global level) appeared to have a statistically significant negative impact on the business-sentiment indicators in the manufacturing and services sectors in the United States. As in the short period (Section 4.1), an adverse business-sentiment reaction was observed in the extended period. The regression models for the COVID-19 impact on consumer-and business-sentiment indicators in China are provided in Table 6.
The results in Table 6 (t-values, p-statistics, and R-squared) allows us to state that, when taking into account the period of January 2020 to January 2021, in the case of China: In sum, it can be stated that, although the COVID-19 pandemic had a rapid negative short-time effect on consumer-confidence and business-sentiment indicators (in the manufacturing and services sectors), the evidence for a long-term effect was not so unambiguous. On the one hand, (i) in the Eurozone case, the spread of the COVID-19 pandemic had no statistically significant impact on CCI, or affected this index negatively in the United States and China, and (ii) the PMI in the service sector was significantly negatively affected by the fatality rate or mortality risk of COVID-19 infection. On the other hand, the impact on PMI in the manufacturing industry appeared to be mixed, because this indicator demonstrated a positive reaction to the increase of COVID-19 cases and deaths.

Conclusions
Our research showed that the lowest level of consumer confidence was observed in the Eurozone. This can be explained by the fact that the COVID-19 pandemic has affected Italy very strongly, and a pessimistic mood spread all over the region. However, the volatility of consumer confidence and uncertainty among households was higher in the United States compared with results for the euro area. The other aspect we would like to point out is that consumer-confidence volatility was lowest in China. This was because consumers in China were affected, and evaluated the possible consequences for a more extended period and did not change their minds about the future. This fact is crucial for future economic growth because the main driver of growth is consumption. Our studies showed that the most pessimistic period for consumers and the biggest shock to their mood and expectations in the euro area and the United States was observed in April 2020, when the first peak of new cases and deaths was observed all over the world. In the case of China, the most pessimistic wave of mood was in February.
Consumers in the euro area were the most pessimistic compared with other regions, but the business segment also demonstrated the most pessimistic mood compared with the US and China. The pessimism was spread among business entities in the manufacturing and service sectors. Those sectors in the euro area experienced the sharpest decline compared with other regions. We also would like to stress that the service sector's expectations were lower than in the manufacturing industry in the euro area and China, while the situation in the United States was the opposite. The lowest volatility of business sentiment in the manufacturing and services sectors was found in the United States. This fact could be explained by government and monetary support for businesses. Every entity needs to be aware of possible help that could lower the negative effects of the COVID-19 pandemic. Regarding the correlation effects, we observed a substantial positive correlation between business sentiment in the manufacturing and services sectors at a country level. We also revealed a robust correlation between PMI (both manufacturing and services sectors) in the Eurozone and the United States, and a substantial correlation between consumer confidence in the Eurozone and the United States.
The correlation analysis between sentiment indicators and COVID-19 variables showed that the Eurozone consumer-confidence index was not correlated with COVID-19 variables. In contrast, the same index in the United States was strongly negatively correlated with total cases of the COVID-19 pandemic confirmed globally. Consumer confidence in China was strongly negatively correlated with the COVID-19 fatality rate both in China and globally. These conclusions could be used for practitioners to model future economic tendencies or make investment decisions in stressful scenarios.
We also observed that business sentiment in the manufacturing sector in all analyzed regions strongly positively correlated with the global spread of COVID-19 (cases and deaths). Simultaneously, the United States manufacturing PMI positively correlated with, while China manufacturing PMI negatively correlated with, country-level COVID-19 indicators. The Eurozone and United States manufacturing PMI negatively correlated with COVID-19 fatality rate both at the country and global levels. The Eurozone and United States service PMIs were negatively correlated with the COVID-19 fatality rate at both the in-country and international levels. In China, the service PMI was positively correlated with the global spread of COVID-19 and negatively correlated with country-level COVID-19 indicators.
Our study showed no statistically significant impact of COVID-19-related indicators on consumer-confidence indicators. The Eurozone PMI in the manufacturing sector demonstrated a positive reaction to the increase in COVID-19 cases and deaths both in-country and globally.
The fatality rate of COVID-19 infections (at the global level) appeared to significantly negatively impact the business-sentiment indicators in the manufacturing and service sectors.
The regression analysis showed that the consumer-confidence indicator in the United States demonstrated an adverse reaction to the growth of deaths caused by COVID-19 infections. The Eurozone PMI in the manufacturing sector showed a positive response to the increase of COVID-19 cases and deaths of both in-country and globally. As in the Eurozone case, the fatality rate of COVID-19 infections (at the global level) appeared to have a statistically significant negative impact on the business-sentiment indicators in the manufacturing and services sectors in the United States. As in the short period, the adverse business-sentiment reaction was observed in the extended period.
The regression models for COVID-19's impact on consumer-and business-sentiment indicators in China showed that the country's consumer-confidence indicator demonstrated an adverse reaction to the fatality rate of COVID-19 infections (both at the country level and globally). In comparison, China PMI in the manufacturing and service sectors demonstrated an adverse reaction to new COVID-19 cases and deaths per month. The number of cases and recent deaths per month proved to have a statistically significant negative impact on China's business sentiment.
Our results showed that we could have entirely different reactions and tendencies, even in such a critical global situation. So, it is essential to pay attention to those factors while making strategic decisions or creating country-level risk limits, or even when considering regional diversification aspects.
However, it is crucial to notice that this research encountered some limitations. As the economic sentiment indicators analyzed in this research are provided every month, these results are from a relatively small number of observations. Furthermore, given the dynamics of COVID-19-related variables and uneven growth of cases and deaths during the analyzed period, it would be appropriate to assess the impact of the COVID-19 pandemic on selected economic-sentiment indicators during different phases of the pandemic (onset of the pandemic, global spread, the second wave of the pandemic, beginning of vaccination, etc.). Nevertheless, the short data series did not allow this to be done. Dealing with these limitations and the inclusion of more economic sentiment indicators (both general and sectorial) could be the direction for future research. quarter, II-2nd quarter, III-3rd quarter, IV-4th quarter. For TCEZt, NCEZt, TDEZt, NDEZt, and FREZt, see Table 1.