Next Article in Journal
Quantitative Assessment of Environmental Sensitivity to Desertification Using the Modified MEDALUS Model in a Semiarid Area
Next Article in Special Issue
Factors Affecting the Development of Metabolic Syndrome in Single-Person Households: A Sex-Stratified Analysis
Previous Article in Journal
Timely Loss Recognition Helps Nothing
Previous Article in Special Issue
Examining the Impacts of the Built Environment on Quality of Life in Cancer Patients Using Machine Learning
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Individual Subjective Well-Being during the COVID-19 Pandemic

Asian Development Bank Institute, Tokyo 100-6008, Japan
Sustainability 2021, 13(14), 7816; https://doi.org/10.3390/su13147816
Submission received: 10 May 2021 / Revised: 29 June 2021 / Accepted: 7 July 2021 / Published: 13 July 2021

Abstract

:
This paper examines how contextual and institutional factors are associated with individual subjective well-being, which is measured by individuals’ happiness, during the early stage of the COVID-19 pandemic. Using data collected in the People’s Republic of China (PRC), the Republic of Korea, Japan, Italy, the United Kingdom (UK), and the four biggest states of the United States (US) in April 2020, we find that the financial effects (represented by employment and income change) and nonfinancial effects (represented by experiencing negative nonfinancial effects including mental health issues and enjoying positive benefits) caused by nonpharmaceutical measures to contain the spread of COVID-19 are associated with individual subjective well-being. Moreover, positive benefits could reduce the likelihood of becoming unhappy for those who have experienced negative nonfinancial effects or those who have lost their jobs. The results also suggest that the degree to which people agree with their government’s approach to dealing with the COVID-19 pandemic is positively correlated with their happiness. The risks associated with the pandemic, however, are only slightly associated with people’s happiness. We also find that the correlation between the above factors and individual well-being varies from country to country.

1. Introduction

Since the first case was identified in Wuhan, People’s Republic of China (PRC), in early January 2020, COVID-19 has quickly spread to more than 200 countries and territories. By early January 2021, according to the statistics collected by the Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE), more than 85 million people had been infected, of whom more than 1.8 million died [1]. Although the mortality rate is comparatively low, the COVID-19 infection rate is high, with the virus spreading faster than other deadly viruses. Due to the speed and severity of infection, the COVID-19 pandemic has emerged as one of the greatest public health crises in recent decades. To deal with this pandemic, governments around the world have enacted a series of policies to slow the spread and prevent transmission, such as closing borders, imposing travel bans, and implementing lockdowns and quarantine. According to [2], by the end of April 2020, there were about 2.6 billion people, or one-third of the total global population, living under some type of lockdown or quarantine. Such policies have profoundly affected the lives of billions of people around the world in financial and nonfinancial ways.
Many studies have examined the effects of this pandemic on economic growth [3], on the livelihoods of various groups of people [4], inequality [5], firm activities [6], labor markets [7,8], and international trade [9]. A growing number of studies examine the impact of COVID-19 on mental health [10,11,12]. These studies show that the COVID-19 pandemic has not only physical but also psychological effects on individuals’ welfare. Thus, mental health, as well as other aspects of well-being considerations, should be important elements of decisions regarding the approach and intensity of nonpharmaceutical interventions a government should pursue.
In addition to negative mental health effects, nonpharmaceutical measures also bring about several positive benefits, such as having more family time and interactions, and having time to enjoy personal interests, which are increasingly difficult to find under “normal” conditions. Moreover, the literature also identifies some environmental benefits, such as a reduction in air pollution, improvement of air and water quality, lower noise levels, and reduced land surface temperature [13,14]. Such positive benefits of lockdown may also influence individuals’ well-being. Most of the current studies, however, focus on the effects of lockdown measures on several aspects of mental health, such as anxiety, boredom, and loneliness, and completely ignore how such mental health issues and positive benefits brought about by the lockdown measures may affect individuals’ well-being.
Against this background, this study aims to fill the gap by examining how positive and negative nonfinancial effects of nonpharmaceutical measures can contain the spread of COVID-19, and COVID-19-induced financial effects jointly affect individual subjective well-being, which is measured by their happiness. We use the data collected during the third week of April 2020 in six countries: the PRC, the Republic of Korea, Japan, Italy, the UK, and the four biggest states in the US. We find that both the financial and nonfinancial effects of the pandemic are associated with individuals’ happiness (and unhappiness). Moreover, positive nonfinancial benefits could help to reduce the likelihood of being unhappy and their level of unhappiness for those who have experienced negative nonfinancial effects and those who have lost their job because of nonpharmaceutical measures to contain the spread of COVID-19. Our empirical results also suggest that the extent to which people agree with their government’s approach to dealing with the pandemic is positively correlated with their happiness. Their perceptions of the effectiveness of containment measures are also positively associated with their well-being. The perceived risks of the pandemic, however, are not strongly associated with people’s happiness.
This paper contributes to the literature in several aspects. First, our study examines not only the effects of negative nonfinancial consequences (including loneliness; boredom; anxiety; troubled sleep; conflicts with friends, family members, and neighbors; and other consequences) of nonpharmaceutical measures but also the effects of positive benefits (including more free time, more time with family, less pollution, less noise, and other benefits) of such measures on individuals’ subjective well-being. In addition, we also examine the soothing role of experiencing such positive nonfinancial benefits in the relationship between happiness (and level of happiness) and experiencing negative nonfinancial consequences, and between happiness (and level of happiness) and having lost a job. Second, we jointly examine the financial effects and nonfinancial effects on individual subjective well-being, while other studies such as [12] investigate the effects of lockdown (and other nonpharmaceutical measures) or the effects of income loss or job loss on mental health issues such as loneliness, anxiety, and boredom. In other words, we examine how psychological effects (including mental health effects) and financial effects jointly affect individual subjective well-being. Third, this study directly links individual subjective well-being with individual-level measures of institutional factors, measured by individuals’ agreement with the government’s action in dealing with the COVID-19 pandemic. Moreover, using individuals’ agreement with government response to the same global event (i.e., the COVID-19 pandemic) makes isolating the effects of particular institutions on individuals’ well-being easier [15].
The paper is structured as follows. Section 2 provides a review of the literature on the factors determining individual happiness. Section 3 documents the data source, our measurement of variables, and empirical approaches. Section 4 reports the empirical results, and these are followed by concluding remarks in Section 5.

2. Literature Review

Since the manual work of Easterlin [16], the study of happiness has gradually shifted into the gamut of economics. Since the late 1990s in particular, there has been a growing number of studies that have analyzed the determinants of happiness in different countries and periods [17,18]. According to Frey and Stutzer [17], studying the economics of happiness is relevant for several reasons. First, people’s well-being is an essential component of the economy and society. Economics, by itself, is, or should be, about individual happiness. Second, studying happiness provides new answers to various issues concerning the welfare effects of income, unemployment, and inflation. Such insights help in the design of appropriate economic policies, which may mitigate the costs, including nonpecuniary costs, of achieving individual welfare. Third, happiness research can help governments develop appropriate institutions.
Previous studies have identified sources of individual well-being, including socioeconomic factors, contextual and situational factors, and institutional factors. With regard to socioeconomic factors, studies have shown that women tend to be happier than men, the young and the old are happier than the middle-aged, and even age and happiness have a nonlinear relationship [19,20]. While some have found that income is also a major determinant of happiness [15,21], others have found that income has only a minor effect or a diminishing effect beyond a certain threshold [22,23,24]. Meanwhile, employment has been found to have a positive impact on life satisfaction, as it is essential for one to work to earn a sufficient living. Clark and Oswald [25] and Winkelmann and Winkelmann [26] (1998) have shown that unemployment is correlated with substantial unhappiness.
Contextual and situational factors, such as particular employment conditions; stress at the workplace; interpersonal relationships with work colleagues, relatives, and friends; and living conditions and health, also affect life satisfaction [15,27]. There is evidence to suggest that personal and social contact and support, the conditions in areas where people live, and private freedoms are good predictors of happiness and life satisfaction [28,29,30].
In the context of the COVID-19 pandemic, a growing number of studies show that the pandemic and measures to contain the spread of COVID-19 bring about nonfinancial effects, which can ultimately affect one’s subjective well-being. Le and Nguyen [11] use US data and show that the pandemic lockdowns are associated with a variety of adverse psychological outcomes, namely anxiety, worry, disinterest, depression, and poor general health perception. Using longitudinal microdata for the UK over the period 2009–2020, Banks and Xu [31] find a substantial effect of COVID-19 on mental health at the population level. Similar results are also found in Japan [32], the PRC [33,34], and Spain [35]. Pedraza, Guzi and Tijdens [12] examined life dissatisfaction and anxiety during the COVID-19 pandemic in 25 countries and found that job-related changes due to COVID-19 and governments’ measures for coronavirus containment were associated with dissatisfaction and anxiety. However, nonpharmaceutical measures also bring about several positive benefits, such as having more family time and interactions, and having time to enjoy personal interests. Moreover, the literature also identifies some environmental benefits, such as a reduction in air pollution, improvement of air and water quality, lower noise levels, and reduced land surface temperature [13,14]. Such positive benefits of lockdown may also influence individuals’ well-being.
Institutional factors such as governance, accountability, and effectiveness of governance also determine individual well-being. Previous evidence suggests that people are happier with their lives in countries that have a better quality of governance [36,37,38,39]. According to Helliwell et al. [38], there are two major sets of institutional characteristics that affect individual well-being. The first is related to the reliability and responsiveness of governments in their design and delivery of services. The second set is the presence and pervasiveness of key features of democratic electoral elections and representation. In the context of an unexpected global shock such as the COVID-19 pandemic, the reliability and responsiveness of the government tend to strongly affect people’s well-being. If a government’s approach to the COVID-19 pandemic is in line with its citizens’ expectations, it may improve their happiness.
The COVID-19 pandemic has provided an excellent context in which to analyze individual well-being and happiness. The measures that governments around the world have implemented to contain the spread of COVID-19 have had substantial and far-reaching impacts on people’s lifestyles, their employment status, the way they work, their environmental surroundings, and the way they interact and communicate with each other. These changes will have significant financial and nonfinancial effects. These effects are expected to be associated with individual happiness.
This current study contributes to the COVID-19 literature in several aspects. First, in contrast to other studies that examine the effects of nonpharmaceutical measures on individual health issues, this study investigates the effects of mental health issues on individual subjective well-being. Second, our study not only examines the effect of mental health but also the effect of positive consequences of such measures such as having more free time and a less polluted environment on individuals’ well-being. Furthermore, we also quantitatively examine the soothing role of enjoying nonfinancial benefits in reducing the likelihood of being unhappy caused by either experiencing negative mental health effects or having lost a job as a result of the outbreak of COVID-19.
Moreover, this study also contributes to the literature on the determinants of happiness in several ways. First, with regard to institutional factors, this study directly links individual subjective well-being with individual-level measures of institutional factors, while previous literature relied on country-level or regional-level institutional factors. These country-level or regional-level factors could be more exogenous to individual well-being than individual-level measures. However, it is more difficult to disentangle the direct effects of institutional factors on individuals’ subjective well-being since country-level and regional-level factors may indirectly affect individuals’ subjective well-being through other macro-variables such as inflation, employment, and income. Second, in cross-country studies, countries usually differ in many ways. Therefore, it is difficult to isolate the effects of particular institutions on individuals’ well-being [15]. This problem is less acute if the institutional variations occur either within a country or during a major global event such as the COVID-19 pandemic. Our institutional variations are based on variations in the responses of different governments to the same global event, which makes them more appropriate for cross-country analysis.

3. Data and Empirical Approach

3.1. Data

Data used in this study were collected in six countries, namely the PRC, Italy, Japan, the Republic of Korea, the UK, and the four largest states in the US (i.e., California, Florida, New York, and Texas). These countries were selected because they were at different stages of the pandemic at the time of the survey, and they had each responded to the pandemic differently. This dataset was collected by Belot et al. [40] with funding from the Creative-Pioneering Researchers Program at Seoul National University, and from the European University Institute. The survey was implemented between 15 and 23 April. Table 1 shows the number of cases in each country by the third week of April and May 2020.
In each country, around 1000 individuals participated in the survey. In total, the sample comprised 6089 respondents. In each country, the samples were nationally representative of age, gender, and household income [40]. The data were collected in the third week of April 2020. Potential participants were drawn from several different samples to which the survey firm had access. Individuals were initially contacted via email to participate in the online survey. The survey team invited new individuals to participate in the survey to ensure the representativeness of the sample in three dimensions: age, gender, and household income. Dang and Nguyen [41] compare the distributions of respondents by gender and age groups in the data set used in this present study and the distributions of these characteristics obtained from the official figures to examine the representativeness of the survey at the country level. They find that while there are some differences in the proportion of respondents in age groups for Japan and the UK, the differences are not large. Moreover, the proportion of respondents in each of five income quantiles is roughly 20%, indicating that the samples are quite equally distributed among income groups. The survey contains information on the basic demographic variables of respondents, employment and living situations, health and diseases, self-reports on economic and noneconomic consequences of the pandemic, behavior, beliefs about the pandemic, and the responses of the governments. The data are available on the website: https://osf.io/aubkc/. (accessed on 20 May 2020)

3.2. Variable Construction

In Table 2, we show how information from the survey is used to construct our independent variables (unhappiness and happiness), mental health issues and positive benefits (i.e., nonfinancial effects), financial effects, and covariates used in our regression analyses.

3.3. Summary Statistics

Table 3 presents some summary statistics for the whole sample. The summary statistics for each country are reported in Supplementary Materials. About 37% of respondents reported not being happy at the time that the survey was conducted. Notably, nearly half of the Italians in the sample reported being unhappy, while about 30% of Chinese respondents indicated they were unhappy. Of these, 6.63% of the respondents stated that they were extremely unhappy. Italy and the PRC were respectively the two countries with the highest and lowest proportion of individuals who felt extremely unhappy. With regard to negative mental health issues, nearly half of the respondents felt boredom (48%) and anxiety (45%). The proportion of respondents experiencing loneliness, having trouble sleeping, and having conflicts with family members was 26%, 24.7%, and 9%, respectively. Overall, about 80% of individuals had experienced at least one negative nonfinancial effect caused by nonpharmaceutical measures to contain the spread of COVID-19. Meanwhile, a high proportion of respondents, 85%, reported that they enjoyed at least one positive benefit brought about by lockdown measures. More specifically, about 42% (51%) of respondents reported that they had more free time for themselves (for family). The proportion of respondents that reported less pollution and less noise was 44% and 30%, respectively. Only 5.37% of the sample had lost their job permanently, while nearly 25% had lost their job temporarily. The data also showed that almost 50% had experienced a decline in income. More than 50% of respondents in the sample agreed with their government’s approach to the COVID-19 pandemic, while only 15% disagreed. Most individuals in the sample stated that they believed the measures to contain COVID-19 were effective. It is interesting to note that more than one fifth of the sample indicated their belief that those who became infected would die.

3.4. Empirical Approach

Based on the existing literature, the estimate equation would be as follows:
P ( U n h a p p y i ) = Φ ( α 0 + α 1 N e g E x p i + α 2 P o s E x p i + α 3 J o b C h n i + α 4 I n c S h o c k i + α 5 B e l i e f i + α 6 G o v P o l i + α 7 X i )
in which U n h a p p y i t is a dummy variable indicating that individual i is unhappy; P o s E x p i and N e g E x p i t are two variables capturing the positive and negative nonfinancial effects of the COVID-19 pandemic; J o b C h n i t and I n c S h o c k i t are two sets of financial effects of the COVID-19 pandemic relating to changes in employment status and household income; G o v P o l i t is a set of five degrees of agreement with the government’s approach to dealing with the COVID-19 pandemic; B e l i e f i t is a set of people’s beliefs regarding the spread of the pandemic in their local area, the risk of getting infected themselves, and the risk of several complications that may arise once a person becomes infected;   and   X i t is a set of socioeconomic factors including individuals’ age, gender, regions, living arrangement (such as living alone, living in a household with others, etc.), and income quintiles.
We also estimated an ordered probit model as follows:
H a p p y i = Z i β + η i
where H a p p y i is an unobserved latent variable, indicating individual I’s happiness; Z i is a set of independent variables, which include all independent variables in Equation (1), and η i is the error term. The observed ordinal value of happiness, H a p p y i takes on the value 1 (extremely unhappy) through 7 (extremely happy), according to the following scheme:
H a p p y i = j     μ j 1 < H a p p y i < μ j   ;   j = 1 7
The probability that individual i will select alternative j is:
p i j = p r o b ( H a p p y i = j ) = p r o b ( μ j 1 < H a p p y i < μ j   ) = Φ ( μ j Z i β ) Φ ( μ j 1 Z i β )
We estimate Equations (1) and (2) using both pooled data (i.e., cross-country) and each country’s data.
We further examine the role of experiencing positive nonfinancial benefits in mitigating the potential negative nonfinancial effects caused by the COVID-19 pandemic as follows:
P ( U n h a p p y i ) = Φ ( γ 0 + k γ 1 k ( N e g E x p i # P o s E x p i ) + γ 2 J o b C h n i + γ 3 I n c S h o c k i + γ 3 B e l i e f i + γ 6 G o v P o l i + γ 5 X i )
where all variables are similar to those in Equation (1). The differences lie in the factor variables between N e g E x p i and P o s E x p i . Since both variables are binary variables, we have four combinations (i.e., k = 4): (i) not experiencing negative nonfinancial effects and not experiencing positive benefits; (ii) not experiencing negative nonfinancial effects but experiencing positive benefits; (iii) experiencing negative nonfinancial effects but not experiencing any positive benefits; and (iv) experiencing both negative nonfinancial effects and positive benefits. Case (i) could be viewed as the normal state. We expect that experiencing positive nonfinancial effects will reduce the effects of negative nonfinancial effects on being unhappy.
Similarly, we revised Equation (4) to quantify the soothing role of positive nonfinancial effects in the relationship between having lost a job and being unhappy, as in Equation (5).
P ( U n h a p p y i ) = Φ ( θ 0 + θ 1 N e g E x p i + k θ 2 k ( P o s E x p i # J L o s t i ) + θ 3 I n c S h o c k i + θ 4 B e l i e f i + θ 5 G o v P o l i + θ 6 X i )
Similar to Equation (4), the two variables, P o s E x p i and J L o s t i , are both binary variables, so we again have four combinations: (i) not experiencing any positive benefits and not losing a job; (ii) experiencing positive benefits but not losing a job; (iii) experiencing positive benefits and having lost a job; and (iv) experiencing positive benefits and having lost a job. State (i) can be viewed as the normal state. We also expect that experiencing positive benefits will reduce the effects of having lost a job on being unhappy.
Before going further, we discuss some caveats that necessitate a cautious interpretation of our empirical results. First, at the time the survey was conducted, each country was at a different stage in the pandemic, and the governments of these countries had taken different approaches to containing COVID-19. For example, by that time, the PRC and the Republic of Korea had appeared to control the spread of COVID-19, while other countries at that stage had only limited control over the virus. In addition, there are other sources of cross-country variations, such as culture, institutions, governments, etc. Therefore, as Belot et al. [40] suggested, the results of cross-country analyses should be interpreted cautiously. Second, the estimation results could suffer from endogeneity bias. We tried to include as many observable variables in our estimation as possible. Some variables, such as education level, were not collected. Unobservable factors such as personality may also affect both outcome variables (happiness) and independent variables, particularly subjective variables such as nonfinancial effects. Reverse causality may even exist. For example, if people are unhappy, their viewpoints on all aspects measured may be negative, including their government’s approach to COVID-19. Therefore, our estimation results should be interpreted as indications of correlation rather than causality.

4. Empirical Results

4.1. Benchmark Results

Table 4 presents our results estimation of Equation (1) using pooled data (column 1), and samples of Chinese, Korean, Japanese, Italian, British, and American respondents (columns 2 to 7, respectively). Our results show that people who have experienced negative nonfinancial effects (including mental health issues) are more likely to be unhappy than those who have not experienced these effects. Meanwhile, enjoying positive nonfinancial benefits is negatively associated with being unhappy. The associations between the nonfinancial effects (i.e., either negative or positive effects) brought about by measures to contain the spread of COVID-19 are observed in all countries, except in the case of the PRC and Japan. Chinese and Japanese people who enjoyed at least one positive nonfinancial benefit were not less unhappy than those who did not. Our estimation results using such indices are qualitatively consistent with the results presented in this section. The results are available upon request.
Financial effects are also associated with happiness and unhappiness. Our estimation shows that experiencing a decline in household income is positively correlated with unhappiness. This association is observed in almost all countries (except for the UK), although the estimated coefficients differed from country to country. A fall in household income increases the likelihood of being unhappy among Korean and Chinese people by 12.2 and 11.5 percentage points, while the figures for Japanese, Italian, and American people vary by only 7.5 to 9 percentage points. The difference may reflect the increased effectiveness of the social welfare net among the latter group of countries compared to the Republic of Korea and the PRC.
Those who enjoyed no change in employment status are less likely to be unhappy than those who had lost their jobs, while no difference could be observed between those who had lost their job temporarily and those who had lost their job permanently. However, the estimated coefficients on job status variables seem to be driven by the Korean and Italian samples, since we do not observe any significant association between employment status and unhappiness in the PRC, Japan, the UK, and the US.
Individuals’ unhappiness is negatively associated with the extent to which they agreed with their government’s approach to the pandemic. For example, those who strongly agreed with their government’s approach were more likely to be happy than those who strongly disagreed, by 22.9 percentage points. This association is observed in all countries, except the PRC, where the extent to which people agreed with the PRC government’s approach is not related to their unhappiness. Our empirical results, however, show that people’s perceptions of the effectiveness of containment measures are not correlated with their unhappiness.
Similarly, our results using a pooled sample also show that belief about the severity of the pandemic where people live and the risk of several kinds of complications are associated with being unhappy, but the magnitude of this correlation is quite small. The associations among these variables and people’s unhappiness also differ from country to country and are somewhat limited.
With respect to other control variables, we find that men seem to be more unhappy than women, especially among Korean and Japanese people. This is partly because the financial burden on Korean and Japanese people seems to be greater since Japanese and Korean women tend to leave the labor force when they get married. During the pandemic, those in the wealthier quintiles are less likely to be unhappy than those in the poorer quintiles. However, this correlation appears to be driven by the US sample since we do not observe an association between the income quintile and unhappiness in other countries.

4.2. Further Analysis

Table 5 details how each indicator of the negative and positive nonfinancial effects/benefits brought about by nonpharmaceutical measures to deal with the COVID-19 pandemic is associated with individual unhappiness. Our results show that almost all negative nonfinancial effects are positively associated with being unhappy. In nearly all countries, experiencing boredom, loneliness, having trouble sleeping, and general anxiety and stress are positively correlated with being unhappy. Having conflicts with friends, relatives, and neighbors is positively associated with unhappiness only among Koreans. Among the positive benefits, we find that those who enjoyed more free time (and more time with family) and a cleaner environment brought about by the lockdown measures are less likely to be unhappy. However, the relationship between each positive benefit and unhappiness differs from country to country. For example, none of the perceived positive benefits caused by the lockdown measures makes Japanese people happier, while among all perceived positive benefits, only enjoying more time with family seems to make Chinese people less unhappy.
Table 6 reports our estimation results for Equation (2), using an ordered probit model and a pooled sample. The results using each country sample are presented in Online Supplementary Materials. Our results show that both the nonfinancial and financial effects of the COVID-19 pandemic are associated with the level of happiness. Those experiencing negative nonfinancial effects have a lower level of happiness than those who did not experience such effects. Meanwhile, enjoying positive nonfinancial benefits such as having more free time and a cleaner environment is positively associated with a higher level of happiness. Those who lost their job permanently due to the pandemic have a lower level of happiness than those whose employment status remained unchanged. Those who agreed with their government’s approach to the COVID-19 pandemic tended to have a higher level of happiness than those who strongly disagreed with the government’s approach. In general, the results obtained from the ordered probit model are consistent with the results using the probit model, using either the pooled sample or each country sample.

4.3. Soothing Role of Positive Nonfinancial Benefits

Table 7 presents our results estimation of Equations (4) and (5) using pooled data (column 1), and the sample of Chinese, Korean, Japanese, Italian, British, and American respondents (columns 2 to 7, respectively). For the sake of brevity, we present only the estimates for the combinations of experiencing positive and negative nonfinancial effects (Panel A) and of experiencing positive nonfinancial effects and having lost a job (Panel B). The reference group in Panel A is the group of those who did not experience both negative and positive nonfinancial effects, while that in Panel B is the group of those who did not experience positive nonfinancial effects and did not lose their jobs. The empirical results corroborate the results presented earlier that enjoying positive benefits (or experiencing negative nonfinancial effects) caused by nonpharmaceutical measures to contain the spread of COVID-19 is associated with a lower (higher) likelihood of being unhappy. Moreover, among those who experienced negative nonfinancial effects, those enjoying positive benefits have a lower likelihood of being unhappy, thereby indicating the soothing role of positive nonfinancial effects in the relationship between enjoying benefits and happiness. However, the soothing role of positive nonfinancial benefits is not observed in all countries in our sample. In the PRC, experiencing positive and negative nonfinancial effects did not affect people’s happiness. Meanwhile in Japan, among those who did not experience negative nonfinancial effects, experiencing positive nonfinancial effects reduced their likelihood of being unhappy. In Italy and in the UK, experiencing positive nonfinancial effects was not related to happiness among those who did not experience negative financial effects. Moreover, among Italians who experienced negative nonfinancial effects, not having experienced any positive nonfinancial effects increased their likelihood of being unhappy.
The results in Panel B, Table 7 also suggest the soothing role of enjoying positive nonfinancial benefits in the relationship between having lost a job and happiness. Among those who did lose their jobs, individuals who experienced positive nonfinancial effects were less likely to be unhappy than those who did not. The soothing role of experiencing positive nonfinancial effects, however, is not observed in all countries. Experiencing nonfinancial effects is associated with being happy among those who did not lose jobs in the Republic of Korea, Italy, the UK, and the US, but not in the PRC and Japan. Meanwhile, among those who lost jobs, experiencing positive nonfinancial benefits was associated with being happier in only Italy and the UK and not in the other four countries.
Table 8 reports the soothing role of positive nonfinancial benefits of nonpharmaceutical intervention measures in the relationship between experiencing negative nonfinancial effects and levels of happiness (Panel A) and between having lost a job and levels of happiness (Panel B). The results are obtained from estimating ordered probit models using the pooled sample. Our results show that, among those who did experience negative nonfinancial effects, those who enjoyed positive nonfinancial benefits were more likely to have a higher level of happiness. We also observed a similar pattern in Panel B, in which experiencing positive nonfinancial effects plays a positive soothing role in the relationship between having lost a job and the level of happiness. Enjoying positive benefits brought about by lockdown measures increases the level of happiness of those who lost their jobs as a result of the pandemic.

5. Conclusions

The COVID-19 pandemic has provided an excellent context in which to analyze individual well-being and happiness. The measures that governments around the world have implemented to contain the spread of COVID-19 have had substantial and far-reaching impacts on people’s lifestyles, their employment status, the way they work, their environmental surroundings, and the way they interact and communicate with each other. Such changes have had significant financial and nonfinancial effects. This paper has examined factors that are associated with the unhappiness of individuals in six countries during the COVID-19 pandemic: the PRC, the Republic of Korea, Japan, Italy, the UK, and the four largest states in the US. Using data collected during the third week of April 2020, we found that, on average, both COVID-19-induced financial effects such as job loss or a decline in income and nonfinancial effects such as enjoying positive benefits or experiencing negative nonfinancial effects (including mental health issues) caused by nonpharmaceutical measures to contain the spread of COVID-19 were associated with individual subjective well-being. We also found that enjoying positive benefits such as more free time or a less polluted environment may reduce the likelihood of being unhappy for those who have experienced negative mental health issues or lost their job.
Being happy is also positively associated with the extent to which an individual agrees with their government’s approach to the COVID-19 pandemic. Meanwhile, happiness was not found to be strongly associated with individuals’ belief in the severity of the pandemic or the risk of complications arising once a person is infected. Our estimation results suggest that the magnitude of the correlation between each of the above factors and happiness differs from country to country. The results were quite consistent regardless of the estimation methods we used (either probit or ordered probit with an ordinal degree of happiness).
The effects, however, were different from country to country. One may argue that the difference may be due to the differences in the exposure to COVID-19 across countries and differences in measures adopted by each country’s government to contain the spread of the virus. In fact, in April 2020, Asian countries were quite successful in containing the spread of COVID-19. Different from the PRC, Japan and the Republic of Korea did not use strict measures such as complete lockdown or nationwide school closures. Meanwhile, European and some US states adopted stricter measures. However, we think that such differences only partially explain the differences in the effects. The differences may also lie in the differences in the structure of the economy and the cultural differences. For example, in Japan, the proportion of single households is quite high, so lockdown measures would change people’s lifestyle only slightly.
Our empirical results have several policy implications. First, governments should not only focus on mitigating the financial and economic effects of the pandemic but also take into account nonfinancial effects, since both financial and nonfinancial effects are strongly associated with individual well-being. Second, having an appropriate set of policies to deal with the crisis is not enough to improve people’s well-being. The government should either provide evidence of the effectiveness of such policies to the public or convey their message more effectively to garner support from their citizens for their policies. This will help to improve individuals’ well-being and thus social welfare.
As mentioned in the empirical approach, the present study had several limitations. First, at the time the survey was conducted, the countries in the sample were at different stages in the pandemic. Moreover, there are other sources of cross-country variation, such as culture, institutions, governments, etc. Therefore, the results of cross-country analyses should be interpreted cautiously. Second, although we tried to include as many observable variables in our estimation as possible, some variables, such as education level, were not collected. Furthermore, unobservable factors such as personality may also affect both outcome variables (happiness) and independent variables, especially subjective variables such as nonfinancial effects. Therefore, the estimation results could suffer from omitted variable biases. Reverse causality may even exist. For example, if people are unhappy, their viewpoints on all aspects measured may be negative, including their government’s approach to COVID-19. Therefore, our estimation results should be interpreted as indications of correlation rather than causality.

Supplementary Materials

The following are available online at https://www.mdpi.com/article/10.3390/su13147816/s1, Table S1: Summary Statistics for Each Country, Table S2: Individual Happiness during the COVID-19 Pandemic (Chinese Sample, Marginal Effects),Table S3: Individual Happiness during the COVID-19 Pandemic (Korean Sample, Marginal Effects),Table S4: Individual Happiness during the COVID-19 Pandemic (Japanese Sample, Marginal Effects), Table S5: Individual Happiness during the COVID-19 Pandemic (Italian Sample, Marginal Effects), Table S6: Individual Happiness during the COVID-19 Pandemic (British Sample, Marginal Effects), Table S7: Individual Happiness during the COVID-19 Pandemic (American Sample, Marginal Effects).

Funding

The APC was funded by the Asian Development Bank Institute.

Institutional Review Board Statement

This study uses the data collected by Belot et al. [40]. Prior to starting data collection, Belot et al. [40] had obtained approval from the ethnic board at the University of Exeter.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study. Please refer to [40].

Data Availability Statement

The original data for this study is available at https://osf.io/aubkc/. (accessed on 20 May 2020).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Wikipedia. Covid-19 Pandemic by Country and Territory. 2020. Available online: https://en.wikipedia.org/wiki/COVID-19_pandemic_by_country_and_territory (accessed on 8 January 2021).
  2. WEForum. Lockdown Is the World’s Biggest Psychological Experiment—And We Will Pay the Price. 2020. Available online: https://www.weforum.org/agenda/2020/04/this-is-the-psychological-side-of-the-covid-19-pandemic-that-were-ignoring (accessed on 25 May 2020).
  3. McKibbin, W.; Fernando, R. The economic effects of COVID-19. In Economics in the Time of COVID-19, VoxEU.org eBook; Baldwin, R., Weder di Mauro, B., Eds.; CEPR: London, UK, 2020. [Google Scholar]
  4. Glover, A.; Heathcote, J.; Krueger, D.; Ríos-Rull, J.V. Health Versus Wealth: On the Distributional Effects of Controlling a Pandemic; Working Paper No. w27046; National Bureau of Economic Research: Cambridge, MA, USA, 2020. [Google Scholar]
  5. Alon, T.M.; Doepke, M.; Olmstead-Rumsey, J.; Tertilt, M. The Impact of COVID-19 on Gender Equality; Working Paper No. w26947; National Bureau of Economic Research: Cambridge, MA, USA, 2020. [Google Scholar]
  6. Inoue, H.; Todo, Y. The propagation of the economic impact through supply chains: The case of a mega-city lockdown to contain the spread of COVID-19. Covid Econ. 2020, 2, 43–59. [Google Scholar] [CrossRef] [Green Version]
  7. Fadinger, H.; Schymik, J. The costs and benefits of home office during the Covid-19 pandemic: Evidence from infections and an input-output model for Germany. Covid Econ. 2020, 2, 107–134. [Google Scholar]
  8. Gottlieb, C.; Grobovšek, J.; Poschke, M. Working from home across countries. COVID Econ. 2020, 8, 71–91. [Google Scholar]
  9. Baldwin, R.; Evenett, S. COVID-19 and Trade Policy: Why Turning Inward Won’t Work, a VoxEU.org eBook; CEPR Press: London, UK, 2020. [Google Scholar]
  10. Brodeur, A.; Clark, A.E.; Fleche, S.; Powdthavee, N. Assessing the impact of the coronavirus lockdown on unhappiness, loneliness, and boredom using Google Trends. arXiv 2020, arXiv:2004.12129. [Google Scholar]
  11. Le, K.; Nguyen, M. The psychological consequences of COVID-19 lockdowns. Int. Rev. Appl. Econ. 2020, 35, 147–163. [Google Scholar] [CrossRef]
  12. Pedraza, P.; Guzi, M.; Tijdens, K. Life Dissatisfaction and Anxiety in Covid-19 Pandemic; GLO Discussion Paper, No. 544; Global Labor Organization: Essen, Germany, 2020. [Google Scholar]
  13. Dang, H.A.; Trinh, T.A. The Beneficial Impacts of COVID-19 Lockdowns on Air Pollution: Evidence from Vietnam; Working Paper No. 13651; Institute of Labor Economics (IZA): Bonn, Germany, 2020. [Google Scholar]
  14. Praveena, S.M.; Aris, A.Z. The impacts of COVID-19 on the environmental sustainability: A perspective from the Southeast Asian region. Environ. Sci. Pollut. Res. 2021. [Google Scholar] [CrossRef]
  15. Frey, B.S.; Stutzer, A. Happiness, economy and institutions. Econ. J. 2000, 110, 918–938. [Google Scholar] [CrossRef]
  16. Easterlin, R.A. Does economic growth improve the human lot? Some empirical evidence. In Nations and Households in Economic Growth; Academic Press: Cambridge, MA, USA, 1974; pp. 89–125. [Google Scholar]
  17. Frey, B.S.; Stutzer, A. Happiness and Economics: How the Economy and Institutions Affect Human Well-Being; Princeton University Press: Princeton, NJ, USA, 2010. [Google Scholar]
  18. Graham, C. Happiness around the World: The Paradox of Happy Peasants and Miserable Millionaires; Oxford University Press: Oxford, UK, 2012. [Google Scholar]
  19. Appleton, S.; Song, L. Life satisfaction in urban PRC: Components and determinants. World Dev. 2008, 36, 2325–2340. [Google Scholar] [CrossRef] [Green Version]
  20. Kusago, T. Rethinking of economic growth and life satisfaction in post-WWII Japan: A fresh approach. Soc. Indic. Res. 2007, 81, 79–102. [Google Scholar] [CrossRef]
  21. Ball, R.; Chernova, K. Absolute income, relative income, and happiness. Soc. Indic. Res. 2008, 88, 497–529. [Google Scholar] [CrossRef]
  22. Boes, S.; Winkelmann, R. The effect of income on general life satisfaction and dissatisfaction. Soc. Indic. Res. 2010, 95, 111. [Google Scholar] [CrossRef] [Green Version]
  23. Clark, A.E.; Frijters, P.; Shields, M.A. Relative income, happiness, and utility: An explanation for the Easterlin paradox and other puzzles. J. Econ. Lit. 2008, 46, 95–144. [Google Scholar] [CrossRef] [Green Version]
  24. Vendrik, M.C.; Woltjer, G.B. Happiness and loss aversion: Is utility concave or convex in relative income? J. Public Econ. 2007, 91, 1423–1448. [Google Scholar] [CrossRef]
  25. Clark, A.E.; Oswald, A.J. Unhappiness and unemployment. Econ. J. 1994, 104, 648–659. [Google Scholar] [CrossRef]
  26. Winkelmann, L.; Winkelmann, R. Why are the unemployed so unhappy? Evidence from panel data. Economica 1998, 65, 1–15. [Google Scholar] [CrossRef] [Green Version]
  27. Rodríguez-Pose, A.; Maslauskaite, K. Can policy make us happier? Individual characteristics, socio-economic factors and life satisfaction in Central and Eastern Europe. Camb. J. Reg. Econ. Soc. 2012, 5, 77–96. [Google Scholar] [CrossRef] [Green Version]
  28. Bjørnskov, C. Social capital and happiness in the United States. Appl. Res. Qual. Life 2008, 3, 43–62. [Google Scholar] [CrossRef]
  29. Lim, C.; Putnam, R.D. Religion, social networks, and life satisfaction. Am. Sociol. Rev. 2010, 75, 914–933. [Google Scholar] [CrossRef] [Green Version]
  30. Sujarwoto, S.; Tampubolon, G.; Pierewan, A.C. Individual and contextual factors of happiness and life satisfaction in a low middle-income country. Appl. Res. Qual. Life 2018, 13, 927–945. [Google Scholar] [CrossRef]
  31. Banks, J.; Xu, X. The mental health effects of the first two months of lockdown during the COVID-19 pandemic in the UK. Fisc. Stud. 2020, 41, 685–708. [Google Scholar] [CrossRef]
  32. Yamamura, E.; Tsutsui, Y. COVID-19, Mental Health, and Domestic Violence: Evidence from Japan. VOXEU. 22 June 2020. Available online: https://voxeu.org/article/covid-19-mental-health-and-domestic-violence (accessed on 25 October 2020).
  33. Liu, S.; Yang, L.; Zhang, C.; Xiang, Y.T.; Liu, Z.; Hu, S.; Zhang, B. Online mental health services in PRC during the COVID-19 outbreak. Lancet Psychiatry 2020, 7, 17–18. [Google Scholar] [CrossRef]
  34. Wang, C.; Pan, R.; Wan, X.; Tan, Y.; Xu, L.; Ho, C.S.; Ho, R.C. Immediate psychological responses and associated factors during the initial stage of the 2019 coronavirus disease (COVID-19) epidemic among the general population in PRC. Int. J. Environ. Res. Public Health 2020, 17, 1729. [Google Scholar] [CrossRef] [Green Version]
  35. Losada-Baltar, A.; Jiménez-Gonzalo, L.; Gallego-Alberto, L.; Pedroso-Chaparro, M.S.; Fernandes-Pires, J.; Márquez-González, M. “We’re staying at home”. Association of self-perceptions of aging, personal and family resources and loneliness with psychological distress during the lock-down period of COVID-19. J. Gerontol. Ser. B 2020, 76, 10–16. [Google Scholar] [CrossRef] [Green Version]
  36. Helliwell, J.F. How’s life? Combining individual and national variables to explain subjective well-being. Econ. Model. 2003, 20, 331–360. [Google Scholar] [CrossRef] [Green Version]
  37. Helliwell, J.F.; Huang, H. How’s your government? International evidence linking good government and well-being. Br. J. Political Sci. 2008, 38, 595–619. [Google Scholar] [CrossRef] [Green Version]
  38. Helliwell, J.F.; Huang, H.; Grover, S.; Wang, S. Empirical linkages between good governance and national well-being. J. Comp. Econ. 2018, 46, 1332–1346. [Google Scholar] [CrossRef]
  39. Ngoo, Y.T.; Tan, E.C.; Tey, N.P. Determinants of life satisfaction in Asia: A quantile regression approach. J. Happiness Stud. 2021, 22, 907–926. [Google Scholar] [CrossRef]
  40. Belot, M.; Choi, S.; Jamison, J.C.; Papageorge, N.W.; Tripodi, E.; van den Broek-Altenburg, E. Six-Country Survey on Covid-19. Covid Econ. 2020, 17, 205–219. [Google Scholar]
  41. Dang, H.A.H.; Nguyen, C.V. Gender inequality during the COVID-19 pandemic: Income, expenditure, savings, and job loss. World Dev. 2020, 140, 105296. [Google Scholar] [CrossRef]
  42. Oswald, A.J.; Wu, S. Well-being across America. Rev. Econ. Stat. 2011, 93, 1118–1134. [Google Scholar] [CrossRef] [Green Version]
Table 1. Number of Cases in Selected Countries.
Table 1. Number of Cases in Selected Countries.
23 April 202022 May 2020
PRC82,80482,971
Republic of Korea10,70811,142
Japan12,82916,519
Italy189,973228,658
United Kingdom138,078254,195
United States862,6051,589,223
Source: Wikipedia [1].
Table 2. Definition and Measurement of Variables.
Table 2. Definition and Measurement of Variables.
VariablesQuestions in SurveyMeasures
Level of happiness (dependent variable)How happy do you feel these days?=1 if the response is “Extremely unhappy” =2 if the response is “Moderately unhappy”=3 if the response is “Slightly unhappy” =4 if the response is “Neither happy nor unhappy” =5 if the response is “Slightly happy” =6 if the response is “Moderately happy” =7 if the response is “Extremely happy”
Being unhappy (dependent variable)Question: How happy do you feel these days?Answer: =1 if the respondent answers either “extremely unhappy,” “Moderately unhappy,” or “Slightly unhappy.”=0 otherwise (i.e., neither happy nor unhappy; slightly happy; moderately happy; or extremely happy)
Financial/Economic effects: Change in employment statusHave you lost your job or has your activity (as self-employed) been stopped as a consequence of the COVID-19 pandemic?=1 if the response is “Yes, permanently”=2 if the response is “Yes, temporarily”=3 if the response is “No”
Financial/Economic effects: Household incomeHave you experienced a fall in household income as a consequence of the COVID-19 pandemic?=1 if the response is “Yes”=0 if the response is “No”
Negative nonfinancial effectsHave you experienced any negative nonfinancial effects from the societal changes occurring due to the epidemic, such as:BoredomLonelinessTrouble sleepingGeneral anxiety and stressIncreased conflict with friends or relatives or neighborOthers?=1 if the respondent has experienced at least one negative effect=0 otherwise
Positive nonnegative effectsHave you experienced any positive nonfinancial effects from the societal changes occurring due to the epidemic, such as (select all that apply)? a Enjoying more free timeEnjoying time with family Reduction of air pollutionReduction of noise pollutionOther (please specify)?=1 if the respondent has enjoyed at least one positive effect=0 otherwise
Agreement with government approachDo you agree with the current approach taken by your government in response to the pandemic?=1 if response is “Strongly disagree”=2 if response is “Somewhat disagree”=3 if response is “Neither agree nor disagree”=4 if response is “Somewhat agree”=5 if response is “Strongly agree”
Effectiveness of containment measuresHow effective do you believe each of these measures is in reducing the spread of the epidemic?Shutting down schoolsShutting down public transportShutting down nonessential businessesLimiting mobility outside homeForbidding mass gatheringsIntroducing fines for citizens that don’t respect public safety measuresRequiring masks to be worn outside by everyoneFor each measure, there are five answering options: Not at all effective; Slightly effective; Moderately effective; Very effective; Extremely effective For each of the seven measures, we assigned a score of 0 if the response was “Not effective at all” and a score of 1 otherwise. To calculate variable relating to effectiveness of containment measure, we sum up our newly assigned score for each measure. The aggregate score ranges from zero to seven. For ease of interpretation, we calculated the z-score
Belief: Fraction of infected peopleWhat fraction of people in your local area do you think are currently infected?Fraction in percentage points (integers 0–100)
Belief: Probability of being infectedWhat do you think is the probability that you are, or have been, infected with COVID-19?Probability in percentage points (integers 0–100)
Belief: Prob. of infection without symptomsWhat do you think is the probability that an infected person develops no symptoms?Probability in percentage points (integers 0–100)
Belief: Probability of serious illness, not requiring hospitalizationWhat do you think is the probability that an infected person develops a serious illness (stays in bed) without requiring hospitalization?Probability in percentage points (integers 0–100)
Belief: Probability of a serious illness, requiring hospitalizationWhat do you think is the probability that an infected person develops a serious illness that requires hospitalization?Probability in percentage points % (integers 0–100)
Belief: Probability of dying if infectedWhat do you think is the probability that an infected person dies?Probability in percentage points (integers 0–100)
GenderWhat is your gender?=1 if male=0 otherwise
AgeHow old are you?=1 if between 18 and 25=2 if between 26 and 35=3 if between 36 and 45 =4 if between 46 and 55 =5 if between 56 and 65 =6 if between 66 and 75 =7 if above 75 =8 if refused to answer
IncomeIn what range is the gross annual income of your household?=1 if the response is in the range of the first quintile =2 if the response is in the range of the second quintile=3 if the response is in the range of the third quintile =4 if the response is in the range of the fourth quintile=5 if the response is in the range of the fifth quintile=6 if the response is “Prefer not to answer”
Living arrangementWhich of the following best describes your current living situation?1 = Live alone in my home (may have a pet) 2 = Live in a household with other people 3 = Live in a facility such as a nursing home that provides meals and 24-h nursing care 4 = Temporarily staying with a relative or friend5 = Temporarily staying in a shelter or homeless
Regions Dummy variables for 22 provinces in the PRC, 20 provinces in Italy, 8 regions in Japan, 16 provinces in the Republic of Korea, 12 regions in the UK, and 4 states in the US.
a This question has been widely used and been validated by previous studies [17,30,42].
Table 3. Summary Statistics: All Countries.
Table 3. Summary Statistics: All Countries.
MeanStd. Dev.
Well-being indicators
Not happy these days37.3048.36
Extremely unhappy6.6324.89
Moderately unhappy11.5031.90
Slightly unhappy19.1739.36
Neither happy nor unhappy33.0147.03
Slightly happy13.4734.14
Moderately happy13.1733.82
Extremely happy3.0517.21
Financial and nonfinancial effects
Experienced negative nonfinancial effects80.5939.56
Experienced positive nonfinancial effects85.6535.06
Permanent job loss5.3722.55
Temporary job loss24.7743.17
No change in employment69.8645.89
Experienced a fall in household income 49.3350.00
Agreement with government approach
Strongly disagree 6.0823.89
Somewhat disagree9.2028.90
Neither agree nor disagree20.5140.38
Somewhat agree32.6746.90
Strongly agree 31.5546.47
Effectiveness of containment measures
Shutting down schools66.3847.24
Shutting down public transport63.4448.16
Shutting down nonessential businesses62.7048.36
Limiting mobility outside home66.7447.12
Forbidding mass gatherings79.9040.08
Introducing fines for citizens that do not respect public safety measures63.8548.05
Requiring masks to be worn outside by everyone64.6747.80
Beliefs and perceptions
Fraction of infected people21.4123.19
Probability of infected people without symptoms34.6627.09
Probability of serious illness without requiring hospitalization32.1825.54
Probability of a serious illness that requires hospitalization32.6827.65
Probability that infected people would die21.8026.36
Male48.4349.98
Note: Unit %. Source: Author’s calculation.
Table 4. Being Unhappy During the COVID-19 Pandemic (Marginal Effects).
Table 4. Being Unhappy During the COVID-19 Pandemic (Marginal Effects).
(1)(2)(3)(4)(5)(6)(7)
All PRCRep. of KoreaJapanItalyUKUS
Negative nonfinancial effects0.273 ***0.236 ***0.332 ***0.116 ***0.368 ***0.358 ***0.307 ***
(0.017)(0.056)(0.049)(0.035)(0.041)(0.040)(0.041)
Positive nonfinancial effects−0.099 ***−0.119−0.101 **−0.049−0.169 **−0.121 ***−0.104 **
(0.018)(0.117)(0.050)(0.033)(0.070)(0.041)(0.041)
Job loss
Temporary job lost−0.0340.092−0.206 **−0.032−0.116 **0.010−0.050
(0.029)(0.068)(0.088)(0.094)(0.057)(0.068)(0.060)
No change in employment status−0.081 ***0.072−0.306 ***−0.066−0.102 *−0.071−0.085
(0.028)(0.070)(0.086)(0.087)(0.058)(0.067)(0.060)
Experienced a fall in household income 0.093 ***0.119 ***0.118 ***0.079 **0.092 ***0.0390.077 **
(0.013)(0.032)(0.032)(0.032)(0.034)(0.034)(0.033)
Agreement with the current government approach
Somewhat disagree−0.057 *0.034−0.085−0.090−0.081−0.0990.020
(0.032)(0.236)(0.089)(0.059)(0.069)(0.084)(0.074)
Neither agree nor disagree−0.193 ***−0.206−0.207 ***−0.194 ***−0.175 ***−0.303 ***−0.174 ***
(0.028)(0.225)(0.076)(0.051)(0.068)(0.077)(0.066)
Somewhat agree−0.198 ***−0.187−0.270 ***−0.180 ***−0.153 **−0.318 ***−0.171 ***
(0.027)(0.220)(0.076)(0.056)(0.059)(0.071)(0.059)
Strongly agree −0.224 ***−0.256−0.240 ***−0.166*−0.204 ***−0.312 ***−0.168 ***
(0.028)(0.220)(0.077)(0.088)(0.063)(0.073)(0.060)
Effectiveness of containment measures (aggregate, z-score)0.001−0.0030.011−0.003−0.027−0.0110.019
(0.006)(0.017)(0.018)(0.013)(0.018)(0.016)(0.015)
Belief
Fraction of infected people−0.001 **−0.002 **−0.001−0.001−0.001 *0.0010.000
(0.000)(0.001)(0.001)(0.001)(0.001)(0.001)(0.001)
Probability of being infected0.000−0.0010.0000.001 *0.002 **−0.0000.001
(0.000)(0.001)(0.001)(0.001)(0.001)(0.001)(0.001)
Prob. of infection without symptoms0.0000.0010.0010.000−0.000−0.001−0.000
(0.000)(0.001)(0.001)(0.001)(0.001)(0.001)(0.001)
Probability of serious illness not requiring hospitalization0.001 **0.003 ***−0.001−0.0010.0000.001−0.000
(0.000)(0.001)(0.001)(0.001)(0.001)(0.001)(0.001)
Probability of a serious illness requiring hospitalization−0.001 ***−0.001−0.000−0.001−0.000−0.001−0.001
(0.000)(0.001)(0.001)(0.001)(0.001)(0.001)(0.001)
Probability of dying if infected0.001 ***0.0010.002 **0.003 ***0.0010.0010.000
(0.000)(0.001)(0.001)(0.001)(0.001)(0.001)(0.001)
Gender (male = 1)0.027 **0.0260.062 **0.087 ***−0.041−0.0170.034
(0.012)(0.028)(0.031)(0.030)(0.030)(0.029)(0.031)
Income groups
Second quintile−0.015−0.0030.0170.0060.004−0.024−0.053
(0.020)(0.049)(0.050)(0.046)(0.051)(0.047)(0.050)
Third quintile−0.038 **−0.042−0.0650.0500.032−0.036−0.132 ***
(0.019)(0.049)(0.048)(0.046)(0.050)(0.047)(0.049)
Fourth quintile−0.041 **−0.090 *−0.029−0.0300.061−0.048−0.115 **
(0.020)(0.047)(0.050)(0.047)(0.050)(0.049)(0.050)
Fifth quintile−0.033−0.069−0.012−0.0060.025−0.034−0.095*
(0.021)(0.052)(0.053)(0.052)(0.056)(0.051)(0.056)
Preferred not to say−0.003 −0.0130.0180.1560.003−0.102
(0.042) (0.076)(0.089)(0.108)(0.110)(0.116)
N60859949621011104210151054
Note: Dependent variable is being unhappy, a dummy variable that takes value of 1 if the individual reported being “Extremely unhappy,” “Moderately unhappy,” or “Slightly unhappy.” Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01. Negative nonfinancial effect and positive nonfinancial effect are two dummy variables if individuals experience negative feeling or enjoy positive activities during the pandemic. The reference group for job loss variables is groups of those who lost their job permanently because of COVID-19. Experienced a fall in household income is a dummy variable indicating those who experienced reduction in household income because of the pandemic. The reference groups for degree of agreement with government approach are those who strongly disagree (with the government approach). Effectiveness of containment measures is the number of measures that respondents think is effective in containing the virus. We calculate z-scores for this variable. The reference group for income groups is the group of those who are in the first quintile (poorest). In all specifications, we controlled for individual age, living arrangement, whether they had contact with any infected person, whether they had any COVID-19 symptoms, whether they accessed open-air areas in their living areas, whether they had traveled abroad since January 2020, whether they kept a distance from their colleagues and the city/province/prefecture and where they live. Source: Belot et al.’s [40] data and author’s estimation.
Table 5. Nonfinancial Effects and Being Unhappy (Marginal Effects).
Table 5. Nonfinancial Effects and Being Unhappy (Marginal Effects).
All PRCRep. of KoreaJapanItalyUKUS
Boredom0.079 ***0.119 ***0.060 *−0.0090.077 ***0.079 ***0.100 ***
(0.012)(0.028)(0.031)(0.032)(0.029)(0.027)(0.029)
Loneliness0.149 ***0.091 ***0.185 ***0.156 ***0.196 ***0.148 ***0.081 ***
(0.013)(0.029)(0.038)(0.043)(0.029)(0.030)(0.030)
Trouble sleeping0.096 ***0.073 **0.0520.0830.104 ***0.141 ***0.104 ***
(0.014)(0.030)(0.045)(0.052)(0.030)(0.028)(0.030)
General anxiety and stress0.160 ***0.164 ***0.167 ***0.051*0.242 ***0.159 ***0.157 ***
(0.012)(0.027)(0.029)(0.030)(0.027)(0.028)(0.029)
Conflict with friends, relatives, or neighbors0.042 **0.0030.135 ***0.0590.0140.0710.050
(0.020)(0.048)(0.052)(0.054)(0.048)(0.049)(0.049)
Other negative feelings0.030 0.058−0.0230.0680.128−0.019
(0.044) (0.114)(0.095)(0.121)(0.124)(0.084)
Enjoying more free time−0.077 ***−0.008−0.115 ***0.021−0.080 ***−0.130 ***−0.147 ***
(0.012)(0.032)(0.033)(0.033)(0.030)(0.028)(0.029)
Enjoying time with family−0.060 ***−0.062*−0.011−0.028−0.045−0.108 ***−0.080 ***
(0.013)(0.036)(0.032)(0.033)(0.030)(0.030)(0.030)
Reduction of air pollution−0.0130.028−0.065 **−0.028−0.002−0.0060.001
(0.014)(0.031)(0.032)(0.044)(0.038)(0.032)(0.032)
Reduction of noise pollution−0.031 **−0.031−0.073 *0.076−0.068 **0.009−0.061 *
(0.015)(0.033)(0.040)(0.055)(0.033)(0.034)(0.035)
Other positive effects−0.0590.195−0.0100.048−0.104−0.094−0.127
(0.045)(0.214)(0.130)(0.111)(0.115)(0.101)(0.082)
Other covariatesYes Yes Yes Yes Yes Yes Yes
Number of observations60859939621011104210151054
Note: Dependent variable is being unhappy, a dummy variable that takes value of 1 if the individual reported “Extremely unhappy,” “Moderately unhappy,” or “Slightly unhappy.” Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01. This table is similar to Table 4 (please refer to note to Table 4), except that we replace two dummy variables, indicating negative and positive nonfinancial effects with indicators, indicating that individuals experience each of six negative feelings and enjoy each of five positive activities. Source: Belot et al.’s [40] data and author’s estimations.
Table 6. Individual Happiness during the COVID-19 Pandemic (Pooled Sample, Marginal Effects).
Table 6. Individual Happiness during the COVID-19 Pandemic (Pooled Sample, Marginal Effects).
Extremely UnhappyModerately UnhappySlightly UnhappyNeither Happy nor UnhappySlightly HappyModerately HappyExtremely Happy
Negative nonfinancial effects0.084 ***0.087 ***0.073 ***−0.021 ***−0.066 ***−0.113 ***−0.044 ***
(0.005)(0.005)(0.004)(0.003)(0.004)(0.006)(0.003)
Positive nonfinancial effects−0.039 ***−0.040 ***−0.034 ***0.010 ***0.030 ***0.052 ***0.020 ***
(0.005)(0.005)(0.004)(0.002)(0.004)(0.007)(0.003)
Job loss
  Temporary job lost−0.015−0.014 *−0.010 *0.0070.011 *0.016 *0.005 *
(0.009)(0.008)(0.006)(0.005)(0.006)(0.009)(0.003)
  No change in employment status−0.028 ***−0.027 ***−0.021 ***0.010 **0.021 ***0.033 ***0.012 ***
(0.009)(0.008)(0.006)(0.005)(0.006)(0.009)(0.003)
Experienced a fall in household income 0.027 ***0.028 ***0.023 ***−0.007 ***−0.021 ***−0.036 ***−0.014 ***
(0.004)(0.004)(0.003)(0.001)(0.003)(0.005)(0.002)
Agreement with the current government’s approach
  Somewhat disagree−0.046 ***−0.030 ***−0.011 ***0.031 ***0.024 ***0.026 ***0.006 ***
(0.014)(0.009)(0.003)(0.010)(0.007)(0.008)(0.002)
  Neither agree nor disagree−0.085 ***−0.065 ***−0.034 ***0.052 ***0.051 ***0.064 ***0.017 ***
(0.013)(0.008)(0.004)(0.009)(0.006)(0.007)(0.002)
  Somewhat agree−0.093 ***−0.075 ***−0.042 ***0.055 ***0.058 ***0.076 ***0.022 ***
(0.012)(0.008)(0.003)(0.009)(0.006)(0.007)(0.002)
  Strongly agree −0.108 ***−0.094 ***−0.060 ***0.055 ***0.072 ***0.102 ***0.032 ***
(0.013)(0.008)(0.004)(0.009)(0.006)(0.008)(0.003)
Effectiveness of measures (aggregate, z-score)−0.004 **−0.004 **−0.003 **0.001 **0.003 **0.005 **0.002 **
(0.002)(0.002)(0.001)(0.000)(0.001)(0.002)(0.001)
Belief
  Fraction of infected people−0.000 **−0.000 **−0.000 **0.000 **0.000 **0.000 **0.000 **
(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)
  Probability of being infected 0.0000.0000.000−0.000−0.000−0.000−0.000
(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)
  Probability of infected people with no symptoms−0.000−0.000−0.0000.0000.0000.0000.000
(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)
  Probability of serious illness without requiring hospitalization−0.000−0.000−0.0000.0000.0000.0000.000
(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)
  Probability of a serious illness that requires hospitalization−0.000−0.000−0.0000.0000.0000.0000.000
(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)
  Probability that infected people would die0.000 ***0.000 ***0.000 ***−0.000 ***−0.000 ***−0.000 ***−0.000 ***
(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)
Male0.0050.0050.004−0.001−0.004−0.007−0.003
(0.003)(0.003)(0.003)(0.001)(0.003)(0.004)(0.002)
Income groups
  Second quintile−0.012 **−0.011 **−0.009 **0.004 **0.009 **0.014 **0.005 **
(0.006)(0.006)(0.004)(0.002)(0.004)(0.007)(0.002)
  Third quintile−0.016 ***−0.016 ***−0.013 ***0.006 ***0.012 ***0.020 ***0.007 ***
(0.006)(0.005)(0.004)(0.002)(0.004)(0.007)(0.002)
  Fourth quintile−0.021 ***−0.022 ***−0.018 ***0.006 ***0.016 ***0.027 ***0.010 ***
(0.005)(0.006)(0.004)(0.002)(0.004)(0.007)(0.003)
  Fifth quintile−0.021 ***−0.021 ***−0.017 ***0.006 ***0.016 ***0.027 ***0.010 ***
(0.006)(0.006)(0.005)(0.002)(0.005)(0.007)(0.003)
Preferred not to say−0.003−0.003−0.0020.0010.0020.0040.001
(0.013)(0.012)(0.009)(0.005)(0.009)(0.014)(0.005)
Number of observations6085
Note: Dependent variable is ordinal degree of happiness ranging from extremely unhappy to extremely happy. Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01. All covariates used in this table are similar to those in Table 4. Please refer to note to Table 4. Source: Belot et al.’s [40] data and author’s estimations.
Table 7. Soothing Role of Enjoying Positive Nonfinancial Effects.
Table 7. Soothing Role of Enjoying Positive Nonfinancial Effects.
(1)(2)(3)(4)(5)(6)(7)
All PRCRep. of KoreaJapanItalyUKUS
Panel A: Negative and positive nonfinancial effects
Negative No; Positive Yes−0.133 ***−0.145−0.115 *−0.099 *−0.165−0.013−0.100 *
(0.027)(0.109)(0.067)(0.053)(0.127)(0.053)(0.060)
Negative Yes; Positive No0.211 *** 0.275 ***0.0720.325 **0.447 ***0.289 ***
(0.032) (0.084)(0.050)(0.147)(0.068)(0.071)
Negative Yes; Positive Yes0.132 ***0.0610.193 ***0.0500.1810.285 ***0.176 ***
(0.026)(0.108)(0.061)(0.045)(0.125)(0.052)(0.058)
Job loss
  Temporary job lost−0.0290.065−0.202 **−0.020−0.109 *−0.009−0.035
(0.028)(0.072)(0.087)(0.092)(0.057)(0.068)(0.059)
  No job lost−0.073 ***0.040−0.304 ***−0.054−0.091−0.073−0.072
(0.028)(0.073)(0.085)(0.085)(0.058)(0.067)(0.059)
Panel B: Job loss and positive nonfinancial effects
Negative experience0.272 ***0.243 ***0.330 ***0.114 ***0.369 ***0.362 ***0.311 ***
(0.017)(0.055)(0.049)(0.035)(0.040)(0.039)(0.041)
Job loss No; Positive Yes−0.089 ***−0.179−0.108*−0.039−0.174 **−0.083*−0.092 **
(0.021)(0.132)(0.060)(0.034)(0.081)(0.047)(0.047)
Job loss Yes; Positive No0.141 *** 0.1080.111−0.0440.288 ***0.163
(0.050) (0.107)(0.108)(0.142)(0.110)(0.106)
Job loss Yes; Positive Yes−0.047*−0.1570.020−0.013−0.168 **−0.034−0.057
(0.024)(0.133)(0.069)(0.053)(0.085)(0.056)(0.055)
Number of observations60859949621011104210151054
Note: Dependent variable is being unhappy, a dummy variable that takes value of 1 if the individual reported “Extremely unhappy,” “Moderately unhappy,” or “Slightly unhappy.” Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01. All covariates used in this table (not presented for brevity) are similar to those in Table 4. Please refer to note to Table 4. Source: Belot et al.’s [40] data and author’s estimations.
Table 8. Soothing Role of Enjoying Positive Nonfinancial Effects: Levels of Happiness.
Table 8. Soothing Role of Enjoying Positive Nonfinancial Effects: Levels of Happiness.
(1)(2)(3)(4)(5)(6)(7)
Extremely UnhappyModerately UnhappySlightly UnhappyNeither Happy nor UnhappySlightly HappyModerately HappyExtremely Happy
Panel A: Negative and Positive Consequences
Negative No; Positive Yes−0.023 ***−0.039 ***−0.054 ***−0.037 ***0.025 ***0.081 ***0.047 ***
(0.005)(0.007)(0.009)(0.006)(0.005)(0.013)(0.007)
Negative Yes; Positive No0.078 ***0.080 ***0.063 ***−0.030 ***−0.063 ***−0.097 ***−0.030 ***
(0.010)(0.009)(0.008)(0.007)(0.007)(0.012)(0.005)
Negative Yes; Positive Yes0.033 ***0.040 ***0.039 ***−0.003−0.031 ***−0.056 ***−0.020 ***
(0.005)(0.007)(0.008)(0.002)(0.005)(0.011)(0.005)
Job loss
  Temporary job lost−0.013−0.012−0.0090.0060.0100.0140.005
(0.009)(0.008)(0.006)(0.004)(0.006)(0.009)(0.003)
  No job lost−0.025 ***−0.024 ***−0.019 ***0.009 **0.019 ***0.030 ***0.011 ***
(0.009)(0.008)(0.006)(0.004)(0.006)(0.009)(0.003)
Panel B: Job lost and positive nonfinancial effects
Negative experience0.083 ***0.086 ***0.072 ***−0.021 ***−0.065 ***−0.112 ***−0.043 ***
(0.005)(0.005)(0.004)(0.003)(0.004)(0.006)(0.003)
Job loss No; Positive Yes−0.038 ***−0.037 ***−0.028 ***0.014 ***0.028 ***0.045 ***0.016 ***
(0.007)(0.006)(0.004)(0.004)(0.004)(0.006)(0.002)
Job loss Yes; Positive No0.075 ***0.047 ***0.018 ***−0.047 ***−0.038 ***−0.044 ***−0.011 ***
(0.024)(0.013)(0.004)(0.016)(0.010)(0.011)(0.003)
Job loss Yes; Positive Yes−0.028 ***−0.026 ***−0.018 ***0.012 ***0.020 ***0.030 ***0.010 ***
(0.008)(0.007)(0.005)(0.004)(0.005)(0.008)(0.003)
Number of observations6085608560856085608560856085
Note: Dependent variable is the ordinal degree of happiness ranging from extremely unhappy to extremely happy. Standard errors in brackets; * p < 0.1, ** p < 0.05, *** p < 0.01. All covariates used in this table (not presented for brevity) are similar to those in Table 4. Please refer to note to Table 4. Source: Belot et al.’s [40] data and author’s estimation.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Long, T.Q. Individual Subjective Well-Being during the COVID-19 Pandemic. Sustainability 2021, 13, 7816. https://doi.org/10.3390/su13147816

AMA Style

Long TQ. Individual Subjective Well-Being during the COVID-19 Pandemic. Sustainability. 2021; 13(14):7816. https://doi.org/10.3390/su13147816

Chicago/Turabian Style

Long, Trinh Q. 2021. "Individual Subjective Well-Being during the COVID-19 Pandemic" Sustainability 13, no. 14: 7816. https://doi.org/10.3390/su13147816

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop