## 1. Individual Characteristics and Overall Government Performance

## Source details

**Canonical URL:** [1. Individual Characteristics and Overall Government Performance](https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021053-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2021/english/wpiea2021053-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2021/english/wpiea2021053-print-pdf.pdf.json)

---

### I. Introduction — Key findings
- Survey timing: All fieldwork was conducted between April 15 and April 23, 2020.
- Sample: 13,200 respondents in 11 countries: Canada, China, France, Germany, India, Japan, South Korea, Mexico, Saudi Arabia, UK and the U.S.; each country has around 1,200 observations.
- Main patterns documented:
  - Systematic differences in perceptions by gender, employment status, tele-workability of jobs, region, and income.
  - Women are more likely to have an unfavorable view of government responses on both health and economic fronts.
  - Non-college educated, the unemployed, and those in jobs that cannot be done from home perceive government performance as less favorable.
  - Urban individuals and those employed in larger firms report more favorable perceptions of government responses to the health crisis.
  - Age is generally not a significant determinant of perceptions, though the young have a less favorable perception of the health response in emerging market economies.
  - Richer individuals have a more favorable perception of government responses in advanced economies, and in countries where less stringent containment measures were adopted; conversely, the rich are less supportive in countries with higher initial levels of inequality.
- Beliefs about the future:
  - Older individuals, the economically vulnerable, and urban residents are more worried about the health and economic well-being of vulnerable groups.
  - Individuals personally affected by the crisis (job loss, contact with sick people, or personally symptomatic) hold particularly pessimistic beliefs about the future.
  - Women are more pessimistic about prolonged job losses in countries with more stringent containment measures.
  - Concerns are stronger overall for individuals in emerging market economies, despite lower incidence of COVID-19 cases at survey time.
- Preferences for redistribution:
  - Roughly two-thirds of respondents believe the poor have been unfairly burdened by the pandemic and that redistribution is necessary.
  - Those who believe the poor are unfortunate are more likely to support pro-equality redistributive policies.
  - Strong systematic support for financial assistance to vulnerable businesses and households among the old, college-educated, and individuals in tele-workable jobs, even controlling for income.
  - Those who suffered adverse health impacts are less willing to support redistributive policies.
  - Urban residents express strong concerns for the vulnerable but are less supportive of government interventions.
  - Individuals with a favorable assessment of government responses are more willing to provide social safety nets for those hit hardest.

### I. Introduction — Policy implication emphasized
- Policies should take into account the unequal labor market and health impacts of the pandemic to quell economic anxieties and social discontent.
- Effective government actions on health and economic fronts can promote support for redistributive policies.

### II. Data and Methodology — Survey design and measures
- Survey instrument: 2020 Edelman Trust Barometer Spring update (standardized, representative individual-level data).
- Representativeness: Nationally representative based on age, region, gender, and additionally in the UK and U.S. by race/ethnicity. Income and education are not nationally representative for the emerging market countries in the sample.
- Government performance assessment:
  - Three measures: Overall, Health and Economy.
  - All measures defined on an index from 1 to 5; higher index values indicate a more favorable assessment.
  - Overall: "Meeting your overall expectations of how the government should be responding."
  - Health: “Ensuring that medical supplies and good quality medical treatment are available even in the poorest areas."
  - Economy: “Taking the necessary actions to keep the economy from collapsing”.
  - Overall, 46 percent of respondents believed the government was doing “well” or “very well” (i.e., at least 4 on scale of 5).
- Support for the poor and future concerns:
  - Four measures: Vulnerable more affected; Capitalism is harmful; Prolonged job losses; Threat of automation.
  - All measures defined on an index from 1 to 9; higher values indicate stronger agreement.
  - Overall, 67 percent of respondents believe the vulnerable are being unfairly burdened.
  - More than half of respondents: worried about long-term COVID-related job losses; worried about threat from automation; unfavorable view of capitalism.
- Policy expectations (binary indicators with 1 = expect leading role):
  1. Give Financial relief: “Giving financial relief and support to people who have lost their jobs or livelihood”
  2. Help small businesses: “Helping small businesses survive until they can reopen”
  3. Manage the spread: “Develop and put in place the necessary policies and procedures to slow and control the spread of the virus”
  4. Prep for reopening: “Plan and prepare for how the country will get back to normal after the worst of the pandemic is over.”

### II. Data and Methodology — Empirical approach
- Baseline model: fixed effects linear probability model (Equation 1) with country fixed effects 휸௖, weighted by public weights, errors clustered at the country level.
- Outcome variables: converted to binary indicators.
  - Government assessment: dummy = 1 for rating ≥ 4 on 1–5 scale.
  - Concerns about the future: dummy = 1 for rating ≥ 7 on 1–9 scale.
  - Policy preferences: binary as recorded in survey.
- Individual controls include: Female dummy, young dummy (1 if < 30 years), college dummy, rich dummy (1 if household income above the 75th percentile).
- Occupational controls (working-age subsample): Unemployed dummy, tele-workability dummy, small firm dummy (firm size < 100 employees).
- Directly impacted measures: Health impact (infected or knew someone close infected); Economic impact (lost job or furloughed due to COVID-19).
- Heterogeneity by country-level variables (Equation 2): interact individual variables with country dummies 휃௖ for Stringency Index, timing of COVID-19 spread, Gini Index; create dummies for advanced vs emerging economies.

### III. Results — Assessment of government performance
- Aggregate and subgroup patterns:
  - Overall average: 46 percent rate government “well” or “very well”.
  - Country variation: Japan lowest perceptions; Saudi Arabia and India highest perceptions.
  - Unemployed, women, the less educated, and those in firms with fewer than 100 employees hold less favorable views.
  - Urban, young (below 30 years), and tele-workable job holders report more favorable assessments.
  - For most countries (except South Korea and Saudi Arabia), perceptions of economic response are generally more favorable than health response.
- Detailed determinants:
  - Gender: women have less favorable assessments across overall, health, and economic dimensions; about 7 percent of the gender gap is explained by household income.
  - Employment: those who lost their jobs due to COVID-19 have less favorable perceptions of health response; employees in small firms have less favorable health-view; tele-workable sector employees have more favorable overall views.
  - Education: college-educated more favorable for economic performance; around 50 percent of the education gap is explained by household income.
  - Income: rich have more favorable assessments in advanced economies; rich more approving where less stringent measures were adopted; rich less approving in countries with higher initial inequality.

### III. Results — Support for the poor and concerns about the future
- Aggregate measures:
  - 67 percent believe vulnerable are disproportionately impacted.
  - More than half worried about long-term job loss, automation, and hold unfavorable view of capitalism.
- Determinants of concerns:
  - Gender: women more likely to believe vulnerable are more affected; effect disappears when controlling for tele-workability; in high-stringency countries women report heightened anxieties about prolonged job losses.
  - Age: older individuals more likely to believe vulnerable are more affected; view capitalism as harmful; more concerned about automation; in emerging markets older people more likely to report concerns about prolonged job losses.
  - Personally affected: those experiencing job loss or exposure to infection are more sympathetic toward vulnerable and more pessimistic about future across all concern measures.
  - Employment: unemployed more concerned about prolonged job losses; unemployed respondents in countries with early surge followed by downward trend show amplified anxieties.
  - Tele-workable jobs: less worried about automation and prolonged job losses.
  - Education: education not statistically significant in baseline once controls included; college educated less anxious in countries with downward trends in cases.
  - Urban residence: consistently more pessimistic across concerns.
  - Income: no baseline relationship with concerns; richer households less pessimistic in advanced economies.
- Synthesis:
  - Economic anxiety amplified early in crisis among economically vulnerable: unemployed, non-tele-workable workers, and those experiencing illness/job loss.

### III. Results — Policy preferences
- Baseline patterns:
  - No statistically significant differences across gender and income in baseline for preferences on government support and interventions.
  - Majority expect government to play leading role on financial relief, help small businesses, manage spread, and prep for reopening.
- Determinants:
  - Age: older people favor financial support for households and small businesses relative to young.
  - Personally affected: those who lost jobs show no significant increase in support for government; those with adverse health impacts show a negative and statistically significant effect (less supportive).
  - Employment: unemployed favor government preparing for reopening; small-firm employees support financial relief; tele-workable workers support government role in managing crisis and liquidity support.
  - Urban: urban residents are less supportive of government interventions in baseline, though this reverses in emerging markets and high-stringency countries.
    - Gelbach (2016) decomposition: being young reduces the negative urban effect on provision of financial support by 10 percent; tele-workability weakens it by 5 percent; much of urban-rural divide remains unexplained.
  - Income: higher household incomes more likely to support social safety nets and liquidity support; upper two quartiles and middle two quartiles show significant positive effects.
  - Perception of government: individuals with more favorable assessment of government performance expect government to take leading role in liquidity support, addressing health crisis, and preparing for reopening.
- Robustness:
  - Results robust to Gelbach (2016) decomposition, probit specifications, restricting sample to advanced economies, excluding early-peak Asian countries, and dropping China/Saudi Arabia where applicable.
  - Alternative survey items for economic and health assessments and for policy preferences yield consistent results.

### IV. Conclusion
- Perceptions of government handling of the pandemic depend on gender, age, region, employment, socio-economic status, and recent adverse experiences.
- Economically vulnerable and those hardest hit view government responses less favorably; personal illness or job loss heightens pessimism about prolonged job losses and automation risk.
- Policies that directly reduce economic hardship and anxiety (e.g., direct transfers, social safety nets) are important to contain pandemic fallout and to build support for redistribution.
- Preferences for redistribution vary by personal experience and views about the poor; favorable assessments of government performance are associated with greater willingness to support financial relief for individuals and firms.
- Limitations: single time period early in crisis; framing of survey questions may affect results; longer panel datasets needed to track changes over time.

*Source: wpiea2021053-print-pdf*

### 1. Individual Characteristics and Overall Government Performance ______________________ 8

### 1. Individual Characteristics and Overall Government Performance ______________________ 8

### 1. Individual Characteristics and Overall Government Performance
- Section title: "Individual Characteristics and Overall Government Performance"
- Page: 8

### 2. Individual Characteristics and Socio-Economic Concerns
- Section title: "Individual Characteristics and Socio-Economic Concerns"
- Page: 9

### 3. Individual Characteristics and Role of Government
- Section title: "Individual Characteristics and Role of Government"
- Page: 10

### 4. Coefficients of Individual Characteristics on Government Responses
- Section title: "Coefficients of Individual Characteristics on Government Responses"
- Page: 13

### 5. Coefficients of Individual Characteristics of Socio-Economic Concerns
- Section title: "Coefficients of Individual Characteristics of Socio-Economic Concerns"
- Page: 14

### 6. Impact of Region and Age on Preferences for Redistribution (marginal effects)
- Section title: "Impact of Region and Age on Preferences for Redistribution (marginal effects)"
- Page: 16

### 7. Marginal Effect of Income and Age on Policy Preferences in Advanced Economies
- Section title: "Marginal Effect of Income and Age on Policy Preferences in Advanced Economies"
- Page: 18

*Source: wpiea2021053-print-pdf - 1. Individual Characteristics and Overall Government Performance ______________________ 8 — https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021053-print-pdf.pdf*

### References  __________________________________________________________________________________  21

### I. Introduction

### Key findings
- The paper uses an individual-level, cross-country survey conducted by the Edelman Trust Barometer in mid-April 2020 to examine perceptions of government performance, beliefs about the future, and attitudes about redistribution.
- Survey timing: All fieldwork was conducted between April 15 and April 23, 2020.
- Sample: 13,200 respondents in 11 countries: Canada, China, France, Germany, India, Japan, South Korea, Mexico, Saudi Arabia, UK and the U.S.; each country has around 1,200 observations.
- Main patterns documented:
  - Systematic differences in perceptions by gender, employment status, tele-workability of jobs, region, and income.
  - Women are more likely to have an unfavorable view of government responses on both health and economic fronts.
  - Non-college educated, the unemployed, and those in jobs that cannot be done from home perceive government performance as less favorable.
  - Urban individuals and those employed in larger firms report more favorable perceptions of government responses to the health crisis.
  - Age is generally not a significant determinant of perceptions, though the young have a less favorable perception of the health response in emerging market economies.
  - Richer individuals have a more favorable perception of government responses in advanced economies, and in countries where less stringent containment measures were adopted; conversely, the rich are less supportive in countries with higher initial levels of inequality.
- Beliefs about the future:
  - Older individuals, the economically vulnerable, and urban residents are more worried about the health and economic well-being of vulnerable groups.
  - Individuals personally affected by the crisis (job loss, contact with sick people, or personally symptomatic) hold particularly pessimistic beliefs about the future.
  - Women are more pessimistic about prolonged job losses in countries with more stringent containment measures.
  - Concerns are stronger overall for individuals in emerging market economies, despite lower incidence of COVID-19 cases at survey time.
- Preferences for redistribution:
  - Roughly two-thirds of respondents believe the poor have been unfairly burdened by the pandemic and that redistribution is necessary.
  - Those who believe the poor are unfortunate are more likely to support pro-equality redistributive policies.
  - Strong systematic support for financial assistance to vulnerable businesses and households among the old, college-educated, and individuals in tele-workable jobs, even controlling for income.
  - Those who suffered adverse health impacts are less willing to support redistributive policies.
  - Urban residents express strong concerns for the vulnerable but are less supportive of government interventions.
  - Individuals with a favorable assessment of government responses are more willing to provide social safety nets for those hit hardest.

### Policy implication emphasized
- Policies should take into account the unequal labor market and health impacts of the pandemic to quell economic anxieties and social discontent.
- Effective government actions on health and economic fronts can promote support for redistributive policies.

---

### II. Data and Methodology

### Survey design and measures
- Survey instrument: 2020 Edelman Trust Barometer Spring update (standardized, representative individual-level data).
- Representativeness: Nationally representative based on age, region, gender, and additionally in the UK and U.S. by race/ethnicity. Income and education are not nationally representative for the emerging market countries in the sample.
- Government performance assessment:
  - Three measures: Overall, Health and Economy.
  - All measures defined on an index from 1 to 5; higher index values indicate a more favorable assessment.
  - Overall: "Meeting your overall expectations of how the government should be responding."
  - Health: “Ensuring that medical supplies and good quality medical treatment are available even in the poorest areas."
  - Economy: “Taking the necessary actions to keep the economy from collapsing”.
  - Overall, 46 percent of respondents believed the government was doing “well” or “very well” (i.e., at least 4 on scale of 5).
- Support for the poor and future concerns:
  - Four measures: Vulnerable more affected; Capitalism is harmful; Prolonged job losses; Threat of automation.
  - All measures defined on an index from 1 to 9; higher values indicate stronger agreement.
  - Overall, 67 percent of respondents believe the vulnerable are being unfairly burdened.
  - More than half of respondents: worried about long-term COVID-related job losses; worried about threat from automation; unfavorable view of capitalism.
- Policy expectations (binary indicators with 1 = expect leading role):
  1. Give Financial relief: “Giving financial relief and support to people who have lost their jobs or livelihood”
  2. Help small businesses: “Helping small businesses survive until they can reopen”
  3. Manage the spread: “Develop and put in place the necessary policies and procedures to slow and control the spread of the virus”
  4. Prep for reopening: “Plan and prepare for how the country will get back to normal after the worst of the pandemic is over.”

### Empirical approach
- Baseline model: fixed effects linear probability model (Equation 1) with country fixed effects 휸௖, weighted by public weights, errors clustered at the country level.
- Outcome variables: converted to binary indicators.
  - Government assessment: dummy = 1 for rating ≥ 4 on 1–5 scale.
  - Concerns about the future: dummy = 1 for rating ≥ 7 on 1–9 scale.
  - Policy preferences: binary as recorded in survey.
- Individual controls (examples):
  - Female dummy (1 if female), young dummy (1 if < 30 years), college dummy (1 if at least college educated), rich dummy (1 if household income above the 75th percentile).
- Occupational controls (working-age subsample):
  - Unemployed dummy (1 if unemployed), tele-workability dummy (1 for sectors above the median tele-workability), small firm dummy (1 for firm size < 100 employees).
- Directly impacted measures:
  - Health impact: respondent infected or knew someone close infected.
  - Economic impact: lost job or furloughed due to COVID-19.
- Heterogeneity by country-level variables (Equation 2): interact individual variables with country dummies 휃௖ that equal 1 if country is above the sample median in Stringency Index, or experienced early surge then declining cases, or above median in Gini Index; also create dummies for advanced vs emerging economies.

---

### III. Results

### Assessment of government performance
- Aggregate and subgroup patterns:
  - Overall average: 46 percent rate government “well” or “very well”.
  - Country variation: Japan lowest perceptions; Saudi Arabia and India highest perceptions.
  - Unemployed, women, the less educated, and those in firms with fewer than 100 employees hold less favorable views.
  - Urban, young (below 30 years), and tele-workable job holders report more favorable assessments.
  - For most countries (except South Korea and Saudi Arabia), perceptions of economic response are generally more favorable than health response.
- Detailed determinants:
  - Gender: women have less favorable assessments across overall, health, and economic dimensions; about 7 percent of the gender gap is explained by household income.
  - Employment: those who lost their jobs due to COVID-19 have less favorable perceptions of health response; employees in small firms have less favorable health-view; tele-workable sector employees have more favorable overall views.
  - Education: college-educated more favorable for economic performance; around 50 percent of the education gap is explained by household income.
  - Income: rich have more favorable assessments in advanced economies; rich more approving where less stringent measures were adopted; rich less approving in countries with higher initial inequality.

### Support for the poor and concerns about the future
- Aggregate measures:
  - 67 percent believe vulnerable are disproportionately impacted.
  - More than half worried about long-term job loss, automation, and hold unfavorable view of capitalism.
- Determinants of concerns:
  - Gender: women more likely to believe vulnerable are more affected; effect disappears when controlling for tele-workability; in high-stringency countries women report heightened anxieties about prolonged job losses.
  - Age: older individuals more likely to believe vulnerable are more affected; view capitalism as harmful; more concerned about automation; in emerging markets older people more likely to report concerns about prolonged job losses.
  - Personally affected: those experiencing job loss or exposure to infection are more sympathetic toward vulnerable and more pessimistic about future across all concern measures.
  - Employment: unemployed more concerned about prolonged job losses; unemployed respondents in countries with early surge followed by downward trend show amplified anxieties.
  - Tele-workable jobs: less worried about automation and prolonged job losses.
  - Education: education not statistically significant in baseline once controls included; college educated less anxious in countries with downward trends in cases.
  - Urban residence: consistently more pessimistic across concerns.
  - Income: no baseline relationship with concerns; richer households less pessimistic in advanced economies.
- Synthesis:
  - Economic anxiety amplified early in crisis among economically vulnerable: unemployed, non-tele-workable workers, and those experiencing illness/job loss.

### Policy preferences
- Baseline patterns:
  - No statistically significant differences across gender and income in baseline for preferences on government support and interventions.
  - Majority expect government to play leading role on financial relief, help small businesses, manage spread, and prep for reopening.
- Determinants:
  - Age: older people favor financial support for households and small businesses relative to young.
  - Personally affected: those who lost jobs show no significant increase in support for government; those with adverse health impacts show a negative and statistically significant effect (less supportive).
  - Employment: unemployed favor government preparing for reopening; small-firm employees support financial relief; tele-workable workers support government role in managing crisis and liquidity support.
  - Urban: urban residents are less supportive of government interventions in baseline, though this reverses in emerging markets and high-stringency countries.
    - Gelbach (2016) decomposition: being young reduces the negative urban effect on provision of financial support by 10 percent; tele-workability weakens it by 5 percent; much of urban-rural divide remains unexplained.
  - Income: higher household incomes more likely to support social safety nets and liquidity support; upper two quartiles and middle two quartiles show significant positive effects.
  - Perception of government: individuals with more favorable assessment of government performance expect government to take leading role in liquidity support, addressing health crisis, and preparing for reopening.
- Robustness:
  - Results robust to Gelbach (2016) decomposition, probit specifications, restricting sample to advanced economies, excluding early-peak Asian countries, and dropping China/Saudi Arabia where applicable.
  - Alternative survey items for economic and health assessments and for policy preferences yield consistent results.

---

### IV. Conclusion

- Perceptions of government handling of the pandemic depend on gender, age, region, employment, socio-economic status, and recent adverse experiences.
- Economically vulnerable and those hardest hit view government responses less favorably; personal illness or job loss heightens pessimism about prolonged job losses and automation risk.
- Policies that directly reduce economic hardship and anxiety (e.g., direct transfers, social safety nets) are important to contain pandemic fallout and to build support for redistribution.
- Preferences for redistribution vary by personal experience and views about the poor; favorable assessments of government performance are associated with greater willingness to support financial relief for individuals and firms.
- Limitations: single time period early in crisis; framing of survey questions may affect results; longer panel datasets needed to track changes over time.

*Source: wpiea2021053-print-pdf*

### References

### wpiea2021053-print-pdf - References

### References
- Adams-Prassl, Abigail, Teodora Boneva, Marta Golin, and Christopher Rauh. 2020a. “Inequality in the Impact of the Coronavirus Shock: Evidence from Real Time Surveys.” IZA DP No. 13183.  
- Adams-Prassl, Abigail, Teodora Boneva, Marta Golin, and Christopher Rauh. 2020b. "The Impact of the Coronavirus Lockdown on Mental Health: Evidence from the US," Working Papers 2020-030, Human Capital and Economic Opportunity Working Group.  
- Algan, Yann, P. Cahuc and M. Sangnier. 2015. “Trust and the Welfare State: The Twin Peaks Curve.” The Economic Journal 126(593): 861-883.  
- Algan, Y., G. Sergei, S., Papaioannou and E. Passari. 2017. “The European Trust Crisis and the Rise of Populism.” Brookings Papers on Economic Activity: 309-382.  
- Alesina, Alberto, and Eliana La Ferrara. 2000. “Participation in Heterogeneous Communities.” Quarterly Journal of Economics 115: 847-904.  
- Alesina, Alberto and Giuliano, Paola. 2011. “Preferences for Redistribution.” Handbook of Social Economics, A Bisin and Benhabib, J, 93-132. North Holland.  
- Alesina, Alberto, Stefanie Stantcheva, and Edoardo Teso. 2018. “Intergenerational Mobility and Preferences for Redistribution.” American Economic Review 108(2): 521 – 554.  
- Alon, Titan M, Matthias Doepke, Jane Olmstead-Rumsey, and Michele Tertilt. 2020. “The Impact of COVID-19 on Gender Equality.” Working Paper 26947. National Bureau of Economic Research.  
- Bartik, Alexander W., Marianne Bertrand, Zoë B. Cullen, Edward L. Glaeser, Michael Luca, and Christopher T. Stanton. 2020. “How Are Small Businesses Adjusting to COVID19? Early Evidence from a Survey.” NBER Working Paper 26989.  
- Binder, Carola. 2020. “Coronavirus Fears and Macroeconomic Expectations.” Review of Economics and Statistics: forthcoming.  
- Brodeur, A., Grigoryeva, I., & Kattan, L. 2020. “Stay-at-Home Orders, Social Distancing and Trust.” IZA Discussion Paper No. 13234. Institute of Labor Economics.  
- Brück, T., Ferguson, N., Justino, P., & Stojetz, W. 2020. “Trust in the Time of Corona.” IZA Discussion Paper No. 13386. Institute of Labor Economics.  
- Brussevich, Maria Era Dabla-Norris and Salma Khalid. 2020. “Who will Bear the Brunt of Lockdown Policies? Evidence from Tele-workability Measures Across Countries.” IMF Working Paper WP/20/88.  
- Bui, T. T. M., Button, P., & Picciotti, E. G. 2020. “Early Evidence on the Impact of COVID-19 and the Recession on Older Workers.” NBER Working Paper No. 27448. National Bureau of  
- Christian, Cornelius, Lukas Hensel, and Christopher Roth. 2019. “Income Shocks and Suicides: Causal Evidence From Indonesia.” Review of Economics and Statistics 101(5): 905–920.  
- Cowell, Frank and Erik Schokkaert. 2001. “Risk Perceptions and Distributional Judgments.” European Economic Review 45(4): 941-952.  
- Coibion, O, Y Gorodnichenko and M Weber. 2020. “The Cost of the COVID-19 Crisis: Lockdowns, Macroeconomic Expectations, and Consumer Spending.” BFI Working Paper 2020-60.  
- Era Dabla-Norris, Frederico Lima and Hibah Khan. "What Determines Social Distancing? Evidence from Advanced and Emerging Market Economies.” IMF Working Paper (forthcoming).  
- Daniele, Gianmarco, Andrea F.M. Martinangeli, Francesco Passarelli, Willem Sas, Lisa Windsteiger. 2020. “Wind of Change? Experimental Survey Evidence on the COVID-19 Shock and Socio-Political Attitudes in Europe.” CESifo Working Paper No. 8517.  
- Durante, Ruben, Louis Putterman and Joël J. Van der Weele. 2014. “Preferences for Redistribution and Perception of Fairness: An Experimental Study.” Journal of the European Economic Association, vol. 12(4), pages 1059-1086.  
- Fetzer, T, L Hensel, J Hermle and C Roth. 2020. “Coronavirus Perceptions and Economic Anxiety.” VoxEU.org, 21 March.  
- Gelbach, Johah. 2016. “When Do Covariates Matter? And Which Ones, and How Much?” Journal of Labor Economics 34(2): 509 – 543.  
- Giuliano, Paola and Spilimbergo. Antonio. 2014. “Growing Up in a Recession.” The Review of Economic Studies 81(2): 787–817.  
- Gozgor, Giray. 2020. “Global Evidence on the Determinants of Public Trust in Governments During the COVID-19.” CESifo Working Paper No. 8313.  
- Hanspal, Tobin, Annika Weber and Johannes Wohlfart. 2020. “Exposure to the COVID-19 Stock Market Crash and its Effect on Household Expectations.” CEBI Working Paper Series, No. 13/20.  
- Haushofer, Johannes and Jeremy Shapiro. 2016. “The Short-term Impact of Unconditional Cash Transfers to the Poor: Experimental Evidence from Kenya.” The Quarterly Journal of Economics 131(4): 1973–2042.  
- International Monetary Fund (IMF). 2020. “Global Prospects and Policies.” World Economic Outlook Chapter 1. October 2020. International Monetary Fund, Washington, DC.  
- Kuziemko, Ilyana, Michael I. Norton, Emmanuel Saez and Stefanie Stantcheva. 2015. “How Elastic Are Preferences for Redistribution? Evidence from Randomized Survey Experiments.” American Economic Review 105(4): 1478 – 1508.  
- Margalit, Yotam. 2013. “Explaining Social Policy Preferences: Evidence from the Great Recession.” American Political Science Review 107(1): 80-103.  
- Klemm, Alexander, and Paolo Mauro. 2021. “Pandemics and Progressivity.” IMF Working Paper (forthcoming).  
- Mongey, Simon, Laura Pilossoph, and Alex Wienberg. 2020. “Which Workers Bear the Burden of Social Distancing Policies?.” CEPR Covid Economics 12.  
- Sands, Melissa. 2017. “Exposure to Inequality Affects Support for Redistribution.” PNAS first published January 9, 2017; https://doi.org/10.1073/pnas.1615010113  
- Stevenson, Betsy and Justin Wolfers. 2011. “Trust in Public Institutions over the Business Cycle.” American Economic Review 101(3):281–87.

### Annex — Tables and Notes
- Table 1: Summary Statistics of Individual Covariates by Country  
  - Note: Mean/Median for the sample weighted by the survey weights.
- Table 2: Assessment of Government Performance – Baseline  
  - Note: Results from Equation (1) where the dependent variable is an indicator that equals 1 for favorable assessment of the government performance (at least 4 on the scale of 1-5 in the survey). Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Country-level sample/population statistics excerpt (as presented):  
  - United States 0.49 0.48 473862566643240 0.09 0.15 0.59 0.44  
  - Canada 0.50 0.49 484158439614000 0.11 0.13 0.42 0.54  
  - China 0.51 0.53 4538159900406580 0.02 0.06 0.46 0.17  
  - Japan 0.49 0.49 5248546181756020000 0.03 0.02 0.49 0.48  
  - South Korea 0.50 0.50 484442184810452809340 0.04 0.04 0.67 0.45  
  - India 0.52 0.52 37275148522352750 0.03 0.24 0.82 0.11  
  - United Kingdom 0.49 0.49 484127635308000 0.09 0.04 0.42 0.42  
  - Germany 0.49 0.49 514629290279020 0.04 0.04 0.31 0.27  
  - France 0.48 0.48 494223378217760 0.07 0.09 0.45 0.32  
  - Mexico 0.49 0.48 38291468772352260 0.08 0.05 0.70 0.19  
  - Saudi Arabia 0.58 0.55 353073877936900 0.07 0.06 0.53 0.22  
  - Variables listed: Unemployed College-educated Male Median age Median Household income (in local currency)
- Table 3: Assessment of Government Performance by Personally Affected  
  - Note: Results from Equation (1) where the dependent variable is an indicator that equals 1 for favorable assessment of the government performance. Individual controls include gender, age, urban, education, employment status and household income dummies. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 4: Assessment of Government Performance – Advanced vs Emerging Market Economies  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for favorable assessment of the government performance and interaction term 퐸푀 is a dummy that equals 1 for emerging economics and 0 otherwise. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 5: Assessment of Government Performance by Stringency of Containment Measures  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for favorable assessment of the government performance and interaction term Stringent is a dummy that equals 1 for above the median stringency index in OxCGRT. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 6: Assessment of Government Performance by Timing of COVID-19 Spread  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for favorable assessment of the government performance and interaction term Down is a dummy that equals 1 for countries that saw an initial surge followed by a downward trend in COVID-19 cases at the time of survey. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 7: Assessment of Government Performance by Initial Level of Inequality  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for favorable assessment of the government performance and interaction term Gini is a dummy that equals 1 for above median inequality by Gini Index. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 8: Beliefs about Social and Economic Concerns– Baseline  
  - Note: Results from Equation (1) where the dependent variable is an indicator that equals 1 for strong concerns about socio-economic problems (at least 7 on the scale of 1-9 in the survey). Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 9: Beliefs about Social and Economic Concerns–Effect of Being Personally Affected  
  - Note: Results from Equation (1) where the dependent variable is an indicator that equals 1 for strong concerns about social problems. Individual controls include gender, age, urban, education, employment status and household income dummies. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 10: Beliefs about Social and Economic Concerns– Advanced vs Emerging Market Economies  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for strong concerns about social problems and interaction term EM is a dummy that equals 1 for emerging economics and 0 otherwise. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 11: Beliefs about Social and Economic Concerns by Stringency of Containment Measures  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for strong concerns about social problems and interaction term Stringent is a dummy that equals 1 for above the median stringency index in OxCGRT. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 12: Beliefs about Social and Economic Concerns by Timing of COVID-19 Spread  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for strong concerns about social problems and interaction term Down is a dummy that equals 1 for countries that saw an initial surge followed by a downward trend in COVID-19 cases at the time of survey. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 13: Beliefs about Social and Economic Concerns by Initial Level of Inequality  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for strong concern about social problems and interaction term Gini is a dummy that equals 1 for above median inequality as measured by the Gini Index. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 14: Policy Preferences, Baseline  
  - Note: Results from Equation (1) where the dependent variable is an indicator that equals 1 for expectation for government playing a leading role in the specified policy measures. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 15: Policy Preferences by Personally Affected Individuals  
  - Note: Results from Equation (1) where the dependent variable is an indicator that equals 1 for expectation for government playing a leading role in the specified policy measures. Individual controls include gender, age, urban, education, employment status and household income dummies. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 16: Policy preferences by Assessment of Government Performance  
  - Note: Results from Equation (1) where the dependent variable is an indicator that equals 1 for expectation for government playing a leading role in the specified policy measures. Overall performance is an indicator that equals 1 for favorable assessment of overall government performance during COVID-19 crisis. Individual controls include gender, age, urban, education, employment status and household income dummies. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 17: Policy Preferences, Advanced vs Emerging Market Economies  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for expectation for government playing a leading role in the specified policy measures and interaction term 퐸푀 is a dummy that equals 1 for emerging economics and 0 otherwise. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 18: Policy Preferences by Stringency of Containment Measures  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for expectation for government playing a leading role in the specified policy measures and interaction term Stringent is a dummy that equals 1 for above the median stringency index in OxCGRT. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 19: Policy Preferences by Timing of COVID-19 Spread  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for expectation for government playing a leading role in the specified policy measures and interaction term Down is a dummy that equals 1 for countries that saw an initial surge followed by a downward trend in COVID-19 cases at the time of survey. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 20: Policy Preferences by Initial Level of Inequality  
  - Note: Results from Equation (2) where the dependent variable is an indicator that equals 1 for expectation for government playing a leading role in the specified policy measures and interaction term Gini is a dummy that equals 1 for above median inequality as measured by the Gini Index. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1

### Robustness Checks (selected)
- Table 21: Assessment of Government Performance – Probit model  
  - Note: Probit regressions results from Equation (1) where the dependent variable is an indicator that equals 1 for favorable assessment of the government performance (at least 4 on the scale of 1-5 in the survey). Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 22: Beliefs about Social and Economic Concerns– Probit model  
  - Note: Probit regression results from Equation (1) where the dependent variable is an indicator that equals 1 for strong concerns about socio-economic problems (at least 7 on the scale of 1-9 in the survey). Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 23: Policy Preferences, Probit model  
  - Note: Probit regression results from Equation (1) where the dependent variable is an indicator that equals 1 for expectation for government playing a leading role in the specified policy measures. Errors are clustered at the country-level and regressions are weighted by public weights provided in the survey. ***p<0.01, **p<0.05 and *p<0.1
- Table 24–29: Additional robustness samples and exclusions (only Advanced Economies; dropping Asian countries)  
  - Notes preserve the same regression specifications, clustering, weighting, and significance notation: ***p<0.01, **p<0.05 and *p<0.1

*Content based on the References and Annex sections of wpiea2021053-print-pdf*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021053-print-pdf.pdf_