Using new cross-country survey and experimental data, we investigate how beliefs about intergenerational mobility affect preferences for redistribution in France, Italy, Sweden, the United Kingdom, and the United States. Americans are more optimistic than Europeans about social mobility. Our randomized treatment shows pessimistic information about mobility and increases support for redistribution, mostly for “equality of opportunity” policies. We find strong political polarization. Left-wing respondents are more pessimistic about mobility: their preferences for redistribution are correlated with their mobility perceptions; and they support more redistribution after seeing pessimistic information. None of this is true for right-wing respondents, possibly because they see the government as a “problem” and not as the “solution.” (JEL D63, D72, H23, H24, J31, J62)

This paper presents a model of optimal labor income taxation where top incomes respond to marginal tax rates through three channels: (1) standard labor supply, (2) tax avoidance, (3) compensation bargaining. We derive the optimal top tax rate formula as a function of the three corresponding behavioral elasticities. The first elasticity (labor supply) is the sole real factor limiting optimal top tax rates. The optimal tax system should be designed to minimize the second elasticity (avoidance) through tax enforcement and tax neutrality across income forms. The optimal top tax rate increases with the third elasticity (bargaining) as bargaining efforts are zero-sum in aggregate. We provide evidence using cross-country times series macro-evidence and CEO pay micro-evidence. The macro-evidence from 18 OECD countries shows that there is a strong negative correlation between top tax rates and top 1% income shares since 1960, implying that the overall elasticity is large. However, top income share increases have not translated into higher economic growth. US CEO pay evidence shows that pay for luck is quantitatively more important when top tax rates are low. International CEO pay evidence shows that CEO pay is strongly negatively correlated with top tax rates even controlling for ˝rm characteristics and performance, and this correlation is stronger in firms with poor governance. These results are consistent with bargaining effects playing a role in the link between top incomes and top tax rates. If bargaining effects in fact exist, optimal tax rates should be higher than commonly assumed. The share of total pre-tax income accruing to upper income groups has increased sharply in the United States. The top percentile income share has more than doubled from less than 10% in the 1970s to over 20% in recent years (Piketty and Saez, 2003). This trend toward income concentration has taken place in a number of other countries, especially English speaking countries, but is much more modest in continental Europe or Japan (Atkinson, Piketty, Saez, 2011 andAlvaredo et al. 2011). At the same time, top tax rates on upper income earners have declined sharply in many OECD countries, again particularly in English speaking countries. ThomasWhile there have been many discussions both in the academic literature and the public debate about the causes of the surge in top incomes, there is not a fully compelling explanation.Most explanations can be classified into market driven changes vs. institution driven changes.The market driven stories posit that technological progress and globalization have been skilledbiased and have favored top earners relative to average earners (see e.g., Gabaix and Landier (2008) for CEOs and Rosen (1981) for Winner-Take-All theories for superstars). Those pure market explanations cannot account for the fact that top income shares have only increased modestly in a number of advanced countries (including Japan, Germany, or France) which are also subject to the same technological forces. The institution driv...

and Yale. We thank Pierfrancesco Mei, Davide Taglialatela, Raphael Raux and especially Leonardo D'Amico for outstanding research assistance. Alesina and Stantcheva are grateful to the Pershing Square Fund for Research on the Foundations of Human Behavior and the Wiener Center for generous support. Harvard IRB approval IRB17-1154. This study is registered in the AEA RCT Registry and the unique identifying number is: AEARCTR-0003342. The views expressed herein are those of the author and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peerreviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

IMPORTANCE Data from the coronavirus disease 2019 (COVID-19) pandemic in the US show large differences in hospitalizations and mortality across race and geography. However, there are limited data on health information, beliefs, and behaviors that might indicate different exposure to risk. OBJECTIVE To determine the association of sociodemographic characteristics with reported incidence, knowledge, and behavior regarding COVID-19 among US adults. DESIGN, SETTING, AND PARTICIPANTS A US national survey study was conducted from March 29 to April 13, 2020, to measure differences in knowledge, beliefs, and behavior about COVID-19. The survey oversampled COVID-19 hotspot areas. The survey was conducted electronically. The criteria for inclusion were age 18 years or older and residence in the US. Data analysis was performed in April 2020. MAIN OUTCOMES AND MEASURES The main outcomes were incidence, knowledge, and behaviors related to COVID-19 as measured by survey response. RESULTS The survey included 5198 individuals (mean [SD] age, 48 [18] years; 2336 men [45%]; 3759 white [72%], 830 [16%] African American, and 609 [12%] Hispanic). The largest differences in COVID-19-related knowledge and behaviors were associated with race/ethnicity, sex, and age, with African American participants, men, and people younger than 55 years showing less knowledge than other groups. African American respondents were 3.5 percentage points (95% CI, 1.5 to 5.5 percentage points; P = .001) more likely than white respondents to report being infected with COVID-19, as were men compared with women (3.2 percentage points; 95% CI, 2.0 to 4.4 percentage points; P < .001). Knowing someone who tested positive for COVID-19 was more common among African American respondents (7.2 percentage points; 95% CI, 3.4 to 10.9 percentage points; P < .001), people younger than 30 years (11.6 percentage points; 95% CI, 7.5 to 15.7 percentage points; P < .001), and people with higher incomes (coefficient on earning Ն$100 000, 12.3 percentage points; 95% CI, 8.7 to 15.8 percentage points; P < .001). Knowledge of potential fomite spread was lower among African American respondents (−9.4 percentage points; 95% CI, −13.1 to −5.7 percentage points; P < .001), Hispanic respondents (−4.8 percentage points; 95% CI, −8.9 to −0.77 percentage points; P = .02), and people younger than 30 years (−10.3 percentage points; 95% CI, −14.1 to −6.5 percentage points; P < .001). Similar gaps were found with respect to knowledge of COVID-19 symptoms and preventive behaviors. CONCLUSIONS AND RELEVANCE In this survey study of US adults, there were gaps in reported incidence of COVID-19 and knowledge regarding its spread and symptoms and social distancing behavior. More effort is needed to increase accurate information and encourage appropriate Key Points Question How do reported incidence, knowledge, and behaviors regarding coronavirus disease 2019 vary across sociodemographic characteristics in the US? Findings In this survey study, the largest differences in coronavirus disease 2019-rela...