key: cord-0934991-vc89cnur
authors: Topazian, Rachel J.; Hatton, C. Ross; Barry, Colleen L.; Levine, Adam S.; McGinty, Emma E.
title: Public support for U.S. social safety net policies throughout the COVID-19 pandemic
date: 2021-11-09
journal: Prev Med
DOI: 10.1016/j.ypmed.2021.106873
sha: 34a0c73ab0d2f414604ca94f47ac759e4aaa1abb
doc_id: 934991
cord_uid: vc89cnur

COVID-19 has stretched the U.S. social safety net and prompted federal legislation designed to ameliorate the pandemic's health and economic impacts. We surveyed a nationally representative cohort of 1222 U.S. adults in April 2020 and November 2020 to evaluate changes in public opinion about 11 social safety net policies and the role of government over the course of the pandemic. A majority of U.S. adults supported six policies at both time points, including policies guaranteeing two weeks of paid sick leave; enacting universal health insurance; increasing the federal minimum wage; and increasing government spending on construction projects, business tax credits, and employment education and training. From April to November 2020, public support was stable for nine of the 11 policies but declined nearly 10 percentage points for policies guaranteeing two weeks paid sick leave (from 76% support in April 2020 to 67% support in November 2020) and extending unemployment insurance benefits (51% to 42%). Declines in support for these two policies were concentrated among those with higher incomes, more education, in better health status, the employed, and those with health insurance. The share of respondents believing in a strong role of government also declined from 33% in April to 26% in November 2020 (p > 0.05). Despite these shifts, we observed consistent majority support for several policies enacted during the pandemic, including guaranteeing paid sick leave and business tax credits, as well as employment-related policies.

The COVID-19 pandemic has tested the limits of the U.S. social safety net, spurring a wave of federal policies designed to provide immediate relief, as well as ongoing discussion about long-term reforms. In March 2020, Congress passed the Families First Coronavirus Response Act (FFCRA), which expanded access to paid sick leave and unemployment benefits (Long, 2020) , and the Coronavirus Aid, Relief, and Economic Security (CARES) Act, which included increased unemployment benefits and payroll tax credits for businesses (Snell, 2021) . In a survey of a nationally representative sample of U.S. adults conducted by our research team in April 2020, we found majority support for several health and economic policies included in FFCRA and CARES and an increase in the share of adults who believe in a strong role of government compared to pre-pandemic (Barry et al., 2020) .

Although CARES and FFCRA received widespread public support (Jones, 2021) , there is little data on if and how the public's views towards social safety net policies evolved throughout the pandemic. Early support might reflect unusually high acceptance of government intervention during a crisis, whereas attitudes might differ as policies became politicized and the severity of health and economic impacts changed. Strong public support increases the likelihood of policy enactment, especially if it endures over time (Page and Shapiro, 1983 ), yet we lack longitudinal data on views towards social safety net policies during COVID-19. In this paper, we evaluate changes in public support for social safety net policies between April and November 2020 to examine implications for current and future policy discussions.

We fielded the longitudinal Johns Hopkins COVID-19 Public Health and Civic Life Survey in a nationally representative cohort of U.S. adults in three waves in 2020: April 7-23, July 7-22, and November 11-30. The survey was administered online by NORC using the AmeriSpeak® panel, which is designed to be representative of the U.S. adult population based on geography, income, and ethnicity. The panel is sourced from NORC's area probability sample and a U.S. Postal Service addressbased sample covering 97% of U.S. households (Dennis, 2021) . There were 1468 Wave 1 respondents (70.4% completion rate), 1337 Wave 2 respondents (91% completion rate), and 1222 Wave 3 respondents (92% completion rate). This paper reports data from the 1222 respondents who completed Waves 1 and 3. Appendix A shows comparisons of the study population to national statistics. Participants received small cash incentives to participate. This study was approved by the Johns Hopkins Bloomberg School of Public Health Institutional Review Board.

In Waves 1 and 3, respondents rated their support for 11 social safety net policies spanning three categories: health, income, and unemployment (Appendix B contains survey questions). Health policies included: two weeks paid sick leave; universal health insurance; single-payer health insurance; publicly funded paid sick leave; and publicly funded paid family leave. Income policies included increased federal minimum wage and universal basic income. Unemployment policies included increased government spending on construction projects, tax credits to businesses, employment education and training, and extension of unemployment benefits. Respondents rated support for publicly funded paid sick leave and family leave by indicating support for workers receiving pay from the federal government or the state government. Respondents rated support for all other policies using a 5-point Likert scale ranging from strongly favor to strongly oppose. We collapsed the scale into dichotomous measures ("strongly favor" and "somewhat favor" were coded as supporting, and "neither favor nor oppose", "somewhat oppose", and "strongly oppose" were coded as neutral or not supporting).

We obtained sociodemographic variables, including gender, race/ ethnicity, age, income, education, and political party through the NORC panel. We asked respondents about their income, health status, and insurance status in Wave 1. In July 2020, we asked respondents retrospectively about their employment status in March 2020.

Participants rated their trust in science using a 4-point scale (a lot, some, not much, or not at all) in Wave 1. In Waves 1 and 3 we asked participants to rate their views on the role of government ranging from 1 ("the government should do only the things necessary to provide the most basic government functions") to 5 ("governments should take active steps in every area it can to try and improve the lives of its citizens"). We compared our results to an identical question asked in a September 2019 Gallup survey (n = 1525) (Jones and Saad, 2019).

We conducted data analysis using Stata Version 16, using survey weights to calculate nationally representative estimates (Stata Statistical Software, 2019). We tested for differences in policy support using chisquare tests. We conducted multivariate logistic regression as a secondary analysis to examine predictors of stopping support for two weeks paid sick leave and extended unemployment benefits between Waves 1 and 3. (Appendix C).

A majority of U.S. adults supported six of the 11 social safety net policies in both April and November 2020 (Fig. 1) . Two health-related policies received majority support: two weeks paid sick leave (76% in April, 67% in November) and universal health insurance (60%, 54%). 58% and 55% of our sample supported increasing the federal minimum wage in April and November, respectively. Three unemployment-related policies were supported by a majority in both waves, including tax Notes: Those responding that they strongly or somewhat favor a policy were coded as supporting. Those responding neither favor nor oppose, somewhat oppose, and strongly oppose were coded as neutral or not supporting. Error bars indicate 95% confidence intervals. Asterisks indicate a statistically significant difference between Wave 1 and Wave 3. credits for businesses (67%, 65%), construction projects (71%, 72%), and employment education/training (68%, 67%). Support for most policies remained consistent from April to November. However, we observed statistically significant declines of nearly 10 percentage points for two policies: two weeks paid sick leave (p < 0.001) and extension of unemployment benefits (p = 0.001). A majority of adults continued to support two weeks paid sick leave in November (support fell from 76% to 67%) whereas support for extending unemployment benefits fell below 50% (51% to 42%).

We observed declines in support for two weeks paid sick leave across all subgroups except for those reporting fair or poor health (Table 1) . We observed statistically significant declines among multiple subgroups; the two groups with the largest declines were those with excellent health status and Republicans. There were declines among those with better health status (excellent: 77.8% to 61.5%, p = 0.006; very good: 73.1% to 64.1%, p = 0.022; good: 78.7% to 66.5%, p = 0.008) but support was consistent among those with fair/poor health status (76.8% to 78.9%). Support fell 16.1 percentage points among Republicans, from 60.2% to 44.1% (p = 0.002), and 8.1% among Democrats, from 91.9% to 83.8% (p = 0.024). Support was consistent among Independents (73.5% to 67.3%, p = 0.124). Support for extended unemployment benefits declined among all subgroups except Hispanic respondents. The two groups with the largest declines were Black and Republican respondents. Support among Black respondents fell 16.7 percentage points (73.4% to 56.7%, p = 0.026). Support declined among White respondents (49.8% to 38.8%, p < 0.001) but was consistent among Other, non-Hispanic respondents (53.3% to 39.7%, p = 0.224) and Hispanic respondents (39.8% to 45.9%, p = 0.436). There was a decline of 14.3 percentage points among Republican respondents (34.9% to 20.6%, p = 0.005). Support also fell among Independents (47.0% to 38.1%, p = 0.032) but there were no statistically significant changes among Democrats (68.8% to 63.4%).

The percentage of respondents believing that "the government should take active steps in every area it can to try and improve the lives of its citizens" fell from 33.4% to 26.4% (p = 0.106).

Our study documents majority public support for six social safety net policies in both April and November 2020, the early and mid-stages of Table 1 Characteristics of adults supporting two weeks paid sick leave and extended unemployment benefits, April and November 2020. Notes: *p < 0.05, **p < 0.01, ***p < 0.001 statistically significant difference from reference category (first row in category) within Wave 1 or Wave 3. The gender, race/ethnicity, age, household income, education, and party variables are baseline data collected in the NORC AmeriSpeak panel. The employment variable was collected retrospectively in Wave 2. The health status, insurance status, trust in science, and role of government variables were collected in Wave 1.

the pandemic in the U.S. Our findings suggest that most Americans supported several policies that federal policymakers are currently considering as part of post-COVID-19 economic recovery legislation. In April 2021, President Biden introduced the American Jobs Plan, a $2 trillion infrastructure proposal that would expand the social safety net alongside economic and infrastructure investments (Tankersley, 2021) . The bill and a companion $3.5 trillion budget proposal include policies similar to those supported by a majority of our sample, including construction projects and employment education/training (Memorandum for Democratic Senators -FY2022 Budget Resolution.pdf, 2021 Vella and Castronuovo, 2021) . While our findings indicate consistent and robust public support for these proposals, views on the role of government appear more fluid. Although the share of individuals supporting a strong role of government rose nearly 10 percentage points from September 2019 to April 2020 (24% to 33%) (Barry et al., 2020) , those gains were nearly erased by November 2020 (26%). This decline is not statistically significant, but the downward trend may reflect a return to pre-pandemic views. In this context, our finding of sustained support for six policies is all the more striking. However, if support for a strong role of government decreases, or as the pandemic's effects wane and policy priorities evolve, it may affect policy views.

Although paid sick leave and extended unemployment benefits were central elements of early pandemic legislation, we observed statistically significant declines for both policies. This could be driven by the declining support for strong government intervention, or reactions to the pandemic's shifting impact on health and the economy. The mandatory paid sick leave benefits included in FFCRA expired in July 2020, in between Waves 1 and 3 (Hill et al., 2021) . The extended unemployment benefits in the CARES Act also expired in July 2020 (Garner, 2021) and the unemployment rate, which peaked at 14.8% in April 2020, had fallen to 6.7% in November (Civilian unemployment rate, 2021). It is possible that these policies, which offered near-term relief, were viewed as more essential early on, but declined in perceived importance as the pandemic evolved. The nature of these declines varies: a large majority of respondents still supported two weeks paid sick leave in November, whereas support for extended unemployment dropped below 50%. However, declines in support for these policies were concentrated among individuals with advantageous health or economic status, whereas policy support remained comparatively stable among groups more likely to experience adverse health and economic impacts of the pandemic, including racial minorities, the unemployed, and those with poorer health.

Our findings also offer insight into ideological factors associated with declines in support for paid sick leave and unemployment benefits. We observed significant decreases in support for two weeks paid sick leave among Democrats and Republicans, and significant declines in support for extended unemployment benefits among Republicans and Independents. We also observed statistically significant declines in support for both policies among those who trust in science a lot or some, a surprising finding given that trust in science might motivate individuals to view COVID-19 as a serious threat that merits a policy response. While we found downward trends among those not trusting in science, the declines were not significant, potentially due to small sample size. Levels of support for both policies remain higher for those who trust in science compared to those who do not, as well as for Democrats and Independents compared to Republicans. While we lack exact comparisons for our survey questions pre-pandemic, this may reflect elevated support in the early days of the pandemic, which has since leveled out.

Our study has several limitations. Respondents' interpretation of survey items could vary given that several items asked about general policies such as "longer and higher benefits from unemployment insurance". We lack exact comparisons prior to the pandemic for the 11 policies, limiting our ability to assess the impact of COVID-19 compared to baseline views. The timing of our Wave 3 data collection, which occurred immediately following the polarized 2020 presidential election, may have affected levels of support, especially for policies that are highly politicized. Although the AmeriSpeak® panel uses probability-based recruitment aligning with best-practice survey research standards, results may be vulnerable to sampling biases. The AmeriSpeak® panel has a relatively low recruitment response rate (34%) but a high retention rate (85%), which may offset recruitment challenges. In addition, small sample size may limit our ability to detect statistically significant differences among certain subgroups (e.g. the unemployed) and limit the comparability of small subgroups across survey waves. However, design of our survey allows for comparisons within our panel and provides longitudinal data on views throughout the pandemic.

We found enduring public support throughout the COVID-19 pandemic for multiple policies designed to strengthen the U.S. social safety net and promote economic recovery. Declines in support for two weeks paid sick leave and extended unemployment benefits over this period may reflect the perceived need for immediate individual relief early in the pandemic, whereas the sustained support for other policies may reflect a perception that broader economic and infrastructure investments are needed to address the pandemic's long-term effects.

Public support for social safety-net policies for COVID-19 in the United States

Technical Overview of the AmeriSpeak Panel NORC's Probability-Based Household Panel

New COVID-19 Unemployment Benefits: Answering Common Questions | U.S. Department of Labor Blog

Congress Allows FFCRA Leave to Expire But Extends Tax Credits for COVID-19 Related Leave in Certain Circumstances. The National Law Review

Bipartisan Support for COVID-19 Rescue Legislation

Gallup poll social series: governance

The Families First Coronavirus Response Act: Summary of Key Provisions. KFF

Memorandum for Democratic Senators -FY2022 Budget Resolution.pdf. Published online August 9

Effects of Public Opinion on Policy

What's Inside The Senate's $2 Trillion Coronavirus Aid Package. NPR. org

Stata Statistical Software

Biden Details $2 Trillion Plan to Rebuild Infrastructure and Reshape the Economy. The New York Times

Pelosi says House working to pass infrastructure bills by Oct. 1. TheHill

The authors have no conflicts of interest to disclose. This work was supported by The Johns Hopkins Bloomberg School of Public Health and the Johns Hopkins University Alliance for a Healthier World's 2020 COVID-19 Launchpad Grant. These entities funded data collection, but were not involved in the design and conduct of the study; management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication. Rachel Topazian receives tuition and stipend support from a grant from the U.S. Centers for Disease Control and Prevention: National Institute for Occupational Safety and Health to the Johns Hopkins Education and Research Center for Occupational Safety and Health (award number T42 OH0008428). This entity was not involved in data collection; design and conduct of the study; management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ypmed.2021.106873.