State and local policy responses to COVID-19: Lessons from evidence
To learn more about the evidence below or to discuss how to apply evidence from J-PAL North America’s randomized evaluations in your context, contact Rohit Naimpally.
To download this note in pdf form, click here.
The COVID-19 pandemic has caused enormous health, social, and economic challenges. J-PAL North America has curated a set of randomized evaluations that tested specific policy options related to the social safety net, health, and education. Our goal is to provide actionable evidence to inform state and local leaders’ policy responses to challenges during the COVID-19 pandemic. Please note that we do not have research evidence specifically for responding to a pandemic. We generalize from existing evidence in different contexts to provide policy recommendations and considerations for state and local governments responding to this situation.
Via our State and Local Innovation Initiative, we are able to provide pro bono technical assistance to state, local, or tribal leaders who would like to tailor and apply evidence in their local context. Our team can help you analyze whether evidence is likely to be relevant in your context and provide implementation details from the original studies. Please contact Rohit Naimpally to learn more.
Increasing access to the social safety net
The federal government expanded social insurance programs in response to the COVID-19 pandemic, but expansion of these programs is not enough to ensure eligible individuals claim these benefits. State and local governments can implement evidence-based policies to ease barriers to access important social insurance programs like the Supplemental Nutrition Assistance Program (SNAP), Earned Income Tax Credit (EITC), and Temporary Assistance for Needy Families (TANF). This section provides lessons from evidence on increasing access and take-up of social benefit programs.
Provide salient information and reminders to increase the take-up of critical social safety net programs.
Many individuals may not access the critical social benefits, such as the EITC, that they are eligible for during this time. The 2019 tax filing deadline extension from April 15th to July 15th due to COVID-19 gives states an opportunity to encourage individuals who could be eligible for the EITC to file their taxes.
A randomized evaluation from Day Manoli and Saurabh Bhargava, in partnership with the US Internal Revenue Service tested the effectiveness of different messages to taxpayers who were eligible for, but did not claim, the Earned Income Tax Credit and found that targeted messages increased take-up of the EITC.
Social benefit programs like SNAP and TANF are also crucial for vulnerable families and individuals, particularly during economic downturns. Yet millions of SNAP-eligible households do not enroll and miss out on assistance that can help address food insecurity. A randomized evaluation from researchers Amy Finkelstein and Matt Notowididgo, in partnership with Benefits Data Trust, found that sending eligible households informational mailings nearly doubled SNAP enrollment, while combining these informational mailings with application assistance tripled enrollment.
It is worth noting that sometimes text messages and/or letters alone are not enough to increase take-up of social insurance programs and tax credits. A recent evaluation by the California Policy Lab found that text messages and letters alone were not enough to increase the take-up of the state and federal EITCs. If possible, states should simplify enrollment processes while also providing targeted outreach to increase take-up of social insurance and tax credits.
- Recommendation: State and local governments can implement mailings or text messages that provide relevant program information for those who may be eligible for benefits, provide assistance, and simplify application processes to increase the likelihood of eligible participants claiming benefits.
- For consideration: States can streamline enrollment for programs like SNAP, TANF, and Unemployment Insurance by implementing presumptive enrollment, or the ability to enroll those who appear eligible, rather than requiring a full application. This mechanism can ease administrative barriers and provide expedited support to those who need it.
For more on this topic, please see J-PAL North America Scientific Director Amy Finkelstein and J-PAL North America Work of the Future Initiative co-chair Matt Notowidigdo’s op-ed on easing access to the safety net, or contact Rohit Naimpally.
Increasing access to health insurance and care during a pandemic
Without a tested vaccine or identified cure, preventive measures and access to medical treatment are critical to combat the spread of COVID-19. Public health authorities looking to ensure residents are covered by health insurance and able to seek medical care can look to the following evidence-based recommendations for guidance.
Reducing barriers to information and enrollment can increase insurance take-up.
In response to the coronavirus pandemic, several states have extended or re-opened the enrollment period for health insurance on state health insurance exchanges. However, despite insurance being available, many eligible Americans remain uninsured. People who are uninsured will likely face barriers to accessing testing and treatment for COVID-19 symptoms. In an effort to mitigate the costs of seeking COVID-19 testing and treatment, the CARES Act expanded the number of tests and services insurers must cover at no cost. However, there are gaps in these protections that may expose patients to unexpected medical bills, and insurance can help mitigate these costs. Experts found that informational interventions improved take-up of plans.
- Recommendation: Letters advertising enrollment deadlines and personalized information such as projected subsidies and plan ratings can increase take-up of insurance, even among healthier individuals. Increased take-up of health insurance will not only allow more Americans to access and pay for healthcare, but also decrease insurance costs in the long-run. The evidence also suggests that simplifying enrollment processes—such as through reminder emails with easy links to enrollment or auto-enrollment—could increase enrollment, more than through subsidies alone.
- For consideration: Under the National Emergencies Act, states have been granted the option to expand Medicaid coverage and reduce barriers to enrollment by applying for Medicaid Disaster Relief State Plan Amendments. One strategy to streamline enrollment is to implement presumptive eligibility, thus allowing hospitals, clinics, and other entities to screen individuals for Medicaid eligibility, and to temporarily enroll those who appear eligible.
Simple nudges and reminders can induce people to take up Medicaid.
Lack of insurance coverage during a global pandemic will cause even greater financial strain on low-income populations. Despite access to free or significantly subsidized health insurance through the Medicaid program, many eligible individuals and families remain unenrolled. A study of low-cost improvements to outreach efforts to Medicaid-eligible residents found these nudges to be particularly effective among hard-to-reach vulnerable populations. It was especially beneficial if outreach efforts were coordinated for times of the month when low-income populations were likely to be responsive.
- Recommendation: Behaviorally-informed nudges with specific next steps, such as postcards, emails, and automated calls reminding people of upcoming deadlines, can increase enrollment in the Medicaid program. Given that many states have authorized shelter-in-place and other stay-at-home orders, letters and other forms of electronic or phone communication will be critical to remind residents of key deadlines and key steps for securing Medicaid coverage.
For more on this topic, please see J-PAL’s note on “Increasing Adherence to COVID-19 Guidelines: Lessons from Existing Evidence.” Policymakers interested in learning more on the evidence presented on health insurance take-up and access are encouraged to contact Hannah Reuter.
Improving virtual and online learning while schools are closed
The physical closing of schools is among the unprecedented COVID-19 disruptions across the United States. Many districts and schools have rapidly adapted by virtualizing learning for their students. It is important to note that the use of many education technology solutions to deliver lessons hinges upon students’ access to reliable internet and internet-enabled devices. While internet access alone is not sufficient to improve learning outcomes, it is in many cases necessary to utilize effective computer-assisted learning programs. If these students are unable to access this necessary technology, the achievement gap will likely widen.
In this section we describe evidence-based approaches that can support learning via education technology informed by a J-PAL North America review of 126 randomized evaluations examining the effectiveness of many different kinds of education technology interventions. Recommendations on tutoring programs are drawn from a forthcoming J-PAL review of 96 randomized evaluations of K-12 tutoring programs.
Educational software, or computer-assisted learning (CAL) programs, shows promise in improving learning outcomes, particularly in math.
Under shelter-in-place and other orders to stay home, students will likely receive classroom instruction and any supplemental exercises online. Effective CAL programs that have been rigorously evaluated share a few common features. Students can advance through exercises at their own pace, allowing them to work incrementally until they have mastered the material. Evidence suggests that simply watching videos is insufficient, no matter how stimulating the content. School leaders should provide educators with guidance on monitoring and rewarding progress for advancing through instructional videos.
- Recommendation: Educational software that adapts instruction to meet student needs or that offers timely feedback to students and reports data on student performance to teachers can support teacher instruction at this time. For example, ASSISTments, a free online math homework platform that offers students feedback as they solve problems, has been evaluated through multiple randomized evaluations and shown to have positive results even when students used the program for less than an average of ten minutes per night, three to four nights per week.
Online learning is more productive when platforms create interactive lessons and offer incentives to learn.
Students in online only courses tend to perform worse than students in in-person-only courses. Studies of Massive Open Online Courses (MOOCs) show that persistence is low and few students get past the first few lessons. It is possible that students taking online courses may struggle with the lack of accountability or miss out on motivating relationships with instructors and peers.
- Recommendation: Engage students in video and online learning by creating assignments that ask students to engage with material and share their reactions through online discussions to increase the likelihood that students process the material.
Tutoring programs have shown promise in improving learning outcomes for disadvantaged students.
Evidence-based education non-profits, including Saga Education, are now delivering their tutoring model virtually to support at-home learning. Saga Education partners with public school districts to provide personalized tutoring during the school day for students who are falling behind. Randomized evaluations of Saga’s model have shown that the program results in large academic gains for students. Saga is offering virtual tutoring services to students in New York City while schools are closed. While this online implementation has not been evaluated, our assessments of SAGA’s approach and evaluations of tutoring-style programs delivered virtually suggest that these supports could be valuable to students who are struggling academically.
- Recommendation: Tutoring programs offering two-to-one support can help students get back on track when schools reopen and will help fight the widening of the achievement gap that is likely to result from school closures.
- For consideration: Evaluations of summer youth employment programs have found these programs offer considerable promise to alleviate the social costs of youth violence. Several cities have been forced to cancel these programs due to COVID-19. While no formal evaluation has been done on the topic, districts might consider partnering with summer youth employment programs to offer youth jobs as virtual two-to-one tutors for younger students needing remedial support during the summer months. For more on summer youth employment programs, see J-PAL North America’s Policy Bulletin: Stopping a Bullet with A Summer Job.
Messaging platforms and behavioral nudges can support parents and students.
Evidence suggests that with short, actionable directions and suggestions for engaging activities, parents are better equipped to support learning at home. Evidence-based family engagement apps are a proven method of providing parents with these behavioral nudges. There is also evidence that suggests timely and specific reminders, like text message reminders about tasks required for matriculation can help students enroll in college at higher rates. Behavioral nudges delivered through text messages may be an inexpensive way for districts to support both parents and students during school closures.
- Recommendation: Districts can adapt learnings from behavioral science and develop timely, actionable text messages for parents to encourage and support their students while learning at home.
For more on this topic, see J-PAL North America Education Technology and Opportunity Initiative Co-Chair Phil Oreopoulos’ op-ed “Stumped by How to Best Serve Students with At-Home Learning? Follow the Evidence” or contact Kimberly Dadisman.
Resources for further reading
Increasing adherence to COVID-19 guidelines: Lessons from existing evidence (J-PAL): Resource compiled by the J-PAL Health Sector to provide recommendations for responding to the COVID-19 pandemic. This note summarizes evidence generated by J-PAL affiliates and is not a summary of all the rigorous evidence on the discussed topics.
How Data Governments Already Have Can Guide the Pandemic Response (Governing): Op-ed on how state and local governments can leverage data collected for operational purposes (administrative data) to help them deal with the economic devastation of the coronavirus.
Easing Access to the Safety Net, and Why We Need to Do It Now (Governing): Op-ed in Governing by J-PAL North America Scientific Directory Amy Finkelstein and J-PAL North America Work of the Future Initiative co-chair Matt Notowidigdo outlining how state and local governments can leverage administrative data to guide policy responses to COVID-19.
Challenges of Equitable Rapid Response Cash Payments (Econofact): Econofact brief by J-PAL affiliated researcher Lisa Gennetian about the need for infrastructure to ensure delivery of economic impact payments or other cash stimulus.
How You Can Protect Your Community, Not Just Your Own Health (The New York Times): Op-ed in the New York Times by J-PAL affiliated researchers Kate Baicker and Oeindrila Dube on lessons learned from the Ebola crisis on garnering trust in communities to spur behavioral change to manage the pandemic.
Recommendations for a Metropolitan COVID-19 Response (Johns Hopkins School of Public Health): Special report for city leaders on public health recommendations for managing the coronavirus pandemic in a metropolitan context.
COVID-19 Resources (Results for America): Suite of resources for state and local governments, including resources from Bloomberg Philanthropies and the National League of Cities.
Coronavirus: What You Need to Know (The National Governors Association): Suite of resources for governors and states including memos on public health interventions, the federal response to the pandemic, and links to the CDC and other public health authorities.