Read the paper here

As the Covid-19 coronavirus batters the global economy, United States policymakers are considering a range of responses to keep families afloat. In addition to targeted policies like payroll tax cuts, paid sick leave, and small business loans, a simple approach is gaining traction: sending checks directly to families. Democrats and Republicans in the House and Senate are joining prominent economists in calling for immediate cash transfers to ensure all Americans can meet basic needs in these dire times.

Many are gravitating specifically to universal basic income (UBI), which would provide the same amount to everyone. This is attractive today because restricting based on income would involve looking to tax returns for the year 2018, which might not represent families’ current finances, or 2019, many of which have not yet been filed. Relying on the IRS could also delay the delivery of this badly-needed money and require special treatment for nonfilers.

If we choose UBI, how large should it be? And who should be included?

To answer these questions, I focused on a simple objective: stopping the poverty rate from rising, as it did in the Great Recession. Nobody knows how much Covid-19 will shrink incomes throughout 2020, so I simulated a range of decreases from 1 to 20 percent. For each scenario, I calculated the UBI necessary to stabilize the poverty rate, depending on who is included, from everyone to only adult citizens.

The results, which can be found in my new paper, reveal that a modest UBI can stabilize poverty — if we include everyone. For example, a $1,000 annual UBI would more than offset a 10 percent fall in incomes — roughly the drop from 2007 to 2011. If we exclude children and noncitizens, however, many would continue to slip into poverty, and the checks would have to exceed $1,500 per recipient.

The higher amount required swamps the savings from sending fewer checks. Giving to everyone means sending 42 percent more checks, but excluding children and noncitizens requires each check to be 83 percent larger to stop the poverty rate from rising.

On a net basis, stabilizing poverty is 23 percent cheaper with a UBI that includes everyone, compared to one that’s limited to adult citizens. Including everyone is also cheaper than all other eligibility options I modeled, such as including all adults but only giving a half share to children.

These results also hold regardless of how low incomes fall; inclusive UBIs are consistently more cost-effective.

Many questions remain: For example, how do these results change if we learn that service workers are disproportionately harmed? How much of the gap could safety net programs fill, especially if they’re expanded? How do UBIs compare to alternative policies on a per-dollar basis?

Poverty stabilization is also far from the only outcome to consider. Paid sick leave might help to contain the infection; payroll tax cuts might keep people employed; small business loans might keep doors from shutting. But each of these is likely to be regressive compared to UBI, and to have a smaller anti-poverty effect. Keeping poverty from rising should be a high priority, but it shouldn’t be the only distributional goal when designing pandemic relief programs.

UBI is well positioned to help all people with the urgency required of this crisis. Cash can be distributed immediately, especially if it’s done broadly. Unlike in-kind benefits like food stamps, it can be used to pay any bills that arise. And as my new paper shows, UBI is a cost-effective tool to prevent a rise in poverty. To maximize the anti-poverty impact, we should send checks broadly, generously, and quickly.