In the early morning hours of Wednesday, March 25, negotiators in the Senate landed on a $2.2 trillion deal to rescue the economy as it falls into recession (with 3.3 million new jobless claims in one week) amid the coronavirus pandemic. In an almost unheard-of display of unity on a major bill, the measure passed 96-0.
It’s a flawed package, but overall a shockingly ambitious measure from a Republican legislature — and one that can and should be made stronger in the coming weeks as the country sinks deeper into recession.
To think through this stimulus, it helps to think back to the last economic crisis and the stimulus we got then.
In late 2008, as the US was spiraling further into its worst economic downturn since the Great Depression, Christina Romer, the UC Berkeley economist whom President-elect Barack Obama had selected as his chief economist, prepared a memo of stimulus options for Obama to review.
To fully repair the economy by the first quarter of 2011, she estimated, would require a “combination of spending, taxes and transfers to states and localities … costing about $1.8 trillion over two years.”
The memo did not make it to Obama. According to Noam Scheiber’s book The Escape Artists, when Romer showed her estimate to fellow Obama adviser Larry Summers, he dismissed the number as too high. She then prepared a memo where the most expensive option was $1.2 trillion. Summers still thought it too high, telling her, per Scheiber, “$1.2 trillion is nonplanetary.” That was his way of saying “ridiculous” or “out of this world.” Congress eventually passed, and Obama signed, a $787 billion package.
Summers wanted the $1.8 trillion in stimulus as much as Romer did — he just thought Congress would reject it as a joke, and prevent the Obama administration from getting anything. This was his calculation when Democrats were set to have 59 out of 100 Senate seats and a massive House majority.
Fast-forward nearly 12 years. The Senate has just passed (and the House is poised to follow suit) a stimulus package costing a whopping $2.2 trillion. And the spending is set to occur within a single year, not spread over two as Romer envisioned in 2008.
So given that background, how should we think about the stimulus emerging from Congress? If it’s compared to the most ambitious stimulus and recovery packages proposed in the runup to its passage, or to the most ambitious ones passed elsewhere in the world, then it leaves a lot to be desired.
Its cash checks are a one-off, with no automatic payments to kick in if and when unemployment spikes. The increase in unemployment benefits is similarly temporary and not tied to objective economic metrics that could keep them from being withdrawn early. The benefits to businesses are opaque and might be inferior to just having the government temporarily take over payroll for struggling companies, the way Denmark has done as its crisis response. There’s too little aid to state and local governments to forestall budget cuts, and no mandate for postal voting in the 2020 election to be sure it’s conducted safely if the country is still under quarantine.
But how does the bill stack up to the last stimulus? It’s definitely an improvement. Even granting that the coronavirus recession is likely to be deeper than the 2008-’09 downturn, the huge increase in scale is an incredibly promising development.
Something about this crisis turned nonplanetary ideas into planetary ones. There are some decades where nothing happens, and some weeks where decades happen. We are living through many such weeks right now, right in a row, and the result is a political environment that would have been unrecognizable mere months ago.
The important thing is that Congress has taken advantage of this rapid ideological movement and passed this bill. Now it needs to pass more like it in the weeks and months to come.
What the stimulus package gets right
The key question that Romer faced in putting together a stimulus package in 2008 was how big the “output gap” was: How much economic activity could exist if the economy were firing on all cylinders but didn’t because of the recession?
That’s one of the major questions macroeconomists have to answer, because it responds directly to recessions like the one that hit in 2008. That recession, like many, was essentially a crisis of expectations. There were just as many workers, just as many factories, just as many retail storefronts, etc. in 2008 as there had been a year earlier, when unemployment was low and the economy was growing.
The “real resources” of the economy hadn’t diminished at all. But the amount of work being done by those workers, factories, stores, etc. suddenly dropped dramatically. It became a vicious self-perpetuating cycle. Businesses didn’t want to hire or invest because they were worried they wouldn’t have enough customers. Customers were nervous to spend because times were so tight, and because so many were unemployed.
The key to the 2008-’09 stimulus was to generate a surge of spending (“demand” in econ parlance) that would enable businesses to start hiring and investing again, so that the economy could return to where it was just a couple of years previous.
The 2020 recession is not like that. The real resources of the economy have diminished, dramatically. Many workers, like bartenders or cooks or waitstaff or retail personnel, are physically barred from working. Physical locations — factories, stores, etc. — that were productive in January are not being staffed in March by government fiat. Workers who can work from home are, but many others can’t because their work isn’t suitable for telecommuting.
In that situation, the near-term goal is not to stimulate economic activity per se. It’s to ensure that absolutely essential real resources — like food, shelter, and medical care — are being adequately provided, and that households have the resources they need to pay for those necessities.
The ”phase three” (following phases one and two) stimulus package that the Senate passed Wednesday night doesn’t do as much as I’d like toward reaching that goal. But it does an awful lot, and frankly more than I ever expected a Republican president and Senate would accept.
The @BudgetHawks have THE summary of the $2.2 trillion just-passed CARES Act. Read it. Use it. Retweet it. Ask Questions. https://t.co/EfPR0PVvPh pic.twitter.com/R68bN79WPD— Marc Goldwein (@MarcGoldwein) March 26, 2020
The most headline-grabbing thing it does is offer $1,200 per adult and $500 per child dependent to most households as a blank check, phasing out at high incomes (above $75,000 for singles and $150,000 for married couples). That’s an immediate influx of money that can pay bills for low-income people, including people with no incomes. I’ve covered cash policy for a long time, and the mere fact that a Republican Congress accepted payments to Americans with no taxable income, the poorest of the poor, is remarkable to me.
More important, though, is the UI section: four months of unemployment insurance of at least $600 per week to everyone put out of work due to coronavirus. That means UI that actually exceeds the wages of people paid $15 an hour or less. Plus, the benefit is available to part-time and contract workers. This is a cash stimulus measure, too, but one limited to people directly disemployed by this crisis. It’s a sensible way to target assistance to the hardest-hit and make sure they can afford food and shelter. The provision is so generous that four GOP senators — Tim Scott (SC), Rick Scott (FL), Ben Sasse (NE), and Lindsey Graham (SC) — objected on the grounds that it’s too generous, threatening to derail the bill. (In the end, all four voted for the bill.)
The most expensive parts of the bill are its sections “bailing out” various businesses by offering $366 billion in loans to small businesses and $504 billion in loans to larger businesses, states, and local governments. It’s hard to know what these programs will do without seeing them implemented. But in the event that Treasury Secretary Steven Mnuchin enforces the provisions barring layoffs for companies benefiting, this could serve as an adjunct to the UI expansion, enabling businesses to avoid payroll reductions and providing more money, overall, to workers than the government will directly.
Finally there’s the question of paying for health care. The bill offers $100 billion to hospitals and $150 billion in emergency aid to state and local governments, the latter premised on being used for coronavirus (indicating it will be used for, say, Medicaid before paying teacher salaries). I don’t know if this will be enough to cover uncompensated care for the uninsured during this crisis, or if this is the best way to structure such aid. But it’s a significant investment.
How the stimulus package could be improved
Overall, the scale and generosity of the package is far above what I expected from a Senate led by Mitch McConnell, who tried his hardest to sink the much smaller stimulus of around $800 billion that Obama wound up proposing in early 2009.
But it could still be improved, and instead of recessing, the Senate should be working on ways to improve it.
The biggest problem is that the package is too short-term. The UI provisions expire in four months. The loans are a one-off package. The state and local aid doesn’t re-up despite states like Ohio saying they’re facing budget cuts of 20 percent. The cash checks only go out once, and not as quickly as they could (ideally through EBT cards, or Social Security checks to existing recipients, for people who didn’t file 2019 or 2018 taxes because they lacked sufficient income). Indeed, some tax filers with no direct deposit information could be waiting months for their checks.
A logical fix to the package is to apply a Sahm rule to it all. Named after Claudia Sahm, an economist at the Washington Center for Equitable Growth who has established herself as one of the most essential voices on coronavirus response, a Sahm rule is a policy where stimulus measures are executed automatically on the basis of objective economic data, not congressional discretion. As my colleague Matt Yglesias explains, the Sahm rule reflects the fact that “historically, when the unemployment rate over the past three months rises at least 0.5 percentage points above the average over the past 12 months, a recession is imminent.” You can see this in action in the data here.
Sahm’s original idea was that the government should just send out checks to every American when this indicator kicks in. We’re past that point in the coronavirus crisis: We know we need these benefits to kick in. What we need instead is to adapt Sahm’s idea so that we extend all or most of the benefits in the congressional stimulus package until the economy is recovered.
Here’s one way to do a Sahm rule: We keep sending out $2,000 checks, financed by the Federal Reserve, for each adult and child (inspired by California Rep. Maxine Waters’s plan) every month until the employment to population ratio for working-age people reaches 80.5 percent again, its level when this crisis started. Once it hits that level, the checks can start phasing out with each 0.1 point increase in the ratio.
You could do the same with aid to state and local governments. Start by offering 50 percent of their 2019 state revenue as a federal grant, taking advantage of the ultra-low interest rates the federal government currently enjoys. Phase down the grant amounts as we approach full employment again. Make the process automatic, reflecting actual economic data, not Congress’s whims. And, while you’re at it, push the Fed to buy state and local debt.
Unemployment insurance is a little different. Paying more than 100 percent of lost wages for low-income workers makes a lot of sense right now when we’re trying to keep people out of unnecessary work for public health reasons. But it may deter people from reentering the job market once the public health emergency is over. So in the medium-run, an automatic, Sahm rule UI program would be pegged to, say, 100 percent of wages up to a certain limit, rather than a flat $600 amount, until employment is back to normal, so there’s still an economic benefit to jumping back into the labor force.
Sen. Michael Bennet (D-CO) has a bill already that would work a bit like this. It would also push states to set up “short-term compensation” or work-sharing programs, which allow companies to use UI funds to keep workers on payroll. That way people can stay attached to their work through this crisis while still getting paid.
We don’t need to pass all these measures immediately. But Congress should pass them this spring so we have automatic policies in place going into the election season, and into the what could be the worst of the pandemic this summer. We cannot rely on legislators to always mobilize aggressively. We need automatic policies that get support to people whether Congress gets its act together or not. And paradoxically, the only way to do that is to have Congress pass automatic stabilizer legislation, and to pass it soon. Just because a $2.2 trillion bill has been passed doesn’t mean Congress’s work is done.
Sign up for the Future Perfect newsletter and we’ll send you a roundup of ideas and solutions for tackling the world’s biggest challenges — and how to get better at doing good.
Future Perfect is funded in part by individual contributions, grants, and sponsorships. Learn more here.