No one knows if the government can save the economy

The coronavirus returns us to the still-unsettled stimulus debate of the late 2000s

It has begun to dawn on lawmakers that the COVID-19 outbreak is very, very serious. They are now looking to contain the virus’s spread, and to counteract the economic damage it could wreak if countless Americans are asked to stop going to work, minimize their shopping trips, avoid public events and keep their kids home from school.

Some relief measures are indisputably called for, even beyond medical help. If people are staying away from work on account of the virus, the government needs to keep them afloat financially. And yes, if industries go bankrupt because of the virus, and not because of their own mismanagement, bailouts make a lot of sense.

But what about the broader, longer-term measures under consideration, such as eliminating payroll taxes for the rest of the year? Can the government save an economy that will otherwise fall ill by injecting money into it this way? That’s an old debate, and one that has never really been resolved. We saw it play out most recently in the late 2000s, when two different presidents enacted stimulus measures in response to the Great Recession: George W. Bush sent everyone checks in 2008, and Barack Obama let loose a massive flood of government spending the next year. (Remember Solyndra and “shovelready projects”?)

The conservative line at the time, especially once Mr. Obama was in charge, was that stimulus doesn’t do much of anything except drive up our debt. There’s a solid case for this view in economic theory, which Brian Riedl makes ably and in great detail elsewhere on this site. The gist of it is that in order to put money into the economy, the government first has to borrow money out from the economy, under the false assumption that the state can find better uses for it than its private owners can. A common analogy is that stimulus is like taking water out of one end of a pool and dumping it into the other.

Most economists believe stimulus can do something good, though. Mr. Riedl himself notes estimates holding that the government can get at least some bang for its buck this way, even if boosting the economy by a dollar requires borrowing and spending more than that. And in a 2014 survey of economists, the overwhelming majority thought Mr. Obama’s stimulus resulted in a lower unemployment rate than otherwise would have been obtained.

The theory here traces its roots to the liberal economist John Maynard Keynes. The idea, basically, is that when the economy is moving slowly, the government can speed things up by taking money that isn’t doing much and putting it to productive use, such as by funding infrastructure improvements or simply giving it to people who are very likely to spend it. If there’s ever a time to do this, the COVID-19 crisis might be a good one because the government can borrow at incredibly low interest rates at the moment.

One problem is that even if the Keynesians are right, we don’t have too clear a picture of how much stimulus delivers how much growth — or, for that matter, how much growth we’ll need to make up once the virus takes its toll. The key statistic is the

“fiscal multiplier,” with a value of one meaning that each dollar in government spending boosts the economy by a dollar. And there’s little consensus on where exactly the multiplier lies or how it varies depending on economic conditions; estimates vary by a factor of two or more. There’s also no consensus on where the outbreak is headed. Similar uncertainty reigned the last time around, when Obama-administration economists predicted that unemployment would peak at 8% with the stimulus and 9% without it, only for unemployment to reach about 10% with it.

But if we accept the idea that stimulus works, we do have a decent sense of how to set it up so that it gets spent as desired. Checks in the mail seem better than tax cuts, being spent about a quarter of the time instead of an eighth, but the money from both is most frequently saved or used to pay off debt. (Checks in the mail may also be better politically; recall that Republicans got little credit for the 2017 tax law in part because it boosted each paycheck slightly rather than resulting in bigger refunds.) Assistance to low-income households and infrastructure investments also seem to create more jobs than do grants to states for higher education. In general, poorer people are more likely to spend an additional dollar, so targeting them with aid is more likely to increase consumer spending.

Uncertainties aside, though, none of this bodes all that well for a temporary payroll-tax cut. Such a measure would slightly boost paychecks instead of giving people a windfall to spend, give more money to the rich than to the poor, neglect the retired and unemployed, and of course fail to target the people most affected by COVID-19. Whatever we do, this shouldn’t be it.

In the near term the government needs to help people directly affected by the pandemic. In the longer term lawmakers probably won’t be able to resist further stimulus. But what they’ll get for it, no one knows.