Credit: Getty Images/iStockphoto/Kameleon007

With on-again, off-again plans for a COVID relief bill currently in "off" mode, it's an opportune moment to step back and ask some basic questions about fiscal policy. Fiscal policy during a pandemic is fundamentally different from during a typical financial crisis, and that should shape the federal government's response. Short-term aid would be better before rather than after the U.S. election, but it needs to be targeted in the right way.

The first classic argument for fiscal stimulus, dating from Keynes, is to increase aggregate demand. That argument doesn't quite apply right now. Movie theaters, airlines and restaurants do face big demand problems, but government spending can't fix that. What these businesses need is for their customers to feel safe against the risk of infection.

A second rationale for fiscal policy is to address problems of supply. The U.S. still has an uneven capacity for rapid COVID-19 testing, and more money for testing would help carry it through what is likely to be a difficult winter. Better testing also would help schools reopen, allowing more parents to get back to work and to higher levels of productivity.

In addition to immediate issues of humanitarian aid, and supporting essential services from state and local governments, the next question is what other problems possible legislation might beneficially address.

The greatest potential problem facing the U.S. economy right now is a loss of organizational capital, most of all in small businesses. As they fold, jobs will be lost and the knowledge and efficiencies embedded in those businesses will disappear. In many cases it will be difficult to reconstitute that knowledge quickly, and so the markets for both labor and goods will be operating at less than full capacity.

Consider performing arts groups in New York City, many of which took a long time to assemble and train their talent. If they go under, many of these people will move away, if only because they and their employers have run out of money. They will opt for alternate career paths, and the arts groups may never be reconstituted. Supply-side grants and loans, however, may help see them through until America approaches some semblance of normalcy again.

That argument makes good sense, but note its limitations. The government need not give further aid to relatively well-capitalized larger businesses. Nor should it subsidize businesses that are not coming back soon no matter what. If Americans are going to remain unwilling for the next several years to trap themselves on cruise ships, that sector should be allowed to shrink radically.

When targeting aid to small businesses, government officials should think about which ones have built up special or hard-to-replicate organizational capital. If most of the labor is relatively unskilled, it's an easier case to make to let those businesses go under — and then see new, reconstituted versions arise in six months or a year from now. Unlike in, say, the performing arts, in the service sector it is easy to find suitable new workers, and so the returns to government assistance are lower.

Most likely the hardest hit areas in the U.S., in economic terms, will be San Francisco and New York City. New York is characterized by extreme population density, and the performing arts and tourism are major parts of the economy. San Francisco and the Bay Area rely on high-paying technology jobs, many of which have now turned into remote work. The case for aid to New York City is relatively strong, whereas the case for San Francisco depends on whether you think that remote work will last. If it will, and those tech jobs stay away from California forever, then the prospect of a permanently downsized San Francisco needs to be seriously considered, as well as which new sectors or industries may help it grow.

If you are very optimistic about vaccines and anti-covid therapeutics, as I am, this strengthens the case for short-term aid. If you believe that more or less normal times for many small businesses will arrive by April or May, that argues for tiding them over in the meantime.

If you are more pessimistic, however, and expect a high level of COVID-19 cases for the next few years, then the argument for aid is weaker. Those endangered small businesses cannot be kept on "life support" for that long; better for them to know that aid is not forthcoming. The economy would then continue to adjust into fewer restaurant jobs and more food-delivery jobs.

The two real arguments for fiscal policy in the U.S. right now are the need for more and better COVID-19 testing, and the gains from acting quickly to preserve America's organizational capital. It remains to be seen whether our politics can quickly remedy these two problems. Nevertheless, any proposals should be judged on how close they come.

Cowen is a Bloomberg Opinion columnist. He is a professor of economics at George Mason University and writes for the blog Marginal Revolution. His books include "Big Business: A Love Letter to an American Anti-Hero."