There have been two big bankruptcy stories in the last week: clothing retailer J. Crew, which filed on Monday, and car-rental agency Hertz, which narrowly avoided the same fate by negotiating an 11th-hour reprieve from creditors.

It might be a little surprising that Hertz got the forbearance, since on paper, Hertz is arguably in a worse position than J. Crew. Two-thirds of the car-rental industry’s business comes from air travelers, and it seems safe to say that most people will be slipping back into chinos and espadrilles before they’ll be cramming themselves into airplanes again.

Car rental is a labor- and capital-intensive business; Hertz and its competitors buy a lot of Detroit’s output, and that generates hefty bills for financing, storage and maintenance. With such high fixed costs, even companies with less debt than Hertz are going to be hard-pressed to stay solvent until air travel recovers — that is, unless you expect a COVID-19 cure by September.

Yet it might not be an accident that the car-rental company was able to buy time for a restructuring while the clothing retailer was not. In the lingo of viral attacks, Hertz is a healthy patient experiencing an acute crisis, while J. Crew had some severe preexisting conditions, notably the steady decline of the malls that for so long formed the backbone of its business.

That distinction is fairly obvious to anyone who thinks about the two companies for a moment. What’s less obvious is what lenders and policymakers should do about it as we enter what’s really the second phase of coronavirus policymaking.

Two months ago it seemed possible, if barely, that we could sort of stop economic time for a bit while we fought the virus, freezing workers and firms in place, then starting up again as soon as we had the virus under control. That was essentially the logic of the first few rounds of bailouts: shovel cash onto balance sheets to prevent financial obligations from dragging companies under, keep workers attached to their employers and paper over the gaps with extra-generous unemployment benefits.

This was never sustainable for more than a few months; it is simply not realistic to keep 60% of the country inside drawing salaries or unemployment benefits while the other 40% does the hard and dangerous work of keeping the grocery stores stocked, the lights on and the hospitals open. But if we’d been more stringent about keeping everyone who wasn’t absolutely needed inside so that the caseloads fell as fast as possible, while building up the kind of aggressive testing/tracing/central quarantine regimes that we’ve seen work in Asia, we might have made people feel safe enough in crowded public spaces to reopen quickly and salvage much, if not all, of the economy.

Or perhaps that was always impossible, the dream of deluded technocrats. At any rate, it certainly didn’t actually happen.

After six weeks of social distancing, daily deaths from COVID-19 have leveled off.

Meanwhile, social distancing is breaking down at both the individual and political levels — heck, apparently, we liberty-loving Americans can’t even ask everyone to wear a mask while they go to the store. So as governors lift movement restrictions, we’re at high risk of new flare-ups. Yet our testing capacity is well below where it needs to be, never mind about the contact tracing or central quarantine parts.

If the disease roars back, then no matter what governors say, people will stay home and more companies will go under — too many to bail out. Which means that markets, lenders and government policymakers are all going to have to start drawing a line between the companies they can help and the ones they should let fail.

One way you could draw that line puts the companies that are doing practically no business at all on the “fail” side, while those that still experience some demand for their products get a lifeline. But that might suggest bailing out the mall retailer with a declining market niche over the travel company that has a bright post-coronavirus future. So it would be better to draw the line in a way that supports the companies that we’ll want and need 18 months from now, or whenever we get a vaccine, even if there’s nothing for them to do right now.

That better strategy still has two major problems, however. The first is that, in the short term, it could actually make the economic numbers worse, because the firms thus helped will be adding even less to national production and employment than the others might. The other, and far larger, problem is that if our public health efforts remain this underpowered, we probably won’t be able to sustain even fundamentally healthy firms long enough for them to reopen.

Follow Megan McArdle on Twitter, @asymmetricinfo.