What to make of the labor shortage talk

What to make of the labor shortage talk

By Matthew Yglesias. SlowBoring.com. 

April 20, 2021. 

A 'Now Hiring' sign hangs in front of a Popeye's restaurant on February 04, 2021 in Miami, Florida

(Photo by Joe Raedle/Getty Images)

In “The Coming Unemployment Insurance Bottleneck,” I warned that the impressive success of the vaccine rollout was likely to lead to a spring surge in demand for hospitality industry workers who might not want to come back to work since the Rescue Plan extends $300/week bonus Unemployment Insurance through the beginning of September.

It’s good to be ahead of the curve, but in that case, I think I was a little too far ahead of the curve since the piece didn’t generate much discussion. Yet now two weeks later, talk of a UI bottleneck is everywhere, with the anecdotes pouring in from annoyed business owners across the country.

Conservative elites are also increasingly talking about this — see Henry Olsen’s column in The Washington Post for a representative example — and critics of the Biden Rescue Plan see it as vindication of their views.

Since I predicted this bottleneck would become a big deal, I’m half-inclined to claim prescience and take a victory lap. The truth, though, is that the evidence for a strong UI effect is just not that compelling at this point. I like a good hot take as much as the next guy, but my main take on this is pretty lukewarm — the data just isn’t there yet, in part because the data reports with a lag. I think I have a policy fix that’s a good idea whether or not this is a big problem, so Congress ought to implement the fix. But is it really a big problem? I think it’s not clear.

The basic issue

As you may recall back in March of 2020, congressional Democrats were anticipating widespread layoffs as businesses shut down due to the pandemic and they wanted to put something into the CARES Act to support workers. Many foreign countries adopted a system where the government paid companies to keep workers on their payrolls even if they were not working. Congress decided to let non-working workers be unemployed, and simply take America’s threadbare Unemployment Insurance system and make it more generous.

The most natural thing to do would have been to boost the replacement rate such that UI benefits would equal 100% of a given person’s normal wages. Or if you want to preserve the principle that work should be more lucrative than non-work, you could have boosted the replacement rate to 95%.

However, states’ creaky UI systems apparently could not do that. So instead, congressional Democrats pushed to boost everyone’s UI checks by $600 per month — which would mean that the median worker was at a 100% replacement rate. But it turned out that the workers hardest hit by layoffs were much more downscale than the overall national workforce, so the CARES Act created a situation where the typical unemployed person was earning a higher income unemployed than employed. This was then offset by a bunch of administrative difficulties in accessing benefits and by the fact that the program, though very generous, arbitrarily expired in late summer.

Now thanks to the rescue plan, Bonus UI is back — albeit at only $300/week — but as Peter Ganong from the University of Chicago explained to me, there are still many people for whom UI is more generous than their old jobs. That’s particularly true because, for at least some workers, the COBRA provisions of the law make it more like a $400/week supplement.

two graphs showing the replacement rate with $300 and $400 supplements

Now you may dimly recall that there was a lot of hubbub about this last year when it was $600/week, but that no problem materialized. You probably read articles that cited Ioana Elena Marinescu, Daphné Skandalis, and Daniel Zhao such as “Job Search, Job Posting and Unemployment Insurance During the COVID-19 Crisis,” which demonstrated there was no disemployment impact of the CARES Act. And if $600/week didn’t cause disemployment, then why would $300/week?

Well, there’s a nuance here. What Marinescu et. al. found was that the bonus UI did reduce job search intensity, but that the number of open positions fell so much faster that there was no reduction in successful matches or job creation.

Today the situation is different and things are opening up, not shutting down, so there at least might be a disemployment effect — even though the bonus amount is smaller.

Employers say they are having trouble

The news is now full of anecdotes of restaurant owners, in particular, saying they want to add staff for the great reopening but are having trouble hiring people.

Here’s a bunch of anecdotes from Rochester, NY.

And another bunch from Syracuse, NY.

Here’s a story about a McDonald’s in Florida offering people $50 just to fill out an application.

In Boise, you can even hear an Idaho Department of Labor official make the case instead of relying on a local business owner.

Now the problem is that there are something like 14,000 McDonald’s franchises in America, so a super-rare problem impacting 0.1% of them could still generate hundreds of anecdotes. That’s especially true because this kind of thing can copycat around the country: someone in Syracuse reads a story about Rochester and then decides to do his own story. And of course, people struggling with recruiting workers are going to be inclined to blame the government rather than their own shortcomings.

I asked Nick Bunker — the Economic Research Director for North America at the Indeed Hiring Lab — what he makes of this, and he’s skeptical.

“We don’t have any indication it’s an impediment,” he tells me. “I could see it maybe slowing things down a bit later in the summer once things are more open.”

One issue this raises is just the question of what exactly we’re talking about. The economy added about a million jobs in March. If we see a big slowdown in April, I’d be comfortable looking at that and saying “this is probably the impact of the Bonus UI.” But are we going to see a big slowdown? I don’t actually see many people forecasting that. Instead, they are saying in a kind of nonspecific way that job growth is just going to happen more slowly than it otherwise might have happened. Which could be true! But it’s hard to say without a more clear hypothesis.

Why not just pay more?

Now one thing that inevitably comes up in this discourse is the “why don’t you try paying more” rebuttal to any labor shortage concern.

And in general, I agree. But if you want to understand what’s happening in the economy right now, I don’t think it’s that hard to see why you’d rather not hand out a permanent increase in wages in response to a bonus UI program that’s going to expire in September.

I also think you can see the impact of the labor market situation on business decisions. A contact in the burrito industry advised me that they’d looked at opening up indoor dining at the Chipotle near my house but decided against it. Basically, you’d need more staff to keep the tables clean, and it’s just not worth it right now.

It’s striking that Chick-fil-A is keeping their dining rooms closed for now. That seems like a smart decision to me — they have a solid drive-through business; it does continue to be the safer choice; and it’ll be a lot easier to add staff in September when bonus UI expires than it is right now.

A lot of restaurants and other businesses closed during the pandemic. There’s a Peet’s Coffee near my house that closed its doors about nine months ago. My assumption had long been that it would either reopen as a Peet’s or else as some other coffee place in due time. But if owners of existing businesses are struggling even a teensy tiny little bit to recruit workers, that would probably put you off launching a new restaurant right now. Especially because you know the UI situation is going to look different in a few months.

Public health vs. the economy, yet again

Another thing to say about this is it’s not always totally clear to me where the disagreement between contending forces lies.

I spoke at the beginning of the week to some Democrats who work on UI issues on Capitol Hill, and their point of view on this is that just because Republican Party politicians got over COVID-19 months ago doesn’t change the fact that normal people are wary of risking their health to cook you dinner.

The conservative point of view is that this is correct — UI is making it possible for people to stay out of the workforce, and that’s slowing the Great Reopening.

Which in turn is just to say that we’re not really arguing about labor market economics at all; we’re just continuing the argument that started in April 2020 over whether the policy emphasis should be to make economic life as-normal-as-possible given the pandemic or to do everything that’s economically viable in order to slow the spread of the virus.

In a weird way, I think everyone ended up being mostly wrong about this. The public health measures that Democratic officials have been inclined to implement — mostly school closures and limitations on indoor dining — do not appear to have been particularly effective at halting the spread of the virus. But the economic harms that Republican officials warned of have overwhelmingly not materialized — it’s been very affordable for the government to borrow lots of money and be more generous to unemployed people, and then send the rest of the not-rich population a stimulus check.

Strictly as a position, though, this is where more data will clarify things down the road. On the liberal interpretation that fear of the virus is a dominant factor, we should expect job growth to pick up as a result of the vaccine rollout — especially in May and June, by which time anyone interested should have had ample time to get their shots. If we add significantly more than a million jobs in May, I think that’s a big win for the liberal view that labor shortages are all about health fears. If we add significantly fewer, I think that’s a big win for the conservative view that labor shortages are all about bonus UI being too generous. I keep thinking about whether this is something I’d put a bet on and I think my honest answer is…no, I wouldn’t.

That’s especially because I know the labor market economics people are arguing about this with a different, larger argument in the back of their heads.

The future of Unemployment Insurance

The COVID-19 recession has been really unusual on a number of levels. But during the Great Recession years, one thing I learned is that one of the biggest disagreements between the Democratic Party’s economic advisors and the GOP’s is their view of Unemployment Insurance.

To the Dem wonks, UI is one of the very best and most effective forms of fiscal stimulus because unemployed people tend to spend every dollar you give them on just keeping their basic lives together. So during the long, slow recovery, the Democrats frequently fought to keep UI benefits flowing for longer. In part that’s a humanitarian issue — jobs were very scarce and people needed a way to get by. But it also reflected a very sincere conviction that a dollar spent on UI is a very sound investment in stimulating the economy.

The GOP wonks see it exactly the other way. Their view is that when the economy is flat on its back, the last thing in the world that you need is a negative shock to the labor supply side.

Here I think Democrats have the much better side of the argument. And a pretty exciting proposal from Ron Wyden and Michael Bennet came out earlier this month designed to make the UI program better for the long term. The headline feature is that they want to raise the minimum replacement level to 75% of ordinary wages and make extended duration of state UI an automatic thing that happens based on either the state or the country's economic conditions. Their proposal would also extend UI coverage to a broader set of workers, avoiding the current practice of excluding part-timers in most states.

They also introduce the idea of a “jobseekers’ allowance” that would provide financial support to people like new graduates and caregivers returning to the workforce after a long absence.

To me, the black fly in the chardonnay here is they don’t go all the way toward simply having the federal government run the program. The entire thrust of these reforms is to reduce state discretion over the design, duration, and generosity of benefits while also increasing the federal role in supervising the technological infrastructure. It would be easier to accomplish all that by centralizing the whole thing under the federal government.

I’m also just not conceptually sold on joint state/federal programs. I tend to think there are a few areas — notably transportation projects — where the federal government’s role should just recede. Then there are other areas like UI and Medicaid that ought to be federalized. When Elizabeth Pancotti was on The Weeds, she explained to me that the joint nature of UI was originally meant to address some potential constitutional objections that the Roosevelt administration was worried about, but that’s not relevant today. The way the Wyden-Bennet proposal works is that UI duration would be a function, in part, of state-level labor market conditions. That kind of differentiation makes sense, but it doesn’t require administrative decentralization.

Now the Democratic aides I’ve talked to about this say that it would be a really heavy lift to draft the legislative frameworks and create the bureaucracy to have a federally administered UI system. And I don’t disagree with that. If you think there’s a realistic chance that a major UI expansion will happen in the relatively near future, this is an easier path to getting that done. But frankly, I’m skeptical. The theory is that after some version of the American Jobs Plan passes, there’s going to be a third major Biden initiative — perhaps called the American Families Plan — that will address child care and other welfare state issues. UI expansion could then be part of that plan.

Whether that’s at all plausible is sort of between God and Joe Manchin at this point, but I’m skeptical. I think we’re in pie in the sky territory here, so it’s actually a good time to try to think through all the details of a federalized UI program.

Meanwhile, for the short term, my pious hope is that Congress should actually fix this problem rather than yelling at each other.

Turn FPUC into a re-employment bonus

Here is some real talk for various stakeholders:

Real talk for Republicans — you’re not going to secure any political advantage by hitting Biden on the economy during a period of time when everyone agrees the economy will be rapidly improving, so you may as well do something helpful.

Real talk for business owners — no matter how much you whine, there is zero chance that Congress is going to go cut FPUC payments.

Real talk for Democrats — the Great Reopening is happening thanks to vaccines and there’s really no point in maintaining the pretense that we need to “pay people to stay home” at this point.

And everyone should note that while the million jobs added in March sounds impressive, we need about a million jobs per month every month for over a year to get back to the pre-pandemic trend.

graph predicting when payrolls will return to pre pandemic trends

So what’s the solution? It’s simple — Congress should pass a law saying that if you have X weeks of Federal Pandemic Unemployment Compensation eligibility left and you get a job, then you get X times $300 from Uncle Sam as a re-employment bonus.

People who can’t find work will still get their FPUC money. But people who can find work won’t be financially penalized for saying yes to a job sooner rather than later. And you don’t need to think of it exclusively in incentive terms either. The re-employment bonus can help you address the barriers (child care, etc.) to work that may have arisen over the past year.

There is no downside at all to doing this, so small business owners ought to call their members of Congress and tell them to do it. And then the members ought to do it. As I’ve said throughout this piece, I think it’s very hard to tell how big of a problem the UI bottleneck really is. But in this case, I really don’t think we need to decide. There’s a solution available that’s robust to different accounts of what’s happening, and that’s the kind of solution you want.