This is Going to Suck, Part 2 (the Economic Impact of COVID-19)

Ten days ago, I wrote an initial analysis of the coronavirus pandemic.

At the time, there were about 350,000 total cases, 15,000 deaths, and 100,000 instances of recovery around the world.

That was March 23. As I write this on the evening of April 2, (ten days later) the world has:

• 1,015,000 total cases (190% increase)
• 53,000 deaths (253% increase)
• 212,000 recoveries (112% increase)

The worst aspects of these data points is that deaths continue to increase at a faster rate than either total cases or recoveries, and recoveries are increasing at the slowest rate of the three major statistics.

One small cause for hope, however, is that the rate of new infections appears to be slowing, at least globally. Domestically (in the U.S.), it might be a different story.

My last article looked closely at statistics from the United States, which at the time had the third-most cases in the world (35,075). Since March 23, U.S. infections have surged far past those of any other country. According to Worldometers, the U.S. currently has:

• 245,066 total cases (up 599% in 10 days)
• 6,075 deaths (up 1,226% in 10 days)

This represents an average daily increase of 22% in the number of cases diagnosed in the U.S., and an average daily increase of 30% in the number of deaths in the U.S. attributable to COVID-19. Here's what this looks like on a standard daily chart:

This is the classic hockey-stick exponential curve, which is exactly the sort of situation policymakers and healthcare professionals have been trying to avoid. To better understand how rapidly the caseload has grown, we should look at the chart on a log scale:

A logarithmic scale gives us a better sense of how rapidly the caseload is growing by compressing the chart into a nonlinear format. Each mark on this chart represents a 10x increase over the previous mark.

Exponential growth, when graphed on a log scale, looks like a steady increase, but it's not.

In late February, there were fewer than 100 cases; in early March, fewer than 1,000; by mid-March, over 10,000; by late March, over 100,000.

What we want to see is a significant flattening or downturn before the line hits the one-million mark. I don't think we will for another month or so. This is an estimate, like my three-to-six-week estimate earlier, and I'm not a medical professional, so I could be quite wrong. I just hope I'm not wrong on the low end.

Many of us now know someone affected by COVID-19, or worse, someone who's died as a result of COVID-19. Yesterday, my mom told me that one of her cousins, whom I last saw at my grandmother's funeral in 2017, died of COVID-19 last weekend in New York City. No one knew he was infected until after he died.

I believe the number of deaths will continue to grow faster than the number of infections as the recently deceased are increasingly tested and found to be coronavirus carriers, for at least another three to six weeks. This speaks to the failure of pre-emptive testing efforts in the U.S. as much as to the risks of COVID-19 itself.

If current statistics are accurate, COVID-19 now has a mortality rate of 2.5% in the United States, nearly double the 1.3% mortality rate I recorded on March 23. The global mortality rate has risen from 4.4% to 5.2% since March 23.

Here's the log scale of COVID-19 deaths in the U.S.:

You can see that the log-scale chart of COVID-19 deaths is rising at an even faster rate than the total-case chart. This is common in the early stages of exponential log-scale charts, and I hope it flattens out over time. As I've already said, though, I expect deaths to outpace total cases for at least three to six more weeks.

There are three possible explanations for this spike in mortality, none of which are particularly reassuring. The first explanation is that it's simply taken several weeks for many of the infected to begin dying from COVID-19. The second explanation is that we initially underestimated the disease's mortality. The third explanation is that we're beginning to see large numbers of people dying -- like my mom's cousin -- of an unknown coronavirus infection, only diagnosed after death.

My early estimates, based on the infection rate and the then-current mortality rate of 1.3%, projected about 330,600 deaths, if the growth in total infections slows to 10% per day. Since then, the growth in infections has averaged 16.4% per day, although it has dropped to a growth rate of 12% to 14% over the past five days.

Based on this new data, we could see anywhere between 14.6 million and 16.9 million total infections by the end of April. A 1.3% mortality rate would result in anywhere from 189,000 to 220,000 deaths. A 2.5% mortality rate would result in anywhere from 365,000 to 423,000 deaths.

My high-end mortality estimates are above those of most other models, which range from 36,000 to over 240,000 American deaths to COVID-19 in the current pandemic. I hope they remain so, but we've yet to see the impact of social distancing and quarantine measures, most of which were only implemented on a state-by-state within the past two weeks. However, my low-end mortality estimates now line up closely with the high-end estimates coming out of the White House.

If anyone in the United States doubted the importance of social distancing and other hygiene measures two weeks ago, the rapid growth in cases and deaths, and the apparent increase in mortality rate, should convince them otherwise.

We are still early on the infection curve here. That's a terrifying prospect. The hope is that our draconian shutdown of the economy and enforced social-distancing measures will flatten the curve, reduce infections and deaths, and eventually allow us to bring the coronavirus pandemic under some degree of control.

However, there's another terrifying prospect for those of us who'll survive...

What will our economy look like once we finally get this coronavirus under control?

Emergency-braking the economy

Early estimates looked for about 2.4 million new U.S. jobless claims last week.

Instead, we had over 3.3 million new jobless claims.

Today, we learned there were 6.6 million new jobless claims this week.

In two weeks, nearly 10 million people have filed for unemployment. That's 6% of the entire workforce, which works out to an unemployment rate of about 9.5%.

To put that in context, the unemployment rate peaked at 10% in 2009, and has only surpassed 10% once before during the postwar era, in late 1982 and early 1983.

It took the U.S. economy two and a half years to deteriorate from its prior low in unemployment (4.4%, reached in May 2007) to its November 2009 unemployment peak of 10%. We're now on track to roar up from a 3.5% unemployment rate, as recorded in February, to (and probably above) a 10% unemployment rate, in roughly two months.

If you thought the two-plus years of the Great Recession sucked, imagine how much it's going to suck when all that pain is compressed into two months.

This has never happened before.

What's happening today amounts to an emergency-braking of the national (and global) economy in a desperate effort to control a rapidly-spreading deadly pandemic.

Everyone knows that this is not normal.

However, many people may be too optimistic about the outcome of these emergency-braking measures. The government is throwing out trillions of dollars to backstop the economy, and few people expect the pandemic lockdown to last for a very long time. As a result, there's a fairly broad consensus (at least among white-collar types with the luxury of sitting at home and analyzing these numbers on their computers) of a quick return to normal.

I'm here to disabuse you of that notion. I don't think it's going to happen.

The death of the economic body

The economy is not like a machine that can be turned off and left idle until we need it again. It won't roar back in the same strong condition it was before.

The economy is more like a human body, with money as its lifeblood.

When you stop the flow of blood through the body, things start going catastrophically wrong very quickly, causing irreversible damage everywhere.

Without blood, your brain dies within four to six minutes. However, it only takes about three minutes after blood flow stops for brain damage to begin. Other organs can survive longer without new blood, but that means nothing if there's no functioning brain to control those organs once they start up again.

This analogy comes from Talking Points Memo's Josh Marshall. It's potent, powerful, and I believe it's entirely accurate. Keep it in mind as you read through the rest of this article.

The economy is likely to remain frozen until the end of April, and possibly until the end of May or even into June. We don't know how long we'll all be stuck inside, because the U.S. federal government has done such an abysmal job thus far of controlling the pandemic.

A late-April restart is highly optimistic. It assumes transmission has already ground to a virtual halt, which it clearly hasn't based on the rapid rise in diagnoses and deaths. A late-May restart risks reoccurrence of the pandemic, which incubate in the body for a long time asymptomatically and is both airborne and long-lived on hard surfaces.

We may have to shelter in place for months until new cases drop towards zero and we no longer endure new COVID-19 deaths on a daily basis. We may also wind up enduring repeated lockdowns if movement restrictions are eased too rapidly.

Most projections I've seen expect the economy to suffer a short, sharp drop through the summer and recover by year-end. I believe this is optimistic, and so does McKinsey:

The A4 scenario is the one most people currently hope for. A1 through A3 represent other "less bad" outcomes. All of them involve the economy returning to a state more or less resembling normal by the end of the year. B1 through B5 represent worse alternatives, which we'd all like to avoid if at all possible.

Unfortunately, the U.S. doesn't appear to have a rapid and effective control over the spread of the coronavirus. At best, given the federal government's incompetence and indifference, and the importance of individualism and independence to the American character, we're more likely to see moderately effective healthcare responses, but recurring pamdemics. I don't believe we'll see broad public-health failures. I certainly hope we don't.

Effective policy interventions, in this instance, also assume that we'll more or less go back to normal after those interventions run their course. I don't believe this will be the case, and most of the rest of this article will explain why I think this is too optimistic. Some blame for the failure of policy interventions must lie with federal and state governments. However, there are a large number of unknown variables in the economy for which "business as usual" may not be possible after a multi-month shutdown. Pumping blood through our economic body in the summer won't bring it back to life after three months of brain death.

I believe scenarios B2 and A1 to be the most likely outcomes.

Much of the country is basically stuck inside indefinitely for the time being. Combine this with the fact that so many have already been furloughed or laid off -- and millions more are soon likely to join them -- and we face a problem on two fronts.

First: how many businesses can survive the coming months without steady revenues?

Second: how many laid-off employees will be able to return to work once pandemic restrictions end and the economy eventually does begin returning to normal?

Who's paying the mortgage?

Let's start with the largest sector of the economy: real estate.

When most of us think about real estate, as it relates to this pandemic, we're probably thinking, "how am I going to pay my mortgage or rent next month?"

But there's someone on the other side of that transaction.

The real estate rental and leasing industry contributes about $2.9 trillion to the U.S. economy. The finance and insurance industries -- including mortgage servicers overseeing roughly $15.8 trillion in loans -- contribute another $1.6 trillion.

Altogether, about 8.9 million people work in the "finance, insurance, and real estate" sector.

Now, let's talk about the $2.2 trillion pandemic stimulus bill.

Most people know it explicitly excludes the cruise industry from bailout money. Many people may know the stimulus allows homeowners to delay mortgage payments on federally-backed loans for up to a year. The bill does not offer any similar relief for mortgage servicers.

The Politico article linked in the prior paragraph explains why that's bad:

When individuals stop making payments on their home mortgages, the companies that handle the loans... mortgage servicers, are still on the hook: They're legally obligated to keep sending money to insurers and investors in mortgage-backed securities, the giant bundles of home loans that are packaged and sold on the securities markets.

A few months of missed payments by a quarter of borrowers could cost mortgage servicers $40 billion. A two-month shortfall, given the rapid rise in cases and deaths, seems optimistic to me. Unfortunately, the government only has some kind of plan in place for about two months, after which "there's going to need to be a bigger solution," according to FHFA Director Mark Calabria.

The economy was stronger at the start of 2020 than it was before the Great Recession. But if you don't have any income, you can't pay your mortgage, regardless of how strong your finances were three months ago.

There were about 6.4 million foreclosures during the Great Recession. Keep in mind that the same number of people lost their jobs during the entire two and a half years of the Great Recession as just lost their jobs this past month.

I certainly hope we don't see as many foreclosures in 2020 as we saw from 2007 through 2009. Provisions in the stimulus bill will help keep the number down. But we could very well see hundreds of billions of dollars worth of mortgage delinquencies across the country before this pandemic plays out.

Multifamily property owners may be in the worst shape. Some complexes are already experiencing rent strikes. People who own multiple properties as a source of income -- particularly if these are vacation or short-term Airbnb-style rentals -- could see an exodus of renters and guests as people attempt to consolidate their finances and minimize travel.

The last crash was primarily driven by mortgage foreclosures. In this crash, the foreclosures could just be a secondary impact of the main problem: the end of consumer demand.

Choking the travel industry

The clearest consequence of lockdown society can be seen right outside your window, especially if you live in a city.

Streets are practically empty. Shoppers aren't shopping. Services aren't provided, food isn't made, stuff isn't sold, and every business that depends on foot traffic of any sort is being choked of its financial lifeblood.

What you can't see in the sky right now is another indicator of collapse.

This is a chart of air traffic in the EU. As you can see, it's down to the bare necessities. Air travel in the U.S. isn't faring much better. Domestic airlines like American have cut up to 80% of their flights through May, and the flights still operating are doing so in a way that's made American's crew uncomfortable over the risk of infection.

Travel and tourism will be devastated for the rest of the year, and possibly for far longer. The psychological impact of the pandemic may linger in public consciousness at least as long as it takes to develop an effective coronavirus vaccine, which could take up to two years. Given the resources pouring into vaccine development, I'd be surprised if it took two years, but a shot in the arm can't counteract months of enforced isolation right away.

Airlines will be supported as "economically necessary" businesses, but many travel-related businesses, such as hotels, event promoters, and vacation-booking companies, won't have similar protections. Millions of travel and tourism jobs could be lost permanently.

Worldwide, up to 75 million jobs, and $2.1 trillion in economic output, could be lost in the travel and tourism industry alone.

In the U.S., travel and tourism accounts for 7.8 million jobs and $600 billion in GDP.

How long can we keep doing this?

Discussing the risk to every industry one by one would make this article many pages longer, but as I already said, you can see the impact right outside your window.

It's not hard to imagine many thousands, and potentially millions, of businesses simply shuttering by the time we're allowed to venture out as normal once again.

Some will fail because their business models couldn't deal with the dramatic shifts in consumer behavior we may see once the pandemic passes. Many more will fail because they've been badly managed, unable to evolve, or simply too dependent on a daily flow of consumer money to survive very long in a shutdown environment.

Even if we can prevent the worst of these failure scenarios, and a relatively small number of businesses go under, we'll still have millions of furloughed or unemployed people in need of work. Some of them can go back to their old jobs. Many won't be able to, won't want to, or simply won't need to because they've found new employment.

The longer the shutdown drags on, the less likely it'll be for businesses surviving on skeleton crews or stimulus funds to easily restaff, and that brings with it a host of other problems.

Turnover costs a lot, and any business that has to replace laid-off workers to regain a "normal" state of operations will deal with some form of turnover.

There were 9.9 million new unemployment claims over the past two weeks. If everything goes back to normal and 90% of these recently unemployed people return to work when this is all over, employers would still have 990,000 positions to fill, at a cost of up to one-third of those positions' salaries.

Let's assume the average salary across those 990,000 jobs is $45,000. That's a $15 billion direct turnover cost, and it's probably a conservative estimate. Many more people will lose their jobs before this is over, and many other impacts will hit the labor market before it goes back to anything resembling the "normal" of January 2020.

Many people could be out of work for so long that their skills deteriorate, which is a major reason most employers won't hire anyone who's been out of work for at least six months.

Let's imagine millions of people sit idle for six months. Some will use the time to polish their skill sets and become more valuable members of the workforce. Many won't. We can't expect everyone claiming an elevated amount of unemployment money to pursue education and training opportunities on their own during the lockdown, nor should we.

Employers and recruiters will be absolutely overwhelmed with job applications for the rest of this year, if not longer. As with any recession, there will likely be many more qualified applicants than there are available jobs for quite some time. Unlike other recessions, most of these applicants will all be chasing the same jobs at the same time -- most likely once increased unemployment payments wind down in about four months.

I wouldn't want to be a human resources professional in August. There's going to be an avalanche of applications for every open role in the country.

Ultimately, a forced nationwide quarantine could wind up creating a long-term class of underskilled people who struggle to find employment in an extremely competitive job market. This is the same problem that drove the growth of the gig economy after the Great Recession, as many people decided they'd rather drive for Uber than jump through all the hoops of the typical employment process.

Ironically, the pandemic could wind up causing lasting damage to the gig economy, either by flooding services with more gig workers than their economies can support, through increased demands for employee benefits and protections, or some combination of both.

The changes ahead

The modern economy is a consumer economy. And let's be honest with ourselves, we really don't need a lot of what we buy.

When you come out of lockdown, will you resume buying all the same junk you bought before? Will you have the same desire for a nice new car once your employer finally realizes all those meetings really could've been emails or videoconferences after all? Will you want to stuff random junk into every crevice of your house after you've been forced to really live in every square inch of your space, 24 hours a day and 7 days a week, for weeks at a time?

McKinsey already anticipates a drop of 40% to 50% in U.S. consumer discretionary spending as a result of the pandemic. Such a massive decline will cost the U.S. 10% of its GDP, not counting the second- and third-order effects of eliminating half of discretionary spending, such as lost jobs, reduced B2B spending, mortgage defaults, and so on.

Most post-crisis economic models tend to anticipate society bouncing back to a state more or less resembling normal. I'm not sure this is the right approach.

The postwar and post-pandemic economy of the 1920s looked dramatically different from the economy of the late 19th and early 20th century. Over twice as many U.S. homes were electrified in 1929 as were electrified in 1919 after the Spanish Flu had passed. Vehicle registrations soared from less than 500,000 to nearly 27 million. The urbanization, electrification, and vehicle-ization of the American economy was transformative, but these trends were driven (and also helped drive) the consumerization of the country.

Now, let's look at where we are with the economy in 2020.

Automation has already replaced millions of manufacturing jobs. In a post-pandemic economy, it's easy to envision automation taking over in many consumer-oriented jobs, from transportation to retail. If there was a strong push for consumer-level automation before, imagine how much stronger it'll be now that human contact carries an elevated risk of death.

You don't even have to imagine it, because 41% of surveyed executives are already committed to greater automating of their businesses now that everyone's on lockdown.

Many jobs at high risk of automation are heavily consumer-dependent:

Food preparation, construction, cleaning, healthcare, hospitality, teaching, driving, manufacturing. Even if they're not strictly consumer-focused jobs, these roles all share one thing: they tend to involve close physical contact with many other people. Many other roles are at risk for different reasons, but the risk is no less real.

When you automate millions of jobs, you need to do something with those millions who no longer have jobs, or you end up with widespread social unrest.

Millions of service jobs took on workers displaced by manufacturing automation in the 1920s and afterwards, but we don't have the same outlets for new employment outlets today. "Learn to code" isn't good advice for most people at risk of displacement, for two reasons: 1. coding isn't something all that many people are good at, and 2. the world doesn't really need millions of new coders.

The future of work prioritizes creativity, imagination, personality, and originality -- all qualities the American public education system works quite hard to suppress.

Even if the education system miraculously reorients itself around 21st-century values, it's still failed to teach these values to anyone in the current economy. Will they retrain themselves on their own? Probably not. Will most businesses help these left-behind workers reskill? Probably not, since they certainly haven't made any effort to do so before.

The new normal

Major recessions tend to be turning points for the economy. This, almost certainly, will be a major recession, of a kind unlike any faced before in modern economic history.

You might think that the artificial nature of this crisis, driven as it is by an intense government response to rapidly rising pandemic cases and deaths, could allow us to come out on the other side with relatively minimal damage.

But the economic body doesn't work like that.

Many businesses will die. Millions of people will be jobless. There won't be any easy, quick way to reintroduce millions of furloughed and laid-off people back into the workforce after several months of idleness. And millions of jobs are at greater risk of automation than ever before.

The American economy depends on wage-earning consumers going out and buying stuff, most of which they simply don't need. To buy this junk, millions of people work mind-wasting bullshit jobs that could (and eventually will) be done by machines. Machines will take over many bullshit jobs, and many people will realize they've been spending too much money on meaningless crap.

I wish I knew how this will all play out. I don't know how bad it's going to be.

I just know it's going to be bad. I believe it'll be worse, perhaps much worse, than many people want to believe. And I know we won't wind up with an economy just like this one on the other end of everything.

I wish I knew what that new economy going to look like, but what I've seen isn't particularly encouraging. And I know I'm not the only one who thinks so.

For the past 40 years, capitalism-at-all-costs has been the driving force behind American policymaking and behind the increasing commodification of American life.

When this is over, I think we need to decide if this fundamental drive still makes sense in an increasingly automated, part-time, side-hustle economy.

If it doesn't -- and it shouldn't -- we need to figure out what comes next.

Or we won't like what comes next, because it's not going to be any better than this.

I know this is far from a comprehensive review of the possibilities to come, and I welcome your feedback. Do you agree with me? Do you think I'm wrong? Do you have anything you think I need to add for depth, detail, or accuracy? Please let me know what you think in the comments!