TLDRUpFront: The second of a four-part series as the United States, and the world, begins tipping-over from COVID-19. The next nine days are going to be hairy. This InfoMullet focuses on the intermediate economic impact preparing the mental model of readers for what’s going to become apparent between now and March 31st .
My first article in the series, Ring Around the Rosies, dealt with the virus itself over the next few days leading up to March 31st. As that’s occurring, there’s going to be a corresponding economic reality unfolding over the same time frame which I’ll try and go over here. The forecasts below are limited to a short-to-intermediate stage, of 3 months. For reasons explained in full near the end, the ultimate economic impact over a longer 6-18-month horizon is going to depend on factors outside of our control that have to do with the nature of COVID-19 itself.
Developed Economies are Consumer Economies
To understand the economic forecasts below it’s important to understand at a high level what makes up the “engine” of economies in developed countries. I’ll be using the United States for reference. Specific values for other developed economies may differ but the relative ratios are similar. For example, US Consumer Spending is close to 70% and the OECD average is 60%. (1) is ~68.4% the United States economy.
Consumer Spending: ~70%
Government Spending: ~16%
Business Spending: ~16% (2)
Much of the Consumer Economy is Grinding to a Halt (and so is Business)
To understand the impact of COVID-19 on the 86% of the US economy made up by business and consumer spending it helps to break these major components down. The chart below expands each of the three major components into subcomponents along with each share they represent of the main component. The color coding indicates the short to intermediate or 3-month impact to the economy of COVID-19 because not all areas are going to be impacted equally. Red indicates a significant slow-down, if not halt in some cases, to spending in that category over the next three months. Amber significant turbulence. While green components may actually see increased spending as consumer habits shift under Fire Break or Mobilization & Lock Down national strategies.
It is unlikely however that the shift in spending to new areas is going to offset the lost spending in others, at least in the intermediate. Part of this is because consumer spending is in large part driven off consumer confidence. And consumer confidence has already fallen off a cliff during the month of March and will be in for a shock over the next 10 days. This will lead to a shift of frugality and savings with uncertainty on what lies ahead, drying up even “online” or “virtual” service purchase after the initial spate of buying activity as self-isolating households ramp up digital purchases.
Part of this halt is that consumers will simply be unable to spend in Lock Down conditions. CalculatedRiskBlog tracks Restaurant Year over Year sales and produced this chart to show the calamity facing restaurants in Seattle, San Francisco, and New York City as their year over year sales dropped by as much as 100%.
On Thursday the Depression Arrives
And then it gets worse.
Every Thursday the Department of Labor releases its weekly report of new unemployment claims from the last week as well as “continued claims.” The two numbers are important, the first shows the inflow of newly unemployed and the second how long they persist. This is because in normal economic times a job loss usually leads to another job. But sustained continued claims are hallmarks of depressions and recessions. For perspective, the all time historic high of seasonally adjusted weekly new unemployment claims was 695,000 in October of 1982. By some estimates however, the numbers this week could reach several million new claims. And at least in the intermediate term, those jobs are not going to come back through shifts in the economy to new sectors. This means they will persist as “continued claims.” The chart below tracks weekly “continued claims” over time, with recessions marked in blue bands. The red arrow is added to show how this week’s numbers may take us half-way or more to the unemployment, in raw numbers, of the previous high during the 2009 financial collapse.
The States are not Ready for This
When the consumer and business economy slows State and Federal governments often provide a temporary backstop of support during the time of crisis. Although there are valid arguments of what constitutes a ‘crisis’ around much Federal spending, a global pandemic certainly seems to fall on the side of qualifying. At the Federal level, with its ability to borrow unlimited amounts and print money on demand, these policy responses are easier. Add a few trillion zeroes in the spreadsheet and you’re off to the races backed by the full faith and credit of the United States government, which thankfully still exists because we didn’t default on that debt during any recent political games-of-chicken.
But States face greater difficulties, they can’t print their own money and investors want a higher premium for the risk of investing in them. Add to that general incompetence and mismanagement and despite a boom economy for over a decade many States are not prepared for the depression that is about to begin and this is going to show in two ways: the ability to meet unemployment payments from existing funds versus borrowing to pay and the sustainability of pension and healthcare funds that States manage on behalf of civil workers. In both cases many States are simply not prepared for what is about to happen.
As of a February 2020 estimate by the Department of Labor, only 31 States and Territories are at or above the minimum solvency level for funding unemployment funds.(9) Unemployment operates like an insurance. Workers and businesses both contribute to pay “premiums” into a general fund that then pays out claims in the event they are laid off. Like most insurance mechanisms this requires two criteria to work. First the premiums must be wisely managed and invested in order to have sufficient reserves to pay claims when presented. Second, sufficient reserves have to be maintained for unexpected surges in claims during a natural disaster, such as a global pandemic bringing the economy to a halt.
The DoL tracks minimum solvency based on these and other factors and awards states that are above the solvency line with interest free borrowing to aid in cushioning hardships during unusual conditions. States that don’t meet that threshold will have to borrow from the Federal government, at interest, to meet payouts. To be clear – state solvency of the UI program doesn’t mean that UI checks won’t be paid. But they’ll have to be paid with borrowed funds adding further long-term strains to state budgets that will already be faced with critical funding needs.
Bipartisan UI Insolvency (9)
The States and Territories below this solvency threshold are New Hampshire, Alabama, Wisconsin, Tennessee, Minnesota, Arizona, Wisconsin, Rhode Island, Maryland, Missouri, Colorado, Delaware, New Jersey, Pennsylvania, Kentucky, West Virginia, Indiana, Connecticut, Ohio, Massachusetts, Illinois, Texas, New York, California and the Virgin Islands. That list is sorted “best positioned to worst positioned” in terms of relative distance from the minimum solvency threshold. Unfortunately, three of the largest states by population, in or approaching lock-down conditions driving economy hardship, are also the worst off in terms of preparedness for UI claims: California, New York and Texas. (9)
State pensions are the trust funds set aside for often large civilian workforces. Although they’ve received less attention than corporate pensions, the reality is that there are fewer corporations offering pensions and those that do tend to be better funded at large scale. And the funding levels as a % to solvency across states vary widely. In 2017, the latest data I could find this morning, eight states had funded their obligations to 95%, but 20 of the lowest-funded plans saw their funding levels at 56% on average.
This often occurs during boom-times, where plush taxes rolling in are seen as ‘easy money’ and not set aside for rainy days. When an economic crisis such as that caused by COVID-19 hits it causes a trifecta of pain. First, the stock assets that many of these pensions have invested in have been reduced by as much of a third over the last month. Second, taxes from businesses will slow to a trickle as the economy halts. Third, expenditures on crisis and unbudgeted services will increase as States mobilize to respond.
The map below illustrates where, at least by 2017, States stood relative to funding their pension and healthcare obligations.
Why this won’t become Mad Max – or at least we’ll be the last to go Thunderdome
As a researcher of violence and instability I’m always interested in how the public views societal collapse, and what causes it. Many have a mental model of hunter-gatherer or warbands roving the desert in convoys of cars chasing after water, fuel, and penicillin while shooting fire from guitars and embracing gothic fetish gear. And we can understand why, because that’s what Hollywood has fed us that “societal collapse” looks like.
But those scenarios always discount several key things, first being the real-world experience we have in economic calamity, even war time, and what happens to societies under those conditions. I often direct Doomsday Preppers that if they want to understand what an apocalyptic landscape looks like, and how to prepare for it, don’t look to the Walking Dead but to the populations of Syrian cities such as Aleppo or Homs during the Syrian Civil War. Having studied the civil war extensively I don’t recall fire-belching guitars making an appearance. Most maintained improvised healthcare, education, daycare, and logistics systems to provision needed materials even under constant aerial bombardment, artillery shelling, chemical attacks, and being cut-off from supplies. As bad as COVID-19 is going to get, its not as bad as unrestricted aerial bombardment of a civilian population.
Combined with the historical examples to the contrary of calamity we can think of economic hardships in terms of how many years they set us back. It’s reasonable to say for example that the Financial Collapse of 2009 set us back 10 years of progress in some ways. But because of Hollywood, most people are willing to jump back 2,000-4,000 years to roving warbands. But they don’t jump back even 100 years to consider the Great Depression, which despite being bad, didn’t feature prominent adoption of gothic-fetish gear and an emergence of feudal society.
Another argument to weigh against the Mad Max scenario comes from economic studies of recession and depression severity, and what causes them. The charts below both compare relative employment levels between depressions and recessions. The first compares select depressions around the world with the US Great Depression and Great Recession. (6) And the chart below that compares all US recessions since WWII against the Great Recession that began with the financial collapse. (7)
Note the two shapes of the line patterns in both charts? In both depressions and recessions that are one form of impact that looks like a sharp V, rapid decline, reaching a turning point and returning rapidly to previous levels. But there is another form, including both the Great Depression and Great Recession where the return to previous levels is much more prolonged.
These two modes are caused by two types of shocks to the system: a financial crisis and an exogenous shock. A financial crisis often originates in misallocation of debt instruments and lending practices. In the Great Depression margin-calls on stocks allowed investors to effectively borrow multiple times what they could afford and in the Great Recession collateralized debt obligations packaged bad mortgages into “assets” that could then be leveraged hundreds of times over. When these collapses occur, credit evaporates – and it is credit that often allows demand to re-emerge quickly after a crisis, as both consumers and businesses borrow to return to economic activity. This is why financial collapses have such a slow recovery.
An exogenous shock crisis behaves differently. Consider a snowstorm that keeps everyone inside for three days. No business occurs while everyone is locked away. But it’s not for lack of demand, or credit. The exogenous factor of the blizzard prevents normal activity. The demand that would be satisfied over those three days “builds up” until the blizzard subsides, the snows melt, and people re-emerge. Now all that pent-up demand expresses itself in buying all the things they would’ve bought during the blizzard. Of course, this doesn’t hold true for everything. People continue to eat during a blizzard, so there is no “pent-up demand” for restaurants even as customers return, they’re still only eating that one lunch not three at once. And the longer the exogenous shock occurs, without support from the government, the more businesses and households will enter into bankruptcy conditions. That’s the big unknown, at least to me at the time I’m writing this, is how healthy corporations are in terms of debt load. When the economy turn it’s like the tide receding on the beach and we can see who isn’t wearing any swim trunks. Corporations that took out leveraged positions or have heavy debt loads may fold and collapse during the next three months, and if enough of them do it may create a rolling financial crises.
But this is why the Federal Reserve has stepped in to offer eye-popping loan guarantees to banks. To keep an exogenous crises from turning into a financial one. One thing to keep in mind with these guarantees, which last I checked is up to $1T/week and $1T/day is these are not actually funds that have been issued yet. They are offers of backstop, and as much about confidence signaling as an actual line of credit. As far as I know, of this morning, no banks had tapped into these measures. Instead this is a measure to prevent the kind of cascading financial and monetary collapse that marked the days of August in 2009 after the collapse of Bear Stearns and Lehman Brothers in rapid succession.
The COVID-19 pandemic is an exogenous shock condition of the worst kind, but it is still an exogenous shock at this time. This means that despite what is likely to be a depression, the road to recovery may be quicker in the long term than is to be expected during these gloomy days of March.
Ultimately though it’s the Virus that Decides
Lest the above paint too rosy a picture the economic outlook is still pretty bleak. As mentioned in Ring Around the Rosies, just as the pandemic is hitting everywhere at once – the global depression that follows in is wake will hit everywhere at once as well. Worse, the nature of the depression we’re entering: it’s length, severity, impact, and time to recovery, is going to depend on three things we have no control over:
- Does the virus grant immunity after having it once?
- Will there only be one “wave” of the virus?
- Will vaccines created against the current virus be effective against future forms of the virus?
If the answer to all three questions are yes, we’re going to look back in a year at the proclaimed transformative power of COVID-19 to alter society, laugh awkwardly and continue on our way in a world that looks very much like the one we live in now, with a few major exceptions. Much like the post 9/11 world was much like the pre 9/11 world, with some major exceptions. Life will return, and continue, as it had as we dig ourselves out of a global depression.
If the answer to any of these three questions is no however , we’re in for a pretty rough ride. If the virus can reinfect, or there are subsequent catastrophic waves, or it mutates while passing through tens if not hundreds of millions of people, then we may be in for an extended lock-down of 6-12 months or periodic recurrent lockdowns each time a new wave surfaces. And that means a fundamental reordering of the society may be necessary on top of the depression conditions.
And it’s because we don’t know yet the answer definitively to those three questions that this forecast is limited to the short and intermediate level of the next 3 months.