Lockdown and The Interconnected Economy: The Country of Shmo

The fact that we have interconnected economies should be so obvious and intuitively understood no reason should exist why people fail to understand that the observation in [1] that unemployment was rising in a state without stay at home orders “SAH” as evidence of the isolated signal of the pandemic is anything but… the authors appear to have failed to factor into the analysis State to State inter-economic dependencies in the least as they pertain to “lockdown” orders, and certainly Global interdependencies.

From the abstract:
“Ignoring cross-regional spillovers, this finding implies that, of the 16 million UI claims made between March 14, 2020 and April 4, 2020, only 4 million are attributable to the Stay-at-Home directives.” [1 – Page 1]

Translation: They ignored economic interdependencies.

"It is worth pointing out that in our analysis we are unable to distinguish between competing mechanisms by which SAH policies might ultimately increase UI claims—for example, either through the collapse in the supply of goods that require on-site labor and/or through the drop in demand for high-contact, nonessential goods and services (e.g. live concert musical performances). We leave this for future research."[1 – Page 4 – Par 3 ]

"The direct ( emphasis added ) effect of SAH orders on unemployment is therefore small relative to the aggregate increase in UI claims, suggesting that a large majority of the increase in unemployment may have occurred in the absence of SAH orders."[1 – Page 4 – Par 5]

This implies the indirect effect of SAH orders has significant potential to describe the aggregate rise in UI Claims - and yet it was hand waived out of existence in the analysis without any explanation.

What is truly disturbing about this is they are saying that the rise in UI in States without SAH orders, hints that such rise is a signal of the pandemic and not instead that the signal represents the strength of inter-economic dependence among state and global economies. They have quickly decided to forget everything they learned in economics about networks of economies. This is patently obvious to the entire field of economics - so much so that the creator of the formal language through which these interconnections are understood "Wassily Leontief" [2] was awarded the Nobel Prize in Economics for its development in 1973.

Picture an entire economic system at work in a steady state condition with no external demand. What this means is we are going to ignore that portions of Economies A, B, C, D… output that in reality act as inputs to external systems ( i.e. the Global Network). The system is categorized as a Closed Economy; This means what is produced remains in the system. Economies A, B, C, D… are inputting/outputting (feeding and eating ) Economies A, B, C, D… An important note is that Economy A is both eating from and feeding itself as well as feeding other economies. Basically, Economies use a portion of their output to sustain themselves.

So we are rolling along in steady state economic conditions and a factory in State A that produces say “tractor tires”, suddenly shuts it doors do to COVID lockdowns. Unfortunately, the assembler of the tractors in another State B not under the lockdown order can no longer finish assembling the tractors in its assembly line. The economy of State B, not under any lockdown order ( in the absence of available inputs ) also lays off employees as an indirect effect of the SAH orders in State A. But it does not stop there. Some portions of Economies C, D, E…that rely on equipment to farm experience a decline in economic output as well. The ripple will go on to causally effect all the interconnected economies in a ways we could barely imagine without a systematic approach( this is at least one reason why it is common practice to keep multiple suppliers for each input in the supply chain…competition for business is another) However, when multiple economic centers in the network are being shut down simultaneously, and more economies are competing for the same business, demand can easily exceed supply. This results in negative economic scaling for anyone left in the game, because supply does not have the ability to rapidly grow to meet the new demand. Industries are run at near maximal capacity for reasons of efficiency, and maximization of profits for the demand that exists ( i.e. they don’t have much excess capacity because growing and maintaining unused space is expensive). If a COVID lockdown that causes a certain economy in the network to suddenly shrink and stay down while simultaneously moving through the network itself locking down potential suppliers of the “tractor tires” in other states, or the supplier can’t meet the new demand, the ENTIRE economic network will shrink regardless of where the lockdown orders are in place.
So how do we account for this? Back to the Leontief input-output model. A simple representation that the Authors of [1] could have made ( which I’m sure there is data for ) if they were truly trying to “isolate” the economic pandemic signal would be an aggregate 50x50 Matrix that captures approximately the economic relationships between the states. If you looked at the actual goods production Matrix, it would become computationally massive rather quickly. The goal in the aggregate would be to determine what are the fractions of economy coming into State A, from A,B,C… ect for every unit of economy State A outputs. Here is a simple 3 State Closed System Network. States A,B,C.

Continued on next post...

Cont...

Sample Analysis: The Country of Shmo

The values that follow are the individual elements of column matrix that together compose the Leontief Matrix L. They represent the fractions of economic input from state A, B, C that go into producing one unit of State A’s economic output. So lets say that State A needs 50% of its own output as an input, 0% of State B’s output as an input, and 20% of State C’s output and input to produce 1 unit of State A output.

Column A = 0.5; 0; 0.2

Note: Column A has no dependency on State B’s economy.

Fill in the other columns as you wish such that Column B has no dependency on State A’s Economy, and C has no dependency on B.

My L ( for Leontief) matrix is as follow

Col. A = 0.5; 0; 0.2
Col. B = 0; 0.6; 0.1
Col. C = 0.2; 0; 0.3

During Steady State we had a certain amount of economic demand from each of the State economies A,B,C to fill the needs of each of the State economies A,B,C. We will call this “D” ( 3x1 Column Matrix) ).

The net supply “S” of economic output “X” from States A,B,C is given by:

S = X – L*X

In economic equilibrium the supply equals the demand S = D

D = X - L*X

Thus the economic output matrix “X” from states A,B,C is given by:

X = (I – L)^(-1)* D

Lets do some computation:

Lets suppose the initial Demand matrix was before lockdown was:

D = 10; 6; 7 ( State A,B,C respectively )

In this case the economic output of each state is approximately:

X = 28; 15; 20

Let’s say that suddenly, we cut the ability of State B to meet their portion of demand by placing them in lockdown. What could happen?

Some things to note:
1) State B contributes least to the overall economic demand and output out of the three states.
2) State A’s economy is not directly dependent on state B’s economy ( i.e in Col A, Row B of L the value is 0 )
3) State C ‘s economy is not directly dependent on state B’s economy either ( i.e in Col C, Row B of L the value is also 0 )

You might think since there is no direct dependence on B’s economy as input to the economies of A and C we can vary B’s economy at will without effecting any change in the system. So lets cut B’s economy in half to 3 and see what pops out of the matrix algebra…

D = 10; 3; 7

X = 27.5; 7.5; 19

As you can see, I tried to make the direct dependence as weak as possible and use the smallest economy and I STILL effected negative change throughout every economic entity in the system. The economic output of B dropped by half (from 15 to 7.5) and both A and C economies shrank as well by smaller, but definitive amounts. This is a 3x3 economy where 1/3 of the interdependencies are zero! The real US State economy is massively more integrated and entangled. As I said, there would be 2500 elements in the array L, and hardly any of the interdependencies would be zero. Also, cities are the largest economic entities in the US; they were also the first to adopt lockdown measures. Overall economic output plays substantial roles in this as well. If I start shuttering State A’s economy by half (the Largest in the model) the entire country of Shmo will feel it much more strongly (23% Decrease in total output vs 15% change with State B), and so did the United States. The economic powerhouse of the US; California, New York, Illinois, Pennsylvania, New Jersey ect… were the early adopters of SAH orders. Why the unemployment rose over the entire nation regardless of SAH orders being a direct effect is due to indirect effects of an interconnected economy.

Summary:
I do not believe the authors of [1] have effectively isolated anything. If you want clear evidence of the pandemics isolated economic signal look no further than Sweden. Feb UR 7.5% - April UR 7.9%.[3]. No government SAH. Right next door in Norway: Feb UR 2.2% - April UR 9.6% [4]. It experienced a peak near 14.7% UR according to [5]. Norway adopted strong SAH orders.

P.S. I have no credentials in economics ( just good old fashion common sense ), so everything I’m saying is obviously open for debate. Which I why I have posted it.

[1] ChaeWon Baek, Peter B. McCrory, Todd Messer, and Preston Mui. (2020). “Unemployment Effects of Stay-at-Home Orders: Evidence from High Frequency Claims Data”. IRLE Working Paper No. 101-20. http://irle.berkeley.edu/files/2020/04/Unemployment-Effects-of-Stay-at-Home-Orders.pdf
[2] https://en.wikipedia.org/wiki/Wassily_Leontief
[3] https://countryeconomy.com/unemployment/sweden
[4] https://fred.stlouisfed.org/series/LMUNRRTTNOM156S
[5] https://www.reuters.com/article/health-coronavirus-norway-unemployment/norways-unemployment-soars-to-record-14-7-of-work-force-idUSS8N2BJ039

@joe-shmo said
Cont...

Sample Analysis: The Country of Shmo

The values that follow are the individual elements of column matrix that together compose the Leontief Matrix L. They represent the fractions of economic input from state A, B, C that go into producing one unit of State A’s economic output. So lets say that State A needs 50% of its own output as an input, 0% of State ...[text shortened]... oronavirus-norway-unemployment/norways-unemployment-soars-to-record-14-7-of-work-force-idUSS8N2BJ039

As I have said, I have no credentials in economics ( just good old fashion common sense ), so everything I’m saying is obviously open for debate. Which I why I have posted it in the Debates Forum. I look forward to constructive criticism on the content in both the technical aspect ( the mathematics of the model ) and my verbal description of interconnected economic systems preceding it. What do people agree, disagree with about what I have presented as a reasonable economic model?

@joe-shmo said
As I have said, I have no credentials in economics ( just good old fashion common sense ), so everything I’m saying is obviously open for debate. Which I why I have posted it in the Debates Forum. I look forward to constructive criticism on the content in both the technical aspect ( the mathematics of the model ) and my verbal description of interconnected economic systems ...[text shortened]... it. What do people agree, disagree with about what I have presented as a reasonable economic model?

@no1marauder said
You haven't. You just artificially plugged some numbers you thought up into an equation.

You ignored the evidence already presented of significant slowdowns in economic activity in the US prior to any government mandated "lockdowns".

@no1marauder said
You haven't. You just artificially plugged some numbers you thought up into an equation.

You ignored the evidence already presented of significant slowdowns in economic activity in the US prior to any government mandated "lockdowns".

Don't pretend that there isn't a phycological aspect to this either. The entire world watched the entire world panic unfold. The panic of others is far more catching than COVID, and media platforms threw trinitroglycerin onto its spread.