Here’s a very compressed summary and some links on standard economic theory around recessions. Of course economists argue about this stuff to no end, so take it all with a grain of salt.
First, there’s a high-level division around what causes recessions. Two main models:
Real shocks: a hurricane, war, virus, etc directly decreases economic output.
Sticky prices + volatile currency: contracts are denominated in dollars, so if the value of a dollar goes up relative to everything else, lots of debtors/employers/etc are unable to pay.
The former is the domain of real business cycle theory (RBC), the latter includes most of both Keynesian and monetarist models—MRU has a good set of intro-level videos on all that. If you want a perspective with more analysis and less politics, I recommend jumping straight into recursive macro models—though this does require a strong math background.
Real-world recessions can involve either or both of these causes. The textbook example of a real shock would be the 1970′s oil drama—though textbooks over the next few decades will likely use coronavirus as an example instead. The 2008 meltdown and the Great Depression, on the other hand, are generally seen as primarily monetarily-driven recessions.
Coronavirus-Specific
One pattern of major concern today is that a real shock induces a minor recession, the value of a dollar goes up relative to most other things due to the real shock (i.e. people have trouble paying their debts), but then the Fed reacts too slowly/too little to bring the value of a dollar back down (i.e. by “printing money”) and a monetary recession results. The Fed’s announcements over the past few weeks have been very good on that front—it remains to be seen whether they’re enough, but qualitatively they’re clearly doing the right sort of things, and they’re signalling willingness to do more if needed. We’ve come a long way since 2008.
As long as the monetary situation continues to look good, we’ll mainly want to use an RBC-style model. Things to place more/less emphasis on:
In an RBC model, we don’t worry so much about “financial contagion”, bank runs, etc. That’s not to say the Fed shouldn’t worry about any of that; rather, those are the problems which can be avoided if policymakers are on-the-ball.
Heterogeneous goods: a real shock will reduce production of some goods but not others. An oil shock looks different from a potato crop failure or a pandemic. Monetary recessions all center on a “shortage” of the same “good”—i.e. money—so we’d expect them to look more similar to each other, whereas real shocks (absent a monetary problem) will look more different from each other.
Loss of capital goods: a lot of study goes into the extent to which real shocks have long-lasting effects, vs a rapid economic bounce-back. The main mechanism for lasting effects, in most models, is that production of capital goods slows due to the shock, but existing capital goods continue to break down. To see how relevant that will be, I’d look at how the virus is impacting production of the major capital sinks: construction, oil wells & pipelines, data infrastructure, power plants & the electric grid, roads & railroads, etc.
To see how relevant that will be, I’d look at how the virus is impacting production of the major capital sinks: construction, oil wells & pipelines, data infrastructure, power plants & the electric grid, roads & railroads, etc.
One interesting point on this front is that the cost to road work and infrastructure improvements is lower now than it normally is, so if you figure out a way to do construction work safely, you could justify above-baseline investment in some major capital sinks. (It’s unclear to me how licensing restrictions come into play here; you have millions of unemployed, but you might not be able to use them to build and repair bridges and roads.)
Here’s a very compressed summary and some links on standard economic theory around recessions. Of course economists argue about this stuff to no end, so take it all with a grain of salt.
First, there’s a high-level division around what causes recessions. Two main models:
Real shocks: a hurricane, war, virus, etc directly decreases economic output.
Sticky prices + volatile currency: contracts are denominated in dollars, so if the value of a dollar goes up relative to everything else, lots of debtors/employers/etc are unable to pay.
The former is the domain of real business cycle theory (RBC), the latter includes most of both Keynesian and monetarist models—MRU has a good set of intro-level videos on all that. If you want a perspective with more analysis and less politics, I recommend jumping straight into recursive macro models—though this does require a strong math background.
Real-world recessions can involve either or both of these causes. The textbook example of a real shock would be the 1970′s oil drama—though textbooks over the next few decades will likely use coronavirus as an example instead. The 2008 meltdown and the Great Depression, on the other hand, are generally seen as primarily monetarily-driven recessions.
Coronavirus-Specific
One pattern of major concern today is that a real shock induces a minor recession, the value of a dollar goes up relative to most other things due to the real shock (i.e. people have trouble paying their debts), but then the Fed reacts too slowly/too little to bring the value of a dollar back down (i.e. by “printing money”) and a monetary recession results. The Fed’s announcements over the past few weeks have been very good on that front—it remains to be seen whether they’re enough, but qualitatively they’re clearly doing the right sort of things, and they’re signalling willingness to do more if needed. We’ve come a long way since 2008.
As long as the monetary situation continues to look good, we’ll mainly want to use an RBC-style model. Things to place more/less emphasis on:
In an RBC model, we don’t worry so much about “financial contagion”, bank runs, etc. That’s not to say the Fed shouldn’t worry about any of that; rather, those are the problems which can be avoided if policymakers are on-the-ball.
Heterogeneous goods: a real shock will reduce production of some goods but not others. An oil shock looks different from a potato crop failure or a pandemic. Monetary recessions all center on a “shortage” of the same “good”—i.e. money—so we’d expect them to look more similar to each other, whereas real shocks (absent a monetary problem) will look more different from each other.
Loss of capital goods: a lot of study goes into the extent to which real shocks have long-lasting effects, vs a rapid economic bounce-back. The main mechanism for lasting effects, in most models, is that production of capital goods slows due to the shock, but existing capital goods continue to break down. To see how relevant that will be, I’d look at how the virus is impacting production of the major capital sinks: construction, oil wells & pipelines, data infrastructure, power plants & the electric grid, roads & railroads, etc.
One interesting point on this front is that the cost to road work and infrastructure improvements is lower now than it normally is, so if you figure out a way to do construction work safely, you could justify above-baseline investment in some major capital sinks. (It’s unclear to me how licensing restrictions come into play here; you have millions of unemployed, but you might not be able to use them to build and repair bridges and roads.)