Everyone is fretting about how severe the downturn that is now starting and causing the worst plunge in the stock market since the 1929 crash is going to be. Much of the discussion has turned on whether the cause of the downturn is a supply shock or a demand shock. Some, perhaps many, seem to think that if the shock is a supply, rather than a demand, shock, then there is no role for a countercyclical policy response designed to increase demand. In other words, if the downturn is caused by people getting sick from a highly contagious virus, making it dangerous for people to gather together to work, then output will necessarily fall. Because the cutback in the supply of labor necessarily will cause a reduction in output, trying to counteract supply shock by increasing demand, as if an increase in demand could prevent the reduction of output associated with a reduced labor force after the onset of the virus, seems like an exercise in futility.
The problem with that reasoning is that reductions in supply are themselves effectively reductions in demand. The follow-on reductions in demand constitute a secondary contractionary shock on top of the primary supply shock, thereby setting in motion a cumulative process of further reductions in supply and demand. From that aggregate perspective, whether the initial contractionary shock is a shock to supply or to demand is of less importance than ensuring that the cumulative process is short-circuited by placing a floor under aggregate demand (total spending) so that the contraction caused by the initial supply shock does not become self-amplifying.
The interconnectedness of the entire economy, and the inability of any individual to avoid the consequences of a social or economic breakdown by making different (better) choices — e.g., accepting a cut in wages to retain employment — was recognized by the most orthodox of all Cambridge University economists, Frederick Lavington, in his short book The Trade Cycle published in 1922 in the wake of the horrendous 1921-22 depression from which the profound observation that serves as the title of this post is taken.
It’s now 60 years since John Nash defined an equilibrium as a situation in which “each player’s mixed strategy maximizes his payoff if the strategies of the others are held fixed. Thus each player’s strategy is optimal against those of the others.” If the expectations of other agents on which other agents are conditioning their strategies (plans) are sufficiently pessimistic, then an unemployed worker may not be able to find employment at any wage, even if it is only a small fraction of the wage earned when last employed. That situation is not the result of a diminution in the productivity of the worker, but of the worsening expectations underlying the strategies (plans) of other agents.
To call unemployment “voluntary” under such circumstances is like calling the reduced speed of drivers in a traffic jam “voluntary.” To suppose that the intersection of a supply-demand diagram provides a relevant analysis of the problem of unemployment under circumstances in which there are massive layoffs of workers from their jobs is absurd. Nevertheless, modern macroeconomics for the most part proceeds as if the possibility of an inefficient Nash equilibrium is irrelevant to the problems with which it is concerned.
There are only two ways to prevent that cumulative decline from taking hold. The first is to ensure that there is an immediate readjustment of all relative prices to a new equilibrium at which all agents are able to simultaneously formulate and execute optimal plans by buying and selling at market-clearing equilibrium prices. Such an immediate readjustment of relative prices to a new equilibrium price vector is, for a multitude of reasons which I have described in my recent paper “Hayek, Hicks, Radner and Four Equilibrium Concepts,” an extremely implausible outcome.
If an immediate adjustment to an unexpected supply shock that would return a complex economy back to the neighborhood of equilibrium is not even remotely likely, then the only way to ensure against a cumulative decline of aggregate output and employment is to prevent total spending from declining. And if total spending is kept from declining in the face of a decline in total output due to a supply shock, then it follows, as a matter of simple arithmetic, that the prices at which the reduced output will be sold are going to be correspondingly higher than they would have been had output not fallen.
In the face of an adverse supply shock, a spell of inflation lasting as long as the downturn is therefore to be welcomed as benign and salutary, not resisted as evil and destructive. The time for a decline in, or reversal of, inflation ought to be postpone till the recovery is under way.
Uneasy Money 