Silas poses the following question in the comments (link)

Let’s say a few scientists have announced that they’ve discovered an effectively unlimited/costless energy source (like the cold fusion situation). A few of their friends have gotten it to “work” too, in that it appears they’re really on to something — perhaps energy problems are about to be solved forever.

On the other hand, most claims like these turn out to be fraudulent or mistaken. And even if they’re doing everything properly, it could be that they’re not tabulating the energy flows correctly (i.e. not realizing that something critical is consumed in the process, meaning it requires more resources and energy input to keep it going indefinitely, which they hadn’t accounted for on deeming it costless). But still, there is much evidence for and against, and the truth or falsity of the claims really matters.

In this situation, it would be very wise not to commit resources to narrow uses, as the eventual unveiling of the truth about the energy claim will obviate many kinds of economic activities. If it turns out to be true, huge sectors of the economy become worthless: the drilling equipment for oil, energy prospecting equipment, coal mines, natural gas pipelines, equipment for maintaining these pipelines, transportation networks that exist because all of this, and, of course, all of the specialized workers and knowledge sets so related. Investing *more* in these will mean you’re “caught with your pants down” when the claims are verified to be true.

On the other hand, investment in the *opposite* direction will be painfully obviated if the claims turn out to be false. So all investments and most purchases carry an additional risk-cost, and so holding off on doing so is a rational response to the economic realities involving high uncertainty about the usefulness of particular goods. In this situation, if you got people to spend and invest for the sake of propping up spending and investing, you’re causing a huge waste of resources: since people don’t yet know which course of action is right, many more goods than otherwise are going to be committed to wasteful production structures.

First let me say that I haven’t come up with any satisfying answers to Silas’ question. I don’t know how to think about it properly, so I will be more humble.

There are a few reasons why the idea that a general reduction in spending due to increased money demand is a desirable signal:

  • A strong signal to avoid investment in affected sectors is already present, people who invest before the uncertainty is resolved will make worse investment decisions than people who don’t.
  • There are lots of reasons why money demand can increase and many of them do not have to do much with uncertainty like this. For example, some new financial technology might require lots of transactions and people need money in order to conduct the transactions.
  • A general spending reduction sends a very blunt signal. A shortage of money is likely to cause people to reduce their spending on many different margins, most of which will not be relevant to the sector. For example in the example you gave, the uncertainty is around one sector (energy) which may have little effect on some other sectors, so it likely makes sense to shift resources that cannot be transferred into the future into those sectors (say entertainment and housing)
  • The signal is fleeting, as prices adjust, the signal will go away even though the need for transfering resources into the future may continue. In a world with no sticky prices this signal would not exist; nominal spending would fall, but real spending would be stable.

I think one of the key issues is the time frame over which this plays out. I am not sure what kind of time frame Silas had in mind: ~2 months or ~2 years.

Over a time period of ~2 months, I am especially unsure how to think about the issue, but keep in mind that most proposals for nominal income targets focus on income expectations ~2 years out, and would only try to stabilize short term expectations to the extent that they affected longer term expectations (and only by trying to stabilize longer term expectations). I am unsure about the desirability of stabilizing income over  short time scales, but in any case, I don’t think it is possible.

Over a time period of ~2 years, I think I understand the issue a little bit better. Over the course of 2 years the appropriate response for the economy is to shift production away from investment in affected sectors and toward resources that can be stored until the uncertainty is resolved investment in unaffected sectors, consumption and leisure. It is important to note that there are many kinds of resources that you cannot transfer into the future easily, for example lots of kinds of labor. The proper response in these cases, is generally to do something else with them rather than do nothing with them. Prices are probably mostly flexible over this time period, so allowing nominal income to rise or fall will not affect resource allocation a lot, but will affect the dynamics of the transition.