We're Not Just Time-Shifting Anymore

In my last post on macroeconomics, I presented a simple model in which businesses and individuals can shift spending across time, and the point of macroeconomic policy is to spread out the spending so as to avoid excessive inflation, on the one hand, or excessive unemployment, on the other.  Now I'll tweak the model slightly.  Both fiscal and monetary policy are capable, not just of time-shifting spending, but of creating spending where it otherwise wouldn't have existed.

This is easy to see in the case of monetary policy.  Imagine that there is a business project that will return 5% on the initial investment.  If prevailing interest rates are over 5%, then it is not a profitable investment - it can't generate enough cash to compensate investors, and so it can't attract capital in the first place (unless people are deluded or misled as to its prospects).  But if the central bank lowers interest rates below 5%, then the project may become profitable.  So it's not that the project was going to happen at time T2, but thanks to low interest rates it will now happen at time T1.  The project wasn't going to happen at time T2 either, barring a drop in interest rates.  The project isn't being time-shifted at all, it's being made feasible by a low-interest-rate environment.  Regardless, this fits our basic story:  the central bank is lowering the interest rate to generate economic activity, which draws resources into productive use and reduces unemployment (assuming the economy isn't already at full employment - if it is, then we would expect a low-interest rate policy to be inflationary).

Fiscal policy is a little more complicated.  In one sense, of course, it's easy to "create" spending that otherwise wouldn't have existed.  But if we are talking about temporary stimulus spending (as opposed to a new, permanent program), then the spending is probably some kind of useful project that would have happened eventually anyway.  Bridges get repaired in time T1 that would have needed repairs in time T2, that kind of thing.  It's hard to come up with useful projects at time T1 that wouldn't have been useful at time T2 or T3 or whatever.  But this isn't a hard-and-fast rule:  because both real resources and government borrowing are cheap (that is, there are resources going unused, and the government can borrow cheaply), some kinds of spending that ordinarily wouldn't be cost-justified can be good candidates for stimulus spending, and in those cases you are really creating spending out of thin air, not just time-shifting it.

So although time-shifting is a good first cut, I think it makes sense to think of stimulus as both shifting spending across time and also generating spending that otherwise wouldn't exist.  In my next post, I will spell out a hypothetical that shows how an economy can get into trouble when the economy generates "the wrong kind of spending."