The Implausibility Of The Lower Bound Escape Clause

Simon Wren-Lewis discussed a recent debate about the fiscal policy of the U.K. Labour Party, in "Will taxes have to rise?" I am not interested in the internal debates of the U.K. Labour Party, so I am largely agnostic about the underlying issues being debated. I just want to react to some statements about the use of fiscal rules and the MMT/neoclassical squabbling. I think the neoclassical position is on thin ice -- and the fiscal rules championed by Wren-Lewis are questionable -- and relying on the "lower bound" to avoid obvious problems with fiscal rules does not really work.

The passage of interest is the following:

What about deficits? Isn’t good fiscal policy all about trying to achieve sustainable deficits (deficits that stabilise the government debt to GDP ratio)? As Jonathan Portes and I have argued, those rules should not apply when interest rates are stuck at their lower bound. When interest rates are at their lower bound, fiscal policy should become the primary tool for macroeconomic stabilisation. When interest rates are at their lower bound, standard academic macro and MMT should be on the same page.

The first thing to note that pretty well any fiscal policy will stabilise the the debt-to-GDP ratio eventually. The debt-to-GDP ratio can only go to infinity in a completely implausible model, as I discussed in a recent post. Any model that suggests that the debt-to-GDP ratio can go to infinity should be rejected.

If we put that aside, we see the somewhat familiar story that neoclassical macro and MMT are on the same page at the lower bound for interest rates. (One may note that Wren-Lewis buries some condescension into what appears appears to be conciliatory text, by contrasting "academic macro" and "MMT", implying that the academics working on MMT are not in fact academics.)

However, this defense of fiscal rules is implausible. In order for a fiscal rule to have any meaning, it has to affect policy in the present. Saying that the rule will be applied on an ever-receding horizon means that the rule is not a rule, and is at best an aspiration. Aspirations are nice, but a waste of time.

It is not possible to just suspend the rule if the policy rate hits 0% (or whatever the central bank decides the "effective lower bound" is). The implication is that central bankers could jerk around the entire apparatus of the state by hiking the policy rate by 25 basis points. Instead, we need to have some fuzzy notion of interest rates being "low," and so the fiscal rule needs to be relaxed if there is a danger that tightening fiscal policy would result in a reduction of activity that will force the policy rate back to the lower bound.

If one takes a look at time series of the previous cycle, that situation described the entire cycle. Furthermore, there is no reason to expect that to change any time soon. So we are supposed to take seriously a "rule" that is probably going to be suspended on almost any plausible forecast horizon?

The final issue is that the fixation on deficits within Wren-Lewis' work is mistaken. The only plausible economic concern of loose fiscal policy is higher inflation, and as 2020 is likely to demonstrate, the linkage between fiscal deficits and inflation is tenuous.

There are three obvious concerns with fiscal rules.

1. They solve a non-existent problem (the debt-to-GDP ratio going to infinity).
2. They require structural changes to the economy to actually be applicable, and there is no obvious quantitative metric when the rule should be "turned on."
3. Since the real worry is inflation, any rules should be in reference to inflation. There is no solid link between deficits and inflation.

(c) Brian Romanchuk 2020