For the past 4 years I’ve been arguing that we should generally assume a zero fiscal multiplier, except under very unusual circumstances. Brad DeLong makes a similar point in a recent post:
In trotting around the country giving versions of DeLong and Summers (2012), “Fiscal Policy in a Depressed Economy”, I have found that a point that seemed completely obvious to us is not obvious at all to many.
Here is the point: an optimizing central bank that cares only about inflation and unemployment because it does not find itself at the zero nominal lower bound and does not fear engaging in nonstandard monetary policy will engage in full fiscal offset: it will take care to make sure that if fiscal policy becomes more stimulative then it will make monetary policy less stimulative by the same amount.
Then DeLong provides a mathematical proof, before concluding:
The important point here is that m and g cannot enter into the objective function h directly, but only indirectly through their effects on inflation and unemployment.
For this reason this argument breaks down at the zero nominal lower bound. At the zero lower bound the central bank does [not?] care only about inflation and unemployment. It cares as well about the magnitude of the non-standard monetary policy measures it must take in order to achieve its net monetary policy impetus value m.
I think this is a very important point. So let’s review the implications:
1. Hydraulic Keynesianism is pretty useless—it tells us almost nothing about the monetary policy reaction function. And that includes the “paradox of thrift” as well as models that evaluate the prospects for NGDP growth by looking at each sector (C, I, G, NX) in isolation.
2. Empirical studies of fiscal stimulus that rely on cross-sectional data—particularly the local or state-level response to fiscal stimulus, are extremely misleading, as they fail to address the monetary policy reaction function. Ditto for national studies within the euro. They will grossly overestimate the aggregate fiscal multiplier. This includes studies cited approvingly by Krugman and DeLong.
3. Empirical estimates of fiscal multipliers are nothing more than estimates of central bank incompetence. This means that “the” multiplier will depend on whether Bernanke had a bad nights sleep, or whether a member of the Board of Governors has a world class ability to be open-minded. No mathematical model or empirical study can ever produce a reliable fiscal multiplier estimate.
4. Because it is objectively false that the “costs and risks” of unconventional monetary policy are greater than the costs and risks of fiscal stimulus, most central banks have the wrong policy, and need to be instructed to target AD more aggressively. In particular, the GOP should block any fiscal stimulus until the Dems agree to instruct the Fed to do all it can to maintain on-target AD. And after that happens the GOP should block any fiscal stimulus since it would be useless. (Yes, I know that the GOP doesn’t even favor an appropriate track for AD, I’m just saying that if they ever did come to their senses, then they should continue to block fiscal stimulus and instruct the Fed to “do the right thing.”)
5. If the Fed uses QE to keep core inflation in the 1.7% to 2.0% range, but would allow higher than 2% inflation if it could be achieved with fed funds rate cuts, then fiscal stimulus might be effective. But it also might not be effective. For example, if the Fed does more or less QE as required to keep inflation within the 1.7% to 2.0% range, then marginal changes in fiscal stimulus might have little or no effect, although large changes might lead to higher than 2% inflation, and hence higher AD. It depends on the type of unconventional monetary policy, and how the perceived cost of that unconventional policy varies with fiscal stimulus. For instance, a Woodfordian “communication policy” might be seen by the Fed as having a fixed cost. Or it might not.
6. In countries where the central bank holds back due to public criticism of above target inflation (I’m looking at you Britain) fiscal stimulus and austerity will have no effect on RGDP. If employment hits record highs while RGDP lags, then “it’s the productivity, stupid.”
I hope other bloggers will adopt DeLong’s enlightened approach. Then we might finally be able to have a sensible debate over fiscal policy, instead of a inane shouting match between intellectually bankrupt “paradox of thrift” arguments and empirically unfounded “crowding out” arguments.
PS. I assume there was a typo in the DeLong quotation, which is why I added “not?”
HT. Lars Christensen