Understanding NGDP targeting
NOMINAL GDP targeting is not a new idea. It has an intellectual pedigree that goes back well before the crisis, but even if we just focus on the recent debate over changing Fed policy to targeting growth in the level of nominal output, we're talking about nearly 3 years' worth of public discussion. Scott Sumner started his blog in early 2009, was linked by Tyler Cowen just a few weeks later, and had the economics blogsphere debating intensely by the end of that year. It has taken a while for non-economist elites to notice, but the debate has been bubbling for a while. Neither has the American debate proceeded in isolation. Central banks pay varying amounts of attention to the path of nominal output, and some—among them the Bank of England—put quite a lot of weight on NGDP. But Kevin Drum writes that it's important to get the NGDP debate out in the open. I suppose that's right; I just figured that's what we'd all been doing for the past 30 months.
The trigger for Mr Drum's post was a recent Wall Street Journal op-ed which purported to call into question claims made on an NGDP target's behalf. What it mainly demonstrated was that quite a lot of journalists haven't paid attention to the debate over NGDP targeting. I supose that's to be expected. Those who have simply must do a better job explaining the contours of that debate to others.
Now is as good a time as any for me to do a little of that explaining. Mr Drum writes:
Matt [Yglesias] is right that one of the theoretical virtues of NGDP targeting is that it combines both employment and inflation into a single metric, which would make this question moot for policymakers, but it unquestionably does imply that during recessions the Fed would tolerate higher inflation. I think that's a good thing (as does Matt); [WSJ writer Evans doesn't]. But it's certainly a key issue that deserves plenty of public discussion.
Let's slow down here. Does an NGDP target imply greater inflation in recessions? Were the Fed to adopt an NGDP level target right now, most supporters of the policy would recommend that the Fed allow for a period of "catch-up", during which the economy would expand at an above-trend rate in order to make up some of the ground lost during the recession. This isn't a feature unique to NGDP targeting; advocates of price-level targeting would call for something similar. During a period of catch-up, inflation would probably run above the desired rate, as would real output growth. This actually isn't even inconsistent with an inflation-rate target. A central bank actually targeting an inflation rate should react to deviations above and below target similarly, suggesting that the Fed should be no more aggressive in fighting above-normal inflation than it was in fighting below-normal inflation; 2% on average is good enough. A period of catch-up NGDP growth and inflation is really only inconsistent with a policy of steady opportunistic disinflation or, to the extent that the two are different, of central bank incompetence. One way of understanding the push for an NGDP target, I think, is as a means to get the central bank to take its mandate more seriously.
Of course, it's worth asking why there is so much ground to be made up in the first place. One of the strongest points in favour of NGDP targeting, in my view, is that it implied a need for far more action from the Fed far earlier in this business cycle. People remember how aggressively the Fed intervened to prop up the financial system in the fall of 2008, but they forget how slow the central bank was to react to what was obviously a precipitous decline in the macroeconomy. The fed funds rate stayed at 2% from April until October of 2008. The Fed didn't ramp up its initial asset purchase programme above $1 trillion until March of 2009, at which point the economy had already lost some 6m jobs. Why the delay? One data point worth noting: the monthly core inflation rate was positive throughout 2008 and 2009. NGDP growth, by contrast, was already negative in the third quarter of 2008, and was sharply negative in the fourth quarter of that year, when total spending in the economy shrank at an 8.4% annual pace. A central bank with an explicit NGDP level target would have faced (appropriately) intense pressure to do much more much sooner than one with the Fed's present, vague focus on an inflation target as a means to broader macroeconomic stability.
Now, in a situation in which a central bank has credibly established an NGDP target, recessions would by definition be due to real shocks. In those cases, maintaining the target would mean higher inflation to go with lower real growth. So if the American economy is hit by a real shock, an NGDP target might mean inflation at 5% and zero real growth, rather than what we might observe today—inflation around 3% and a drop in real growth of perhaps 2%. I'm happy to have a debate about which Americans are likely to prefer, provided that we stipulate that in the meantime, the NGDP target is also preventing major episodes of cyclical unemployment. It's worth mentioning that given a positive productivity shock, an NGDP target would imply real growth above normal levels and inflation below normal levels. An inflation-targeting central bank, by contrast, might respond by adding more stimulus to an economy, potentially inflating bubbles.
Mr Drum continues:
Evans's other two points are worth thinking about too. It's true that the Fed has to pick a target no matter what it's doing, but NGDP is a new one with no track record. That makes it trickier to get a consensus about what the right figure should be, and consensus is important since the whole point of NGDP targeting is that everyone has to believe the Fed is really, truly committed to its target.
NGDP is not a new one with no track record. Countries have been keeping track of nominal output for ages, and most central banks, including the Fed keep a close eye on the path of NGDP either explicitly or implicitly, by putting weight on both inflation and real output in making policy decisions. This isn't some crazy new variable that's been dreamt up. Consensus and commitment to an NGDP target are no more or less important than they are for an inflation or price-level target.
And the question of whether the Fed can hit an arbitrary NGDP target is critical. Central banks have pretty time-tested mechanisms for hitting inflation targets, but growth targets are something different. There are plenty of economists who are skeptical that monetary policy alone can accomplish this.
Luckily for us all, the mechanisms available to hit an NGDP target are exactly the same ones used to hit an inflation target. The Fed will communicate its policy goals, then use the levers at its disposal to move NGDP to the desired level. When the fed funds rate isn't stuck near zero, that means the standard change in the fed funds target rate and corresponding open-market operations. If NGDP is expected to be too high, rates go up; too low, and rates go down. At the zero bound, the tools are the same ones the Fed has been using or saying it could use for the past three years. If you think the Fed can affect inflation, you think the Fed can affect NGDP; that's all there is to it. Now, you might argue that the Fed can't affect inflation, but that's an extremely difficult position to reconcile against recent data.
Mr Drum closes:
Finding some kind of mechanical monetary rule that automatically produces stable growth is sort of the Holy Grail of monetary economics, and we should subject any new proposed rule to plenty of tough questioning.
Perhaps there are people making extraordinary claims for NGDP targeting. In general, I think that most of its supporters consider it to be part of the evolution of monetary economics toward a greater understanding of how the central bank can best achieve macroeconomic stability. I don't consider NGDP targeting to be a panacea or a holy grail. I simply think it's likely to perform better as a policy goal than inflation over the long run, and much better in the rare but very costly economic disaster. And I tend to believe that the idea has grown in popularity not because of unreasonable claims made on its behalf, but because of the strength of the arguments in favour of it.
