Tuesday, October 12, 2010

A Bombshell in the FOMC Minutes

Contrary to what some observers are saying, the FOMC minutes released today reveal a big change in terms of policy options being discussed by the Fed.  For the first time the FOMC has discussed the possibility of targeting the level of NGDP.  Here is the key excerpt (my bold):
With short-term nominal interest rates constrained by the zero bound, a decline in short-term inflation expectations increases short term real interest rates (that is, the difference between nominal interest rates and expected inflation), thereby damping aggregate demand. Conversely, in such circumstances, an increase in inflation expectations lowers short-term real interest rates, stimulating the economy.Participants noted a number of possible strategies for affecting short-term inflation expectations, including providing more detailed information about the rates of inflation the Committee considered consistent with its dual mandate, targeting a path for the price level rather than the rate of inflation, and targeting a path for the level of nominal GDP.
This is huge.  The FOMC is now discussing a NGDP level target, a topic that has has been promoted in the economic blogosphere. For those of us who have been advocating this approach for some time (e.g. here and here) this is incredibly refreshing.  Maybe they are listening to us after all.  Lest you think I am reading too much into the above excerpt from the FOMC minutes, here is what David Pearson, a money manager who follows the FOMC closely, had to say about it:
[T]he Fed is careful in how it chooses its words in both the statements and minutes. Minority views are often referred to in the context of “a few participants noted…” or, “one participant commented that…” Today’s minutes tell us that, “Participants noted a number of possible strategies…” Notice the lack of a qualifier. That this was used to describe the discussion of both price and NGDP level targeting implies a surprising degree of consensus and openness around something never floated before in the minutes. Further, when someone pounds the table and says, “no way”, the minutes usually will say something like, “one participant expressed a concern that adopting such strategies might risk unanchoring expectations…” To do otherwise would be to pretend an strong objection was not raised.

So apparently they all sat around a table discussing that price and NGDP level targeting might be a good thing, and no one blew a gasket. This, I believe, is news.
Unfortunately, we will have to wait until the transcripts are released in six years to learn who actually was discussing NGDP level targeting.


  1. "Unfortunately, we will have to wait until the transcripts are released in six years from now to learn who actually was discussing NGDP level targeting."

    Meanwhile, we can play a guessing game on who in the FOMC is promoting this view.

  2. What is the significance of the Fed calling out the difference between "targeting a path for the price level" and targeting"the rate of inflation"?

    And is there any significance that they said they'd target "...a path for the level of nominal GDP" but didn't speak about the rate of GDP growth?

  3. sdwinkler:

    A price level target has "memory" while an inflation target does not. Thus, if the Fed has a price level target and it falls beneath its target it will create enough extra inflation to catch up with the target and vice versa. On the other hand, if the Fed has an inflation target and it falls beneath the inflation target for a period there is no subsequent catch up in the price level. Only the inflation rate returns to its target. Michael Woodford has a nice OpEd on this distinction today in the FT.

    The same idea applies for a level NGDP target versus a growth rate NGDP target. Note that a NGDP target is nothing more than a total current dollar spending target. Thus, a level NGDP target would aim for a target dollar level for the economy each period. And when it falls beneath target there is a catch up to the level target. A NGDP growth rate target would only return to its target growth rate if there were a departure. It would not return NGDP to its previous level path.

  4. Excellent commentary. Keep up the good work.

  5. If a general gives the command to her troops to "take that hill," but does not have any idea how the hill should be taken, in all likelihood, it will not be taken. If the Fed decides to target nominal GDP but does not know how to achieve its target, in all likelihhod, its target will not be achieved. It is not clear to me the Fed knows how to achieve a nominal GDP level or growth rate.

    Paul Kasriel

  6. David,

    I understand from Bloomberg that Eggerston attended the Sept. FOMC meeting. I wonder if he gave a presentation on targeting...

    Reading his 2003 paper, it seems that his main messages are, "QE has little direct effect, even if the central bank buys risk assets." And so, "do more targeting, less volume of bond purchases." Do I have this right?

    I find this important because people like Hoening are upset over the ad-hoc nature of QE, as well as the open-ended commitment to ZIRP. Targeting is a way to address both concerns. Not that Hoening can be convinced -- more that his, and Fisher's, objections would be partly addressed, and thus blunted, with targeting. I find it interesting that Janet Yellen also recently bemoaned the distortions caused by ZIRP. So again, moving away from rate/QE targets would be welcomed by many on the FOMC.

    Friday's speech, given the forum, offers an opportunity for Bernanke to focus on targeting. Perhaps we will see the "old" Bernanke again.

  7. David, I do hope you took Paul Kasriel's comment to heart. There is no thing except a very loose connection,working through unpredictable expectational effects, with long and variable lags,between any Fed instrument and nominal GDP.
    You are dangerously ignoring the Hayekian warnings of pretense of knowledge I fear.
    Of course, in a vacuum such as we are in now, where nobody has a clue how to pull America out of its funk, many ideas will clamor for policymakers attention. The best that can be said of the NGDP target group is that it may not be as damaging as some.

  8. Paul Kasriel,
    NGDP target will be achieved because economic agents will know that the Fed will set the interest rates and money supply to the levels consistent with the desired NGDP level path.
    At this time nobody knows what criteria the Fed is using when it sets the size of Fed's balance sheet and when it determines the time interest rates will be increased.

  9. We're looking at politics influencing economic decisions right now, that's all.

  10. Richard A:

    Maybe we can learn if Bernanke was one of them from his Friday speech.

    David Pearson:

    I think Eggertson's key point is that expectations matter. I agree. If the Fed would state an explicit NGDP level target and declare it would do whatever is needed to maintain it, such language would change expectations and require fewer asset purchases than would otherwise be needed.


    FDR in the early 1930s was able to spur a recovery using unconventional monetary policy in far worse economic circumstances. Yes, it is only one data point but at least know it can work. Ultimately, though it would help if the Fed help set up a NGDP futures market and promised to target NGDP futures contract.

    Paul Kasriel:

    I suspect a clearly communicated message of the Fed strategy--whatever its target may be--would go a long way in forming a successful charge on the hill.

  11. I would question whether these ideas should be taken so seriously. For reasons good or bad, the Fed wants to ease more aggressively, but the inflation mandate makes that awkward, so they look for something else that gives them cover for doing what they want. Do you think that the Fed would stick to a price level target if it meant a period of unpopular restriction to unwind an inflationary overshoot? Or stick to an NGDP target in a boom when the cry would be that the economy had entered a "new paradigm" and the Fed should "give growth a chance"?

    The Fed get more respect from thoughtful commentators than they deserve. They are as much politicians as technicians.

  12. DB: Yes, if you're talking about QE2: One of the most convincing things in all of econ history is the graph in Eichengreen that shows a country's recovery in the 1930s coincides with its jettison of the Gold Standard.
    So yes monetary policy can have a powerful one time effect. But we need to consider that was an expectations shattering event precisely because it was against the grain of monetary orthodoxy at the time.

    Fast forward 80 years to a world where markets have become hooked on monetary expansion and its not clear we'll get the same effect (this after all is what Friedman said in 1968)
    Also concerning is that yet another exercise of the Bernanke put will ramp up excessive risk-taking setting the scene for another crisis in the future.

    Thoroughgoing financial reform is the only way to cure what ails the US economy. Have you looked at Kotlikoff's Limited Purpose Banking? Seems like a good guideline for discussion.
    I took it that a key Austrian theme is that we should favor institutions that limit the extent of damage resulting from the inevitable error in people's plans.