Wednesday, July 23, 2014

Insure Against Central Bank Incompetence: My Reply to the New Keynesian Strikeforce

Apparently, it is open season on market monetarists. The red dots you see on our chest are from the laser scopes on the new keynesian guns of Tony Yates, Simon Wren-Lewis, and Paul Krugman. These individuals have been firing away with their critiques of market monetarism. Scott Sumner and Nick Rowe have already responded to many of them. Here I want to focus on what I view as one of their better criticisms: even if we are correct that monetary policy alone can end the slump, central banks have not shown themselves willing to do so. So why argue for them to do more? Here is Simon Wren-Lewis making this point:
MM agrees that fiscal stimulus will work unless it is actively counteracted by monetary policy. Nick says we can't always rely on fiscal policymakers being able and willing to do the right thing. But since at least 2011 we have not been able to rely on monetary policymakers in the Eurozone to do either the right thing, or consistently the wrong thing.
I think this is a fair point. The Fed failed to unload both barrels of its gun--by signalling its monetary injections were to be temporary--over the past five years while the ECB actually tightened monetary policy in 2011. Given this reality, the new keynesians want to know why not use fiscal policy? They contend that if monetary policy is too timid or tight, surely fiscal policy could make up for its shortcomings. 

My view, and one that I believe is shared by most market monetarists, is that fiscal policy will not matter much as long as these two central banks are committed to their inflation targets. Any surge in aggregate demand created by fiscal policy would be sterilized by the central bank if it pushed inflation too high. And by all accounts both the Fed and ECB take their low inflation targets seriously. 

In the case of the Fed, it seems to be aiming for core PCE inflation to fall between 1% and 2%. Any fiscal stimulus that pushed inflation outside this corridor would probably be offset. That is why I argued that had there been no American Recovery and Reinvestment Act of 2009 the Fed probably would have taken more expansionary steps back in 2009. It also why the Fed offset the effects of fiscal austerity of 2013. The Fed, then, appears to be doing just enough to maintain its corridor inflation target, which is nowhere near enough to close the output gap. Fiscal policy is bound to be offset in such an environment.

So what is needed is a better way to do macroeconomic policy. One that would allow monetary policy to close the output gap and, in its absence, allow fiscal policy to do the same. I have a proposal does just that. It is a two-tiered approach to NGDP level targeting:
First, the Fed adopts a NGDP level target. Doing so would better anchor nominal spending and income expectations and therefore minimize the chance of ever entering a liquidity-trap... [I]f the public believes the Fed will do whatever it takes to maintain a stable growth path for NGDP, then they would have no need to panic and hoard liquid assets in the first place when an adverse economic shock hits.

Second, the Fed and Treasury sign an agreement that should a liquidity trap emerge anyhow [say due to central bank incompetence] and knock NGDP off its targeted path, they would then quickly work together to implement a helicopter drop. The Fed would provide the funding and the Treasury Department would provide the logistical support to deliver the funds to households. Once NGDP returned to its targeted path the helicopter drop would end and the Fed would implement policy using normal open market operations. If the public understood this plan, it would further stabilize NGDP expectations and make it unlikely a helicopter drop would ever be needed.
The nice thing about this proposal is that it provides insurance against central bank incompetence. Scott Sumner initially did not like this proposal, but as he loves to say the fiscal multiplier is nothing more than an estimate of central bank incompetence. That is, the fiscal multiplier is large only when the central banks fail to properly stabilize aggregate demand. Helicopter drops are fiscal policy and in this proposal they would be applied only when the Fed failed to stabilize demand. Therefore, it is a perfect fit. Employ fiscal policy only when it packs a punch and do so in a manner to preserve a NGDP level target. This should make both market monetarists and new keynesians happy.

Note that this approach with its NGDP level targeting implies a commitment to permanent monetary injections, if needed. It, therefore, holds up against the critiques of helicopter drops that Paul Krugman, Scott Sumner, myself, and others have raised. It would also provide a more systematic approach to monetary policy, a big improvement over the current ad-hoc approach of the Federal Reserve. This increased certainty by itself would be a boon to the economy. Finally, it would eliminate the need for the politically-charged fiscal stimulus spending programs. There is much to like about this proposal on both the political left and right.


  1. As an amateur coming at it from the left, I'm more sympathetic to MM than the New Keyensian strike force. Like Rowe I don't want to let central bankers off the hook. I could go with the two-tiered proposal but would be in favor of more government spending. I'd be in favor of an NGDP level target and a futures market as a priority but then as a lefty would want government to allocate more of the new money and demand. Today Sumner says there wasn't a housing bubble, just tight money and I sort of disagree. This might align me with DeLong. The Fed was overly concerned that inflation wouldn't moderate in 2007. The bubble was deflating but then the credit system froze. If you read the Summers memo to Obama on the ARRA stimulus which lists economists recommendations, "senior Fed authorities" recommended an $800 billion fiscal stimulus. They didn't talk about offset. They're uncomfortable with QE which is why I believe they ended prematurely.

    But the amount of government spending employed to reach the NGDP level (in concert is a political decision and can be reached democratically. The Central Bank can accommodate. If done well government spending can work against the tendencies of capitalism to increase volatility and inequality.

    I'd argue deregulation (and the rise of an unregulated shadow banking system) and the over-reliance on monetary policy led to the housing bubble. Yes better Fed policy could have allowed the bubble to deflate without a panic, which it had started to do. (See the tech stock bubble) But still the economy should be made not to be so top-heavy and fragile that we need a super competent Fed. Dodd-Frank and more government spending done well could make it more robust.

  2. David, I love this approach. MM as long as it works, helicopter drop if it doesn't, it's beautiful.

    How do we make this happen? I am hoping for a plan from you and/or Sumner about how to get traction for these ideas.

    Kenneth Duda
    Menlo Park, CA

  3. David you write:

    "The Fed failed to unload both barrels of its gun--by signalling its monetary injections were to be temporary..."

    But if recessions are interpreted to be caused by temporary shocks to AD, won't any resultant offsetting monetary response from the CB also be expected to be temporary?

  4. "This should make both market monetarists and new keynesians happy."

    Isn't it convention to capitalize the latter, as in New Keynesian (or at least new Keynesian)? Or is this a case of 'if you won't capitalize us, then we won't capitalize you'? ;)

  5. To Tom Brown: the question is really: what is temporary? Is it the amount of bonds the CB buys? Is it the amount of bonds it's buying? Or is it the change in the amount of binds it's buying? Unfortunately the fed signaled it's the first of these (i.e. that in time it will sell them again), which is the least stimulating position, and has barely any traction (though fortunately not nothing.) It should at least have signaled that it was intended to either keep the bonds indefinitely (to maturity) or even keep buying the same amount, at least till the zero bound had receded to a memory.

    1. This comment has been removed by the author.

    2. Christiaan, I did not have the impression that the Fed would sell the bonds before they matured. (But then what does it matter what I thought? I'm nobody important). Still, can you point to where the Fed signaled that? After six years It's still buying isn't it? Tapering, but still buying. Isn't that a signal on its own? Even if they did originally signal what you say, surely they've signaled a different message by now haven't they?

      If we hypothesize that the concrete observable actions (OMOs) of the CB are not sufficient on their own for effective monetary policy, but instead must be accompanied by factor X (call it the X hypothesis), then we have to be able to independently observe X outside of looking to see if monetary policy is effective, if the X hypothesis is to be testable. Otherwise X could be anything and the hypothesis would "work" but it would also be useless. Is there a way we can measure the sufficiency of the permanence conveyed by the CB *outside of* observing whether or not monetary policy is effective?

      I know of at least one example of a hypothesis that attempts to explain the effectiveness of monetary policy across different economies and time periods which does have an independently measurable X which allows the hypothesis to be tested (I can provide a link if you're interested), so I know this kind of thing can be done.

  6. In your post where you talk about the fed signaling monetary injections were to be temporary the chart you post shows some portion of injections are expected to be permanent, the curve is upwards sloping:

    Are you saying that under ngdplt the curve would be more vertical?

    Under ngdplt what is to stop the fed from targeting a corridor of 2-4 % gdp growth sort of like it does now with 1-2% corridor inflation?

  7. ah, so very American. Pop the magic pill, target that GDP and all will be well!
    If only it were so simple. It never is.

  8. "Second, the Fed and Treasury sign an agreement that should a liquidity trap emerge anyhow [say due to central bank incompetence] and knock NGDP off its targeted path, they would then quickly work together to implement a helicopter drop."

    A central bank that did not do the right thing in 2008 due to inflation targeting is not going to agree to this. Might as well propose a plan to:
    Have the Fed sign an agreement that should a liquidity trap emerge anyhow [say due to central bank incompetence] and knock NGDP off its targeted path, they would then quickly drop their inflation target and do the right thing to bring NGDP back to level target.

    When it's not working it's silly to propose solutions that assume away the original flaw.

  9. David advocates coordination between Fed and Treasury at the zero bound. There’s a thoughtful article by Scott Fullwiler written a year ago claiming that such coordination is irrelevant. I think Scott’s article is flawed – for reasons I’ll probably put on my own blog in a day or two. But if anyone else has thoughts on Scott’s article, I’d be interested. See: