Tuesday, June 23, 2015

Scott Sumner Takes it Down a Notch

     Wow. Sumner actually goes there-the 'C' word. Not conservative, we already know he is conservative. Rather conciliatory-he writes a post that seems to clearly be an attempt at something of a peace offering.

     I have to presume that his benefactor Ken Duda got to him a little. I have to admit finding Duda somewhat fascinating. Yes, he's barnkrolling Scott's experiment but he's also donated to and on most policy issues he and I probably agree quite a lot of the time.

   Also, recall that when I grow up I want to be a liberal billionaire.

    "Seriously, I too dream of being a liberal billionaire. And though I am still very, very far from this goal, those of you who know me a little-because you've read me for some length of time-know I'm in much better shape than I was during the years 2009-2014 of living in my parents basement and desperate for any sort of gainful employment."

   "If I were a LB tomorrow I could go to places like Greg's Georgia, Texas, or South Carolina and turn the place upside down overnight simply by making a $50 million dollar contribution to the state Democratic party so they can campaign to retake the state. Well, starting with SC might be a little ambitious but Wisconsin or Ohio could be done quite rapidly and in Georgia and similar Southern states probably quicker than you could imagine. Money really does talk. As Marx himself said it has miraculous qualities."

     I'd also donate millions to Planned Parenthood in Texas, Kansas, and North Carolina-the states where a woman's right to choose is really imperiled. In Texas, her right to an abortion may still hypothetically exist but there are at best 2 clinics left in the entire state that offer abortion services. 

    Duda is an almost romantic figure; a guy with deep pockets who wants to better understand and improve our world. 

   It strikes me that Sumner is lucky to have him as a benefactor beyond the obvious reason that he's loaded. Despite the many negative things I've said about Sumner over the years-not necessarily undeserved-I've never been blind to his virtues. 

  One of them has never been diplomacy. Like most conservatives, Sumner seems not to have any use for a State Department-everything should be handled through Defense, apparently. Duda though has diplomatic skills in spades. 

  He seems to have finally convinced Sumner that a little diplomacy might work. So Sumner now recommends a stabilization commission. This is a major concession for him right away because it admits that when discussing the optimal stabilization program you can't simply ignore politics no matter how 'independent' the Fed is alleged to be. 

  "I know, you are gagging on the title of the post.  I hate commissions too.  But there’s already a lot of discussion about a commission to re-evaluate the Fed’s goals and tactics.  And the current proposals are both too much and too little.  Too much because there are some tactical questions that the Fed itself can resolve better than any commission.  But there are also some questions that the Fed currently cannot answer, and where a commission could be very useful to the Fed. I believe the biggest issue now is what to do about stabilization policy in a world that frequently hits the zero bound during recessions. That’s not the world of the past 50 years, but I believe it’s quite likely to be the world of the next 50 years."

  "Both liberal and conservative economists agree on these basic facts:

   1.  When trend NGDP growth rates are lower, the economy will hit the zero bound more often.  One option is to raise the inflation target.  The Paul Krugman solution.

   2.  Another option is to do something like NGDPLT.  My preferred solution.

   3.  Another option is to keep the Fed’s current policy framework, 2% PCE inflation, growth rate targeting, and unemployment near the natural rate.

    "Economists also agree that option three may require some hard choices.  These include:

   "a.  Pursuing QE to the limit in a liquidity trap.  Allowing the Fed to buy whatever it takes, even if they have to move beyond Treasury debt.  Telling the Fed not to worry about capital risk, the Treasury has them covered.  My second preference."

  " b.  Constraining the Fed to buy securities of no more than a specific amount, say 50% of GDP, to avoid excessive risk.  Other options are also possible here, such as more aggressive cuts in IOR, perhaps to negative levels.  Then just live with a slow recovery.  Similar to current policy."

  "c.  Same as option b, but have an implicit agreement that once the Fed hits its QE limit, fiscal stimulus will take over.  The Larry Summers solution, Krugman’s second preference."

   "Policy is currently hindered by the fact that the Fed doesn’t know exactly how aggressive it should be, partly because Congress is not even aware of these “hard choices.”  So we don’t have any sort of clear policy regime, rather we drift in a sort of limbo, where the Fed doesn’t really know how much others want it to do.  Or whether it would be scolded for large capital losses on its balance sheet if rates rose sharply.  Or whether Congress would support the Fed if it shifted its target higher in order to keep interest rates above zero.  The Fed knows that politicians are concerned that rates are low for savers, but doesn’t know if that concern implies they’d favor higher interest rates that are caused by higher inflation."

    "I don’t think this commission is politically feasible until January 2017, but at that time it just might work. I’m assuming the Dems will again win the presidency and the GOP will retain the House.  Gridlock will make fiscal policy impossible unless an agreement can be reached.  If you put sensible conservatives like Taylor, Mankiw and Hubbard on the committee, with sensible Keynesians, they are all going to understand the trade-offs I discussed above.  The GOP economists can explain to GOP politicians “look, it’s inflation or socialism, take your choice.  If we don’t have a bit more inflation then interest rates will fall to zero, and the Fed will keep expanding its balance sheet, bigger and bigger.”  Or we’d get fiscal stimulus, another option the GOP doesn’t like.  The liberal members of the commission can explain to Democrats “look, it’s better if the Fed handles stabilization policy, and fiscal resources are utilized for pressing social needs, not economic stabilization. And in any case, the GOP will never let us do the amount of fiscal stimulus we need, or they’ll insist on tax cuts that ‘starve the beast’.”

   "Krugman and I may not get our way.  Maybe the commission will compromise on a monetary/fiscal mix, where the Fed takes the lead, but the fiscal authorities act if the Fed ‘s balance sheet hits X% of GDP.  If I lose the battle I’ll stop objecting to fiscal stimulus.  I’ll stop claiming the multiplier is zero.  I’ll stop claiming there is monetary offset.  If that’s clearly the regime, and it’s all spelled out, then so be it. At that point I’ll argue that payroll tax changes are the best form of stimulus."

     I think his political assumptions are right. It will be Hillary and the GOP House in 2017 and obviously nothing will get done until the election is over. I think he premise is also right that even if the Fed could fix it by itself-do NGDPLT or a 4% inflation rate-or even declare publicly that it needs to hit 2% inflation as the rate is barely over 1 it won't if it doesn't feel it has explicit permission from Congress. As independent as it is suppose to be it is a creature of Congress with the Chairman serving at the pleasure of the President and it never forgets this. 

     "But right now there is great uncertainty about who is in charge, and what is expected of the Fed.  This stuff really needs to be clarified for the zero bound environment.  Or at least discussed.  I’ll bet the Fed would be thrilled if Congress told them exactly what their responsibilities were in terms of capital losses, instead of leaving it quite vague."

   "What would Congress decide in the end?  One possibility is keeping the 2% inflation target, and a continual role for fiscal policy.  That’s very possible.  Or Congress might ask the Fed to study options for preventing the zero rate bound from hamstringing monetary policy, and they might buy into a technical fix like level targeting and/or NGDP targeting. I don’t know.  But politics goes in cycles.  After so many years of gridlock, 2017 might be a good time for a compromise.  To make this happen we all have to starting talking up the idea right now—assuming anyone agrees with me."

     The most optimistic part of the scenario is that Congress will compromise. It's hard to imagine the GOP House ever compromising as long as they have the power. However, if Hillary does win as both Sumner and I presume then maybe this might sober them just a little bit. They may realize that they could be in the wilderness for awhile if they are never seen as being constructive. 

   To get an idea of Ken;s mindset I find this comment he left here on Sunday telling:

   "NGDPLT is so beautiful. It appeals to my software engineer sense that the right algorithm structure can simply solve problems that are intractable when you have the wrong structure. Inflation targeting is the wrong structure. It leads to disasters like the NGDP collapse of 2008-2009. It didn't have to happen. NGDPLT would have prevented it, or at least signaled that it was coming, and the Fed would have pulled out all the stops and then gone to congress with quantitative fiscal stimulus requests."

  So maybe the way forward is to be found in a software engineer's sense. I do think there is truth that often the solution to seemingly intractable and complex problems is counter-intuitively simply. 


  1. Thanks, Mike. I appreciate the kind words.

    You didn't mention what I thought was most interesting in Scott's post:

    > Maybe the commission will compromise on a monetary/fiscal mix, where
    > the Fed takes the lead, but the fiscal authorities act if the Fed ‘s
    > balance sheet hits X% of GDP. If I lose the battle I’ll stop
    > objecting to fiscal stimulus. I’ll stop claiming the multiplier is
    > zero. I’ll stop claiming there is monetary offset. If that’s clearly
    > the regime, and it’s all spelled out, then so be it.

    I think Scott lays out his position on fiscal stimulus more clearly than I've seen him elsewhere. It's not that it can never work. It's that it can only work if the markets are confident that the central bank will not tighten as a consequence, i.e., if fiscal stimulus is effectively incorporated into monetary policy.

    I continue to believe the fight between Scott and the Keynesians is mostly unfortunate misunderstanding. We have essentially the same model of the economy and the same goals. The difference in policy prescription reflects different in assumption about how the Federal Reserve and markets will respond to fiscal stimulus. No one can tell whose assumptions are right because the Federal Reserve's actual behavior is impossible to predict, and the behavior of markets is at the limits of human intelligence to comprehend (as required by the EMH of course). That's part of why NGDPLT is so important. With NGDPLT, we would all know what the Fed behavior would be, and we could all see the best forecast of market behavior possible, in real time. NGDPLT creates a framework where MM'ers and Keynesians would agree on when fiscal stimulus is required: whenever the prediction markets indicated that the Fed would fall short of its NGDP level target regardless of its policy instrument settings.


  2. Yeah, Ken, I had meant to include your quotes. I too thought that was very important. My omitting it was a mistake-I thought I had.