Friday, September 27, 2013

Kocherlakota: A Puzzle

For those of us who knew Narayana Kocherlakota as an academic, his recent behavior is a puzzle. As a practicing economic researcher, Kocherlakota taught us some useful things - about asset pricing, about optimal taxation, and about monetary economics, for example. He advanced economic science in important ways, and it's fair to say that we are much better off as a result.

When Kocherlakota was appointed President of the Minneapolis Fed in October 2009, I think we all had high hopes. While a research group in any institution - academic or non-academic - can be fragile and highly-dependent on good leadership, a research department in a Federal Reserve Bank is potentially even more fragile. A Fed president has a lot of power, and Fed employees do not have tenure. Thus, the President in a regional Fed can, if he or she chooses, play a large role in shaping the direction of research, and the choice of personnel to do that research. In October, 2009, I think the consensus opinion was that the Minneapolis Fed under Narayana Kocherlakota was in good hands. It was expected that he would preserve the excellent research operation that had been built up over the course of close to 40 years - an operation that played some role in 4 Nobel prizes in economics, for example. As well, we expected big things from a top-of-the-line researcher, who could apply his skills in new and fruitful ways to the science of monetary policymaking.

I'll leave it to the people who actually work for the Minneapolis Fed to tell you what that's like. As for the contributions of Kocherlakota to monetary policy, I think it's fair to say that he has set us back rather than pushing us forward. Case in point is his latest speech. To summarize, here is what Kocherlakota said in his speech:

1. The state of the labor market represents a problem. The unemployment rate remains high but, more importantly, the employment/population ratio is very low and should be increasing.

2. Monetary policy can do something to solve this problem. Why? Because there exists a Phillips curve tradeoff. The inflation rate is currently lower than the Fed's 2% long-term target, and so we can have more aggregate economic activity and more inflation, and have the best of both worlds.

3. We can learn from the behavior of the Fed during the Volcker era. Kocherlakota argues that, in 1979, the FOMC embarked on the right kind of monetary policy for the time, which he characterizes as "goal-oriented," i.e. the Fed decided to reduce the inflation rate, through whatever means possible. What should the Fed do now? Reduce the unemployment rate and increase employment through whatever means possible.

Basically, Kocherlakota thinks that any lack of success the Fed has had in changing the state of the labor market is due to public perception that the Fed is not serious about it. He thinks the public does not believe the Fed is committed to the second part of the dual mandate, i.e. "maximum employment."

Since Volcker-era monetary policy plays an important part in Kocherlakota's argument, it's useful to review what happened in that period. First, it was widely-accepted by the late 1970s that it was feasible for a central bank to control the rate of inflation. It was also well-understood that the Fed was capable of reducing the inflation rate from levels that were deemed too high, to something lower. The debate was over how costly such an undertaking would be, and how long it would take. Some thought that a "cold turkey" approach was optimal, and "gradualists" thought disinflation should proceed slowly. Some thought that a key problem was the slow adjustment of inflation expecations; some thought that the central bank might have a hard time credibly committing to disinflation. Ultimately, the costs of disinflation turned out to be smaller than the gradualists expected. The 1981-82 recession was relatively severe, but not so long-lived, and inflation was brought down quickly. The Fed did not seem to have suffered an inability to commit.

Kocherlakota likens the Fed's current predicament to the one it faced in 1979, and advocates for a similar "goal-oriented policy." There are plenty of problems with that argument. First, the goal in this case hard to define. What does the Fed want and what should it want, in terms of labor market performance? Kocherlakota is unhappy with the current state of the labor market and thinks we should do something about it. But what ails the labor market exactly? There are more people unemployed and less employed than was the case in early 2007. Kocherlakota should be able to tell us how we'll know the state of the labor market is good, so that, when the time comes, we and the Fed can stop worrying about it. He is on the record as supporting a 5.5% unemployment rate as a threshold for future monetary policy actions, but in his talk he seems to want to focus on the employment/population ratio. What ratio does he "like" under the current circumstances, and why?

Second, in contrast to a goal of reducing the inflation rate in 1979, changing the state of the labor market may not be feasible for the Fed under the current circumstances. Kocherlakota told the Rotary Club in Houghton Michigan that feasibility was obvious. Why? Because the inflation rate is below 2%. Kocherlakota's notion seems to be that there is a policy, not already attempted, that can both increase the inflation rate and increase aggregate economic activity. That policy is forward guidance - forward guidance that is somehow so clearly and forcefully stated that everyone recognizes the Fed's commitment to its goal, and this becomes self-fulfilling. The problem is that the Fed has already tied itself in knots with its forward guidance. Forward guidance has now been changed enough, and is written in sufficiently vague and complicated terms, that those words in the FOMC statement have become meaningless. This is why the Fed had trouble with its communications, beginning in May of this year. Faced with something more tangible - talk about "tapering" - markets responded.

Third, the goal itself may make no sense. As I discuss in this post, more than 5 years after the Lehman collapse, it's highly unlikely that much or any of the low level of employment in the U.S. is explained by wage and price stickiness in the face of the financial crisis shock that we suffered. If there's an inefficiency that explains why we are seeing low labor market activity, it's something other than price/wage stickiness - something that we understand imperfectly. Whether or not monetary policy can do anything about that is something we don't know.

The Fed does indeed have a credibility problem. That credibility problem comes in part from implementing policies - quantitative easing (QE) in particular - the effects of which are imperfectly understood by economists inside or outside of the Federal Reserve System. We don't know what mechanism is at work, we don't have any idea what the quantitative effects are, and yet Fed officials confidently support the use of QE in public, as if they knew exactly what is going on. Further, it is well-known that the real effects of monetary policy are at best temporary, but Fed officials like Kocherlakota seem to want to argue that the failure of policy to "cure the problem" is just a license for doing more.

The interesting part of the comparison to 1979, is that the tendencies we see in someone like Kocherlakota are similar to the ones that got us into the predicament that the Fed set out to solve in the Volcker era. But now, as I argued in this post, the policy error in 2013 may be that the Fed creates too little inflation rather than too much.


  1. If I could ask Kocherlakota one question, it would be:

    -The Fed reduces its UE threshold to 5.5%.
    -With UE stuck at 7%, inflation exp. hit 3%.

    What is the impact on the economy and markets of the Fed tightening to contain inflation?

  2. "If there's an inefficiency that explains why we are seeing low labor market activity, it's something other than price/wage stickiness"

    Perhaps the answer is overly tight money. This is evidenced by the lowest Core PCE readings over the past four years that the US has seen in six decades. Wage growth has been slow.

    I'm not sure I'd agree with you that one should either look only at the unemployment rate or only at the employment to population. They both give slightly different information. I'd also disagree that the problem with "goal oriented" monetary policy is that goals in terms of inflation and unemployment are hard to define. Kocherlakota didn't have any problem.

    1. "Perhaps the answer is overly tight money."

      So, what would easy money be, then?

  3. I think the 1981-1982 recession was short-lived in part because of the rapid decline of oil prices after 1982. Had they stayed at their 1980s level, I am not sure how things would have worked out. Kocherlakota is underestimating the importance of luck in the form of positive supply shocks.

    In any case, the premise among the Fed's leadership these days seems to be that it does not matter if the Fed can do anything to meet its full-employment objective, what matters is that people believe that it can. And, Steve, with posts like this you are not helping! :p

  4. How did it come to pass that a simplistic 3-equation model was somehow all we need to know about monetary policy and the economy ? Remarkable. Perhaps the new generation of input-output models based on Long-Plosser will provide a serious microfoundation for aggregate issues

    1. Somebody could write a book about that. Macro is going in many different directions, and I think that is good. I don't think we will have, or should have, one model to address all aggregate issues.

  5. Talking about research that originated at the Minneapolis Fed, the other day I run into this:

    I quote from the Abstract: "There is no need to use inefficient policies such as wasteful public spending or future commitments to low interest rates"

    So it seems that even within the New-Keynesian framework, there are policies that have not yet been tapped, which can circumvent the zero lower-bound. At the same time, writing this on the first day of the shut-down, one must wonder if the tax reforms suggested in the paper are politically feasible. I think economists are working hard to design new tools, but politicians are incapable of using them.

  6. Those who cannot change their minds cannot change anything.

  7. Presumably this justifies your skepticism about the actions of the Fed post-2009:

    Seems that following a drastically flawed paradigm - New Keynesian econ which is Old K dressed up in new garb - is ultimately the reason for the Fed's failed policies

  8. Steve:
    We need the deets on the NK/PK showdown. We are anxiously awaiting your next post.