Thursday, November 14, 2013

A limited central bank

Philadelphia Fed president Charles Plosser gave a noteworthy speech, "A limited central bank." It's especially noteworthy in the context of Janet Yellen's nomination, discussion between Congress and Fed about how the Fed should be run, the Fed's focus on unemployment, and the current state of the hawks vs. doves debate.

We find out what he thinks of micromanaging the taper based on monthly employment reports:
The active pursuit of employment objectives has been and continues to be problematic for the Fed. Most economists are dubious of the ability of monetary policy to predictably and precisely control employment in the short run, and there is a strong consensus that, in the long run, monetary policy cannot determine employment....

When I talk to Fed types about this, the usual answer is a version of "well, yes, we don't really have that much effect on employment, but employment is in the toilet, we have to do what we can, no?"

Charlie has a good answer to that, along the way blasting his colleagues who want the Fed to continue to fiddle with long term bond markets, mortgage rates, credit spreads, credit "availability" and perceived bubbles:
When establishing the longer-term goals and objectives for any organization, and particularly one that serves the public, it is important that the goals be achievable. Assigning unachievable goals to organizations is a recipe for failure. ...

...We have assigned an ever-expanding role for monetary policy, and we expect our central bank to solve all manner of economic woes for which it is ill-suited to address. We need to better align the expectations of monetary policy with what it is actually capable of achieving.

...Even though the [Fed's] 2012 statement of objectives acknowledged that it is inappropriate to set a fixed goal for employment and that maximum employment is influenced by many factors, the FOMC’s recent policy statements have increasingly given the impression that it wants to achieve an employment goal as quickly as possible.
What should the Fed do?
I have concluded that it would be appropriate to redefine the Fed’s monetary policy goals to focus solely, or at least primarily, on price stability.
The speech is very thoughtful about independence. In a democracy, an agency can only be independent if it has limited powers. An agency that writes checks to voters, allocates credit to favored businesses and industries, cannot be politically independent.

The current deal for independence is written in part in the Federal Reserve act which sets up the current "dual mandate," but
The act doesn’t talk about managing short-term credit allocation across sectors; it doesn’t mention inflating housing prices or other asset prices. It also doesn’t mention reducing short-term fluctuations in employment.
You're getting a sense of what genies Charlie would like to put back in their bottles. It's a bit remarkable for a Fed president to essentially say that Fed policy is not only unwise, but stretching the Fed's legal authority.  Yet independence is a good thing:
Even with a narrow mandate to focus on price stability, the institution must be well designed if it is to be successful. To meet even this narrow mandate, the central bank must have a fair amount of independence from the political process so that it can set policy for the long run without the pressure to print money as a substitute for tough fiscal choices

Such independence in a democracy also necessitates that the central bank remain accountable. Its activities also need to be constrained in a manner that limits its discretionary authority. 
... in exchange for such independence, the central bank should be constrained from conducting fiscal policy... [yet] the Fed has ventured into the realm of fiscal policy by its purchase programs of assets that target specific industries and individual firms.
What would Charlie do to draw some lines in the sand? One, by reinstating traditional limits on what assets the Fed can buy:
One way to circumscribe the range of activities a central bank can undertake is to limit the assets it can buy and hold. My preference would be to limit Fed purchases to Treasury securities and return the Fed’s balance sheet to an all-Treasury portfolio. This would limit the ability of the Fed to engage in credit policies that target specific industries.
Rules are important,
A third way to constrain central bank actions is to direct the monetary authority to conduct policy in a systematic, rule-like manner. It is often difficult for policymakers to choose a systematic rule-like approach that would tie their hands and thus limit their discretionary authority.  
And for more reasons than usual: if the bank is following a rule, it's much less open to political criticism and able to preserve its independence:
Systematic policy can also help preserve a central bank’s independence. When the public has a better understanding of policymakers’ intentions, it is able to hold the central bank more accountable for its actions. And the rule-like behavior helps to keep policy focused on the central bank’s objectives, limiting discretionary actions that may wander toward other agendas and goals
...assigning multiple objectives for the central bank opens the door to highly discretionary policies, which can be justified by shifting the focus or rationale for action from goal to goal.
Charlie agrees: you can't have effective forward guidance without precommitment, and you can't have precommitment and discretion. Here is the slam at how taper talk roiled bond markets
My sense is that the recent difficulty the Fed has faced in trying to offer clear and transparent guidance on its current and future policy path stems from the fact that policymakers still desire to maintain discretion in setting monetary policy. Effective forward guidance, however, requires commitment to behave in a particular way in the future. But discretion is the antithesis of commitment and undermines the effectiveness of forward guidance. Given this tension, few should be surprised that the Fed has struggled with its communications.
In some sense, arguing about the dual mandate is the last war. The Fed is now the Gargantuan Financial Regulator, and the "mandate" includes "financial stability," and detailed discretionary direction of credit flows. Charlie:
Some have even called for an expansion of the monetary policy mandate to include an explicit goal for financial stability. I think this would be a mistake.

The Fed plays an important role as the lender of last resort.... the role of lender of last resort is not to prop up insolvent institutions. However, in some cases during the crisis, the Fed played a role in the resolution of particular insolvent firms that were deemed systemically important financial firms. .. by taking these actions, the Fed has created expectations — perhaps unrealistic ones — about what the Fed can and should do to combat financial instability.
In fact, the bigger the fire house, the more the chance of fires:
I can think of three ways in which central bank policies can increase the risks of financial instability. First, by rescuing firms or creating the expectation that creditors will be rescued, policymakers either implicitly or explicitly create moral hazard and excessive risking-taking by financial firms. For this moral hazard to exist, it doesn’t matter if the taxpayer or the private sector provides the funds. What matters is that creditors are protected, in part, if not entirely.

Second, by running credit policies, such as buying huge volumes of mortgage-backed securities that distort market signals or the allocation of capital, policymakers can sow the seeds of financial instability because of the distortions that they create, which in time must be corrected.
I would add, if you prop up prices in bad times, you kill the incentive for people to keep some cash around to buy in the next "fire sale."
And third, by taking a highly discretionary approach to monetary policy, policymakers increase the risks of financial instability by making monetary policy uncertain. Such uncertainty can lead markets to make unwise investment decisions — witness the complaints of those who took positions expecting the Fed to follow through with the taper decision in September of this year.
"You can keep your bonds if you like them?"

The whole speech is good, I hope my excerpts get you to go to the real thing.


  1. How do you feel about the NGDP school, that thinks that NGDP targeting combines limited central banking, economic stability, and reducing unemployment all in one?

    I don't understand the criticism that reducing unemployment, as a goal, has to set achievable targets; I got the impression that most economists think central banks can reduce unemployment but few think it can precisely "control" unemployment and there's no contradiction there. Achievable targets is not a necessary condition for making a goal worthwhile, in life and in economics. If you want to get as fit as possible, for example, you don't set milestones to give up after reaching, or give up after not reaching.

    1. The NGDP school has its heart in the right place, but does not deal with the central problem of monetary policy right now: its abject ineffectiveness. OK, you want NGDP higher. Now what should the Fed DO? 3 trillion QE didn't budge the thing. Is the magic bullet 6 trillion QE? 20 Trillion?

      This is like saying "all the Chicago Cubs need to do is to adopt a 'win the world series' target."

    2. NGDP is nominal (that's what the N is for!). Friedman managed to convince economists that the Fed has control over inflation through the money supply (both nominal variables). Real GDP is just NGDP minus inflation, and as long as we think it isn't an inverse function of inflation or something like that, the Fed should be able to hit an NGDP target. The Cubs having a "win the world series target" is more like an unemployment or RGDP target, since those are "real" rather than nominal.

      Regarding how much would be the "magic bullet", Sumner's proposal is an NGDP betting market. The amount of monetary expansion is determined by what the market thinks is necessary to hit the target. And since the market tends to jump like a market monetarist would expect in reaction to news of monetary policy, I expect that would work.

      My view of the Fed's recent track record is different from yours: The economy has been growing (albeit weaker than we'd like) even with more "austerity" than Europe. Mike Konczal & Paul Krugman were seeing we'd see a test of "market monetarism", and it looks like it passed. Why hasn't NGDP growth been even higher? Because there isn't a consensus for that at the Fed.

    3. Friedman lied. The Fed neither produces goods (downward pressure on prices) nor consumes goods (upward pressure on prices). The Fed is a bank that lends money. What happens to the prices of goods after the money is lent is entirely at the discretion of the borrower.

  2. I do find it odd that the Fed is given mandates to keep both (a) prices stable (i.e. ~2% growth/year) and (b) unemployment low. To me, this is sort of reminiscent of giving employees at a call center the twin goals of speed and customer service.

    That being said, though, all the data I've seen indicate that inflation has been running below 2% annually in recent years - so given that unemployment has been stubbornly high, it appears that the Fed is failing on both counts.

  3. It shows contempt for democracy. One way around this is to have the government set out the agenda before the election. If price stability is the main goal, so be it, the central bank obeys the elected government and does, like any other government agency, what it is told to do. If it fails and price stability is still considered a priority by the electorate, well like any other minister, they are out at the next election.

    Monetary policy is less important than something like external or internal security, which is felt able to be entrusted to democratic process. Monetarists, unsurprisingly, have a thing about money.

    Not giving a country's government control over its money supply and the full set of instruments associated with it and making an exception to the democratic process for it does not make any sense at all.

    There is only one institution whose decisions should arguably be independent of government: the courts.

    PS You said earlier people should not read Aristotle. But actually his tutor, Plato was very good on this subject (although he would not agree with me).

  4. "I have concluded that it would be appropriate to redefine the Fed’s monetary policy goals to focus solely, or at least primarily, on price stability."-- Plosser. did that work out in Japan? They had 20 years of very mild deflation...and China emerged as the dominant economic and political power in the Far East. The People's Bank of China, btw, now targets 3.5 percent inflation, but last year they targeted 4.0 percent.They care more about robust growth. Oh, that.

    Japan and the USA can recede from the international economic scene through self-imposed monetary asphyxiation, but I doubt that is an advisable course. Indeed, depending on China's inclinations, that could well be a darker world. And aircraft carriers do not spell influence, when the other guy is waving money around.

    (We could attribute China's incredible and sustained rapid growth to communistic five-year plans and SOEs. Somehow, I think more market-oriented enterprises and a growth-oriented central bank were more important).

    The idea of an independent central bank is already a dangerous one---every public agency develops exalted mission statements, insularity, internal cultures and taboos--and all of those traits are magnified at independent public agencies. In fact, I expect most economists would say the words "independent public agency" are a recipe for failure and ossification.

    The ECB was created in 1997 (I think) with that one mission of price stability. It has been a debacle.

    A supreme irony: The Fed has actually obtained a historically extremely high level of price stability since 2008. Inflation has been very low, and not very variable. We seem stuck between 1 percent and 2 percent. It has been very stable. Plosser says this is not stable enough?

    Is Plosser saying the Fed should somehow "micromanage" its way to even greater price stability? Or is he calling for zero inflation (the Bank of Japan approach)? Or, does Plosser's definition of price stability exclude deflation? That is, deflation is okay?

    To sum up, with its large-scale asset purchase program (LSAP or QE) we have seen price stability, with very little deflation or inflation. We have seen slow growth.

    And Plosser says the inflation side has been bungled, but the growth side "is not our job?"


    1. Come on, read the thing before you sound off. Plosser did not anywhere say the inflation side has been bungled. He said the Fed is intervening in all sorts of markets and taking on all sorts of new and dangerous roles; these are "fiscal policy," political, and will come back to haunt the Fed. But he did not say that "the inflation side has been bungled." Let's not follow our friends on the left by talking about imaginary statements, let's talk about what people actually say.

    2. Okay, you are right, I retract the "bungled" statement.

      But Plosser seems to overlooks the high degree of price stability obtained since 2008. What if QE and "micromanaging" was responsible for that price stability, in averting a deflationary recession?

      I actually agree that the Fed should only buy Treasuries, and the interesting part is what if the Fed holds the bonds indefinitely (rolls over, in other words).

      So far, we are seeing no inflation from QE. There was no inflation in Japan from QE 2002-6.

      Is monetizing the national debt a sin? But if you can get away with it...why not?

      The Fed St.Louis says even amuck larger QE (LSAP) program will result in hardly any inflation (and we could actually use a couple percent more).

      The option is there: Buy back the national debt, give the public cash.

      As a taxpayer...this may not be a bad idea...or, do we let this opportunity get away?

    3. The idea of a single mandate somehow improving Fed's independence is a fantasy. It will actually raise pressure on the Fed because it will be harder for the Fed to rebut calls for action - currently it can hide behind the inherent tradeoff of the dual mandate.

    4. Professor, a proponent of NGDPT might point out to the possibility of helicopter drops. If that sounds too much like an arcane solution, Miles Kimball and Scott Sumner have pretty comprehensive ideas that could be used to implement NGDPT.

      It is unclear to me why what your position on their ideas is.

  5. Further confusion: Some have posited the way out of a recession is wide-scale and sudden liquidation and deflation, in the opening round of a recession. This is self-correcting, and owed wages and prices lead the way to recovery...I don't agree with this assessment, but...

    If the Plosser "stable prices only" strategy is followed, then the independent Fed would act very expansively---all guns blazing---at the start of a recession, to keep prices stable. The Fed would obviate the corrective action of the rapid deflation and liquidation. the proper Fed policy then to stand by in the initial stages of a recession, let prices and wages tumble...but then step in and keep prices "stable," that is, at zero inflation?

    Meaning real estate and equities would never recover old levels, or only very, very slowly?

    What would that mean to banks? Investors? Stockholders?

    No fears of of a downward spiral in this scenario?

  6. Having two ways of imparting stimulus (fiscal and monetary) is ridiculous: you might as well have a car with two steering wheels each controlled by a different person.

    As to interest rate cuts, they impart stimulus just via investment spending. We might as well impart stimulus just via spending on cars, massage parlours and chewing gum. As to QE, that destabilises developing economies and makes rich individuals even richer. And third, the crunch was caused by excessive and irresponsible lending: so we try to cure the problem by . . er . . cutting interest rates so as to encourage more lending.

    Hopefully some time in the near future, we’ll tumble to the fact that the existing system for imparting stimulus is stark raving bonkers.

  7. I dearly respect and admire Mr. Cochrane and Mr. Plosser. But I must say that many key points have been touched in these commentary, somthing which make you hard to disagree as there are many unresolved open questions. What surprise is that no mention is really made explicitly about the risk of deflation, the risk of the financial system going down and stop working, the zero lower bound, and the need to provide liquidity to the system in order to ensure the (non-necessarily proper) functioning of the payment sytem and protect the real sector. I find astonishing that little recognition is given to the fact that the risk of financial collapse is still present, that many private sector agents are still hihgly leveraged, that financial institutions must write-off their losses, that more strict regulation (Dodd-Frank is in place). The US has gone through the major financial crisis since the Great Depression, something which cannot be overlooked. Adopting inflation as the primary objective only makes sense when the economy has gone back to normal conditions: stability in financial markets, fiscal consolidation, abscence of market uncertainty, inflation under control and sustained growth. Otherwise, having a single mandate at the present juncture will not help. The credibility of the Fed is at stake. The Fed has limited instruments and must have a limited number of targets. Certainly independence is most welcome. In sum, the Fed institutional and policy frameworks must be improved, unfortunately it is not the right time. Many difficult challenges remain ahead and we are still sailing through unchartered waters.

  8. "The current deal for independence is written in part in the Federal Reserve act which sets up the current dual mandate,"

    Incorrect. The dual mandate came from an Act of Congress that set no goals for the federal reserve or the federal open market committee -

    By the time the federal reserve / federal open market committee became involved, the dual mandate had morphed into a 4 prong mandate:

    "The Act explicitly instructs the nation to strive toward four ultimate goals: full employment, growth in production, price stability, and balance of trade and budget."

  9. Wow. 100 years later we're still not sure what our central bank is for. Can we do one of those ObamaCare reversal things and put the Fed on hold for a year?

    "If you like your dollar's value, you can keep it."

    "well, yes, we don't really have that much effect on employment, but employment is in the toilet, we have to do what we can, no?"

    I've seen this in my own profession. When you don't know what to do, do something you know how to do.

  10. Most of all, this is sad.

    As Benjamin says, Plosser is making the same mistakes that the Bank of Japan made for 20 years PRICE STABILITY is not enough!

    And once we set up an NGDP futures market, the Fed has a guidance instrument to know what it needs to do to increase spending.

    What we can also do is amend the law to allow the fed to write checks directly to employed Americans.

    As to how much money we need to print and spend to where we want to go, calculate velocity of one of the three money stocks (M1, M2, or MZM divided into recent nominal GDP) and use that as a formula to know how much money to print.

    Is that rules based enough?

    1. Re writing “checks directly to employed Americans”, I don’t think you phrased that too well, but I favour something of the sort. I’d put it this way: “let the government / central bank machine simply create new money and spend it (and/or cut taxes) when stimulus is needed.” Or put another way: “let’s merge monetary and fiscal policy”.

      That would be a big change, and it needs thinking thru. Advocates of Modern Monetary Theory tend to favour that policy, plus this publication in the UK favours the policy:

  11. Do you believe in liquidity traps or do you not think monetary policy can effect real variables like employment?

    1. In case he doesn't answer for himself, Cochrane has a fiscal theory of the price level equation which looks to me like an equilibrium liquidity trap.

  12. Why didn't Plosser just tell us his policy rule back when he had a vote? If rule based policy is what we need more of then lead by example. At least, we wouldn't have to guess, when and why he was dissenting and dissect his public statements afterwards.

    It is easy to argue in favor of rule based policy, before actually proposing and commiting to a rule. Let Plosser offer up the right rule and open it up to public scrutiny. Let economists create models showing when and how the rule fails, and perhaps that will lead to better, more robust rules. If the rule works well, perhaps more governors will commit their votes to rules.

    1. Barro and Gordon produced a model that showed that even when we assume omniscience and benevolence with a strict monetary rule, the monetary authority will still cheat because of the marginal gains that can be made through unexpected inflation on unemployment.

  13. During a liquidity crisis the Fed should be able and ready to step in and do things like buy commercial paper issued by major corporations (GE, as one example). But those measures should be short term steps to stabilize the economy.

    I personally think that the conservative agenda being pushed by the Republicans in Congress is wrong but those Republicans got elected. The Fed, by taking extraordinary measures like the mortgage purchases, is thwarting the effect of the elected Republicans' policies and in the process undermining democracy. QE, including mortgage purchases, carries a significant potential long term hidden cost to the Fed of interest rate risk and default risk.

    The Fed should announce that they have listened to the concerns and preferences of the Republicans and that the Fed is going to stop QE (including the mortgage purchases) cold turkey on January 1, 2014.

    1. Why? the Republicans are in charge of only one half of one third of the government and they are hardly uniform in their opinions of monetary policy.

  14. An awful speech:

    Just awful.

    I'm all for rules based policy, but we have to have the correct rule! The correct policy is not price stability, but income stability. Plosser is a hard money sadist who if it were up to him, would give the U.S. Japans lost decades.

  15. Interesting! And here is Kocherlakota on what should Fed do

    In conclusion he says the following:
    "The FOMC’s test today is to figure out how best to deploy this capacity. The answer lies in taking two steps. The first step is to communicate that our goal is to accomplish a fast return to maximal employment while keeping inflation close to, although possibly temporarily above, the target of 2 percent. The second step is to do whatever it takes, on an ongoing basis, to achieve that goal. A goal-oriented approach to monetary policy greatly reduced inflation in the early 1980s. Adopting such an approach in our own time would improve labor market outcomes."

    I wonder what is going on with him!

  16. I'm confused. When I took my into to econ class in undergrad, we used Greg Mankiw's (awesome) textbook. In it, Mankiw espouses 10 principles of economics. Relevant here are

    9. Prices Rise When the Government Prints Too Much Money, and
    10. Society Faces a Short-Run Tradeoff Between Inflation and Unemployment.

    From those principles, I learned that the Fed determines the price level because it controls the printing press, and that though the price level is irrelevant in the long run, short-term changes in the price level can affect the real economy because prices (especially wages) are sticky.

    Today, the Fed indicates that it has an inflation target of approximately 2%. If we were to adopt Mr. Plosser's position, and have the Fed be a strict inflation targeter, we could certainly critize the Fed's recent performance. But not becaue monetary policy has been too loose! Since 2008, the Fed has consistently failed to acheive its inflation target and allowed inflation to stay below 2%. So, Plosser's critique should be that the Fed is too tight, not that its too loose.

    But, if we place any value at all on improving the current labor market, in which millions of people want to work but can't find a job, then what would be the problem with the Fed targeting a 3% rate of inflation for the next several years, so as to benefit from the principle that society faces a short-term tradeoff between inflation and unemployment? Given the chronic prior undershooting of the Fed's inflation target, we'd only make it part way back to where the price level would have been if there had been no recession in the first place.

    1. Or maybe, some of the principles are wrong. The Fed does not "print money." The Fed buys bonds, always taking in something of equal value to what it gives out. Right now it is taking in red M&Ms, giving blue M&Ms in return, and hoping that helps your diet. Maybe not. If the Fed could engineer higher inflation, it suresly would have done so by now. And maybe the Fed right now has little to do with the labor market; the Phillips correlation is not a curve.

    2. Prof. Cochrane,

      You say "if the Fed could engineer higher inflation, it surely would have done so by now."

      I just don't see it. I don't see *any* fed official advocating for higher inflation. In fact, I just the opposite, for instance, as Ben Bernanke boasts of the Fed's exemplary record on price stability in front of Congress.

      You know where I do see a central bank advocating higher inflation? Japan. It actually seems to be working.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.