The Fed shouldn’t be “solving problems,” their job is to avoid creating problems

The previous post linked to this quotation:

The U.S. economy appears to be mired in problems that monetary policy cannot solve.

I am always a bit disoriented when people talk about whether the Fed should “do something” to solve some sort of “problem” in the economy.  The Fed should always, and I mean always, set policy in a position expected to produce the desired results.  If they do that, then there is never any “problem to solve,” or at least any problem that is solvable by monetary policy. 

Yes, the US has loads of problems that monetary policy cannot solve.  But it also has lots of problems that were created by bad monetary policy, which people often overlook.  In a healthy economy US NGDP grows at about 5%.  In the two years after mid-2008 it grew about 1%, as compared to an expected 10% increase.  That reflects extremely weak AD, and aggravates all sorts of other problems in the economy.  The Fed can certainly create problems.

When I read other economists they often seem to visualize a very strange monetary policy process.  They see the economy just cruising along, minding its own business, when all of a sudden “problems” develop.  Then they start debating whether these “problems” are something the Fed could or should be doing anything about.  As if the Fed had no role in creating the problems.  The ship has veered 400 miles off course, maybe we should see if the captain can solve this “ship problem.”

In an ideal world, Fed policy would always be set at a level expected to produce on target nominal growth, say along a 3% or 5% NGDP growth trajectory.  There would never be a need to adjust policy to solve problems, because the policy target (5% expected NGDP growth) is assumed to also represent the policy goal (5% NGDP growth.) 

The Fed obviously doesn’t behave this way.  It’s like a captain that dozes off at the wheel and then awakes with a start when his ship runs aground on the sandbar.  What do we do now? 

It’s hard for me to tell the Fed what to do, because I don’t know where they are trying to go.  Go back to the ship analogy.  Suppose the captain is rather secretive, and will only indicate that he is aiming for a large port in the northeastern US.   He also says that due to wind and current he expects to end up somewhere around Charleston SC.  What sort of advice should we give him?  First, at a minimum he should turn the wheel to the right, to nudge the ship onto a more northerly course.  And second, he should pick an explicit target, and let his crew know which city he is aiming for.  Then his crew could help him decide how to set the steering mechanism at a position expected to produce on target sailing (accounting for wind and currents.)

Right now the Fed has two problems.  They won’t tell us their goal, and they won’t set policy at position expected to reach their (secret) goal.  We know they’d like inflation and employment to be a bit higher, but we don’t know how much higher.  We know they have set their policy levers at a position where they expect to fail, i.e. they hope NGDP grows a bit faster than their forecasting unit expects, not a bit slower.  But we don’t know how much faster.  If I knew their goal I could give advice on how to achieve the goal.  Instead, all I can recommend is “more.”

Commenters often complain that I am trying to use monetary policy to solve deep structural problems in the economy.  No.  I simply want the Fed to decide where it wants to go, tell us where it wants to go, and then set its policy instruments at a position expected to achieve those goals.  Is that too much to ask?

PS.  I’m running way behind.  I’ll try to get all the old comments answered on Thursday.  I wanted to do an election post, but no time.  Ditto for the monetarism conference I just attended–maybe later.  Tomorrow I teach so any Fed policy analysis will come late in the day.  Congratulations to Morgan and the other Republican commenters.



22 Responses to “The Fed shouldn’t be “solving problems,” their job is to avoid creating problems”

  1. Gravatar of Mikko Mikko
    2. November 2010 at 22:45

    So running forward with the ship analogy. You should also consider the fact that the ship doesn’t have a captain. It has a committee and every committee member has a different view on what port the ship should head towards. Some want the ship to go even further south.

  2. Gravatar of DanC DanC
    3. November 2010 at 04:40

    Change the story to the Fed driving a bus, is the best, smoothest, path achieved by a bus driver who slams on the accelerator, then slams on the brakes, then slams on the accelerator, then….

    I’m not against putting a steady fuel supply in the tank, but wouldn’t you prefer a smooth predictable ride.

  3. Gravatar of JTapp JTapp
    3. November 2010 at 05:12

    Marty Feldstein has an op-ed in the FT critical of QEII today. Most of it is same-old, same-old.

    “Mr Bernanke’s argument for QE is based on the “portfolio balance” theory which stresses that, when the Fed buys bonds, investors increase their demand for other assets, particularly equities, raising their price and increasing household wealth and spending. Equity prices have already risen by 10 per cent since Mr Bernanke discussed this approach. But how much further will equity prices rise and what will that do to GDP?
    Neither theory nor past experience can answer the first question. Much of the share price increase induced by QE may already have occurred based on expectations. An optimistic guess would be another 10 per cent. Since households have about $7,000bn in equities, that would imply a wealth gain of $700bn, raising consumer spending by about one-quarter of one per cent of GDP, a welcome but trivially small effect on incomes and employment…The truth is there is little more that the Fed can do to raise economic activity.”

    I would say QE can’t be expected to do much without the commitment to an explicit target, be it price level or NGDP or final sales, etc. QE in the absence of that clear target is sort of like the captain revving the engines but still not stating the destination, right? We may get somewhere faster but it’s still not clear where “there” is supposed to be.

  4. Gravatar of Paul Zrimsek Paul Zrimsek
    3. November 2010 at 06:12

    In other words, recession is always and everywhere a monetary phenomenon?

  5. Gravatar of JTapp JTapp
    3. November 2010 at 07:02

    Jim Hamilton also has a quote on Bloomberg skeptical of QE this morning.
    “(Bernanke) may risk increasing expectations for higher inflation by too much…That perception alone would bring about a series of immediate challenges, such as a rapid flight from the dollar, commodity speculation and possible under-subscription to Treasury auctions.”

    They could obviously solve that problem by stating their target.

  6. Gravatar of Doug Bates Doug Bates
    3. November 2010 at 07:35

    I think the ultimate problem is that the Fed is neither powerful enough nor quick enough to offset mob psychology. Sure, there are things the Fed can do to make things better [or worse] — or we wouldn’t even want to have a Fed at all. But the main driver of the business cycle is mob psychology. The main driver of inflation is a combination of the velocity of money and inflation expectations — both under the control of mob psychology. The Fed can act directly on interest rates, money supply, etc. It can attempt to influence mob psychology via both direct action and “open mouth” operations. But mob psychology is ultimately outside of any one committee’s control, no matter how much we might kibitz. If I were in charge of the Fed, and could pick my own team to be on the committee, we would have our own set of problems to deal with, and all y’all would be kibitzing about how you could do it better.

    So to put this in terms of your ship analogy — the wind and the sea are far more powerful than the rudder and sails, so no matter which way the captain aims the ship, that’s not the direction it will travel. She has to guess how the wind and the sea will react if she aims the ship at Charleston, and then adjust so that maybe the ship will actually head in her intended direction after the wind and sea have their way.

    And telling the wind and sea, “I’m heading to Charleston at 20 knots,” will not necessarily help.

  7. Gravatar of marcus nunes marcus nunes
    3. November 2010 at 10:05

    Krugman has reversed to form:
    “So here’s my problem: even when people have a clear vision of the problem, they lose their nerve when it comes to proposing solutions that actually address that problem. It’s this kind of diffidence that doomed us to inadequate policies when Obama might have had the ability to get stuff through; and now we’re stuck, with little hope of recovery for many years to come”.


  8. Gravatar of JimP JimP
    3. November 2010 at 10:23

    No level targeting. Forget it – says Ben.

  9. Gravatar of JimP JimP
    3. November 2010 at 10:25

    He had the votes for it and didn’t do it – so clearly this is a policy decision on his part.

  10. Gravatar of W. Peden W. Peden
    3. November 2010 at 10:30

    Doug Bates,

    That assumes that one is dealing with wind and seas, rather than people who modify their behaviour in response to what the Fed says it plans on doing.

  11. Gravatar of marcus nunes marcus nunes
    3. November 2010 at 10:50

    Bernanke today reminded me of the famous line in Gone with the Wind when Clark Gable (the Fomc) tells Vivien Leigh (the “public”): “I don´t give a damn my dear”

  12. Gravatar of Cameron Cameron
    3. November 2010 at 12:02

    Not to inflate your ego or anything, but haven’t the last 2 months given us a massive amount of evidence that you are absolutely 100% right about everything?

    Since people started anticipating QEII…

    The stock market has soared 13%
    Commodities have rallied ~15%
    Inflation swaps and TIPS spreads have increased significantly(but are still below the Fed’s “goal”)
    Long term treasury rates have fallen while corporate bond rates have risen.

    And so far all the economic data from October so far (ISM, Auto sales, ADP employment) has been better than expected.

    What would be an alternative explanation? I can’t think of any “liquidity trap” Keynesian, Austrian or Real Business Cycle explanations for this… what is left?

  13. Gravatar of JTapp JTapp
    3. November 2010 at 12:16

    @JimP It’s not clear what he had the votes for. Remember that they take a preliminary reading (a pre-vote) and then an official vote for the statement that shows solidarity. So, the committee may have been split in reality. Hoenig just refuses to go along with the official vote. We’ll not know for another 6 years what really happened today.

  14. Gravatar of Sumner « Free Radical Sumner « Free Radical
    3. November 2010 at 12:20

    […] November 3, 2010 Free Radical Leave a comment Go to comments Here is Sumner saying kind of the same thing I have been saying but in less alarmist language.  It’s hard for […]

  15. Gravatar of Charles R. Williams Charles R. Williams
    3. November 2010 at 12:46

    There are grocery carts that look like cars complete with steering wheels. Kids sit and turn the wheels while their moms are stuffing broccoli into plastic bags. Maybe if the kid acts crazy enough he can tip the cart over. But ever since someone cooked up this idea, the lawyers and the engineers have teemed up to eliminate the possibility that the kid will tip the cart and hurt himself.

    The fed imagines buying T-bonds and creating more excess reserves that the banks are subsidized to sit on. This will slightly increase bank profits. But why not have Treasury sell more T-bils and fewer T-bonds? Then the fed can subsidize the banks 10 basis points to buy T-bills. Same result.

    What does all this sturm und drang have to do with the economy? It gets people worked up because they think the kid might tip the cart? Or maybe they think the cart will be more stable if mom fills it to the brim with inflated balloons.

    Krugman is right about this (and little else) – it would take an order of magnitude more QE than anybody contemplates to move the economy, except possibly for the spook factor.

  16. Gravatar of libfree libfree
    3. November 2010 at 13:07

    Maybe we need a sextant?

  17. Gravatar of scott sumner scott sumner
    3. November 2010 at 14:50

    Mikko, Yes, that’s a good point.

    DanC, Yes a smooth rise would be better than a steady fuel supply.

    JTapp, I agree that QE by itself is not very powerful.

    Paul, No, the 1974 recession was caused by real shocks, and many recessions (including this one) are partly caused by real shocks and partly caused by nominal shocks.

    JTapp, Regarding Hamilton, I see no danger of excessively high inflation expectations–just the opposite.

    Doug, The Fed has almost infinite ability to boost AD. Remember Zimbabwe?

    Marcus, I think he is right about monetary policy–the Fed doesn’t have the nerve to do what’s necessary. I’m glad Congress lacks the nerve to enact his favored fiscal policy.

    JimP, Yes, Bernanke is far too cautious.

    Cameron, I haven’t really given much thought to the market reaction, because it was no surprise to me. I was used to seeing the same thing happen in the 1930s. No, I’m not that old, but I once read all the NYT from the 1930s, so I recall how markets respond strongly to hints of monetary stimulus when rates are near zero.

    And thanks for further inflating my ego. 🙂

    Charles, Krugman’s right that this isn’t nearly enough, but it’s hard to say how much would be. The problem is that it isn’t linear. At some point QE would start to raise inflation expectations, and then it would immediately become far more expansionary. So his numbers are meaningless. Krugman’s also right that a higher inflation target, level targeting, would be much more powerful.

    libfree, Perhaps a NGDP futures market would be a sort of sextant.

  18. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    3. November 2010 at 15:21

    Charleston reports the ship is going to Palm Beach FL:

  19. Gravatar of Paul Zrimsek Paul Zrimsek
    3. November 2010 at 15:33

    So what’s wrong with calling the AD effect of the real shocks a problem that the Fed has to solve?

  20. Gravatar of Doug Bates Doug Bates
    4. November 2010 at 09:40

    I agree the Fed could go insane and print what seems like an infinite amount of money — this makes it seem extremely powerful and throws all our metaphors out the window. But that is not the same as achieving an inflation or NGDP target using a combination of monetary actions and open mouth operations. If the Fed really decided to spark hyperinflation, they could not precisely target hyperinflation of, say, 1000%. They would probably go well off target — we might have hyperinflation of 500% or 5000%, or maybe even 7000000% once velocity shot up as people scrambled to dispose of their long-term savings overnight and went shopping twice per day to spend each half-day’s paycheck before it became worthless. If the average velocity of money went from 1/year to 2/day that would in itself account for a 700000% increase in consumer prices, no increase in supply needed.

    My point is that precision will always be lacking if the Fed tries to target aggregate economic variables (CPI, NGDP) instead of something very specific, such as short-term interest rates or the price of gold.

  21. Gravatar of ssumner ssumner
    4. November 2010 at 17:39

    123, Who knows where they are going.

    Paul, I agree, they should offset the AD effects of real shocks.

    Doug, You solve that problem by doing level targeting, not growth rate targeting. That anchors long term expectations, and keeps velocity more stable.

  22. Gravatar of Mechanical vs. Intentional Economics « azmytheconomics Mechanical vs. Intentional Economics « azmytheconomics
    7. June 2012 at 07:41

    […] who base their estimates on mathematical models or correlations. On the intentions side, Scott Sumner compares the central bank to a captain of a ship – if the course is off, change the steering […]

Leave a Reply