It’s hard to commit to a lousy policy regime

A few days ago I did a post that was partly a comment on this post by Brad Delong.  But since people keep sending it to me, I’ll make a few more observations.  Consider this passage:

The first of these requires enormous gross asset swaps in order to free up any noticeable amount of risk-bearing capacity. Moreover, it requires that the monetary authority have the power and the will to commit the taxpayers to backstopping it and bearing substantial amounts of default risk.

The second faces all of the problems of the dynamic consistency literature, but in reverse. After the economy has recovered, what reason is there for those at the head of the monetary authority to deviate from their standard policy rule, as their predecessors who are in office now promised they would? And since those currently at the head of the monetary authority cannot bind their successors, why should their promises today be more than cheap talk?

There are costs of letting inflation in the future rise above the monetary authority’s target level. First, there are the consequences for the monetary authority’s credibility as an inflation fighter. Second, there are a variety of costs arising from the tilting of the duration structure of the debt that comes from monetary authority decisions to oppose a current downturn by shifting to a higher target for the price level in the future.

The first paragraph is misleading, as I don’t see any reason for the Fed to buy assets with default risk.  And any losses to the Fed from duration risk would be more than offset by gains to the Treasury (for which T-securities are a liability.)  Indeed that’s why many people fear inflationary policies when governments get deeply in debt.   But even if you assume some weird irrationality, where voters care about Fed losses but not Treasury gains (and it sound ridiculous even writing these words) it’s all beside the point, as easier money doesn’t require a bigger monetary base.

The bigger problem here is that DeLong is describing a lousy monetary regime where the central bank sets out to “fix” problems on an ad hoc basis.  Of course it would be difficult to adhere to that regime.  It’s a disaster.  The central bank shouldn’t try to fix problems, it should try to target NGDP along a specified trajectory, aka “level targeting.”  Under that sort of policy regime the problems mentioned by DeLong vanish into thin air.  There is no need to change policy after exiting the ‘liquidity trap,” as your policy is always exactly the same.

DeLong’s bigger mistake is to assume that the current Fed policy is expansionary, and that even more of the same would be needed for an even more expansionary policy.  Just the opposite is true.  As Bernanke pointed out in 2003, neither low interest rates nor a bloated monetary base mean easy money, the only proper indicators are inflation (lowest since the mid-1950s over the past 3 1/2 years), and NGDP growth (lowest since Herbert Hoover was president over the past 3 1/2 years.)  I agree with Bernanke, money’s been ultra-tight.  That’s why rates are so low in the US and even lower in Japan, where money is even tighter.  And it’s one reason why the base is so bloated (3 times its normal level as a share of GDP.)  If we had NGDPLT (and zero IOR), the base would be far smaller and interest rates would be much higher.

I realize that Bernanke and I are viewed as weirdos for not using interest rates and the money supply as indicators of the stance of monetary policy, and Brad is expressing the conventional wisdom.  All I can say is the conventional wisdom on money/macro issues has almost never been right at any time in human history, so why should this case be any different?

Finally, Brad might argue that we don’t have NGDPLT, so it’s a moot point.  Under the current regime the Fed can’t inflate.  But I see no evidence that the Fed, or any other fiat money central bank that wanted to inflate, was ever unable to do so.  Come back to me when the Fed says like it’s out of ammo and/or acts like it’s out of ammo.


Tags:

 
 
 

19 Responses to “It’s hard to commit to a lousy policy regime”

  1. Gravatar of Cthorm Cthorm
    13. March 2012 at 15:15

    Finally, Brad might argue that we don’t have NGDPLT, so it’s a moot point. Under the current regime the Fed can’t inflate. But I see no evidence that the Fed, or any other fiat money central bank that wanted to inflate, was ever unable to do so. Come back to me when the Fed says like it’s out of ammo and/or acts like it’s out of ammo.

    That’s because our current Fed regime is econo-shamanism plain and simple. The Fed is in paralysis because the FOMC has a majority of members with an incoherent understanding of M-policy. I’m now reading every statement that doesn’t tighten policy as good news. The Fed can’t inflate right now. The Fed is absolutely acting like it’s out of ammo, even if Chairman Bernanke is well aware that they have infinite ammunition. The only thing lacking is the Will to do so, which is itself evidence that rule-based policy is essential.

  2. Gravatar of dwb dwb
    13. March 2012 at 15:31

    on android so i cant link the post, but i am pretty sure if you go back to oct 31, delong mentioned ngdp targetimg as great medicine for a liquidity trap. Romer reminded hi at a department seminar. i once had a boss totally enamored of the last idea je heard, evev if it contradicted yesterdays big idea. deja vu

  3. Gravatar of dtoh dtoh
    13. March 2012 at 15:33

    Scott,
    How does the Fed inflate? I have already posed this question, but given the current situation where the opportunity cost of holding reserves is zero (or close) and where the opportunity cost of holding cash is very low, how does the Fed inflate. Since the opportunity cost of holding is so low and base has become a good substitute for Tbills as a store of wealth, don’t OMP just result in a drop in V which completely offsets the increase in M.

    At some point, an expectation of inflation will presumably result in an increase in V, but how and when does this happen?

  4. Gravatar of Cthorm Cthorm
    13. March 2012 at 15:57

    @dtoh

    The Fed does not really need OMP to inflate. All they need is a microphone. All they need to do is announce an intention to pursue X% inflation or Y% NGDP growth. The opportunity cost of holding cash or T-bills would rapidly increase thereafter, those opportunity costs are not independent variables decided ex ante. If expectations of inflation do not rise, the Fed can just say it will conduct unlimited OMP until it meets his goal. The easiest way to lose a fortune is to bet against a determined central bank.

    Now there is some question of the legality of such a Fed announcement. I think it’s perfectly consistent with their Dual Mandate, but a former professor of mine (and former Fed Senior Advisor) is less certain about the legality. If that is the case Congress might need to instruct/authorize the Fed to change it’s target.

  5. Gravatar of M.R. M.R.
    13. March 2012 at 15:58

    Scott, you wrote: “I don’t see any reason for the Fed to buy assets with default risk.”

    I’ve often wondered: Does your optimal monetary policy regime depend on the existence of a large, riskless public debt?

    Imagine a small government (relative to the size of the economy) that balances its budget. There is no public debt. Can the central bank do NGDP targeting without taking on default risk under this scenario?

    Obviously this is not a practical concern right now, but I’m interested in understanding the necessary conditions.

    Sorry if you’ve answered this one before — you probably keep getting the same questions over and over …

  6. Gravatar of dtoh dtoh
    13. March 2012 at 16:16

    CThorm,
    I understand your argument, but as Scott says, “expectations have to be about something,” and given Fed dithering to date and the political difficulty of dramatically increasing the amount of physical currency, it seems to me that this is not quite as easy as it sounds.

  7. Gravatar of Cthorm Cthorm
    13. March 2012 at 16:36

    @dtoh

    I agree that the probability of the Fed actually inflating right now is very low. But I wouldn’t be concerned about the difficult of dramatically increasing the amount of physical currency. The “about something” could be the Fed unleashing the $2+ Trillion of excess reserves into circulation by cutting IOR to zero. I think that would be plenty to get expectations up.

  8. Gravatar of John Thacker John Thacker
    13. March 2012 at 16:37

    Meanwhile, unlike DeLong, Christine Romer is asking the right questions:

    Given the Fed’s forecast for subdued inflation and “moderate economic growth” over the coming few quarters, Romer can’t help but wonder: “Why aren’t they doing more to help the economy?”

  9. Gravatar of ssumner ssumner
    13. March 2012 at 17:06

    Cthorm, It refused to do more QE because it thinks the recovery is going well, that means it doesn’t think it’s out of ammo.

    dwb, Yes, I recall him talking that way.

    dtoh, It raises NGDP expectations by committing to a higher NGDP target path than the market current expect. Even fiscal supporters like Romer/Bernanke/DeLong agree that works.

    MR, You said;

    “I’ve often wondered: Does your optimal monetary policy regime depend on the existence of a large, riskless public debt?”

    There are lots of other options, but the main point is that it’s not a monetary policy question, it’s a monetary regime question. It’s a question of what the policy goals are. You can’t deal with these questions on an ad hoc basis. If there is no government debt to buy, you still need a monetary system, and you still need to adjust the money supply.

    There are countries where governments own far more assets than the national debt (Singapore, etc) and they can do effective monetary policy. So I don’t see that as a big issue. And I’d add, it’s an issue that has nothing to do with the question of what monetary policy is best in a given situation.

    John Thacker, That’s a good question.

  10. Gravatar of dwb dwb
    13. March 2012 at 18:44

    http://delong.typepad.com/sdj/2011/10/yet-another-note-on-benefits-of-nominal-gdp-targeting-1.html

    “Level nominal GDP targeting also has very nice properties in an economy buffeted by either positive or negative commodity-price supply shocks.”

    you, know, like umm now??

  11. Gravatar of Dtoh Dtoh
    13. March 2012 at 19:34

    MR
    Lot of options even without government debt.

    1. Monetary authority can buy other assets.

    2. MA can make banks buy / create assets.

    3. MA can make banks borrow from the MA.

  12. Gravatar of RebelEconomist RebelEconomist
    14. March 2012 at 03:33

    Scott, you note that inflation has been the “lowest since the mid-1950s over the past 3 1/2 years”. I would be a bit careful of making comparisons like this. The cpi was engineered downwards (by IIRC about 3/4%) by the Boskin Commission from 1996.

    Therein lies an important lesson for would-be designers of monetary policy regimes like you. Whether or not the Boskin reforms were reasonable, an objective independent central bank should have said, “as the cpi is adjusted, we will target a lower rate of inflation”. But the Fed didn’t. And that is a key problem for monetary policy regimes.

    Few central bankers outside Germany like to be party poopers, so they will tend to accept relaxations of the target with no more than token resistance and make use of escape clauses. The Bank of England has provided a compendium of such backsliding (change to cpi, targeting their own forecast, toothless sanctions like letter-writing, etc). As the Boskin example shows, once the dust settles, even seasoned observers of monetary policy like you tend to overlook such slippage. A designer of monetary policy regimes should therefore endeavour to provide as few of these opportunities as possible – even central bankers must be cornered like rats before they will do the right thing!

    If you care about inflation, the weak point of NGDP targeting is that it must embed an assumption of potential real growth; I am sure that that would be used to undermine prudent monetary policy. And if you must advocate NGDP targeting, I would suggest recommending a target that makes a conservative assumption about potential growth (ie 4% not 5%), not least to remove the temptation to adopt the policy for a bit of immediate relief.

  13. Gravatar of Ritwik Ritwik
    14. March 2012 at 05:18

    Scott, let’s say that the Fed ‘does nothing’, meaning that the fed funds rate stays the same, Bernanke makestypical wishy washy, non-committal consensus type statements, etc.

    Let’s say in 2012, US NGDP grows by 5%, with inflation at 3%.

    Let’s say throughout this time you wrote posts criticizing the Fed.

    In this hypothetical world, Karl Smith would have an answer. Tyler Cowen would have an answer. De Long/ Krugman would muster an answer while talking about how the pain of ’08-11 was avoidable.

    What would your answer be?

  14. Gravatar of M.R. M.R.
    14. March 2012 at 05:22

    Scott, you wrote: “There are lots of other options … If there is no government debt to buy, you still need a monetary system, and you still need to adjust the money supply.”

    No argument there. I was just responding to your comment that “I don’t see any reason for the Fed to buy assets with default risk,” and your comment that DeLong had been misleading in suggesting otherwise.

    I interpreted you to be saying that the central bank could successfully execute an NGDP level-targeting regime without straying from (riskless) public debt purchases.

    This may very well be true in the U.S. right now (although how can we be certain?), but it isn’t a priori true.

  15. Gravatar of fmb fmb
    14. March 2012 at 06:23

    I don’t think you really address the dynamic consistency problem he raised. My policy is not to pay kidnappers, but it does not make sense to say that establishing that policy makes the problem of sticking to it in the face of a ransom note vanish into thin air. The policy is cheap talk.

    I can imagine many (including DeLong in his post) thinking something like: sure, you can follow your NGDP targeting rule, but this might lead to uncomfortably high inflation when we exit the trap, are you sure we can stomach that when it happens?

    MM’s may be so comfortable with NGDPLT that they don’t really acknowledge such fears. Basically: NGDPLT -> resulting OMO and inflation outcomes must be a good idea, QED. But, for someone who doesn’t quite accept that, it probably seems non-responsive to an allegation of a dynamic consistency problem.

    I do think that you’re right, though. My interpretation is that in a more traditional flexible inflation targeting regime, when you start to pick up slack you will experience some inflation as idle resources can’t come back online fast enough (I think this is the Karl Smith story). Such inflation should be treated like a temporary supply shock, and as such should not beget tightening.

    It’s not committing to be irresponsible, it’s predicting that the type of inflation that emerges will be the sort where staying loose is the responsible thing, at least for longer than some might expect.

    As you might say, never reason from an inflation change. If all inflation changes are created equal it looks like a dynamic consistency problem, but if they have varying sources requiring different responses then there need not be any such problem.

  16. Gravatar of dwb dwb
    14. March 2012 at 06:59

    News from the Department of ARRRGGH:
    It’s worth pointing out, by the way, that this time the Fed did all that Friedman denounced it for not doing in the 1930s. The fact that this wasn’t enough amounts to a refutation of Friedman’s claim that adequate Fed action could have prevented the Depression.

    arghhhh. how come “we did not do enough” applies to the fiscal stimulus argument and not the the monetary policy argument. arrrgh. I belive the BOJ has done far more QE (as a % of the stock of debt) than the Fed, the primary reason the Fed has held back is peevish fear of inflation… NOT ineffectiveness. arrrrrgh. I feel like a pirate today.

    http://krugman.blogs.nytimes.com/2012/03/14/macro-retrogression/

  17. Gravatar of dwb dwb
    14. March 2012 at 07:00

    … sorry i meant Bank of England not BOJ. oops.

  18. Gravatar of Cthorm Cthorm
    14. March 2012 at 08:08

    Cthorm, It refused to do more QE because it thinks the recovery is going well, that means it doesn’t think it’s out of ammo.

    I suppose I should have said “the Fed won’t inflate right now”. The Fed is acting like they are out of ammo, i.e. they are not willing to act, precisely because they are complacent with the pace of the ‘recovery’. It’s dispiriting when you see the likes of Bullard hand-waving about ‘structural unemployment’ or a superstitious fear of pending hyper-inflation (completely absent in markets). I actually do think hyper-inflation is a possibility, but on a timeline of 20-30 years, if the Federal Government continues to increase deficits. But the Fed has nothing to do with that, they can actually decrease that likelihood by spurring faster nominal income growth.

    Also, I found out through Justin Wolfers that the Fed Reserve has a new twitter account. I immediately asked them why they are ignoring the advice of some of the most prominent economists advising a nominal income target.

  19. Gravatar of ssumner ssumner
    14. March 2012 at 12:05

    Rebeleconomist, You said;

    “If you care about inflation, the weak point of NGDP targeting is that it must embed an assumption of potential real growth; I am sure that that would be used to undermine prudent monetary policy. And if you must advocate NGDP targeting, I would suggest recommending a target that makes a conservative assumption about potential growth (ie 4% not 5%), not least to remove the temptation to adopt the policy for a bit of immediate relief.”

    But I don’t care at all about inflation, nor should anyone else. The objections you raise don’t apply at all to NGDP targeting, indeed it’s another good reason to abandon inflation targeting.

    As you know, I support level targeting, so even 5% NGDP growth is too low right now. Excessive NGDP growth is the least of our problems.

    Ritwik, Pegging nominal rates is not doing nothing, it’s doing a highly unstable monetary policy.

    I’m sorry, but I don’t understand the question you are asking me.

    MR, You said;

    “I interpreted you to be saying that the central bank could successfully execute an NGDP level-targeting regime without straying from (riskless) public debt purchases.”

    I didn’t answer that very well. I should have said the following: It is possible that the Fed would have to buy risky assets, but extremely unlikely under NGDP targeting. It is somewhat more likely thast they would have to do so under the current monetary regime, but still very unlikely. This problem could occur under any regime, including inflation targeting. So I see it as a separate question from whether targeting NGDP is a good idea.

    fmb, Your comment mixes a lot of issues, so it’s hard to respond:

    1. Would stable 5% NGDP growth be politically acceptable, if achieved? I say yes.

    2. If there were deviations from the trend line because of policy errors, would returning to the trend line be politically acceptable? I say probably, especially if the Fed had already made an explicit promise to do so. Doing so on an ad hoc basis (like right now) would be much more controversial.

    3. Is the money supply path required to meet a 5% NGDP growth target politically acceptable? I think it is, partly because the base would be less than 1/2 its current size.

    dwb, His criticism of Friedman is generally pretty feeble. BTW, Friedman didn’t say the Fed doing what he suggested would have prevented the Depression, just made it milder. Was this recession milder than the Great Depression? I’m pretty sure even Krugman would answer yes.

    Cthorm, I basically agree with you, I just interpreted the “ammo” metaphor differently.

Leave a Reply