Zero Rate Bound, RIP

The zero rate bound in economics plays much the same role as black holes (or “singularities”) play in physics.  It is the point where all the laws of macroeconomics break down.  Quantitative easing doesn’t work, the central bank can’t ease policy by cutting rates, increased saving no long leads to more investment, wage cuts don’t boost employment, and the AD curve may slope upward. 

I’ve never believed most of this Keynesian nonsense, but as I argued in the previous post, if other people believe it, especially other people who are central bankers, then it matters.  But within months the zero rate bound will be a thing of the past, merely of interest to antiquarians.  What will cause this momentous change?  The Fed is about to drop the use of the fed funds rate as its short run target (sometimes called “instrument”) of policy.  It will be replaced with the interest rate on excess reserves.  Because that is an administered rate, it really will be an instrument of policy.  And it can be either positive or negative.

Jan. 26 (Bloomberg) — Federal Reserve policy makers are considering adopting a new benchmark interest rate to replace the one they’ve used for the last two decades.

The central bank has been unable to control the federal funds rate since the September 2008 bankruptcy of Lehman Brothers Holdings Inc., when it began flooding financial markets with $1 trillion to prevent the economy from collapsing. Officials, who start a two-day meeting today, have said they may replace or supplement the fed funds rate with interest paid on excess bank reserves.

“One option you might want to consider is that our policy rate is the interest rate on excess reserves and we let the fed funds rate trade with some spread to that,” Richmond Fed President Jeffrey Lackertold reporters on Jan. 8 in Linthicum, Maryland.

The central bank needs to have an effective policy rate in place when it starts to raise interest rates from record lows to keep inflation in check, said Marvin Goodfriend, a former Fed economist. Policy makers are concerned that the Fed funds rate, at which banks borrow from each other in the overnight market, may fail to meet the new target, damaging their credibility and their ability to control inflation as the economy recovers.

‘Extended Period’

The choice of a benchmark is the “front line of defense against inflation, and also it’s at the heart of the central bank being able to precisely and flexibly guide interest-rate policy in the recovery,” said Goodfriend, now a professor at Carnegie Mellon University in Pittsburgh.

The Federal Open Market Committee is likely to maintain its pledge to keep interest rates “exceptionally low” for an “extended period” in a statement at about 2:15 p.m. tomorrow, economists said. The Fed probably won’t raise interest rates from record lows until the November meeting, according to the median of 51 forecasts in a Bloomberg survey of economists this month.

Fed Chairman Ben S. Bernanke, in July Congressional testimony, called interest on reserves “perhaps the most important” tool for tightening credit.

OK, it doesn’t say it’s definite.  But let’s face it; these sorts of highly coordinated “trial balloons” aren’t accidental, especially when Bernanke’s name is attached.

This is a huge policy shift.  Since 1913 the Fed had mostly conducted monetary policy by changing the supply of the medium of account.  (The exception was the 1936-37 RR increase.)  In the US the medium of account is called the monetary base, and is composed of bank reserves and currency held by the public.  It looks like (starting later this year) monetary policy will be implemented through changes in the demand for base money.  The Fed will adjust the demand for base money by adjusting the rate that they pay on excess reserves.  This fulfills one part of Robert Hall’s radical policy proposal of 1983.  The other part is to have the interest rate adjust ex post in such a way as to keep the expected future price level stable.

I think the profession as a whole used to think that Hall’s many “optimal monetary regimes” were a bit silly.  “How can there be more than one,” was a joke I once heard.  Maybe it’s time to take another look at his innovative work on monetary economics. 

Up until now “conventional” monetary policy has been defined as adjustments in the fed funds target rate.  This rate cannot go below zero, otherwise banks would simply hold on to the reserves.  And this explains why people like Paul Krugman often emphasized that conventional policy became ineffective at the zero bound:

I keep seeing economics articles and blog posts that insist that we’re NOT in a liquidity trap (and, of course, that yours truly is all wrong) because the situation doesn’t meet the author’s definition of such a trap. E.g., the interest rates at which businesses can borrow aren’t zero; or there are still things the Fed could do, like buying long-term bonds or corporate debt, or something.

Well, my definition of a liquidity trap is, purely and simply, a situation in which conventional monetary policy — open-market purchases of short-term government debt — has lost effectiveness. Period. End of story.

Now, if you prefer a different definition of a liquidity trap, OK; call our current situation a banana, instead. But changing the name does not change the essential fact — namely, conventional monetary policy has lost effectiveness.

Within a few months Krugman’s statement will become “inoperative,” as “conventional monetary policy” will be adjustments in the rate paid on reserves, which obviously can be negative.  Krugman may reply that this doesn’t eliminate the problem of the liquidity trap.  The public can still hoard currency.  He’s fond of pointing to the fact that in the late 1990s large safes were the most popular consumer durable in Japan.  But rates are near zero in America, and the main problem with hoarding is banks holding excess reserves, not the public holding currency.  I won’t get into all the minutia of this issue; I discussed them in another post.  The bottom line is that negative rates on reserves don’t completely eliminate the need for the central bank to manage expectations, but they make it easier to use their policy lever as a means of doing so.  Alternatively, you might say that it doesn’t eliminate the liquidity trap, but it cuts it in half. 

[Technically it cuts it in half from the monetarist perspective; from the Keynesian perspective it may have no impact, as short term rates stay near zero.  But you know what I think of the Keynesian perspective.]

Part 2.  Dr. Stranglelove and the Doomsday Machine

For some reason I often think of this line from Dr. Strangelove:

Dr. Strangelove: Of course, the whole point of a Doomsday Machine is lost, if you *keep* it a *secret*! Why didn’t you tell the world, EH?
Ambassador de Sadesky: It was to be announced at the Party Congress on Monday. As you know, the Premier loves surprises.

I guess the Fed also like surprises.  But perhaps someone can explain this to me.  Why would the Fed adopt a new policy ideally suited for overcoming the zero rate bound, at the very moment that they are about to embark on a policy of gradually raising interest rates?  Isn’t the whole point of the interest on reserve program that it allows you to cut the policy rate below zero, and prevent banks from hoarding ERs?  (Something that can’t be done with a conventional Fed funds target.)

And for you conspiracy buffs, how about this little nugget from the Bloomberg article I linked to at the top:

The new reliance on reserve interest could also increase the policy clout of Fed governors in Washington at the expense of the 12 regional Fed bank presidents, Reinhart said.

Congress gave only the Fed governors the authority to set the deposit rate. The presidents have historically favored higher rates and voiced more concern about inflation.

“The Federal Reserve Act puts a very high weight on comity,” said Reinhart, now a resident scholar at the American Enterprise Institute in Washington. Using interest on reserves for setting policy “can change the tenor of the discussions, and I don’t know how they get around it.”

If you’re wondering what this means, you might want to go an re-listen to the first 20 minutes of Michael Belongia’s interview with Russ Roberts on Econtalk.

HT:  Daniel Carpenter


Tags:

 
 
 

78 Responses to “Zero Rate Bound, RIP”

  1. Gravatar of Daniel Kuehn Daniel Kuehn
    26. January 2010 at 12:18

    OK, but a zero-lower-bound is only one, post-Japan, manifestation of the broader concept of the liquidity trap.

    ZIRP is a very intuitive way of talking about the problem. But the end of ZIRP doesn’t end the problem itself – Keynes, Hicks, etc. – none of these guys even talked about a zero lower bound explicitly.

  2. Gravatar of Doc Merlin Doc Merlin
    26. January 2010 at 12:52

    Side note: The fact that credit cards have rewards acts similarly to a negative interest rate for people who pay them off every month. They are given short term credit (less than a month) and at the same time paid for it.

  3. Gravatar of Doc Merlin Doc Merlin
    26. January 2010 at 12:53

    I forgot to add, this “effective negative rate” can be substantial as I get between 1 and 3 percent back on purchases.

  4. Gravatar of 123 123
    26. January 2010 at 13:21

    The whole zero rate bound issue is overrated. Negative interest rates sound very unjust to most people. I’ve heard that people in Germany have subtle psychological issues that cause them to have one percent rate bound, and ECB was able to run around this limitation by setting one year interest rate to one percent.

  5. Gravatar of Mark A. Sadowski Mark A. Sadowski
    26. January 2010 at 13:31

    Scott,
    You wrote:
    “But perhaps someone can explain this to me. Why would the Fed adopt a new policy ideally suited for overcoming the zero rate bound, at the very moment that they are about to embark on a policy of gradually raising interest rates? Isn’t the whole point of the interest on reserve program that it allows you to cut the policy rate below zero, and prevent banks from hoarding ERs?”

    The answer is in the Bloomberg article:
    “The central bank has been unable to control the federal funds rate since the September 2008 bankruptcy of Lehman Brothers Holdings Inc., when it began flooding financial markets with $1 trillion to prevent the economy from collapsing. Officials, who start a two-day meeting today, have said they may replace or supplement the fed funds rate with interest paid on excess bank reserves.”

    Perhaps you’re seeing what you want to see. I agree interest on ER opens the door to negative interest on ER but that’s evidently not on the minds of the Fed. Starting with the failure of Lehman the effective fed funds rate veered all over the place. They lost the ability to reliably set the fed funds rate in September 2008 (so there goes your central bank credibility) and they are merely paving the way to raise interest rates in the future. It would seem interest on ER is the tool they are going to use to tighten monetary policy without appearing to have gone completely spastic.

  6. Gravatar of Dan Carroll Dan Carroll
    26. January 2010 at 14:06

    I think the assumption has been all along that the purpose of the Fed paying interest on reserves was to use it as at least a key (if not the primary) tool for controlling the money supply in the future – i.e., to be able to reign in inflation quickly if it were to heat up. By announcing it now, the Fed is both formalizing the rule and signaling that it will be able to prevent an outbreak of inflaton for those who are worried about it. While you may not be concerned about inflation (and neither am I probabilistically), it is within the layout of potential scenarios. Therefore, reducing the probability of inflation in the perceptions of market participants gives the Fed more credibility and therefore more room to expand the money supply if it deems it necessary. If he has a credible exit strategy, then he can take more risks on the expansionary side.

  7. Gravatar of dlr dlr
    26. January 2010 at 15:03

    I believe the Fed still doesn’t have legislative authority to pay or charge interest on vault cash (the recent law only included non-cash reserves). So unless it gets that authority or otherwise changes the rules (like capping vault cash as a % of reserves), it still won’t be able to charge interest on reserves without banks switching to vault cash.

  8. Gravatar of Simon K Simon K
    26. January 2010 at 15:33

    dlr – Interesting point, but don’t banks need to have funds in their reserve accounts in order to lend the excess to one another through the fed funds market?

  9. Gravatar of The Ambrosini Critique » Blog Archive » “Zero Rate Bound, RIP” The Ambrosini Critique » Blog Archive » “Zero Rate Bound, RIP”
    26. January 2010 at 18:58

    [...] filling this one under “William Wallace Watch” (ref). Is there a difference between level targeting and [...]

  10. Gravatar of Bill Stepp Bill Stepp
    26. January 2010 at 19:09

    The Fed has never been able to control perfectly the Fed funds rate. It’s always influenced by the market to varying degrees. As for paying interest on excess reserves, doesn’t that discourage banks from lending, particularly to small businesses, at a time when doing so might lead to more employment and economic recovery?
    If I had a vote for Fed chairperson, it would go to either Alfred E. Newman or Lassie. They couldn’t be worse than the Helicopter Pilot.

  11. Gravatar of Doc Merlin Doc Merlin
    26. January 2010 at 21:56

    @dlr

    “I believe the Fed still doesn’t have legislative authority to pay or charge interest on vault cash (the recent law only included non-cash reserves). ”

    Hrm, if this is true, that explains who has been buying up all those t-bills. A trick to get government debt bought up.

  12. Gravatar of Jon Jon
    27. January 2010 at 00:26

    The Fed cannot pay interest on vault cash. The nose under the tend vis-a-vis, these payments were to compensate for the costs of hold reserves. The ECB does this for instance. Other countries simply eliminated reserve requirements entirely. I don’t make those statements as endorsement…

    Scott: I think you are wrong to think they will adopt negative rates.

    General point of discussion: banks practice yield conversion. Why would the interest on reserves matter? i.e., sure the opportunity cost of holding reserves is X but the rate on even a five year loan is much higher. I just don’t think this sterilization argument makes sense.

  13. Gravatar of scott sumner scott sumner
    27. January 2010 at 05:52

    Daniel. There are two aspects to the liquidity trap problem.

    1. Do zero rates stop central banks from adopting expansionary policies? Vitually everyone, including Krugman says no.

    2. Does the zero rate bound present problems for central banks who rely exclusively on the fed funds target as their policy level and signal? Yes. Going to IOR totally eliminates that problem.

    Are there other ways a central bank could screw up? Absolutely. But it takes one of the most common central bank mistakes off the table.

    Doc, That’s a good example.

    123, That’s a good point. I never believed the zero rate bound was a real barrier. It was psychological. Back around October 2008 the ECB had a target rate of around 4%, and they were having the same problems as the US, AD was falling off the table. So I kept telling people; “Look at Europe; the zero rate bound is not the problem here. Central banks are just way too slow to react to falling AD.”

    Mark, I know. The question was rhetorical (or sacrcastic.) It’s aimed at those people out there (99% of the population) who think “well you can’t blame the Fed, they did all they could.” Indeed there is another answer to my question. If the zero bound was a problem, and the Fed wanted more monetary ease than they were able to implement, then why did they only cut the target to 0.25%, why not to zero? Of course we all know the answer, the Fed thought 0.00% would be too expansionary.

    Dan, I agree, but see my answer to Mark. The Fed may be worried about inflation, but there are actually people in America who would like to see AD grow faster—including most economists and most Obama officials. Believe it or not, a big jump in NGDP is still treated as good news. The 4th quarter will probably come in over 5%, and people will cheer, not say “time to tighten.” But the Fed has different priorities. The Fed would rather see about 5% NGDP growth going forward, the Obama people (and me) prefer a couple years of 7% to partially catch up to trend.

    dlr, That’s not really a problem. If the Fed explained to Congress that this was a powerful tool to get money moving, Congress would approve in a heartbeat. Earlier I explained various ways the vault cash issue could be dealt with, including a cap on vault cash as a share of assets. See the old post I linked to in this post.

    BTW, I don’t mean to suggest that negative rates are my preferred policy option. I agree it is clumsy. My point is it can be done. NGDP futures targeting is far superior.

    Simon, Yes, but that doesn’t address dlr’s point, as the fed funds rate can be above the IOR.

    Bill, Yes, paying positive interest on ERs discourages lending. The Fed adopted the policy in early October 2008. Go figure.

    That’s why I want negative interest on reserves, to encourage lending.

    Jon, You said;

    “Scott: I think you are wrong to think they will adopt negative rates.”

    Didn’t I say the opposite? If they were going to adopt negative rates they would have done so already. That was the point of my Dr. Strangelove analogy. It is very unlikely that they will adopt negative rates. The Fed never even cut them to zero, they cut them to 0.25%, which is important in a liquidity trap, as that rate is still well above the T-bill yield, encouraging hoarding of reserves.

    I said they were going to adopt a target that eliminated the zero bound.

    Regarding your last point. Interest rates don’t matter, except to the extent that they impact the demand for base money. Right now a negative rate on ERs would have a big impact on the demand for base money.

  14. Gravatar of Jeff Jeff
    27. January 2010 at 08:37

    I used to work at the Cleveland Fed, and from that experience I expect the regional Federal Reserve Banks to fight this tooth and nail. Each of the 12 District Banks has a Board of Directors stacked with important local people who have no trouble getting the ear of their Congressmen and Senators. I don’t think the Board can win a political battle with them.

  15. Gravatar of jj jj
    27. January 2010 at 08:48

    We know ex post that the fed wasn’t aggressive enough with conventional monetary policy to even hit the zero bound, so we might assume that they wouldn’t have taken reserve rates negative either, if they had that instrument in 2008. However, if the market knew that there WAS a way to avoid the trap, maybe it wouldn’t have ever been necessary to bring rates that low.

    In other words — How much of the recent fall in NGDP was due to expectations that the fed might hit the zero bound and enter a liquidity trap?

  16. Gravatar of jj jj
    27. January 2010 at 09:03

    I think the effectiveness of rates below 0% is completely determined by the least-cost alternative method of holding cash. Assuming the fed charges an interest penalty on both reserves and vault cash, the lower bound will move from zero, to whatever the public’s cost of holding cash is — maybe -2% or something like that. So this new fed policy isn’t a comprehensive solution to liquidity traps, but it’s an improvement on the current lower bound.

  17. Gravatar of jj jj
    27. January 2010 at 09:08

    “Why would the Fed adopt a new policy ideally suited for overcoming the zero rate bound, at the very moment that they are about to embark on a policy of gradually raising interest rates?”

    Same reason Bernanke introduced paying interest on reserves when he did — because the hawks support it. I think Bernanke is very constrained by the hawks and has to sneak his changes in when it suits them.

  18. Gravatar of Marcus Nunes Marcus Nunes
    27. January 2010 at 09:08

    @jj
    I think you have it backward. The large drop in NGDP was not due to expectations that the fed might hit the ZLB. It was the result of expectations back in july-august-september 08 that the Fed would increase interest rates to “supress inflation”, which pulled down expected future NGDP growth (that impacts present NGDP growth).
    When the Fed tried to “turn the ship around” it was too late. Expectations of hitting the ZLB before the economy picked up was responsible for the introduction of fiscal stimulus.

  19. Gravatar of 123 123
    27. January 2010 at 09:26

    “If the zero bound was a problem, and the Fed wanted more monetary ease than they were able to implement, then why did they only cut the target to 0.25%, why not to zero? Of course we all know the answer, the Fed thought 0.00% would be too expansionary.”
    I think that Fed decided that 0.25 has no significant difference from 0, but allows money market funds to function. Once 0.25 bound was hit, Fed started QE.

  20. Gravatar of Mark A. Sadowski Mark A. Sadowski
    27. January 2010 at 10:26

    Scott,
    You wrote:
    “Mark, I know. The question was rhetorical (or sacrcastic.)”

    I know you know. What’s more I know you know that I know. But a double act doesn’t fully work unless the “stooge” responds to the “wag.”

  21. Gravatar of jj jj
    27. January 2010 at 10:53

    @Marcus Nunes

    The market clearly thought the Fed wouldn’t provide enough money, for whatever reason. To confirm your explanation we need to know the market forecast of fed funds, from the summer of 2008 — do you know where to get that? I’ve been searching but I can’t find the federal funds rate forecasts from that time period.

  22. Gravatar of Phil Koop Phil Koop
    27. January 2010 at 11:09

    “Isn’t the whole point of the interest on reserve program that it allows you to cut the policy rate below zero …?”

    No. The whole point is that it allows you to put a floor on interest rates, the better to control inflation. You have read too much into that article; if you look again, you will see that it is written from the perspective of fighting inflation, not deflation.

    But what if the Fed had the good sense to take your advice venture negative rates? The trouble is that A) once default expectations are accounted for, available lending opportunities have negative expected nominal returns, and B) banks are already too thinly capitalized. It is true that if real rates are sufficiently high it is possible that some lending opportunities might still yield positive real returns. But conventional liabilities are denominated in nominal dollars; the result of following your advice would be to create losses that banks cannot afford. Bankrupting more banks will do nothing to increase lending.

    Even if Fed funds rates turned negative along with the benchmark, that would merely move the problem to other banks. To allow your idea to work would require negative rates at the discount window. Hmm. Now that I say that, it occurs to me that this might actually work. But I think that politically it would be impossible.

  23. Gravatar of Matthew Yglesias » Excess Reserves Matthew Yglesias » Excess Reserves
    27. January 2010 at 11:30

    [...] the Fed could be doing to boost employment, look no further than this article on excess reserves (via): Banks’ excess reserves, or deposits held with the Fed above required amounts, totaled $1 [...]

  24. Gravatar of Doc Merlin Doc Merlin
    27. January 2010 at 11:38

    @Jon
    “General point of discussion: banks practice yield conversion. Why would the interest on reserves matter? i.e., sure the opportunity cost of holding reserves is X but the rate on even a five year loan is much higher.”

    Because the supply of good credit risks willing to sell you promises for future money is limited, and more limited in this recessionary environment.

  25. Gravatar of Eulogy for the Liquidity Trap | The Barstool Economists Eulogy for the Liquidity Trap | The Barstool Economists
    27. January 2010 at 13:16

    [...] http://www.themoneyillusion.com/?p=4002 [...]

  26. Gravatar of Bill J Bill J
    27. January 2010 at 15:03

    Isn’t it pretty much consensus that loan volume is set by loan demand, not by cost of borrowing reserves or supply of reserves?

    Given the state of the economy, there is only a given amount of potential demand for credit. Unless the end-users of credit can pay zero interest or less, how are any of these policies expected to be effective?

    Suppose FFR was negative 5%.
    Banks borrow reserves for free and can make 2% spread by loaning to end users at negative 3%. At a rate of neg. 3%, I can borrow unlimited amounts of money at zero risk and do literally nothing with it, and earn 3% annually. I can invest in 1 year t-bills and pocket their yield as well with zero risk. I can also risk the funds, lose money, and pocket the difference, as long as the loss is between 0 and 3%.

    Demand for credit hinges on the existence of profitable investment opportunities. As it should. If the entire world of businesses and entrepreneurship deem that profitable opportunities are absent, or limited, then nothing the FED does can change that, short of paying firms to borrow and subsidizing their losses. What a novel idea… institutionalize the subsidizing of loss throughout the entire economy, invert the entire profit and loss system. Anything to get credit going again.

    Has it occurred to anyone that the solution might involve clearing away the debt of the economy in some manner other than a poorly engineered inflation of prices?

    When a massive volume of credit is extended against a pool of assets at a price far above their economic value, than those debts must either default and be written off, or the income stream to service them will by necessity have to be DIVERTED from the productive economy to keep them afloat.

    Either the debts incurred for housing will be written down to their economic value, or the rest of the economy will have to divert its normal spending and demand away from goods and services and toward servicing bad debts.

    Just imagine the following scenario…

    10 trillion dollars is loaned into the economy over a period of ten years… the funds are invested in assets who’s yield is expected to cover the debt service of 5% with 2% left over… the assets in reality only yield 3.5%, and the market revalues them to be worth 5 trillion, not 10 trillion. unable to make a profit, and unable to service the debt, the borrower(s) defaults, the assets are sold and the lenders take a 50% loss on the liquidation of the re-valued assets. unless the rest of the economy wants to divert its demand and resources to service the debt, it must default. if it does not default, the money will be sucked out of the rest of the economy, interest AND principal, in a net-credit destruction. new credit will not be created to replace the amount withdrawn to service the massive debt…

    the banking system is net-destroying credit by not replacing the credit it withdraws from the economy in interest and principal… the economy is ‘loaned-up,’ the loans were squandered on assets that cannot in fact support the debt-service, and in order for the debts to be paid, money must be extracted from the healthy economy, creating a net-deflation, a decline in credit/money supply.

    the way out of deflation is to default the bad debts, let the banks take their medicine, let the loan-assets be sold/marked to market and written down, freeing the rest of the economy from an un-payable debt burden. the demand for goods, services and credit too, will revive.

    keeping the banks alive means killing the economy.

  27. Gravatar of Marcus Nunes Marcus Nunes
    27. January 2010 at 18:25

    Scott
    Alan Meltzer does not believe you:
    http://online.wsj.com/article/SB10001424052748704375604575023632319560448.html?mod=WSJ_Opinion_LEFTTopOpinion

  28. Gravatar of Roland Buck Roland Buck
    27. January 2010 at 20:23

    “Keynes, Hicks, etc. – none of these guys even talked about a zero lower bound explicitly.”

    Precisely. Keynes’ liquidity trap was not at zero interest. In Keynes’ liquidity trap, the long term rate (not short-term rates) would reach a floor significantly ABOVE zero below which it would not go because the speculative demand for money would cause the total demand for money become infinite at that point. And Keynes, writing in 1935, after the trough of the Great Depression had passed said that, while this may happen some time in the future, he knew of no instance in which this had happened.
    The validity of Keynes’ model in no way is dependent on there being a liquidity trap. It is based on adjustments in output, rather than in the interest rate, being the factor that makes desired saving and investment equal.

  29. Gravatar of Roland Buck Roland Buck
    27. January 2010 at 20:36

    “I think Bernanke is very constrained by the hawks and has to sneak his changes in when it suits them.”

    This is why Obama has to follow the reappointment of Bernanke with promptly filling the two unfilled slots on the Board of Governors with people who take the Fed’s mandate of achieving maximum employment seriously. (Nobody who thinks the economy is in a liquidity trap and that monetary policy has lost its effectiveness please.)

    These two postitions have been vacant for much of last year. I fail to understand why that is and have tried using Google to find out why it is, but have been utterly unable to find any article that deals with this.

  30. Gravatar of sraffa sraffa
    27. January 2010 at 23:27

    A couple of questions:

    1) Any more thoughts on Sweden and their negative interest rate policy? Has it been a success, failure, too soon to tell, etc. I read the blog occasionally, so maybe you’ve already mentioned this elsewhere.

    2) Will negative interest rates bind? What’s to stop the following scenario: Fed sets excess reserve interest rate to be -0.5%. It tries to do an open-market operation(or buy MBS), but no banks choose to sell their assets as they don’t want to hold cash given negative interest rates. This means that once ZIB binds (or whatever reason excess reserves would start accumulating), open-market operations are ineffective, as these excess reserves yield a negative return.

    I like the idea of negative-interest rates, but I’m not totally convinced how this would work in practice. It seems like there’s some way banks could avoid this.

  31. Gravatar of Marcus Nunes Marcus Nunes
    28. January 2010 at 04:06

    @jj
    I managed to get the FF futures for the period (summer 08). On July 8, the expected FF for the August 8 meeting was 3% (an expected increase from the 2% that was in place since the April meeting). On August 16, the FFF indicated again that rates would rise (to 3%) on the September meeting. Remember that in the August meeting Dallas Fed president Fisher had voted for an immediate rise and that in the Statement it was said that the next move of rate would likely be UP!!!

  32. Gravatar of Jon Jon
    28. January 2010 at 09:37

    Scott,

    Allan Meltzer makes just the same point I do in todays WSJ

  33. Gravatar of TheMoneyIllusion » Guess who discussed negative interest rates on money in 1998? TheMoneyIllusion » Guess who discussed negative interest rates on money in 1998?
    28. January 2010 at 11:25

    [...] few days ago I did what I thought was one of my more important posts, pointing out that the Fed’s likely new operating target will eliminate the problem of the [...]

  34. Gravatar of scott sumner scott sumner
    28. January 2010 at 19:21

    Jeff, Maybe, but I wouldn’t be too sure. Have you listened to the Mike Belongia interview?

    jj, Those are good questions that I have been struggling with all year. The Fed could have done much more in 2008, so the zero bound doesn’t seem like it should have been a problem. And yet almost everyone seems to think it was a problem, so that probably did affect expectations. And of course the Fed never tried to dispel those myths. Why didn’t the Fed ask for legislative authority to adopt negative rates?

    jj#2, Liquidity traps never were a problem. And the policy is a definitive solution to the zero lower bound. My newest post tries to clear up all the confusion here. The concept of the zero lower bound and the concept of monetary policy ineffectiveness are distinct, and should not be mixed up.

    jj#3, Yes, it was a rhetorical question. Even the Fed admitted the intention was contractionary.

    Marcus, I think you are both right. The anti-inflation policy hurt a lot in September, and fear of a liquidity trap hurt later in the fall.

    123, I disagree, and so does James Hamilton. Hamilton argued that a zero rate would have led to excessive inflation once the base doubled. I don’t know if he was right, but I think people at the Fed were afraid that he might be right. The Fed wasn’t willing to risk hyperinflation.

    Mark, I’m dizzy now. But thanks.

    JJ#4, I don’t know either, but go to the St Louis Fred and get 3 month T-bill data. Those yields are a decent forecast of future expected fed funds rates.

    Phil, You should read the post I linked to. It can be set up in a way that doesn’t hurt bank profits. And banks can always invest in Treasuries.

    I do understand your point about fear of inflation, my question was rhetorical, or kind of sarcastic.

    Bill, You said;

    “Isn’t it pretty much consensus that loan volume is set by loan demand, not by cost of borrowing reserves or supply of reserves?

    Given the state of the economy, there is only a given amount of potential demand for credit.”

    First, the purpose of negative rates is primarily to get banks to hold fewer excess reserves, not to get them to lend more.

    Second, the whole point is to improve the “state of the economy” so you can’t hold that as a given. With higher inflation expectations there is more loan demand.

    You really need to read the post I link to, as it is clear you don’t understand the proposal. I am not talking about subsidizing banks to lend money.

    Marcus, Meltzer in just playing right into Krugman’s hands. In a few years Krugman will be saying “the monetarists all predicted high inflation, see how bankrupt their model is.”

    Roland, I recall answering your comment somewhere. Keynes was horribly confused about liquidity traps–people would be better off studying Krugman than Keynes.

    sraffa, Unfortunately, Sweden set it up in a way that can be easily evaded, so it is merely symbolic. In any case, Sweden’s not part of the euro, so they don’t face any liquidity trap problems.

    Jon, I get 100s of comments. You’ve got to remind me what your comment was. In general, I wasn’t impressed with the Meltzer article. I am afraid monetarism is dying out.
    Let’s hope someone like Mike Belongia can revive it.

  35. Gravatar of Doc Merlin Doc Merlin
    28. January 2010 at 20:37

    @Bill J
    ” the entire world of businesses and entrepreneurship deem that profitable opportunities are absent, or limited, then nothing the FED does can change that, short of paying firms to borrow and subsidizing their losses.”

    Thats exactly what the fed has been doing lately with its buy up of housing assets.

  36. Gravatar of Doc Merlin Doc Merlin
    28. January 2010 at 20:40

    “Isn’t it pretty much consensus that loan volume is set by loan demand, not by cost of borrowing reserves or supply of reserves?”

    I believe in finance yes, in economics well… not so much.
    Its a point I have been trying to make here, that the supply of credit is really not very helpful, what is important is the supply of good credit risks who want loans.

  37. Gravatar of 123 123
    29. January 2010 at 02:34

    “I disagree, and so does James Hamilton. Hamilton argued that a zero rate would have led to excessive inflation once the base doubled. I don’t know if he was right, but I think people at the Fed were afraid that he might be right. The Fed wasn’t willing to risk hyperinflation.”
    This is strange, because in Japan central bank balance sheet expansion and ZIRP has created no risk of hyperinflation.

  38. Gravatar of ssumner ssumner
    29. January 2010 at 14:24

    123, Good point, and you might be right. But the expansion happened much faster here, and before rates had fallen to zero. The MB nearly doubled in a month or so. And the fed funds target was still at 2%. Don’t get me wrong, I think there is at least a 50/50 chance you are right, but the Fed wasn’t willing to take that risk.

    Because I favor futures targeting, I don’t worry much about us stumbling into hyperinflation accidently. If it happens, we’ll do it on purpose. But the Fed doesn’t seem to believe in futures targeting.

  39. Gravatar of Roland Buck Roland Buck
    29. January 2010 at 21:23

    “Roland, I recall answering your comment somewhere. Keynes was horribly confused about liquidity traps–people would be better off studying Krugman than Keynes.”

    You answered this in the post int “The Money Illusion.” See my reply to this post. On pages 207 and 208 Keynes lists 4 items that can keep the interest rate too high.

    The liquidity trap is explined in item 2 and my posting above is based on this. Keynes’ exposition there is clear and unambiguous and Keynes asserted that he knew of no instance of when one had occured. My point is that Keynes’ concept of the liquidity trap, as explained in item 2, is very different from Krugman’s.

    The problem with the gold standard you refer to is discussed separately in item 3, and that is clearly not a description of the liquidity trap.

  40. Gravatar of Roland Buck Roland Buck
    29. January 2010 at 21:25

    “The Fed wasn’t willing to risk hyperinflation.”

    Worrying about inflation when the economy is at 10% unemployment is the equivalent of a person lost in the Sahara and dying of dehydration worrying about drowning.

  41. Gravatar of Roland Buck Roland Buck
    29. January 2010 at 21:44

    “Given the state of the economy, there is only a given amount of potential demand for credit.”

    Even with the economy in a bad recession, so that most additional expenditures that are financed by borrowing will not be made, there will always be some expenditures that are just at the margin of being made and not made. A reduction in the relevant interest rates (long-term rates, not the yield on federal funds) will cause these marginal expenditures to be made and therefore the funds to be borrowed to finance them.

    Therefore lower long-term interest rates will increase expenditures. However, in a bad recession it is POSSIBLE that, even at zero interest, not ENOUGH of such additional expenditures would be made to restore the economy to full employment. In that case, rather than in a liquidity trap, monetary policy would lose its ability to restore the economy to potential output(though it would retain the ability to increase output at least PARTIALLY move the economy to potential output).

  42. Gravatar of Bill J Bill J
    30. January 2010 at 15:13

    To Scott and others who replied…

    How can there be any increase in inflation or in inflation-expectations with the supply of credit undergoing net destruction? Deflation of credit is the policy of the banking system. I.e., withdrawing interest and principal from mortgage-debtors that are not covered by the asset in question-the house. Credit is not being extended to replace the credit being withdrawn. If the housing stock behind the loans was capable of yielding enough to support the loans, there would be no drain on AG. Since the mortgage debtors must re-direct demand to debt service, the AG of the economy must shrink. Any large extension of debt that is squandered will by necessity destroy AG, unless the debt is allowed to default and be written down to its real value.

    The money to repay banks has to come from the rest of the economy, since the mass of housing assets were financed at prices that do not reflect their underlying yield. They either must default or be paid by shrinking the rest of the economy.

    The failure of demand in the broad economy is a result of the re-direction of funds originally destined for goods and services, to repayment of mortgage debt.

    Preventing housing loans from being revalued in the market and written down to their economic value is strangling the AG of everyone. With demand shrinking, credit will continue to undergo net contraction. If the banks don’t liquidate their assets at market prices, AG has to shrink to support the debts. That’s why the bailouts were bad policy. They prevented the mortgage-debt from being written down to levels homeowners could afford without dipping into AG.

    Debts, to be sustainable, must be payable out of the real income generated by the financed-asset. Otherwise, the economy must divert real demand to service the debt.

  43. Gravatar of scott sumner scott sumner
    31. January 2010 at 08:00

    Roland, I answered the Keynes point on another thread.

    You said;

    “Worrying about inflation when the economy is at 10% unemployment is the equivalent of a person lost in the Sahara and dying of dehydration worrying about drowning.”

    I agree they should not have been worried about inflation. Having said that, there are many examples throughout history of inflation and unemployment coinciding. But I agree, they shouldn’t have been worried about inflation in 2008, or today.

    Roland#3, Was that last quotation mine?

    You said;

    “Therefore lower long-term interest rates will increase expenditures. However, in a bad recession it is POSSIBLE that, even at zero interest, not ENOUGH of such additional expenditures would be made to restore the economy to full employment. In that case, rather than in a liquidity trap, monetary policy would lose its ability to restore the economy to potential output(though it would retain the ability to increase output at least PARTIALLY move the economy to potential output).”

    Monetary policy can still increase NGDP as much as you’d like, even if nominal rates are zero.

    Bill J. What is “AG”?

    Monetary policy can create hyperinflation even if every bank in the country were closed down and there was no credit at all.

  44. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    31. January 2010 at 11:10

    “123, Good point, and you might be right. But the expansion happened much faster here, and before rates had fallen to zero. The MB nearly doubled in a month or so. And the fed funds target was still at 2%. Don’t get me wrong, I think there is at least a 50/50 chance you are right, but the Fed wasn’t willing to take that risk.

    Because I favor futures targeting, I don’t worry much about us stumbling into hyperinflation accidently. If it happens, we’ll do it on purpose. But the Fed doesn’t seem to believe in futures targeting.”

    I still think that Fed sees no big difference between 0 and 0.25 on reserves now. What is important for Fed is having the tool of paying interest on reserves in future when inflationary pressures actually arise.

  45. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    31. January 2010 at 11:35

    Also, the balance sheet expansion in the US happened after the shock to the credit system that changed the velocity of money, just like in Japan.

  46. Gravatar of Bill J Bill J
    31. January 2010 at 15:59

    Scott,

    Inflation, in our system, has always been caused by credit, not base money supply. How much currency do you suppose the FED would need to create to overcome the contraction in credit being carried out by the banking system?
    Before the crisis, there was less than 1T in currency in the entire global economy, much of it overseas. The dynamics of inflation and deflation were actually credit dynamics, in my opinion, and largely independent of FED control.
    The massive balloon of credit for real estate, which circulated throughout the economy during the boom, and which is now collapsing, is magnitudes greater than the supply of base money. How can fiddling with the monetary base be expected to overcome credit dynamics in the tens of trillions of dollars?

    As to my previous post, AG was supposed to read AD, aggregate demand… if you don’t mind reading it again with corrections. Sorry for the mistake.

    “How can there be any increase in inflation or in inflation-expectations with the supply of credit undergoing net destruction? Deflation of credit is the policy of the banking system. I.e., withdrawing interest and principal from mortgage-debtors… Credit is not being extended to replace the credit being withdrawn.

    If the housing stock behind the loans was capable of yielding enough to support the loans, there would be no drain on demand. Since mortgage debtors must re-direct demand to debt service, the AD of the economy must shrink.

    Any large extension of debt that is squandered will by necessity destroy AD, unless the debt is allowed to default and be written down to its real economic value.

    The money to repay banks has to come from the rest of the economy, since the mass of housing assets were financed at prices that do not reflect their underlying yield. They either must default, or be paid by shrinking the rest of the economy.

    The failure of demand in the broad economy is a result of the re-direction of funds originally destined for goods and services, to repayment of mortgage debt.

    Preventing housing loans from being revalued in the market and written down to their economic value is strangling the AD of everyone. With demand shrinking, credit will continue to undergo net contraction.

    If the banks don’t liquidate their assets at market prices, AD has to shrink to support the debts. That’s why the bailouts were bad policy. They prevented the mortgage-debt from being written down to levels homeowners could afford without diverting AD from the real economy.

    Debts, to be sustainable, must be payable out of the real income generated by the financed-asset. Otherwise, the economy must divert real demand to service the debt.”

  47. Gravatar of Bill J Bill J
    31. January 2010 at 16:25

    Checkout:
    The Fed cannot inflate. Buy Bonds
    http://www.winterspeak.com/2010/01/fed-cannot-inflate-buy-bonds.html

  48. Gravatar of Bill J Bill J
    31. January 2010 at 16:36

    Also see:
    Time for Inflationists to Put Up or Shut Up, by Rick Ackerman
    http://www.rickackerman.com/2010/01/time-for-inflationists-to-put-up-or-shut-up/

  49. Gravatar of scott sumner scott sumner
    1. February 2010 at 07:08

    123, The Fed would still have that tool at 0% or -0.25% IOR. If they are using the reasoning you suggest, then we have just suffered a severe recession from an absurd lapse in reasoning by the Fed.

    Now is not the time to worry about inflation, now is the time to try to create more inflation. If Bernanke is opposed to more inflation, he should not have been reappointed. That’s because if he is opposed to more inflation, then ipso facto he is opposed to more NGDP growth. And that’s just nuts. But I don’t think he is nuts, just not very forceful.

    123#2, Yes, but it didn’t cut base velocity in half. That’s never happened anywhere in a short period. Rather the IOR cut base velocity in half. The financial crisis led to a modest fall in velocity.

    BillJ,

    You said;

    “Inflation, in our system, has always been caused by credit, not base money supply. How much currency do you suppose the FED would need to create to overcome the contraction in credit being carried out by the banking system?
    Before the crisis, there was less than 1T in currency in the entire global economy, much of it overseas.”

    These two statements contradict each other. The lower the demand for cash, the less they have to create to hit any NGDP objective. Becasue cash is usually not a close substitute for other assets, small increases in the base have a huge effect. That’s why it’s called high-powered money. When T-bill yields are near zero, the demand for cash is more elastic, and somewhat larger increases are needed, but not all that large. In any case, the big problem now is ERs. Put a negative rate on ERs and you could get inflation with a much smaller monetary base.

    Regarding your previous post, debt is a zero sum game. The gain to one side is a loss to the other. AD is not directly affected. If a debt crisis causes a drop in nominal rates, then it may increase base demand, which is deflationary if the Fed doesn’t offset it. Unfortunately the Fed dropped the ball in late 2008.

    In my view the debt crisis is not a cause of falling AD, but a symptom. As NGDP falls people have less income to repay loans, this causes defaults to rise. but the root cause of NGDP falling is tight money, a money supply too small to support 5% NGDP growth that was anticipated when the debts were contracted.

    FDR showed it is possible to generate fast economic growth during a severe credit crisis; we should take a look at what he did, which was basically price level targeting.

    The first post you link to shows little awareness of recent work on liquidity traps. Central banks are not powerless during liquidity traps, that idea was discredited long ago. I have dozens of posts that discuss why.

    The second post is right about inflation not being a problem in the near future, but for the wrong reason. The Japanese never used an expansionary monetary policy to try to escape delation. The BOJ is firmly opposed to any inflation at all, and prefers stable or mildly falling prices.

  50. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    1. February 2010 at 13:30

    “123, The Fed would still have that tool at 0% or -0.25% IOR. If they are using the reasoning you suggest, then we have just suffered a severe recession from an absurd lapse in reasoning by the Fed.”
    Bernanke’s previous academic research on credit channel and financial accelerator supports the idea that when banks are bad, there is no big difference between +0.25% and -0.25%. This is the reason he supported the TARP and stress tests, as it is the only way low interest rates can get any traction.

    “Now is not the time to worry about inflation, now is the time to try to create more inflation. If Bernanke is opposed to more inflation, he should not have been reappointed. That’s because if he is opposed to more inflation, then ipso facto he is opposed to more NGDP growth. And that’s just nuts. But I don’t think he is nuts, just not very forceful.”

    Now it is the time to worry about both medium term deflation and long term inflation.

    “123#2, Yes, but it didn’t cut base velocity in half. That’s never happened anywhere in a short period. Rather the IOR cut base velocity in half. The financial crisis led to a modest fall in velocity.”
    The financial crisis has led to a modest fall in M2 velocity. It has also led to a dramatic fall in M0 velocity. 25 bps is just peanuts for banks.

  51. Gravatar of Bill J Bill J
    1. February 2010 at 15:18

    Scott,

    The millions of people delinquent on their mortgages, and the millions more cutting their household spending to the bone to stay current on theirs mortgages, credit card bills, auto loans, and other NON-PRODUCTIVE DEBT, are the SOURCE

  52. Gravatar of Bill J Bill J
    1. February 2010 at 16:13

    Scott,

    I do not agree that ‘debt is a zero sum game.’

    The millions of people delinquent on their mortgages, and the millions more cutting their household spending to the bone to stay current on their mortgages, credit card bills, auto loans, and other NON-PRODUCTIVE DEBT, are the SOURCE of the collapse of aggregate demand and therefore GDP (R or N).

    At the time they incurred these debts, including the massive cash-out refinancing of home equity, I do not believe households expected either A) to reduce their future spending drastically to service their debts, OR B) to see their incomes rise drastically to cover debt service.

    I believe they took on these obligations with the same expectation as the financial sector, that home values and the other assets which constituted their networth would continue to rise at the same pace as in the years prior to their decision. Some were right and early, others were late and wrong, but this was the basic gamble that the household sector and the financial sector took when levering-up their balance sheets on rising asset prices.

    They did this without regard to the actual YIELD of these assets, their rental value, which could not cover the cost of servicing the mortgage WITHOUT price appreciation.

    All this debt that was incurred WITH NO WAY to pay it back absent continued housing price inflation, contributed to AD and GDP. Not just cashing out equity, but ALL the spending, borrowing, & dis-saving that took place as a result of expectations about increasing real estate values.

    Also the cash that was FREED UP by ARMS, teaser rates, neg-am, interest only… as well as ZERO DOWN PAYMENTS, credit card, auto and student debt RUN UP on the basis of home values, or consolidated into second mortgages AGAINST fictitious and impossible home prices.

    WHERE WAS THE MONEY GOING TO COME FROM TO PAY THESE DEBTS, UNDER ANY CIRCUMSTANCES? THE DEBTS WERE NON-PRODUCTIVE, THEY WERE NOT SELF-LIQUIDATING IN ANY SENSE; THEY EITHER FINANCED PURE CONSUMPTION OR PURCHASED ASSETS WHO’S YIELD COULD UNDER NO CIRCUMSTANCES LIQUIDATE THE DEBTS.

    ALL of this spending created artificial demand and artificial growth in GDP! That spending was simply FUTURE DEMAND pulled forward to the present (2003-2007), and could not be sustained for the simple reason that the DEBT INCURRED could only be repaid by draining money from the rest of the economy, to repay the financial sector.

    It is not ZERO SUM, for the simple reason that NET INFLOWS TO THE FINANCIAL SECTOR DO NOT CREATE DEMAND in place of the demand DESTROYED by UNPRODUCTIVE DEBTS. The financial sector simply draws as much as it can from the rest of the economy (the non-defaulting part of the public) which must reduce its spending to a commensurate degree. Those who cannot reduce their spending default on a large scale, which destroys the assets of the financial sector, creating insolvency (when allowed by the government). The PAYMENT of such debts destroys the demand for goods and services of those debtors who decide to service and repay their debts by reducing spending on real things.

    This money simply disappears, it is not replaced by new credit from the financial sector. THAT IS THE DEFLATION; THE NET DESTRUCTION OF CREDIT BY THE FINANCIAL SECTOR when it extends too much lousy and unpayable debt.

    And, of course, this contraction of credit and spending is the source of falling GDP, NOT THE RESULT of falling ‘NGDP.’

    When credit is squandered on useless activity it cannot be paid back, except out of the remaining economic activity… and when this happens on a massive scale relative to overall GDP and overall credit, a collapse in AD is inevitable.

    It is then necessary for the market to clear the debt and adjust the prices and re-adjust overall activity to suit the new reality. OR, government can come in and cancel debts, create money, etc., via fiscal policy. Printing base money and handing it out to banks, while strangling the household sector with debts, will not restore AD.

    The market can do it, or fiscal policy, done right, can do it. Which one is optimal I don’t know.

  53. Gravatar of Bill J Bill J
    1. February 2010 at 20:19

    Scott,

    Try to think about it this way.

    Imagine that every dollar which was loaned out for real estate, above it’s long term average or trend line, was actually unsecured, non-recourse consumer debt. Because that’s actually what it was and is.

    What would the interest rate of that debt be?
    How about 28-29%, or, as much as legally allowed to the banks.

    More to the point: How much of that debt would have been created in the first place, if it were acknowledged to be purely consumer credit with no collateral, no economic value or yield to cover the stream of income a mortgage represents?

    And, if it was extended, what kind of toll would it take on the economy and aggregate demand the moment this credit was cut off and called in?

    What would happen to the market value of assets collateralized by such loans, sitting on the balance sheets of the financial sector?

    Well, if such credit was extended at 4-6%, massively under-pricing the REAL risk profile and payment-potential of such loans, it would sink every bank holding such assets, depending on their leverage and capitalization.

    Finally, what would happen to the macro-economic/credit DYNAMICS of this economy?

    A huge portion of this debt would default outright the moment it could not be refinanced or rolled over. Without additional credit, that portion which continued to be serviced and paid off on time would be withdrawing money from ongoing economic activity. This re-direction of the flow of spending solely to EXTINGUISH DEBT would reduce AD, reduce MV (velocity), and reduce broad money supply.

    Demand for goods and services would contract; so would loan demand and employment demand.

    As long as money was flowing NET TO the financial sector (MONEY DESTRUCTION), to extinguish debts, rather than NET FROM the financial sector (MONEY CREATION), to create credits, deflation would result, with a re-gearing of economic activity and relationships to fit the new picture of demand and income.

    FAILURE TO DEFAULT ANY AND ALL DEBTS NOT SUPPORTED AND SERVICED FROM PROFIT-YIELDING ACTIVITY OR INVESTMENT IS A NET LOSS TO AND A DRAIN ON THE ENTIRE ECONOMY.
    IF THIS MEANS LOSS OF BANKER’S CAPITAL OR BANK LIABILITIES, SO BE IT; CLEAR THE UNECONOMIC DEBTS ON THE MARKET AND ALLOW SOLVENT INSTITUTIONS TO TAKE OVER THE DEPOSITS. IF NATIONALIZATION IS NECESSARY TO PREVENT A TOTAL COLLAPSE OF DEPOSITS, WHICH IS DOUBTFUL, SO BE IT; AT LEAST THE BAD DEBTS ARE CLEARED AND THE SYSTEM CAN BE STABILIZED.

    THE ONLY THING PREVENTING THESE THINGS FROM HAPPENING ARE POLITICS AND ITS WEAPON IDEOLOGY.

  54. Gravatar of Doc Merlin Doc Merlin
    1. February 2010 at 23:07

    Re: debt, zero sum game

    Debt is only a zero sum game when savings = debt.

  55. Gravatar of Bill J Bill J
    2. February 2010 at 12:03

    It’s not zero sum in this sense: money blown today on consumption cannot be recovered through any policy measure, monetary or fiscal. Either the debt can be recouped by the financial sector through shrinking the economy and withdrawing money, or the loss can be borne by the bank’s shareholders and creditors and the debt’s extinguished.

    Trillions of dollars are owed by millions of people to a few state-backed mega-institutions. To recover those debts, which are neither secured with collateral nor servicable from investment, income and demand must be diverted from profitable economic activity to extinguish the debts. The banks are forced by their previous lending to withdraw money from the economy, which cannot be replaced profitably with a new stream of credit. So on NET, credit is destroyed and extinguished in a flow FROM the economy TO the banks.

    That’s the nature of FRB. Inflation and deflation are a credit phenomenon, not a monetary one. Inflation and deflation in prices can swing wildly without a change in base money. That’s why gold didn’t respond for years of credit expansion as long as the money remained relatively fixed. The market understood that the inflation was fueled by credit and would reverse once the flow of credit reversed. All that spending and demand that inflated GDP and various price indexes was future demand pulled forward by unsound debt that was not used to finance investment.

    The best bet the FED has is to print money to fund deficits via payroll/income tax holiday, until DEBT to GDP is reduced to sustainable levels and AD returns… But let the rest of the economy adjust naturally on a micro-level. Stop the moral hazard and bailouts and subsidies to the crooks.

  56. Gravatar of ssumner ssumner
    3. February 2010 at 13:43

    123, Even if I agreed with your view that 0.25% is unimportant, which I don’t when T-bills are yielding 0.1%, I would still disagree with your answer. You are talking about credit demand, Iam talking about money demand. Would Bernanke say a negative rate of 3% would have no impact on ERs? I doubt he even discussed that possibility.

    If you have a link to him talking about a negative 0.25% rate on ERs, I’d love to see it.

    Bill J, I didn’t say money spend on consumption was a zero sum, I said debt was. If the real debt goes up due to deflation, one side gains and the other side loses. The same with default—one side pays less and one side receives less. Again, monetary policy determines the trajectory of NGDP, debt problems are a symptom of NGDP that has falled far below trend.

    Of course it’s true that many heavily indebted people are cutting back on consumption, but the question is why? And my answer is because NGDP has fallen. And that is because money was far too tight.

    Doc Merlin. Dbt and saving are different. If I lend you $100 the total debt in society has increased, but saving is unchanged.

  57. Gravatar of Bill J Bill J
    4. February 2010 at 12:28

    ssumner,

    How is it possible for NGDP not to fall when consumption plunges, credit contracts and demand for credit collapses.

    The Fed doesn’t control GDP, nominal or otherwise. Unless it has some way of handing money over to debtors, who for years fueled GDP by their borrowing and spending, and are now unable to do so, GDP will fall. Much of the growth of the past 8 years was driven by debt accumulation. Unless credit continues to grow at an accelerated rate, NGDP should reverse as debt is paid down rather than increased.

    Buying bonds with printed money should not be expected to increase NGDP, since all it does is accumulate as excess bank reserves and distort the mortgage market, encouraging more malinvestment in real estate to occur. The government including the FED would probably be content to do nothing if the banks weren’t endangered by it. All the fiscal and monetary policy so far has been to bail out the banks. I don’t believe the fed cares about price stability or unemployment much. It safeguards the interests of the banks.

    Price stability over the long term would be the natural consequence of a fixed money supply, allowing productivity gains to gradually lower the price of increased output of final goods, benefiting everyone. Unemployment would rarely become a problem without the macro distortions created by the fed’s control of interest rates and money supply.

  58. Gravatar of Bill J Bill J
    4. February 2010 at 12:54

    But we seem to be talking past each other. Let me instead ask you this. What would you do as a FED chairman. Please explain in some detail what you would do today, or would have done starting in 2000, 2006, or wherever you feel comfortable. I have my doubts that you have any constructive policy idea, with specific details. But maybe Ben Bernanke or the next chairman will read your response or blog posting.

  59. Gravatar of ssumner ssumner
    6. February 2010 at 06:59

    Bill J, I’d announce an NGDP market. Create a NGDP futures market. Subsidize trading in the market. And inject enough cash into the economy until the NGDP futures price was equal to the target. I’d also put a negative interest rate on excess reserves.

  60. Gravatar of Bill J Bill J
    11. February 2010 at 09:23

    ssumner,

    a) do you have an opinion as to why there is no gdp or ngdp futures market after all these years?
    b) when you say ‘inject enough cash into the economy,’ what do you mean? what would you be purchasing, and from who?
    c) what if your pumping failed to elevate ngdp futures to the desired level? would you continue pumping, pumping, pumping, regardless of the distorting effect on whatever asset you were purchasing, such as t-bills?
    d) what if it took a tripling, quadrupling, quintupling, etc., of the base money supply to get speculators to bid your ngdp futures up to your desired levels? and if this required the total mis-pricing of the assets being accumulated, such as t-bills and their interest rates?
    e) how would you manage NOT to overshoot and generate a sudden snap beyond the target, due to over-pumping?
    f) why do you have condifence that the speculating community could accurately project ngdp through the futures market, considering the wild swings of futures market prices for all other commodities and contracts traded on the exchanges?
    g)especially considering all the unpredictable distortions your unlimited monetary pumping would create in the markets and the very uncertainty such an unprecedented policy would engender?
    h) what if, finally, your pumping still failed to create the desired ngdp futures price (whether correct or not)?
    i) put another way, what if reaching your ngdp target required a uselessly inflationary and distorting influence on the markets without positively assisting any re-balancing of relative prices, debt levels and economic activity and relationships to a sustainable equilibrium?
    i.e., what if the money illusion is, in fact, an illusion, where it counts?

  61. Gravatar of ssumner ssumner
    12. February 2010 at 07:00

    Bill,

    a. No, but it’s good there isn’t a private market–it indicates little interest in hedging NGDP risk.

    b. I’d buy T-securities.

    c. Yes.

    d. It wouldn’t take much at all, indeed the base would have been far lower in late 2008 if investors had expected 5% NGDP growth. But if it increased 5-fold, so be it. By the way, the same problem could occur with the Taylor Rule, or inflation targeting, or any other regime.

    e. NGDP futures prices are observable in real time. the point is not to stabilize NGDP, but rather to stablize NGDP expectations.

    f. See previous answer. The better analogy is the gold standard. That price was kept fixed for 54 years (1879-1933), with no wild swings. And all that time there was a completely free market for gold. Anyone could buy and sell gold. My proposal is a gold standard, but replacing gold with NGDP futures. That’s because we care more about NGDP expectations than gold prices.

    g. The economy is unstable when NGDP expectations fluctuate, as in late 2008. My proposal would make the economy more stable.

    h. Then the Fed would own the entire world, and all Americans could retire, living off the hard work of other countries. We would own all the assets in the world.

    i. I don’t care about inflation, I care about NGDP growth. If I cared about inflation, I’d favor targeting infaltion. As a practical matter NGDP growth of 5% will keep inflation fairly well behaved, but that is a side issue.

  62. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    14. February 2010 at 14:29

    “Even if I agreed with your view that 0.25% is unimportant, which I don’t when T-bills are yielding 0.1%”
    When T-bills are yielding 0.1% and are fully substitutable to ERs that yield 0.25% then the rate of 0.1% is more important for understanding of the monetary policy. 0.24% is just a subsidy that some lucky banks receive but it does not change their behaviour.

  63. Gravatar of Bill J Bill J
    14. February 2010 at 15:34

    ssumner, thanks for the clarifications.

  64. Gravatar of ssumner ssumner
    15. February 2010 at 09:16

    123, I disagree. 0.25% is important because it causes a massiv eincrease in ERs. If banks earned minus 2% on ERs, the demand for ERs would be much lower and the price level much higher. Indeed Hamilton claims we would have had high inflation with even a 0% rate on ERs.

    Bill, you are welcome.

  65. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    15. February 2010 at 11:17

    Bernanke in his recent testimony before the congressional committee said:
    “By late 2008, this target reached a range of 0 to 1/4 percent, essentially the lowest feasible level. With its conventional policy arsenal exhausted and the economy remaining under severe stress, the Federal Reserve decided to provide additional stimulus through large-scale purchases of federal agency debt and mortgage-backed securities (MBS) that are fully guaranteed by federal agencies. In March 2009, the Federal Reserve expanded its purchases of agency securities and began to purchase longer-term Treasury securities as well. ”
    and
    “Most importantly, in October 2008 the Congress gave the Federal Reserve statutory authority to pay interest on banks’ holdings of reserve balances. By increasing the interest rate on reserves, the Federal Reserve will be able to put significant upward pressure on all short-term interest rates, as banks will not supply short-term funds to the money markets at rates significantly below what they can earn by holding reserves at the Federal Reserve Banks.”

    I think this testimony is sincere and it was not influenced by the need to reflect the consensus of the FOMC. This testimony supports my view that:
    1. 0-0.25 range is the practical equivalent of zero
    2. When zero interest rate is approached, the next step should be QE, and not the negative interest rates
    3. Payment of interest on reserves will have any effect only when Fed will want to push short term interest rate higher at some time in the future

  66. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    15. February 2010 at 11:39

    This footnote from Bernanke’s testimony is even more interesting:
    “The Federal Reserve believes it is possible that, ultimately, its operating framework will allow the elimination of minimum reserve requirements, which impose costs and distortions on the banking system. ”
    This is also a signal to those who obsess with the role of reserves too much.

  67. Gravatar of scott sumner scott sumner
    16. February 2010 at 08:04

    123, You said;

    “I think this testimony is sincere and it was not influenced by the need to reflect the consensus of the FOMC. This testimony supports my view that:
    1. 0-0.25 range is the practical equivalent of zero
    2. When zero interest rate is approached, the next step should be QE, and not the negative interest rates
    3. Payment of interest on reserves will have any effect only when Fed will want to push short term interest rate higher at some time in the future”

    Your first point confus two completely separate issues. Th zero to 0.25% range refers to the ff target, not the IOR rate. I’ve never denied that the ff target was near the lowest level possible. I was talking about the IOR, which obviously can go negative.

    I don’t deny that you and Bernanke agree, but you haven’t provided any evidence that he is correct. QE is not the proper policy when you are in a liquidity trap. Rather price level of NGDP targeting is required. Bernanke once knew that, but somehow forgot.

    123#2, I am all for eliminating reserve requirements. But we don’t need IOR to do that.

  68. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    17. February 2010 at 13:48

    “Your first point confuses two completely separate issues. The zero to 0.25% range refers to the ff target, not the IOR rate.”
    For money multiplier ff is much more important than IOR rate. If negative IOR had any effect, we would quickly see negative ff rate.

    “I am all for eliminating reserve requirements. But we don’t need IOR to do that.”
    If reserve requirements are likely to be abolished, then how small 25bps payment on reserves can be so contractionary?

    Another point – IOR at any time are much less important than the reaction function of central bank. Market expectations of 3 years of 0.5 IORs would be much more stimulative than market expectations of 1 year of 0.0 IOR.

    “I don’t deny that you and Bernanke agree, but you haven’t provided any evidence that he is correct. QE is not the proper policy when you are in a liquidity trap. Rather price level of NGDP targeting is required. Bernanke once knew that, but somehow forgot.”
    I believe that the only really credible way to have a level targeting in October 2009 was QE.
    BTW FOMC January minutes were released today:
    “The U.S. economy will expand by a range of 2.8 percent to 3.5 percent this year, compared with a median projection of 2.5 percent to 3.5 percent in November, when officials last gave forecasts.”

  69. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    17. February 2010 at 13:54

    Have you seen this by Hamilton:
    “Although banks may be content at the moment to hold a trillion dollars idle as excess reserves each day, one would suppose and hope that this will change as the economy recovers. But the resulting multiple expansion of credit and withdrawal of the reserves as currency would be impressively inflationary– the $1.1 trillion in credits for cash currently held by banks is a bigger number than the cumulative sum of green currency that the Federal Reserve has delivered to banks week after week since its inception a century ago…”
    http://www.econbrowser.com/archives/2010/02/bernanke_on_the_3.html

    So, according to Hamilton, low money multiplier is not caused by 25bps IOR, but by bad economy. IORs will have any effect only when economy recovers and Fed increases IORs.

  70. Gravatar of scott sumner scott sumner
    18. February 2010 at 11:05

    123, Nothing you say here makes any sense to me. The IOR is the opportunity cost of reserves, not the fed funds rate. The reason short term rates matter is not because they effect loan rates, but rather because they affect the demand for base money, and hence the price level. The IOR has a bigger effect on the demand for base money than the ff rate. If the IOR is 8%, banks will have a strong incentive to hoard reserves, even if the ff rate is 0%.

    The only credible way to do price level targeting is to set a price level target. The Fed refuses to do that. Once they set a price level target the market will determine the monetary base need to hit that target.

    123#2, I don’t know what Hamilton thinks about interest on reserves now, but when it was announced he argued that the program was very contractionary. I see no reason to believe he has changed his mind. I agree that when the economy recovers there will be a lot of inflation if the IOR is kept at 0.25%, so that statement by itself means nothing.

  71. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    20. February 2010 at 05:56

    When there are so many excess reserves, reserves and short term T bills, and fed funds market are complete substitutes. Opportunity cost of base money is not the IOR, but a minimum of IOR and fed funds rate. It is just a modern version of Gresham’s law.

    I support the idea of price level targeting. But I am sure that QE is needed in order to quickly correct deviations from price levels during banking panics.

  72. Gravatar of ssumner ssumner
    21. February 2010 at 07:52

    123, Op. cost is the highest valued alternative foregone, that means the highest of the ff rate and the IOR.

    It depends what you mean by QE. You might need to inject a lot of base money to hit a 5% NGDP target, but probably you wouldn’t as demand for base money only rises sharply when NGDP expectations fall far below target. But I’m not trying to be dogmatic here; if QE must supplement NGDP targeting, so be it.

  73. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    21. February 2010 at 14:16

    One of us has a problem with opportunity costs. Since you are a rare economist who understands supply and demand :) (see http://www.themoneyillusion.com/?p=2006), maybe I am wasting our time, but let me try.
    The story from which I have learned opportunity costs went like this: Mr. X has no money, three diamonds (one diamond 20000usd), and two grandmothers (10000000usd each). Mr. X wants to buy a car (20000usd), maybe two cars. Mr. X sells one diamond, hoards his grandmothers and buys a car, opportunity cost of the first car is just $2000, the same for the second car. Nothing changes if we replace Mr. X with a commercial bank A, diamond with cheap T-Bill, fed fund or repo financing, and grandmother with hoarded reserves with 25bps yield. Every bank has access to cheaper liquidity than 25bps reserves, so opportunity cost for banks is the lowest of the ff rate and the IOR. The meaning of IOR is that if Fed raises IOR say to 8%, banks are likely to run out of alternative sources of funds, those alternative sources of funds will get more expensive and opportunity cost will go up. But 25bps is just a tiny amount that changed almost nothing at all.

  74. Gravatar of ssumner ssumner
    22. February 2010 at 07:42

    123, Sorry, the Grandmother example went slightly over my head. I suppose you meant reserves are like the grandmother, in that even though they seem valuable, you can’t use them because you need them. Even in that case I’d say the IOR was a problem, because it made you want to hold them even more.

    I picked my 8% example for a reason, it’s no good to say if the IOR was 8% you wouldn’t be able to borrow in the fed funds market, I am assuming you can. I picked a big number to heighten the contrast, to make it easier to see why its the IOR that is key, when it differs from the fed funds rate. If the fed funds rate was 2% and the IOR was 8%, a banker would be crazy to borrow at 2% in the fed funds market and lend at 7% to a company, even though the 5% spread looks good. Why would it be crazy? Because they could hold the money as reserves at 8%, and not take any risk. That’s why 8% is the op. cost.

    There is a related question which is why didn’t banks arbitrage the difference. That’s something that I can’t answer. But my interest in monetary policy is not the effect it has on interest rates, which is a lousy indicator of policy, but rather the effect on the demand for base money. And for the demand for base money, the IOR is the relevant indicator.

  75. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    22. February 2010 at 16:52

    Reserves are valuable, but you don’t need to use them when cheaper alternatives are available. Hoarding of Susan B. Anthony dollars just after they were introduced did not create deflation, because enough regular dollars were minted so the opportunity cost of dollar stayed low. Payment of 25bps on reserves did not create deflation, because other actions of the Fed added huge amount of liquidity and ffr stayed low. Returning to your example of 2% fed funds rate and 8% IOR, a banker would be crazy to ignore the opportunity to borrow at 2% and lend at 7% to the company, because Gresham’s law says he will never get an opportunity to increase his hoard of reserves.

  76. Gravatar of ssumner ssumner
    23. February 2010 at 06:11

    123, Well I flat out disagree with you on the 7% loan. If he made the loan and I was on the Board of Directors I would have said “are you nuts? lending that out in a risky loan when we can earn a safe 8% letting the money sit in our vaults?”

    And the policy did create a little bit of deflation, despite your assertion. The CPI fell about 2% or 3% in the 6 months after the policy, didn’t it? It certainly created falling NGDP, which is a better measure of deflationary monetary policy. I understand that IOR must be looked at in context, what else are they doing. But my point is that they weren’t doing enough other things to prevent deflation.

  77. Gravatar of 123 – TheMoneyDemand 123 - TheMoneyDemand
    23. February 2010 at 13:44

    If rationality assumption is not relaxed, Board of Directors will reply “But we can’t increase the amount of safe money in our vaults, can we”? In this case it helps to talk about dollars and pencils, and not about interest rates and monetary base. Imagine an office supply manager, who had one dollar, and bought pencils with a cost of one dollar and utility of two dollars for the company. And then a Board Member says: “What a waste of resources, we should have bought one 199x Susan B. Anthony dollar instead of pencils, such dollars have a resale value of $3!”

    I have to agree with you that Fed wasn’t doing enough things to prevent deflation, but IORs were a step in the right direction. IORs were implemented as a part of the enormous change in monetary policy framework. Previous monetary policy framework prevented Fed from expanding balance sheet quickly in order to fight deflationary forces (as there were fears about reversibility of such balance sheet expansion). Current framework gives Fed flexibility in using QE to fight deflation while preserving interest rate instrument as the principal instrument for the future when policy needs to be tightened. I’d happily take 50bps IORs and Fed balance sheet triple the current size (provided that asset side of the balance sheet contains long duration or high credit risk assets).

  78. Gravatar of Potpourri Potpourri
    15. March 2010 at 19:59

    [...] Scott Sumner discusses the Fed’s hints that they will drop the “federal funds rate” as their target [...]

Leave a Reply