Mark Thoma on IOR

Mark Thoma recently made the following comments on the Fed’s interest on reserve program:

There’s been a lot of talk lately about the Fed’s policy of paying interest on reserves with many claiming that this has caused banks to retain reserves that might have otherwise been turned into loans, and thus the policy has depressed aggregate activity. However, paying interest on reserves is a safety net for the Fed that allowed them to do QEI and QEII. If the Fed wasn’t paying interest on reserves, QEI would have likely been smaller, and QEII may not have happened at all.

First, on whether paying interest on reserves is a constraint on loan activity, the supply of loans is not the constraining factor, it’s the demand. Increasing the supply of loans won’t have much of an impact if firms aren’t interested in making new investments. Businesses are already sitting on mountains of cash they could use for this purpose, but they aren’t using the accumulated funds to make new investments and it’s not clear how making more cash available will change that.

Second, I doubt very much that a quarter of a percentage interest — the amount the Fed pays on reserves — is much of a disincentive to lending (market rates have fluctuated by more than a quarter of a percent without a having much of an impact on investment and consumption).

Third, this a safety net for the Fed with respect to inflation. Paying interest on reserves gives the Fed control over reserves they wouldn’t have otherwise, and control of reserves is essential in keeping inflation under control. If, as the economy begins to recover, the Fed loses control of reserves and they begin to leave the banks and turn into investment and consumption at too fast a rate, then inflation could become a problem. 

But by changing the interest rate on reserves, the Fed can control the rate at which reserves exit banks.

I agree with much of what Thoma has to say, but would add a few comments.  Let’s start with the fact that without IOR a massive increase in the monetary base would eventually lead to high inflation.  I think that’s right, but it’s not really a reason for having a positive IOR right now, rather it’s a reason for having an IOR program in place.  (Indeed later in the post Thoma indicates that he holds the same view.) 

In January and March 2009 I published a couple papers that suggested the Fed might want to implement a negative IOR.  The plan was adopted just a few months later.  Unfortunately, it was adopted by Sweden, not the US.  (And the Swedish plan had some loopholes.)

I agree that the existence of IOR made QE1 and QE2 more likely, but I wonder whether Thoma realizes the implication of the argument.  I’ve tirelessly argued that fiscal stimulus was unlikely to be very effective because the Fed probably would have mostly neutralized the effects by doing less QE (and other forms of monetary stimulus such as negative IOR, level targeting, etc.)  So if the Congress had done the $1.3 trillion stimulus that many liberals recommended it seems unlikely that the Fed would have done QE1.  And if they had done no fiscal stimulus, QE1 would almost certainly have been much bigger.  The Fed may be a bit slow on the uptake, but they do eventually notice inflation falling below their target, and take corrective actions.  

I’ve never rigidly argued that fiscal policy can’t work, or didn’t have any stimulative effect in 2009, just that the monetary reaction problem made multiplier estimates highly unreliable.  Now Thoma says we also can’t rely on estimates of the impact of having no IOR, because without IOR there might have been no monetary stimulus.   And he may well be right–it’s essentially the same argument I use against fiscal stimulus.  The difference is that fiscal stimulus is quite costly, so the stakes are much higher.

I’d also point out that my monetary reaction argument is actually stronger for fiscal stimulus, than his argument is for QE1.  That’s because fiscal stimulus was done to boost the economy, precisely the same motivation as monetary stimulus.  On the other hand the original increase in the monetary base (which occurred in the fall of 2008) was aimed not at economic stimulus, but rather at rescuing the banking system by injecting liquidity.  Recall that the Fed decided against cutting the fed funds target on September 16, 2008, because they viewed the risks of recession and inflation as equally balanced.  But they might well have felt a need to rescue the banking system with some liquidity injections even if Congress had not given them authorization for IOR.

Sometimes I like to daydream and imagine a scenario where Congress refuses to authorize IOR, TARP, and all the other initiatives used to bail out the banking industry.  Bernanke would have had to go to the Fed and deliver the awful news:

Ladies and gentlemen, Congress won’t let us bail out the banks by any means other than good old-fashioned monetary stimulus.  We tried hard to get authorization for a program that would rescue banks without also boosting AD, but those morons on Capital Hill won’t go along.  I am afraid we have no choice but to do massive monetary stimulus, which unfortunately will boost NGDP growth.

Yes, they might have found that the only way to bail out the banks was to bail out the economy.  Wouldn’t that have been awful!

PS.  Whenever an economist thinks they’ve discovered a new idea, you can be sure someone else got there first.  I thought I was the first to discuss negative IOR, but Rodney Everson sent me the following quotation for an unpublished monograph written in 2001. 

In essence, Japan will begin to immediately recover if its monetary authority charges each bank an interest penalty each week based on the amount of excess reserves reflected on its books at that time, assuming, of course, that they are farsighted enough to then provide the excess reserves.  This will make the potato “hot” once again, and excess reserves will no longer be held in hand. 
 
Alternatively, they could raise the rate of overnight money to 1 or 2 percent which, of course, is impossible under current theory because it would be considered a “tightening.”  If you, the reader, now understand why such a “tightening” is necessary before an “easing” can be effected, then you have grasped the essence of this monograph.  You also should then understand the immediate danger we face under the current Federal Reserve policy of steadily driving the federal funds rate lower.

The quotation is from the monograph mentioned in this blog post.  BTW,  I think the second paragraph is wrong, but would be interested in what others think.


Tags:

 
 
 

44 Responses to “Mark Thoma on IOR”

  1. Gravatar of marcus nunes marcus nunes
    22. November 2010 at 20:11

    This is awful (from his 2008 post):
    “This country has been on a Japanese trajectory for over 20 years now. Japan had their stock market mania, then their real estate bust and then their lost decade, a decade when the Japanese monetary authorities could not get their money supply to expand”.
    Interestingly, Greenspan managed to get the “Great Moderation” exactly because he managed (as wished by Friedman) to “freeze the base” (in this case total reserves). And as you have frequently written, Japan simply did not want inflation above zero!

  2. Gravatar of marcus nunes marcus nunes
    22. November 2010 at 20:13

    His second paragraph says exactly what Kocherlakota said some weeks ago: “”Zero” interest rates are deflationary!

  3. Gravatar of Full Employment Hawk Full Employment Hawk
    22. November 2010 at 20:20

    “QEII may not have happened at all”
    The operative word is MAY. May does not mean “would not have.”

    If the Fed is reluctant to do the additional quantitative easing that the depressed economy badly needs because it is reluctant to accept additional increases in its asset portfolio, cutting and then eliminating interest payments on excess reserves is an effective way of adding to the stimulus without such an increase in the asset portfolio.

    The Fed having the ability to pay interest on excess reserves is a highly desirable addition to the Fed’s tools for fighting inflation. But with the economy experiencing disinflation, this is not a tool that should be used at the present time any more than raising the federal funds target, which is the most important tool for fighting inflation, should be used at this time.

    “Second, I doubt very much that a quarter of a percentage interest — the amount the Fed pays on reserves — is much of a disincentive to lending”

    While an increase in lending would be very desirable, as long as ending the payment on this interest increases the money supply and lowers market interest rates, it will have a stimulative effect on the economy even if bank lending is not affected. And the 25 basis points is more than the yield on federal funds, short-term treasuries, and short-term commercial paper.

    A major reason that the economy remains depressed is too much energy is being expended on finding arguments of why things that can work won’t. That definitely applies to not paying interest on excess reserves while the economy is depressed.

    The same thing applies to expansionary fiscal policy. I find it implausible that if Obama had quickly filled the vacancies on the BOG with full employment hawks, monetary policy would have been significantly weaker if the stimulus had been larger. Obviously leaving the seats on the
    BOG vacant for an extended period of time was a major Obama adminstration blunder, just as failing to make an all out push for a larger stimulus was. If the stimulus that got past the Republican filibuster could not have been increased, a second stimulus using reconciliation, which only required 51 votes, should have been passed. Only certain types of stimulus could have been passed using reconciliation, but it could have gotten it up to
    Romer’s suggested 1.2 Trillion.

  4. Gravatar of Mark A. Sadowski Mark A. Sadowski
    22. November 2010 at 20:47

    IOER is the elephant in the US QE room. How will any effort to raise AD through MP work now that IOER is set to 0.25%.

    But, if we reduce IOER we might be able to generate more MS with less MB. Given we are now nearly 10% below previous NGDP trend (and increasing) how could that hurt?

  5. Gravatar of scott sumner scott sumner
    22. November 2010 at 21:02

    Marcus, I think he confuses cause and effect with the low interest rate comment. I see low rates as an effect of (previous) tight money.

    But I was impressed that he mentioned negative IOR in 2001.

    Full Employment Hawk, I agree with your IOR comments. The monetary policy counterfactual is tough, all I can say is that if the Fed is doing its job, the fiscal multiplier is zero, Of course there’s a good retort: “If the Fed is doing its job” is a pretty generous assumption.

    Mark, I agree, although I think the QE already did a little bit of good, but there is a danger it may be offset by renewed problems in Europe. I hope they fix their debt problems quickly.

  6. Gravatar of Shane Shane
    22. November 2010 at 21:08

    “The difference is that fiscal stimulus is quite costly, so the stakes are much higher.”

    Frequent commentator Morgan and I can agree on one thing: raising the inflation rate has a real cost. It just costs what I consider the “right” people at the right time. Economics is called the dismal science for a reason–someone always pays.

    Bernanke realizes that one of the effects of QE is to lower the cost of government borrowing AND also to partially monetize the debt. That’s why he’s practically begging for more Fiscal stimulus–it is the other half of the supposedly “free lunch”–and who only eats half of their free lunch?

  7. Gravatar of Mark A. Sadowski Mark A. Sadowski
    22. November 2010 at 21:48

    Scott wrote:
    “Mark, I agree, although I think the QE already did a little bit of good, but there is a danger it may be offset by renewed problems in Europe. I hope they fix their debt problems quickly.”

    I don’t think we should rely to heavily on Europe (or any other geographical entity) to solve our problems.

    I’m meeting with my colleagues on Wednesday for a pre-holiday dinner at my house. We (PhD candidates or graduates from the UD) are from Cote D’Ivoire, Jordan, Japan, the U.S., Malaysia, Romania, Kenya, Latvia, Morocco, etc. etc.

    I have already decided on my theme for my somewhat informal pre-dinner address (wine soaked rant): “How the US will solve the world’s problems by solving its own”.

  8. Gravatar of Doc Merlin Doc Merlin
    22. November 2010 at 21:58

    @Mark:

    ‘”I have already decided on my theme for my somewhat informal pre-dinner address (wine soaked rant): “How the US will solve the world’s problems by solving its own”.’

    When we have our internal crap in order, the world does seem to run better.

  9. Gravatar of Mark A. Sadowski Mark A. Sadowski
    22. November 2010 at 22:15

    Doc Merlin wrote:
    “When we have our internal crap in order, the world does seem to run better.”

    Who knows when? One of these days….

  10. Gravatar of Bogdan Bogdan
    22. November 2010 at 23:48

    When the Fed introduced the Continental European style interest rate on reserves (I think this was done in late September or beginning of October 2009), the federal funds rate was above 2%. It was only in December that the federal funds rate was cut again repeatedly and lowered to the 0-25 bps where it is still today. Consequently, I think it cannot be made responsable for any tightining during the August-September-October period, on the contrary the positive rate on reserves can even be thought of as a de facto cut of the official federal funds rate for the banks, since now the banks practically lent to each other at (efffective)Fed funds-IOR, or ~1% before late October 2008. Am I mistaken?

  11. Gravatar of Bogdan Bogdan
    22. November 2010 at 23:50

    Should read “…I think this was done in late September or beginning of October 2008″! not 2009

  12. Gravatar of Full Employment Hawk Full Employment Hawk
    23. November 2010 at 00:23

    “all I can say is that if the Fed is doing its job, the fiscal multiplier is zero”

    Opposition to the use of fiscal policy is based on two different propositions:

    1. Expansionary Fiscal policy does not increase AD (or NGDP) because it crowds out an equal amount of private spending. Cochran’s position is an example of this.

    2. Expansionary Fiscal policy does increase aggregate demand
    but this will be quickly offset by the monetary policy engaging in monetary policy that moved AD in the opposite direction. Your position is an example of this.

    These are very different proposition which require very different theoretical arguments and empirical evidence.

    I obviously do not accept proposition 1. And the actions of the Fed do not, in my opinion provide me with any reason to accept proposition 2. With the economy stuck in a very deep hole we should use all methods of getting the economy out of it that have promise of working.

  13. Gravatar of Full Employment Hawk Full Employment Hawk
    23. November 2010 at 00:34

    “but there is a danger it may be offset by renewed problems in Europe. I hope they fix their debt problems quickly.”

    The European Central Bank is not helping the problem with its excessively tight monetary policy.

    The time is coming where the ordinary working people of Greece, Ireland, Spain, and Portugal will no longer be willing to tolerate the high unemployment rates the current monetary regieme is imposing on them. (In Ireland the upcoming by elections may well result in the government no longer willing to support the current austerity.) When that happens, the Euro will be in deep, deep trouble. This may even result in the collapse of the Euro system.

  14. Gravatar of Jon Jon
    23. November 2010 at 00:57

    Its also a question of creating a spread between the excess reserve-rate and the required reserve-rate. You want to create an incentive structure that drives bank-lending. Potentially this actually means paying HIGHER rates on the required reserve portion.

  15. Gravatar of Bonnie Bonnie
    23. November 2010 at 00:58

    We had somewhere in the area of $1.3T fiscal stimulus. The major portion of it was in the Recovery Act which was around $800B, and there were additions redubbed jobs bills and the like, and $490B omnibus spending in 2009. I don’t really know why after all of it is totaled it is said that it wasn’t what the Dr. ordered.

    I find it interesting that those who were pushing for fiscal instead of monetary stimulus use the symptom of the real problem, people hording cash, as reason that monetary stimulus won’t work when there is no reason it would not also be a limitation of fiscal stimulus. Sure, the initial distribution may wide, but who is to say that the mountains of cash won’t be sat on afterward? Of course this is just fantastic musing because it’s a mistake to assume that stimulus was actually spent in a way that stimulates instead of going to close budget shortfalls of states, which I suspect the majority of it did (with the exception of getting body scanners at the airports so we can appear in live nude video every time we want to take a flight). And that is exactly the problem with fiscal stimulus. It isn’t necessarily only the size that matters; it’s whether or not it is responsibly and effectively managed to be simulative. The only thing that turned out to be shovel-ready this time around was average private-sector Joe, which is a real shame.

  16. Gravatar of Bill Woolsey Bill Woolsey
    23. November 2010 at 05:23

    Scott:

    What does it mean “inject liquidity into banks” and how is it supposed to help? In understand creating an excess supply of reserves and how that might get banks to lend. In particular, it will get some banks to lend to other banks that might need to borrow, which might held those other banks. How exactly does increasing the supply and demand for reserves so that there is no excess supply of reserves do any good? And why would this called “injecting liquidity into the banks?”

    No, real point of this program was for Fed to allocate credit. Since the banks would have lent the reserves that the Fed is paying interest on (perhaps by purchasing government bonds,) and the Fed is now lending the funds troubled money-center banks, there has been a shift of credit. If this is supposed to be called “liquidity” there is a shift in liquidity.

    But that is my first question. What does this mean?

    Injecting liquidity?

  17. Gravatar of Bill Woolsey Bill Woolsey
    23. November 2010 at 06:14

    Jon:

    The point isn’t to increase bank lending. The point is to end an excess demand for money–or really what would be an excess demand for money if expenditures on final goods and services were on target.

    Increasing the quantity of money and reducing the demand to hold money both work to solve the problem. If the banks hold less reserves and instead spend them, perhaps by making loans, but they can “lend” by purchasing existing bonds too, then the increase in the amount of bonds the Fed must buy is smaller–the conventional money multiplier. But having banks pay lower interest on deposits reduces the demand to hold money. Paying banks to hold excess reserves is counterproductive. It motivates banks to pay higher interest on deposits that require reserves, which happen to be those that serve as money.

  18. Gravatar of Mark A. Sadowski Mark A. Sadowski
    23. November 2010 at 07:10

    Bonnie,
    You wrote:
    “We had somewhere in the area of $1.3T fiscal stimulus. The major portion of it was in the Recovery Act which was around $800B, and there were additions redubbed jobs bills and the like, and $490B omnibus spending in 2009.”

    I’m not going to get into a tit for tat over what I don’t consider to be a very important issue. However sometimes the facts need to be recognized.

    To be technical, the Omnibus Appropriations Act by itself contained almost no new spending. It was the remaining parts of the FY 2009 budget that had failed to pass under Bush.

    After the 110th Congress and President Bush were unable to reach an agreement on nine appropriations bills in 2008, the Congress extended funding for the caninet agencies covered under the bills until March 6, 2009 under a continuing resolution. The Omnibus Appropriations Act of 2009 covered these nine bills and provided funding for the agencies until the end of fiscal year 2009. The act actually totaled $410 billion, which, when added to the appropriations bills that were approved the previous session, was estimated at the time to amount to about a 6.7% increase over spending in 2008.

    According to the CBO’s January 2009 estimate for FY 2009, outlays were projected to be $3,543 billion and revenues were projected to be $2,357 billion, leaving a deficit of $1,186 billion. Keep in mind that these estimates were made before Obama took office, based on existing law and policy, and did not take into account ARRA or any other actions that Obama might implement.

    Now let’s fast forward to the end of FY 2009, which ended on September 30, 2009. According to CBO, it ended with spending at $3,515 billion and revenues of $2,106 billion for a deficit of $1,409 billion.

    So the interesting thing is that when one compares the CBO report on FY 2009 from January 2009 (Bush’s last) and the final report in October 2009, spending for FY 2009 actually declined by $28 billion despite the implementation of ARRA.

  19. Gravatar of Jeff Jeff
    23. November 2010 at 08:07

    Interest on reserves of 0.25 percent could be a much bigger deal than you think. Suppose that banks are extremely risk-averse and so the only forms of lending they will undertake is either (i) holding excess reserves that pay interest, or (ii) buying short term Treasury debt (Tbills).

    If the IOR rate is lowered to zero, the funds now held as excess reserves will instead be used to buy Tbills. So long as Tbills on the secondary market pay any interest at all, they are a better investment than excess reserves that pay nothing. The sellers of those Tbills deposit the proceeds in banks, which again creates excess reserves, although only 90 percent as much as before due to the marginal 10 percent reserve requirement. The banks buy more Tbills and the cycle repeats until you end up with the banks holding ten times as much in Tbills as they did in excess reserves when the IOR rate was positive. And notice also that deposits also increased by the same amount.

    Now let’s put some numbers on this. M2 is currently about $8.8 trillion, and excess reserves are about $966 billion. So if they could, the banks would end up buying roughly $9.6 trillion in Tbills. However, at the end of June 2010, there was only about $7.3 trillion in Treasury debt held by the public, and only about $2.4 trillion of that matured within a year. So if they want to get rid of their excess reserves banks will either have to buy longer-dated Treasuries and/or buy non-government securities (or lend privately) as well.

    The point that no one is talking about is that if banks did manage to rid themselves of all their current excess reserves, M2 would more than double in the process. Even Krugman would be impressed by that.

    Now of course I realize that banks probably wouldn’t get rid of all their excess reserves. Their purchases would quickly drive short term Tbill rates to zero, and medium term rates close to zero. And they would buy some private debt as well. Almost certainly they would stop while they still had some excess reserves. Just where that point is nobody really knows.

    We do know that when inflation and nominal interest rates are both very low, as they are now, money demand skyrockets. Paying interest on reserves plays a part in keeping money supply from expanding to meet that demand. And since people reduce their spending when they want to increase their money holdings, you get the weak aggregate demand we see today.

    This “real balance” channel is how old-line monetarists thought about the effects of monetary policy. It’s not present in most of the DGSE models that have been used in macro for the past thirty years, and so the profession seems to have forgotten all about it. But it follows naturally from Say’s Law, and models without it are missing something important.

  20. Gravatar of Mark A. Sadowski Mark A. Sadowski
    23. November 2010 at 08:33

    Jeff,
    Just an addendum to your comment. The current IOER of 0.25% not only pays more than 1, 3 or 6 month T-Bills, it also pays as much or more than all AA financial and nonfinancial commercial paper.

    http://www.federalreserve.gov/releases/cp/

    So, yes, it is a big deal.

  21. Gravatar of Rod Everson Rod Everson
    23. November 2010 at 08:33

    I wrote the monograph that Scott mentioned in the P.S. and we had some email exchanges over the weekend during which we got a little side-tracked in a discussion over whether the fed funds rate is a market-determined rate or a fed-determined rate. (I am of the opinion the Fed can, within reasonable limits, put the fed funds rate wherever they want it, but that is a secondary consideration.)

    The main point of the monograph was to convince others that the Fed has always possessed what I refer to as a Strong Form of monetary control, i.e., that they have always easily been able to bring about rapid changes in the money supply simply by managing the level of excess reserves. However, if the mechanism operates in the manner that I describe, then they will experience difficulty if they ever make the mistake of either 1) letting the funds rate get to zero, or 2) start paying interest on excess reserves. This difficulty, a hypothetical in 2001, is now reality. Essentially, the Fed has lost control of the money supply and is not likely to regain it until they employ the tools described in the monograph. QE1,2 or 3, or 4, won’t do it but are definitely complicating the Fed’s primary assignment of managing the money supply.

    The monograph is where I try to explain all this. It’s a PDF that is easily found if you go to the blog Scott links to in his P.S. (Don’t bother commenting there as it’s an inactive blog.)

    I followed the Fed very closely for about 25 years, from 1975 to about 2000, and developed my ideas during the time when the weekly swings in the M1 money supply were closely watched and analyzed by Wall Street economists. In the early years of that time frame I was the bond portfolio manager of a state retirement system and had access to most Wall Street research. Had we never introduced NOW accounts in the early 80′s, I’m convinced that the contents of the monograph would now be considered common knowledge, as certain economists were homing in on the data that indicated that the Fed had very strong control over short term swings in the money supply.

    NOW accounts basically trashed the data because of the changes they drove in the level of reserves. The weekly money numbers quickly lost their market influence as a result. Nonetheless, the Fed still retains the powers indicated in my monograph, even if they are unaware of them currently.

    As I said, my blog is inactive. I’ll do any responding here on Scott’s blog. And thanks Scott, for the acknowledgment.

  22. Gravatar of marcus nunes marcus nunes
    23. November 2010 at 09:01

    It´s sad to see Krugman write:
    “When short-term interest rates are up against the zero lower bound, whatever power the Fed has to influence the economy comes largely from its ability to affect expectations. This is true even for Bernanke-style quantitative easing: you can’t really push down longer-term yields unless the market believes that you’re going to keep buying until the rates are where you want them. It’s even more true when it comes to credibly raising expected rates of inflation.

    So if a large political faction begins yelling and screaming as the Fed attempts QE, this will have the effect of undermining the policy’s effectiveness”. And link to this piece:
    http://online.wsj.com/article/SB10001424052748703531504575625121365997204.html

  23. Gravatar of Rod Everson Rod Everson
    23. November 2010 at 09:14

    Jeff,

    You wrote: “Now of course I realize that banks probably wouldn’t get rid of all their excess reserves. Their purchases would quickly drive short term T-bill rates to zero, and medium term rates close to zero. And they would buy some private debt as well. Almost certainly they would stop while they still had some excess reserves. Just where that point is nobody really knows.”

    If the Fed really wants to get the money supply growing again here’s what I think they need to do.

    1. Drop the interest rate on Excess Reserves (IOER) to zero. The interest on Required Reserves (IOR) is a sop to the banking system but is irrelevant otherwise. However, the IOER only frustrates their traditional reserve management.

    2. Drain the 900+ bn of excess reserves until they’ve reached traditional levels of 100mm to 1bn each reserve period. (Do this by selling securities from their portfolio.)

    3. Once the excess reserve levels are back to what was considered normal just a couple of years ago, set a target rate of about 3% internally and add/drain reserves as needed to convince the banks’ fed funds desk managers that the new rate is indeed 3% (2% or 4% would work equally well, incidentally; the point is to get it well above zero.)

    4. Each week for a couple of months, run excess reserve balances of about $1bn and the money supply will quickly begin rising (although the current confusion over what is actually the money supply will cause some difficulty in assessing what is actually happening.)

    Now, you can see why this would run into some objections, since #2 and #3 would almost universally be interpreted as a tightening. My monograph explains why they not only aren’t tightening moves, but are instead necessary moves if the Fed is to regain control of the money supply. The first part of the monograph describes how the process in point #4 worked in the past.

    In short, the Fed already had all the tools it needed to manage the money supply, but instead they’ve gone out and acquired a some new, ineffective, ones while letting the ones that have always worked get dulled and covered with rust from non-use. The result is that they now have a mess on their hands and are proving incapable of performing their basic responsibilities. To return to using the tools that worked in the past, they need to do points 2 and 3 that I described. It won’t be easy, but if they fail to do so we are going to go through ever more interesting times.

  24. Gravatar of Benjamin Cole Benjamin Cole
    23. November 2010 at 09:31

    Is it my imagination, or has the Sarah Palinized right-wing moved from bashing QE2 to bashing airport security?

    QE2 should be out of the news cycle in just a few more days.

    Thank the merciful heavens for the new airport screening devices. They may have saved QE2.

  25. Gravatar of Rod Everson Rod Everson
    23. November 2010 at 09:44

    Jon wrote:

    “Its also a question of creating a spread between the excess reserve-rate and the required reserve-rate. You want to create an incentive structure that drives bank-lending. Potentially this actually means paying HIGHER rates on the required reserve portion.”

    You’re close here, but what I maintain is that it’s a question of creating a spread between the excess reserve rate and the fed funds rate. Ideally, the excess reserve rate (IOER) would be returned to zero and then the Fed would set a target rate for fed funds somewhat above zero (after first removing the nearly one trillion dollars of current excess reserves.)

    Once this is done, all the Fed needs to do is reliably supply a relatively modest level of excess reserves each reserve period for a few months and money will immediately begin to grow. Less than 1bn of excess reserves would be more than sufficient.

    How will money grow, you might ask? Well, when the excess is sufficient and the targeted funds rate is positive then at the end of a reserve period, fed funds circulate like mad, with each funds desk manager attempting to avoid the opportunity cost of holding a non-earning asset overnight. The rapid circulation of fed fund balances results in a similar activity in other overnight markets as rate rapidly fall at the end of the reserve period. Suddenly, a funds desk manager finds he has another 100mm from a recent deposit (via a customer participating in the commercial paper action, for example) and rushes to sell it. The process continues, with the funds rate falling well below the target rate on the last day of the reserve period.

    In the end, all that circulating ends up influencing the level of the money supply. Then, during the next reserve period, the Fed simply supplies the additional required reserves to support the higher money balances and repeats the process. In a matter of weeks, the money supply begins to grow steadily (not a process that should be sustained, obviously.)

  26. Gravatar of Rafael Rafael
    23. November 2010 at 09:53

    Scott,

    I remember I linked a paper by Woodford and Curdia discussing IOR and variations in the size and composition of the central bank’s balance sheet formally in a NK framework. Perharps I missed it, but did you comment that paper on that occasion?

    Their views on price level targeting are not so different form yours (http://www.themoneyillusion.com/?p=2872), but the parts about IOR and Fed´s balance sheet seem to be (at least to me).

    Conventional and Unconventional Monetary Policy

    Authors: Vasco Cúrdia and Michael Woodford

    link: http://www.newyorkfed.org/research/staff_reports/sr404.html

  27. Gravatar of Full Employment Hawk Full Employment Hawk
    23. November 2010 at 10:55

    “I find it interesting that those who were pushing for fiscal instead of monetary stimulus”

    I am definitely not in this camp. The issue is not fiscal stimulus or monetary stimulus. It is fiscal stimulus AND monetary stimulus. When the economy is in as bad a hole as we are in we need to use ALL available methods to increase the rate of growth in NGDP.

  28. Gravatar of Full Employment Hawk Full Employment Hawk
    23. November 2010 at 11:02

    “When short-term interest rates are up against the zero lower bound, whatever power the Fed has to influence the economy comes largely from its ability to affect expectations. This is true even for Bernanke-style quantitative easing: you can’t really push down longer-term yields unless the market believes that you’re going to keep buying until the rates are where you want them.”

    This is based on the assumption that the pure expectations theory of the term structure of interest rates (or the expectations theory with only a liquidity premium) hold. But there are good reasons for believing that short-term and long-term securities are imperfect substitutes, so that risk averse investors have preferred habitats at all spectrums of liquidity. In that aase a decrease in the supply of long-term bonds will directly raise their prices and lower long-term interest rates.

  29. Gravatar of Full Employment Hawk Full Employment Hawk
    23. November 2010 at 11:08

    In the title of the Krugman article referred to above it states
    “Doubts about the central bank’s ability to expand its bond-buying program have driven Treasury yields higher.”

    It would seem more likely that the inflation expectations effect is doing this due to the fact that the quantitative easing has increased inflation expectations. Scott predicted that long-term interest rates would rise.

  30. Gravatar of Liberal Roman Liberal Roman
    23. November 2010 at 11:11

    All our debates are irrelevant right now. Trichet and the ECB are about to destroy the world.

    And there is no way Bernanke does enough QE to shield us from the European mess.

    It’s amazing how quickly things cam go from being positive (late October) to negative (right now) and possibly to the catastrophic (in a few more weeks?)

  31. Gravatar of JimP JimP
    23. November 2010 at 14:20

    Skidelsky says FDR buying gold was a total failure – and so will QE2 be now.

    http://www.project-syndicate.org/commentary/skidelsky35/English

  32. Gravatar of scott sumner scott sumner
    23. November 2010 at 14:26

    Shane, I assume we both want low and stable inflation–the question is how to get there. Right now it is unstable–it is falling.

    Mark, Yes, I often argue we should solve our own problems, not wait for good luck to turn up.

    Bogdan, Yes, the tight money occurred before the IOR.

    Full Employment Hawk, If the Fed is targeting inflation then it obviously offsets fiscal stimulus. That’s why fiscal stimulus fell out of grad textbooks in the 1990s–even Keynesians like Krugman admit that there is no case for fiscal stimulus unless rates are stuck at zero. So it’s not like I’m putting forward some sort of novel argument–it was standard macro circa 2006. Where I differ from people like Krugman is that I assume the Fed continues to target inflation even when rates fall to zero–using tools like QE. I admit that is a more debatable point, but it’s certainly a reasonable argument–indeed Thoma is a Keynesian who favors fiscal stimulus, and yet he employs the same sort of logic I do.

    Jon, Last year I talked about that exact option–much higher rates on RRs and negative rates on ERs.

    Bonnie, I don’t really know the exact numbers.

    Bill, I really don’t know why the Fed felt the banks needed liquidity, but apparently they did. My understanding was that the interbank loan market started to freeze up after Lehman, but it’s not my area of expertise, so I may be wrong. By “liquidity, I simply meant more base money.

    Mark, Good point about the deficit, but here’s what puzzles me. The Obama stimulus did include lots of spending (I seem to recall it was roughly half the total, but the exact amount doesn’t matter.) My question is this; If the actual spending fell short of the Bush proposal, despite passing an $800 billion stimulus plan, how much extra spending would we have gotten for 1.3 trillion? Is it 13/8 times negative $28 billion? That would be about negative $45 billion. And would that have made any difference?

    Jeff, You said;

    “Interest on reserves of 0.25 percent could be a much bigger deal than you think.”

    Then it would have to be a pretty big deal, as I have been one of the biggest complainers about IOR in the entire blogosphere over the past two years.

    Jeff, I agree that no IOR would drive T-bill yields to zero, maybe slightly negative (cost of storing cash).

    If you are serious about stimulus, you make IOR strongly negative, and the reserves go out into circulation.

    Rod, I agree that we need to get the fed funds rate above zero, but the right way to do it is through an expansionary Fed action, and raising the fed funds target is contractionary.

    Marcus, Yes, those pressuring the Fed to tighten are essentially prolonging the recession.

    Benjamin, Going after the TSA would be a much more productive use of Sarah Palin’s time.

    Rafael, I don’t recall the paper. Are there any specific points they made that you want me to comment on? I agree with Woodford on the need for level targeting.

    FEH, Yes, I understand you supported monetary stimulus. Actually, very few liberals opposed it. My complaint was that certain prominent liberal bloggers ignored the issue in 2008 and 2009.

    Liberal Roman, And you haven’t even factored in Korean War II.

  33. Gravatar of Full Employment Hawk Full Employment Hawk
    23. November 2010 at 14:28

    “Trichet and the ECB are about to destroy the world”

    That is a clear and present danger. The meltdown of the Euro could well lead to a world-wide depression.

    People spoke too soon when they thought we had overcome the risk of another Great Depression. We are not out of the woods yet.

  34. Gravatar of scott sumner scott sumner
    23. November 2010 at 14:30

    JimP, Skidelsky is wrong.

  35. Gravatar of Full Employment Hawk Full Employment Hawk
    23. November 2010 at 14:48

    “Full Employment Hawk, If the Fed is targeting inflation then it obviously offsets fiscal stimulus.”

    Only if the Fed is SUCCESSFULLY targeting inflation, which it clearly is not.

    Many economic conservatives argue that expansionary fiscal policy will not increse AD at all because it only crowds out an equal amount of private expenditures. What the Fed does is not an issue for this argument. If I understand you correctly, you do not hold this position. Once one rejects this proposition and accepts that expansionary fiscal policy does increase AD and bases one’s objection to it on the grounds that the Fed will successfully offset this with monetary policy that decreases AD by the same amount, the debate over the effectiveness of fiscal policy becomes a debate of whether one can expect the Fed to act in that way. I agree that if it does act that way you are correct. But I see no reason from the Fed’s actions that it will successfully act in that way, even if it tries, which is also questionable.

  36. Gravatar of Mark A. Sadowski Mark A. Sadowski
    23. November 2010 at 14:59

    Scott,
    You wrote:
    “My question is this; If the actual spending fell short of the Bush proposal, despite passing an $800 billion stimulus plan, how much extra spending would we have gotten for 1.3 trillion? Is it 13/8 times negative $28 billion? That would be about negative $45 billion. And would that have made any difference?”

    That’s not a question as much as a statement. One of the biggest problems with discretionary fiscal stimulus is that it only leads to more spending if there are also no spending cuts. Evidently, given the CBO figures, things didn’t work out exactly as planned.

    Would an even larger fiscal stimulus have “worked”? Not given the above. And, more importantly, it would not have worked because the Fed would have used it as an excuse for doing even less.

  37. Gravatar of Full Employment Hawk Full Employment Hawk
    23. November 2010 at 15:11

    Iknow that your are not an admirer of Keynes, but Keynes would have supported quantitative easing. As he stated on page 206 of the General Theory “Perhaps a complex offer by the central bank to buy gilt-edged bonds of all maturities, in place of the single bank rate for short-term bills is the most important practical improvement which can be made in the technique of monetary management.”

  38. Gravatar of Shane Shane
    23. November 2010 at 16:23

    Regarding the Fed counteracting Fiscal stimulus through doing less, it seems that there are actually two ways of answering the question, in terms of what is plausible, and what you yourself would actually do faced with an equivalent situation that impacted your material well-being directly.

    If you are a hobo and you panhandle up enough change to get a McRib sandwich, it is somewhat plausible that if you save the portion that you don’t eat initially, someone will notice that you still have some extra left over food and refuse to give you additional change.

    Hobo Paul decides to keep the Sandwich, reasoning that it limits his downside risk, while still giving him a chance of the upside of a brand new sandwich.

    Hobo Scott, on the other hand, reasons that if he throws out half of the McRib, he possibly increases his chances of a fresh new sandwich, but at the expense of increasing his downside, which he doesn’t really mind, because he hates leftovers with a passion.

    So yes, maybe we are stuck eating leftovers now, rather than enjoying our fresh new McRib, because we played it safe and kept that old one. But on the other hand, this assumes an idealistic view of human nature in which people are marginally less ready to reward you because of marginal increases in your well-being. In other words, it assumes a fully “autistic” human nature.

    I think most people would tend to act more like Hobo Paul. This not only because it seems safer, but also because we have an intuitive sense that marginal calculations tend to play only part of the role in the decision making process–the question is not just how much people give, but whether they give at all. Hobo Scott is right that people may give him less money if they see he has a half eaten McRib, but he is wrong in assuming that it makes them less likely to give at all.

    We know that people both tend to think in marginal terms sometimes, and that at other times, they think in dichotomous terms, such as when the question is something that is essentially black-or-white, for example, whether to act or whether to do nothing. Bernanke does not seem to me to be acting much like an economist right now–after all, he’s basically spent the last two years doing everything he warned against.

    The lesson, I think, is that people may give you marginally less change based on your current well-being, but this is a secondary effect, especially when people are thinking in black-or-white (i.e., about whether or not to do something at all). I think the decision to give you anything at all is driven by other factors. In this case, it is possible BOTH that Fiscal stimulus made the size of QE1 and 2 marginally smaller, while also having a much stronger effect on whether they occurred at all (which would tend, I think, to vindicate Hobo Paul if the effect was sufficiently strong).

    This is because, as the reaction of Palin et al. demonstrates, stimulus is stimulus as far as the moral calculus is concerned, and Bernanke is as much a politician as an economist now–he needs and wants political cover for stimulus. This is, I think, the correct interpretation of his plea for additional fiscal stimulus–he is saying, in effect, that he is deeply uncomfortable throwing money at the smelly hobo that is the banks while everyone else is standing by and telling him they are undeserving of his charity. If Obama and Congress start pitching in too, I think there is a strong case to be made that he would be more likely to give again rather than less likely, even if the amount might be slightly smaller than if he had just courageously decided to give again on his own without additional cover.

  39. Gravatar of Rod Everson Rod Everson
    23. November 2010 at 21:18

    “Rod, I agree that we need to get the fed funds rate above zero, but the right way to do it is through an expansionary Fed action, and raising the fed funds target is contractionary.”

    But that’s exactly what I said would happen Scott, i.e., that any attempt to raise the FF rate would be viewed as tightening. So, the only way for the FF rate to rise is in reaction to a demand for funds created by other means, some other expansionary move by the Fed.

    The problem at hand, of course, is that they have been trying to do exactly that with no apparent success. What will $2 trillion of excess reserves accomplish that $1 trillion couldn’t?
    They’ve lost control of the monetary mechanism exactly because they lowered the rate too close to zero and until they reverse that error they will not regain control. Ordinarily it would require only a trivial effort for them to raise the rate. Now they have nearly $1 trillion of excess reserves to worry about first, however, something that vastly complicates the situation since draining them will surely be perceived as a tightening.

    Withdrawing them will have the same effect on the overall economy as did adding them in the first place–zero. We need to go back to what was proven to work in the past, and it would be easy to do except for the difficulty in justifying the actions necessary.

  40. Gravatar of Full Employment Hawk Full Employment Hawk
    24. November 2010 at 13:34

    “Withdrawing them will have the same effect on the overall economy as did adding them in the first place–zero.”

    In 1937 the Fed doubled reserve requirement because it was concerned about the large amount of excess reserves that were in the system, and this decreased the money supply and was at least a major factor in causing another recession. One can debate about whether this or the contractionary fiscal policy engaged in that year was the main cause, but what the Fed did played a major role.

  41. Gravatar of Rod Everson Rod Everson
    24. November 2010 at 15:06

    But in 1937 banks were (reputedly) holding those excess reserves to build customer confidence in their solvency. When the Fed increased the reserve requirement, the banks, in an effort to restore the metric that they had been using to assure solvency, strove to again build excess reserves. This would obviously have a contractionary effect.

    So, the question today would be, are banks holding excess reserves to assure solvency, or for some other reason, say the ability to earn interest on them, interest that was not available in 1937.

    I’m still of the opinion that the Fed could reduce excess reserves in the system to historical norms with little effect on the economy, though the financial markets would probably take a little while to sort out what’s going on.

  42. Gravatar of scott sumner scott sumner
    24. November 2010 at 19:40

    Full Employment Hawk, The interest rate effect is not enough to fully crowd out private expenditure. I agree with you there. For complete crowding out to occur you need an inflation targeting central bank.

    The question of whether it must be successful is interesting. I certainly understand your point, but I tend to think in terms of policy expectations. Thus a given fiscal stimulus might be expected to produce X% increase in inflation. If the Fed is targeting expected inflation, it will contract until the expected impact of fiscal stimulus is zero. Whether the actual impact is zero is another question. But policy decisions must be made on the basis of their expected effects.

    Mark, Sorry if I was a bit sarcastic. My sense is that there weren’t a lot of shovel ready projects available. And of course as the feds paid for potholes to be filled, the 50 little Hoovers just used that money to cut back on their own spending.

    FEH, I do admire many things about Keynes. The Tract on Monetary Reform is an excellent book. I agree he probably would have supported QE. But I prefer Fisher’s macroeconomics.

    There are many similarities between Keynes and Krugman, including their intelligence, their writing skill and their sarcasm.

    Shane, Yes, the best argument for fiscal stimulus is that there is some wiggle room within the Fed. They will actively boost inflation if it falls below a certain level, but won’t try to restrain it unless it rises well above the trigger point for QE.

    But that doesn’t answer all my objections, as I think if the Congress hadn’t acted, the Fed would have done much more. Because even with $800 billion in stimulus we are near the bottom trigger point.

    Bernanke isn’t the only player in this game; if he were policy would have been more expansionary. But even the various QEs that he has done partly reflect the responses of the inflation hawks to fiscal stimulus. Do more and the hawks will do less.

    Rod and FEH, Rod is talking about both eliminating IOR and reducing ERs. The net effect of both is ambiguous in my mind–but I’d rather get rid of IOR first, then see how many reserves we needed to boost expected NGDP growth up to desired levels.

    I’ve continued this discussion on a more recent post.

  43. Gravatar of 123 123
    1. December 2010 at 15:25

    Scott, You said:
    “On the other hand the original increase in the monetary base (which occurred in the fall of 2008) was aimed not at economic stimulus, but rather at rescuing the banking system by injecting liquidity.”
    The original increase in the monetary base was aimed at lowering effective fed funds rate to the levels set by FOMC, so it was an economic stimulus, and equivalence argument you make about fiscal stimulus multiplier fully applies to IOR. Bernanke was scared about the banking system only when fed funds rate market broke down (he let Lehman go bankrupt), just like Trichet is content these days to buy Greek bonds only to the extent needed to keep euro money market rates at the target levels.

  44. Gravatar of ssumner ssumner
    2. December 2010 at 18:33

    123, It seems a question of semantics. You say it was stimulus, I say an attempt to prevent tightening. At the September meeting the Fed decided the economy did not need any more stimulus.

Leave a Reply