Calling 123

I’ve frequently argued that the interest on reserve policy was a huge mistake, comparable to the 1936-37 decision to double reserve requirements in the midst of the Great Depression.  I pointed out that both policies sharply increased the demand for base money, and the implementation of IOR was associated with a big stock market crash during the first 10 days of October, 2008.  I don’t know exactly when the stock market understood what was going on, but I read that the decision was made on October 3rd, announced on the 6th, and implemented on the 8th.  Whatever the exact dates, it’s clear that the story became understood at some point during the famous crash.

The title of the post refers to frequent commenter (and blogger) 123, who knows more about the IOR program than I do, and disagrees with my view that it was contractionary.  Indeed he argues it was beneficial.  I’m wondering how he will react to this Louis Woodhill piece from RealClearPolitics:

The available data indicates that it was the Fed’s IOR program, not the collapse of Lehman Brothers on September 15, 2008, that crashed the real economy and sent unemployment skyrocketing. Because the two events were only three weeks apart, many people believe that it was the Lehman bankruptcy that precipitated the worst economic downturn since the Great Depression. However, the market data from that period suggests strongly that the real cause was IOR.

A valid way to gauge whether events are “good” or “bad” for the economy is to look at the stock market’s reaction to them. The day that Lehman Brothers collapsed, the S&P 500 went down 4.71%. Three days later (i.e., at the fourth market close after the event), the S&P 500 was down by a total of 3.61% from its pre-Lehman close.

At the time of the Fed’s IOR announcement, the S&P 500 was down by a total of 12.18% from its pre-Lehman close, 15 trading days earlier. However, the day that the Fed announced IOR, the S&P 500 fell by 3.85%, and it was down by a total of 17.22% three days later.

On October 22, 2008, the Fed announced that it would increase the interest rate that it paid on reserves. The S&P 500 fell by 6.10% that day, and it was down by a total of 11.11% three days later. On November 5, 2008, the Fed announced another increase in the IOR interest rate. The S&P 500 fell by 5.27% that day, and it was down by a total of 8.60% three days later.

I’ve always thought the timing of the implementation of IOR was an interesting coincidence, but hardly definitive.  But for some reason I never got around to closely investigating the program in detail.  I had no idea that the two subsequent IOR rate increases were also associated with mini-crashes.  Again, the evidence is hardly definitive.  But crashes this size are quite rare.  To have all three contractionary IOR decisions (and the only contractionary IOR decisions in all of American history!) each be associated with a once-in-a-blue-moon market plunge, is certainly suspicious.  I’m surprised others haven’t made a big deal of this pattern—people have spun elaborate economic theories based on much flimsier empirical evidence.  And remember, economic theory predicts the decision should have been contractionary.

Over to you, 123.

PS.  There’s an easy way to test this theory.  If Bernanke finds QE2 doesn’t produce the effects he’s looking for, he should try lowering the IOR to 0.15%.  No need to worry about those annoying regional Fed presidents; the Board determines the IOR.  And you won’t have Sarah Palin looking over your shoulder—I can’t imagine the Tea Party would get up in arms over a Fed decision to slightly reduce the Fed’s subsidy to big banks.

Here’s a prediction–the Dow rises 10 points for each 1 basis point reduction in the IOR.

HT:  JimP

Update 11/12/10:  123 has responded on his (or her?) blog.  As I indicated, he knows more about this that I do.  Part of our dispute has been over the “other things equal” assumption.  I look at the IOR issue holding the size of the base constant, he tends to assume that the increase in the base was facilitated by having the IOR program.  He also discusses the problem of increased uncertainty over future fed funds rates, something I know little about.  I do have one question, however.  123 says:

As October 2008 FOMC minutes put it, “In the overnight federal funds market, financial institutions became more selective about the counterparties with whom they were willing to trade.”

Is the implication that IOR solved the counterparty risk problem?  If so, how?  And why is the uncertainty about fed funds rates a big problem, when the level of rates was low thoughout this period, and extremely low if you recall that these are overnight loans.  How much risk did ffr instability actually produce?

PS.  I love that George Bush quotation at the end of his post.


Tags:

 
 
 

52 Responses to “Calling 123”

  1. Gravatar of Mark A. Sadowski Mark A. Sadowski
    11. November 2010 at 20:11

    How could anyone rational look at IOER as anything other than contractionary?

    It pays more than 1, 3 or 6 month T-bills. It pays more than every form of sort term commercial paper. Why would banks chose to invest in anything short term when they can “invest” in cash and earn interest totally risk free?

    As long as IOER is in effect expect the economy to go sideways (or perhaps do even worse).

  2. Gravatar of Doc Merlin Doc Merlin
    11. November 2010 at 21:34

    @Mark A. Sadowski.

    Yah it fubars the short term paper market. Also, if you believe in liquidity traps you could argue that it raises the rate at which liquidity starts being trapped at, so instead of the trap being at 0% it is at the IOR rate.

  3. Gravatar of Mark A. Sadowski Mark A. Sadowski
    11. November 2010 at 21:40

    Doc Merlin,
    You wrote:
    “Yah it fubars the short term paper market.”

    You betchya.

    I don’t believe in the liquidity trap (anymore than I believe in unicorns). But if we are in a liquidity trap it must be because of IOER.

  4. Gravatar of Full Employment Hawk Full Employment Hawk
    11. November 2010 at 23:07

    At a time when banks are sitting on huge amounts of excess reserves it makes absolutely no sense to reward banks for sitting on them. Clearly there should be absolutely NO interest paid on excess reserves as long as the economy needs monetary stimulus. It is totally counterproductive. The time to pay interest on excess reserves is when the time for monetary restraint comes. Paying no interest on them until monetary restrait is needed has the additional advantage of keeping the Fed’s inflation fighting powder dry.

    It is especially shocking that the Fed would engage in such a contrationary monetary policy after the Lehman collapse had created a financial crisis. At the very least these counterproductive actions greatly aggrevated the problem. When the next generation’s counterpart to Friedman and Schwartz write their counterpart to the MONETARY HISTORY OF THE UNITED STATES, this will surely be called a major blunder.

  5. Gravatar of David Pearson David Pearson
    11. November 2010 at 23:10

    Pre-Lehman, the S&P stood at around 1250. By October 12, it had hit 900. From October 12 to November 10, the S&P traded in a range between 900 and 1000. Clearly, if the IOR increase was as important to the markets as you believe, then the S&P would not have picked that exact time period to consolidate (trade sideways) after a steep decline.

    My recollection of events was that the market viewed the increase in the IOR as merely an operational move by the Fed intended to bring the FFR closer to the target rate. The Fed FFR had been trading away from the target by a wide margin, and this had been unsettling markets. In fact the Fed’s press release on 10/22 states as much.

  6. Gravatar of Alan Alan
    12. November 2010 at 00:58

    Scott,

    I’ve been thinking a bit about the IOR program and while I disagree with 123 about it being beneficial, I’m not sure it was all that contractionary. While I think a positive IOR is essentially the same as the Fed engaging in (contractionary) OMOs, its only a shift between various types of public debt, and not between private and public debt, and hence doesn’t resolve the “flight to quality” at the heart of the current situation.

    Even if the Fed lowers the IOR to zero, will there be a switch into private debt? I’m skeptical of this (in fact it should have already happened given that IOR is only 25bps); the main issue is “flight to quality”, not “flight to base money”. I also can’t see how a negative IOR will be all that beneficial, as it wouldn’t resolve the flight to quality issue.

    Anyway I would like to hear your thoughts on this issue.

    Alan

  7. Gravatar of Jeff Jeff
    12. November 2010 at 04:03

    @David,

    Let me rephrase that for you. Money was not as tight as the Fed wanted it to be, and IOR was intended to remedy that.

    Umm, I think that means IOR tightened money.

    There is an argument for IOR that says: The banks were in trouble and needed capital. Paying IOR is like giving them a capital injection equal to the expected net present value of the stream of interest payments.

    How big is the injection? If the rate paid is the risk-free overnight rate, and yield curve usually slopes up, then the amount of reserves on which interest is being paid is an upper bound on the net present value of the payments. Prior to late September 2008, that number was less than $50 billion, not enough to make much of a difference. And the injection is spread across all reserve-holding banks in proportion to their holdings, not targeted, so that further lessens its impact.

    Still, I suspect that at least some of the advocates of IOR were thinking of it as helping out a bit with the financial crisis, while ignoring the much larger negative effect of the monetary tightening it represented. But then, as Milton Friedman could have told you, many at the Fed have never believed that money matters.

  8. Gravatar of Mike Sandifer Mike Sandifer
    12. November 2010 at 05:16

    Scott,

    You mention Lehman and I’ve wondered about something.

    Dean Baker claims that we could have withstood the collapse of the largest financial institutions without anything like the doomsday consequences discussed during the bailouts. But, my thinking’s been that if a failure of a third of US banks and the associated drop in the money supply led to most of the unemployment during the Great Depression, then the effects of letting the largest lenders fail in ’08 might have been far worse. After all, weren’t they responsible for something like 70% of all US lending?

  9. Gravatar of scott sumner scott sumner
    12. November 2010 at 05:42

    Mark, Doc and FEH, I guess I am preaching to the choir.

    David, This is a very odd way to look at stock market data. The stock market is supposed to immediately react to new information–in all three cases is fell sharply. Why would you look at a long term period when doing an event study? In any case, even if you do take a long term period, the stock market was far lower after the 3rd increase, as compared to before the first announcement. On the other hand, Lehman had only a modest effect.

    Mike, I believe the failures would have had a far less severe an impact than people assume, assuming the Fed targeted NGDP, level targeting. Otherwise the effect might have been severe–I don’t know.

    Remember how we were told it would be a catastrophe if GM went bankrupt? Well they went bankrupt, and nothing bad happened. It doesn’t prove anything–but it tells me to be suspicious of claims of catastrophe.

    Having said all that, if I could go back in a time machine I’d bail out Lehman–because we now know the Fed didn’t target NGDP.

  10. Gravatar of James in London James in London
    12. November 2010 at 06:05

    Having said all that, if I could go back in a time machine I’d bail out Lehman-because we now know the Fed didn’t target NGDP.

    You are such an interventionist, and friend of the bankers. I just don’t believe you have any real commitment to structural reform or imposing moral hazard. When I first read your commentary I thought you were a Keynesian socialist (who ends up bailing out bankers) hiding in free market clothing, first impressions look right.

  11. Gravatar of ssumner ssumner
    12. November 2010 at 06:11

    Alan, That’s basically the same as the liquidity trap argument. I have three objections. ERs and other securities may be close substitutes, but they aren’t perfect substitutes.

    The 1/4 point increase is not necessarily trivial, as it is enough to push the rate on reserves above other similar assets.

    The 1937 doubling of reserve requirements also only led to a 1/4 point increase in short term rates, but it sharply reduced the money multiplier and the broader money supply.

    I agree that it is possible the effect is trivial, but my hunch is that the markets would respond positively to an attempt to cut or eliminate IOR as part of a stimulus package.

  12. Gravatar of David Pearson David Pearson
    12. November 2010 at 06:37

    Scott,

    I notice that you don’t think stock market levels are very important compared to changes. My point is that, in reaction to new information, the level of the S&P hardly budged. What people knew on October 12 and what they knew on October 22 both caused them to assess the same value on the market — about 900. So therefore, how can the news of October 22 be considered significant? In your world view, I would imagine the news would have caused a significant deterioration in the level of stock prices relative to its recent trading range.

    I’m not saying the announcement had no impact (although I believe, from memory, the impact was minimal). I am saying that the impact, relative to everything else the markets were digesting at the time, was insufficient to reduce the level of the S&P in a material way.

    For emphasis, let me put it another way. If you say, “markets believed that raising the IOR materially reduced growth prospects in the economy,” then, based on market trading levels, I would have to say the data does not support that thesis.

  13. Gravatar of David Stinson David Stinson
    12. November 2010 at 06:43

    Scott, have you seen this? It may be relevant to this discussion.

    http://www.ny.frb.org/research/current_issues/ci15-8.pdf

  14. Gravatar of David Pearson David Pearson
    12. November 2010 at 06:54

    An example of what I mean in my earlier post. An event study might find that QE announcements from August to November of this year materially reduced long term bond yields. However, 10yr yields now trades above pre-Jackson Hole levels, and the 30yr is materially higher. So the thesis, “Fed policy changes reduced long term interest rates,” is true, or false?

  15. Gravatar of Full Employment Hawk Full Employment Hawk
    12. November 2010 at 07:30

    “I also can’t see how a negative IOR will be all that beneficial, as it wouldn’t resolve the flight to quality issue.”

    If a significant penalty were imposed on excess reserves, banks would shed most of their excess reserves. The best outcome would be if they shed some of them by increasing their lending. But even if they do not, their purchase of securities would create a significant liquidity effect on interest rates, driving interest rates down (only to have that offset by income and inflationary expectations effects resulting from the stimulative effects of the liquidity effect). Since the yield on very short-term interest rates is very low, we should expect a lot of the bond purchases be of longer-term bonds, affecting the long-term interest rates, which are the relevant rates that influence most expenditures.

    Imposing penalties on excess reserves is probably the most effective single way to give the economy the needed stimulus at this time. But this would be controversial. The anti-recovery faction (what Krugman calls the pain caucus) would go ballistic. But at the least, they should stop paying interest on excess reserves.

  16. Gravatar of Full Employment Hawk Full Employment Hawk
    12. November 2010 at 07:34

    ONE way we can know that the economy is not in a liquidity trap is because there is one important interest rate that has not reached the zero floor: The interest rate on excess reserves. (Not that we would be in one if the rate were zero.
    A zero rate on excess reserves is a NECESSARY condition for a liquidity trap, not a SUFFICIENT condition.) But as long as this rate can be reduced, all talk about a liquidity trap is wrongheaded.

  17. Gravatar of Doug Bates Doug Bates
    12. November 2010 at 07:36

    I would be very cautious about imputing aggregate market index price causation to any single news item. Firstly, because there is a normal, arbitrary, ongoing variance in any market index. Secondly, because there may be other market-relevant news items occuring that day. Thirdly, because momentum traders who trade based on what they think everybody else will think inevitably cause short-term overshoot. Fourthly, because both irrational fear and irrational exuberance do exist — so a short-term market reaction is not the same as a rational balancing of realistic risks and expectations. And last, but not least, there are market-policy feedback effects, in which policy is adjusted based on expected market reaction, in a way that is not always predicted by market prices beforehand — in other words, markets may throw an unrealistic pricing tantrum in reaction to a tentative policy decision in order to force a policy change, but if everybody knew the tantrum would force a policy change, then the policy change would happen anyway and the pricing tantrum would never occur.

  18. Gravatar of Mike Sandifer Mike Sandifer
    12. November 2010 at 07:46

    Scott,

    My impression has been that few, if any of the major financial institutions would have failed if NGDP had been maintained at target.

  19. Gravatar of Jon Jon
    12. November 2010 at 08:56

    Great to see you put this out there.

    My preferred policy knob has been IOR for sometime–which is in part why I’ve been soft on QE. QE is a powerful tool but IOR is more powerful because of its effect on the money multiplier.

    Why? Because bank lending liquidates the interbank market just as much as the public market. So the cost of funds and the public market fall in tandem, but a floor on the interbank market decouples the two. So now an expansion of lending to the public collapses the interest-rate margin. Ergo there is no more lending.

    Its a price control on the cost of funds.

  20. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    12. November 2010 at 09:29

    Scott, you said:
    “And why is the uncertainty about fed funds rates a big problem, when the level of rates was low thoughout this period, and extremely low if you recall that these are overnight loans”
    Average level of rates was high and variable – see the second chart in my post.

    You said:
    “How much risk did ffr instability actually produce?”
    The best indicator is 3 month libor (see the third chart in my post).

    You said:
    “he tends to assume that the increase in the base was facilitated by having the IOR program”

    The October 6, 2008 announcement had two items – IOR program and increase in the monetary base. October 22 announcement signaled no additional expansion of monetary base and was contractionary. Here is an excerpt from October 6 announcement.
    “Consistent with this increased scope, the Federal Reserve also announced today additional actions to strengthen its support of term lending markets. Specifically, the Federal Reserve is substantially increasing the size of the Term Auction Facility (TAF) auctions, beginning with today’s auction of 84-day funds”
    and
    “The sizes of both 28-day and 84-day Term Auction Facility (TAF) auctions will be boosted to $150 billion each, effective with the 84-day auction to be conducted Monday. These increases will eventually bring the amounts outstanding under the regular TAF program to $600 billion. In addition, the sizes of the two forward TAF auctions to be conducted in November to extend credit over year end have been increased to $150 billion each, so that $900 billion of TAF credit will potentially be outstanding over year end.”

    You said:
    “”In the overnight federal funds market, financial institutions became more selective about the counterparties with whom they were willing to trade. “
    Is the implication that IOR solved the counterparty risk problem?”

    No. The (knowledge) problem was the lack of a reliable indicator of fed funds market, as only the strongest counterparties traded. The answer to the problem was to use survey-based overnight Libor, and expand monetary base and LOLR support until overnight Libor (and counterparty risk) drops to acceptable levels.

  21. Gravatar of Morgan Warstler Morgan Warstler
    12. November 2010 at 09:40

    123 says,

    “Start of interest on reserves program led to the announcement of October 13, 2008, on that date the Fed promised to supply unlimited quantity of reserves until the policy objectives have been met.”

    Why does he say it was the IOR? Wasn’t the Fed saying “unlimited quantity of reserves” what solved counter party risk?

    Wasn’t that the white flag? The Great Bend Over?

    The banks won’t loan to each other overnight because these assets are fiction… the Fed says, “don’t worry we’ll print whatever we have to” and…

    IOR is now called the green shoots?

    It sounds like 123 thinks zombies are better than death.

  22. Gravatar of StatsGuy StatsGuy
    12. November 2010 at 10:19

    Bernanke has no reason to increase IOR right now – he’s found more effective and discrete ways to “recapitalize” large banks.

    You don’t have to be a conspiracy theorist to accept the notion that there’s a lot of horse trading going on behind closed doors. You just need to believe Coase.

    Finally, the one CONSISTENT principle the Fed has pursued has been to support the financial system. Period. I would challenge you to name a single decision in the last 3 years that hasn’t favored the financial system. Even QEI was not unleashed until the Fed was absolutely sure the threat of rising interest rates had been crushed, and the risk of financial collapse through defaulting loans and asset depreciation had exceeded the risk of financial collapse through loss of bond valuations. AKA, liquidity crunch.

    The current QEII round has directed preferential liquidity through primary dealers, thereby supporting trading activity which has buttressed bank balance sheets (in other words, infused capital) to compensate for the bad loan book.

    You currently view Bernanke et. al.’s endorsement of QEII as an intellectual victory. Consider, for a moment, it simply reflects a closer alignment between the interests of large financial entities, and the broader economy.

  23. Gravatar of Morgan Warstler Morgan Warstler
    12. November 2010 at 11:37

    StatsGuy, I think you can use this handy formula:

    Banks > Republicans > Unemployed (Democrats)

    The real trick is getting the Dems to accept it like this:

    Republicans > Unemployed > Banks

  24. Gravatar of Yglesias » Interest on Reserves Yglesias » Interest on Reserves
    12. November 2010 at 12:29

    […] in exchange for doing so. Then they raised the interest rate. And then they raised it again. Via Scott Sumner, Louis Woodhill makes a very strong argument that this has been a massively underrated […]

  25. Gravatar of Benjamin Cole Benjamin Cole
    12. November 2010 at 14:06

    Yeah, let’s do what Scott Sumner says. I say we go all in.

    We need to stimulate this economy by any means necessary.

    BTW, the unit labor cost report (DoL, 11/4) says unit labor costs are falling by 1.8 percent 3Q y-o-y. Add that to weak AD.

    How does this set up for inflation?

    Inflation right now is holding hands with Jimmy Hoffa. Inflation has as much chance of a comeback as Milli Vanilli. Tiny Tim is more of a threat than inflation.

  26. Gravatar of PJ PJ
    12. November 2010 at 14:49

    So, is there much agreement on why the fed increased IOR in October 2008? I’m with StatsGuy – the only reasonable explanation I come up with is that it was a back door way to recapitalize banks while everyone (including the bankers themselves) tried to figure out who was solvent. But that doesn’t explain why it remains above 0.

  27. Gravatar of Mark A. Sadowski Mark A. Sadowski
    12. November 2010 at 16:47

    Scott,
    I’m starting Monday’s lecture with that Bush quote:

    “If money isn’t loosened up, this sucker could go down”.

    Even our lackluster previous president understood the implications. Too bad that our current president is too nuanced to figure out how much money matters.

  28. Gravatar of Jon Jon
    12. November 2010 at 19:43

    123 writes:

    The best indicator is 3 month libor (see the third chart in my post).

    LIBOR diverged substantially from the commercial paper markets. The non-financial market exhibited little to no rise in yields which eviscerates the notion that there was a liquidity issue.

    There was a counter-party risk problem.

    Moreover, LIBOR diverged from the domestic financial CP market. This indicates some combination of 1) the problem was largely a European one with European banks holding the substantial share of the subprime paper from the US and the survey based approach is useless and too easily manipulated.

    If the CP market data isn’t enough, John Taylor put out several papers using the OIS spread which shows clearly that LIBOR was not a useful indicator.

  29. Gravatar of Richard W Richard W
    12. November 2010 at 19:53

    I am with 123 that IOR are not contractionary and merely assist the Fed in maintaining the target rate. The level of reserves both required and excess is entirely determined by the central bank. Therefore, the various Fed policy initiatives will inevitably lead to a build up in excess reserves. No matter what one individual bank does with its reserves they can’t change the total level of reserves in the banking system. I think people are thinking in terms of the textbook money multiplier. However, that does not apply when the central bank pays interest on reserves.

    On the one hand they want to reduce disruption in the interbank market to prevent banks from contracting their lending. Furthermore, they are also maintaining their target rate. With the various other initiatives that have led to excess reserves they would need to sell bonds otherwise the market interest rate would fall below the target rate. For example, if the Fed did not pay IOR the market interest rate in the interbank market would fall to zero and the market would freeze as the opportunity cost of holding reserves was zero. That would be contractionary as banks would be forced to shrink their balance sheets. As the Fed have said the excess reserves were not created to serve any purpose of their own. They are merely a byproduct of other initiatives.

  30. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    13. November 2010 at 01:48

    Jon said:
    “LIBOR diverged substantially from the commercial paper markets. The non-financial market exhibited little to no rise in yields which eviscerates the notion that there was a liquidity issue.

    There was a counter-party risk problem.”

    Fed funds market is a market that by definition has a counterparty risk premium. If a counterparty risk premium rises, NY Fed has a responsibility to do one of two things – reduce risk-free rates, or reduce the risk premium, so fed funds rate does not rise. NY Fed failed to do this.

    Domestic financial CP market was stabilized by specific Fed interventions, and this market exhibits a participation bias.

    Despite all disadvantages of LIBOR, John Taylor decided to use LIBOR in his Taylor rule when he updated it in 2008.

    Richard W,
    Interbank rates falling to zero would be a good thing. The problem that IOR solved was interbank rates that were too high.

  31. Gravatar of Richard W Richard W
    13. November 2010 at 08:26

    The Fed viewed that as leading to conflicting policy objectives.

    Another way the central bank can eliminate the tension between its conflicting policy objectives is to pay interest on reserves. When banks earn interest on their reserves, they have no incentive to lend at interest rates lower than the rate paid by the central bank. The central bank can,
    therefore, adjust the interest rate it pays on reserves to steer the market interest rate toward its target level.

    http://www.newyorkfed.org/research/staff_reports/sr380.pdf

  32. Gravatar of scott sumner scott sumner
    13. November 2010 at 09:06

    James, I favor abolishing FDIC, TBTF, the GSEs, the CRA, FHA, the mortgage tax deduction, etc, etc. I want laissez faire in housing and banking. You might be right about Lehman, I honestly don’t know.

    David Pearson, There are many reasons why stock prices change, that’s why you look at the immediate response after news hits the market. It is very likely that some positive news hit the markets during those days in October when prices rose, and then prices fell when IOR was started, or increased. That’s a universally accepted way of doing event studies.

    I am not claiming that I have a complete explanation of what was driving stock prices, indeed I’m still not certain IOR depressed prices–it could have been coincidence. But I am confident that we need to look at the effect right after the announcement.

    David Stinson, Yes, I seem to recall it confirms my view that IOR had a contractionary intent,

    David Pearson, As you may know, I am not convinced that easy money reduces interest rates. Indeed the QE announcement raised longer term bond yields (10 year) so your data is no surprise to me.

    Full Employment Hawk, I completely agree with your views on IOR.

    Doug, I have a number of posts that discuss the circularity problem that you mention–and I agree it does make things more complicated. I also have many posts supporting the EMH, so I disagree with you there.

    Mike, I agree, and my newest posts provide more evidence that it was an AD problem.

    Jon, Thanks, I agree.

    123, OK, I see your point about the Libor rate. But I don’t see why IOR was required for the October 13 decision to supply lots more reserves. They are logically distinct. Even if you argue they are politically related, that wouldn’t change my mind, as I would simply disagree with the politics. I have never argued that no IOR and no reserve injections would have been better, I’ve always argued that no IOR with existing reserve injections would have been better. If the Fed can’t separate the issues, then that’s their problem. Since we agree about October 22, I don’t think we are as far apart as we had assumed.

    I’m still a bit confused about why if counterparty risk continued, rates fell close to zero simple because the quantity of reserves increased. But that’s a side issue.

    I see the stock market crash as resulting from continued disappointment that the Fed did nothing effective in the first 10 days of October. The stock market was right–neither IOR or the modest rate cut of the 8th was effective.

    more to come . . .

  33. Gravatar of scott sumner scott sumner
    13. November 2010 at 09:45

    statsguy, You said;

    “Finally, the one CONSISTENT principle the Fed has pursued has been to support the financial system. Period. I would challenge you to name a single decision in the last 3 years that hasn’t favored the financial system.”

    How about the tight money policy of 2008 that caused NGDP to plunge, assets prices to plunge, and loan defaults to soar? I still think there were honest errors made, otherwise what’s the point of my blog?

    Having said that, I do think the Fed is too close to the banks, so you are partly correct.

    You said;

    “You currently view Bernanke et. al.’s endorsement of QEII as an intellectual victory.”

    A small victory. A much bigger victory is that lots of prominent liberal blogs now sound a lot like I did two years ago.

    PJ, The Fed claims they did it to prevent rates from falling below target (2% at the time) while they did massive injections of reserves to provide liquidity.

    Mark, I thought that Bush quote was interesting as well. In fairness to Bush haters, 123 once told me he wasn’t sure if Bush meant monetary policy or fiscal actions like TARP. But the word “loosened’ does suggest monetary policy.

    Jon and 123, You both seem to have good arguments–I really don’t know enough about these markets to be able to offer anything useful that you two don’t already know.

    Richard, You said;

    “I am with 123 that IOR are not contractionary and merely assist the Fed in maintaining the target rate.”

    That’s precisely why it was contractionary. W/o IOR rates would have immediately fallen from the target rate of 2% to 0%. That’s why IOR was contractionary.

    Richard, Yes, I read that and it is precisely the problem–it raises rates.

  34. Gravatar of 123 123
    13. November 2010 at 09:51

    Scott, you said:
    “That’s precisely why it was contractionary. W/o IOR rates would have immediately fallen from the target rate of 2% to 0%. That’s why IOR was contractionary.”
    No, without IOR (other stimulus being equal) the rates would have fallen to 0 sometime in November 2008.

  35. Gravatar of Richard W Richard W
    13. November 2010 at 11:10

    Rates too high in the interbank market = frozen market leading to banks being forced to shrink their balance sheet.

    Market rate falling below the target rate = counter-party risk remains and would lead to a dysfunctional market and banks being forced to shrink their balance sheet.

    IOR provided a floor and a ceiling to bring stability to the interbank market.

  36. Gravatar of Doc Merlin Doc Merlin
    13. November 2010 at 12:01

    @Richard W:

    “IOR provided a floor and a ceiling to bring stability to the interbank market.”

    If by that you mean it propped up the banks? Yes, I agree. If you mean anything beyond that, I would have to disagree.

  37. Gravatar of 123 123
    13. November 2010 at 12:10

    Scott, you said:
    “But I don’t see why IOR was required for the October 13 decision to supply lots more reserves. They are logically distinct.”
    There was a question of commitment – what to do when interest rates fall down to fed funds rate target. The first option was to promise to raise IOR, the second (default) option was to promise to contract the monetary base. The first option is more dovish. The second option is hawkish, and one month ago Plosser publicly said the second option should be used in the future to exit from QE.
    October 6 announcement is a botched version of October 13 announcement. On October 6 the Fed underestimated the quantity of monetary base that needs to be supplied.

    “Even if you argue they are politically related, that wouldn’t change my mind, as I would simply disagree with the politics. I have never argued that no IOR and no reserve injections would have been better, I’ve always argued that no IOR with existing reserve injections would have been better. If the Fed can’t separate the issues, then that’s their problem.”
    They are related because we have the fed funds rate constraint. In a sense it is a political problem. Greenspan would have ignored the FOMC for two months until they agree to zero rates. Perhaps FOMC and Fed board should be suspended for four years so Bernanke can develop a comprehensive framework (level targeting?) that he says is missing now. On the other hand, IOR is the best exit strategy from QE, so I am sure Bernanke would keep IOR.

    “Since we agree about October 22, I don’t think we are as far apart as we had assumed.”
    In my first IOR post in August ( http://themoneydemand.blogspot.com/2010/08/should-fed-stop-paying-interest-on.html ) I said:
    “Current target of federal funds rate is the range of 0.00 – 0.25. The Fed should cut federal funds rate target to zero; and the Fed should stop paying interest on reserves while fed funds rate is zero. The primary benefit of such move is that a signal would be sent to markets that a majority of FOMC members believe they should combat deflationary shocks, thus providing at least a short term support to stock market. Direct benefits of 25bps reduction are not so large”
    I always thought that higher IOR is contractionary to the extent that it causes higher fed funds rates.

    “I’m still a bit confused about why if counterparty risk continued, rates fell close to zero simple because the quantity of reserves increased. But that’s a side issue.”
    This is a very important issue. There are two channels – higher monetary base stabilizes NGDP expectations and counterparty risk declines. The second channel is related to LOLR operations that reduce counterparty risk premium either because they involve a quasi-fiscal subsidy or because they reduce externalities related to illiquidity. Which channel do you think is more important?

    “I see the stock market crash as resulting from continued disappointment that the Fed did nothing effective in the first 10 days of October. The stock market was right-neither IOR or the modest rate cut of the 8th was effective.”
    Yes, the first effective thing was October 13 announcement. Neither the October 6 switch to the floor system nor frequent increases in credit support programs were sufficient to stop the crash. The interest rate cut in 8th October was much more modest than the decrease in 3 month Libor that started after October 13.

  38. Gravatar of Full Employment Hawk Full Employment Hawk
    13. November 2010 at 14:37

    One currently anomolous phenomenon in the monetary system is that the M1 money multiplier is now slightly less than one, that is, it has become a money CONTRACTOR, so that M1 is increasing by less than the monetary base. If payment of interest on bank reserves were discontinued, this would most likely bring it back to more than one. A penalty on excess reserves would significantly increase it.

    M1 is not only the best currently available measure of the means of payment in the economy, but is much more closely related to the variable the Fed controls, the monetary base, than M2. An increase in M1 resulting from an increase in the money multiplier would have a very expansionary effect on the economy.

  39. Gravatar of q q
    14. November 2010 at 05:36

    @123, if i wanted to understand what you are saying from first principles, what would i read?

  40. Gravatar of Scott Sumner Scott Sumner
    14. November 2010 at 09:38

    123, The Fed did IOR to keep rates from falling. Economic theory predicts the IOR would become the opportunity cost of funds. Maybe not to zero, because as you indicated risk was increasing.

    Richard, I’m not a fan of interest rate targets to begin with, and 2008 simply confirmed my view that the Fed never should have been targeting interest rates. But if they were, both the target and the IOR should have been lower.

    123#2, I’ll have to study October 13th a bit more–you might be right. I still have trouble with the claim that the Fed needed to hit their ff target. Even after IOR was instituted they failed to hit the target, if my memory is correct.

    On counterparty risk, are you saying the increase in the base was so massive that the ffr rate fell to the target rate (near zero by december) despite the counterparty risk–implying the risk adjusted rate fell below zero? Then why not hold vault cash?

    Full Employment hawk, You might be right about M1 (I don’t follow that issue closely.) I’d just add that some claim MZM now measures transactions balances best.

    q, You might want to read some monetarist literature, because they emphasize the excess cash balance approach–and also the fact that monetary policy affects many asset prices. On the internet I have articles at Cato, and The American, which explain my views. My early blog posts from February 2009 lay out some of my views, as does the “FAQs” link.

    I’m not sure what monetarist literature is best, I always liked Irving Fisher and Milton Friedman.

  41. Gravatar of Richard W Richard W
    14. November 2010 at 12:15

    The thing is Louis Woodhil is doing what you say not to do and that is reason from a price change. There were lots of other things happening at that time. The coordinated interest rate cut was on the 8th. The Washington G7 meeting 10th – 12th. Initially the markets liked the G7 meeting and the Dow had its biggest ever daily rise. Credit markets were almost unmoved. On the Wednesday the Dow had its second biggest one day fall just two days after a record rally.

    This is what was happening in the interbank market.

    http://4.bp.blogspot.com/_DQf1dkiRNmw/SO5rjBqvDaI/AAAAAAAABQ8/Gw-HE0LS-lU/s1600-h/interbk0810.jpg

    It is important to remember some context of those days. At the G7 meeting Paulson was continually pressed about whether the US would follow the British plan and buy equity stakes in the banks and crucially guarantee all interbank loans. The market if it has one voice was unimpressed with TARP and the conventional view was the British plan was better. Even if equity markets were negative after IOR announcements. The more likely explanation why they were negative is because they were disappointed and wanted interbank guarantees as that is where all the stress was.

  42. Gravatar of Jon Jon
    14. November 2010 at 19:00

    Scott writes:

    Jon and 123, You both seem to have good arguments-I really don’t know enough about these markets to be able to offer anything useful that you two don’t already know.

    I think the data speaks for itself; I’m not sure I know what 123 means when talking about the CP market being limited. Its surely larger than the eurodollar market covered by LIBOR.

    http://lostdollars.org/static/moneycost.png

    The decoupling of financial and nonfinancial CP is a clear support of the claim that counterparty risk not liquidity was the key element of the rising cost of funds. This also means that the banking panic itself was not substantially raising the cost of funds in the direct lending (CP, Bond) markets. Yields therein were still controlled by the FF target.

    Given that the nonfinancial CP rates kept tracking the eff FF, I think Scott has it right: the support of a high eff FF rate by way of the IOR policy kept policy too tight–kept rates above the falling natural-rate level.

    This WAS done to make the TAF policy neutral as 123 claims, but that’s precisely the problem. TAF was attacking an imaginary liquidity issue. The real problem was left unchecked (rates too high). … and ironically I would argue TAF completely failed. What quelled the cost-of-funds problem was a combination of SEC filings and the TARP backstop.

  43. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    15. November 2010 at 06:00

    Jon, low interest rates on nonfinancial CP is an evidence of tight monetary conditions and flight to liquidity.

    I never said that CP market was “limited”

  44. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    15. November 2010 at 06:19

    Scott, you said:
    “The Fed did IOR to keep rates from falling.”
    The Fed did IOR to keep rates from falling below 2% and increasing above 2%. As rates were way above 2%, IOR was beneficial.

    Economic theory predicts the IOR would become the opportunity cost of funds.
    Yes, IOR became the opportunity cost of funds sometime in late October or early November, from that date we should blame the Taylor rule.

    You said:
    “I’ll have to study October 13th a bit more-you might be right.”
    Compare the powerful October 13th announcement to the wimpy October 6th announcement:
    Here is October 6th:
    “In addition, the Federal Reserve and the Treasury Department are consulting with market participants on ways to provide additional support for term unsecured funding markets.”
    Here is October13th:
    “The BoE, ECB, and SNB will conduct tenders of U.S. dollar funding at 7-day, 28-day, and 84-day maturities at fixed interest rates for full allotment. Funds will be provided at a fixed interest rate, set in advance of each operation. Counterparties in these operations will be able to borrow any amount they wish against the appropriate collateral in each jurisdiction.”

    “I still have trouble with the claim that the Fed needed to hit their ff target. Even after IOR was instituted they failed to hit the target, if my memory is correct.”
    At first fed funds were too expensive, as they miscalculated the expansion of the monetary base needed to bring fed funds rate down, later fed funds were too cheap, as Fannie and Freddie do not receive IOR.

    “On counterparty risk, are you saying the increase in the base was so massive that the ffr rate fell to the target rate (near zero by december) despite the counterparty risk-implying the risk adjusted rate fell below zero? Then why not hold vault cash?”
    Overnight counterparty risk fell to very low levels by december. One of the reasons was increased monetary base together with the magic phrase “tenders of U.S. dollar funding at 7-day, 28-day, and 84-day maturities at fixed interest rates for full allotment”. Risk free rate was between 0 and 25 bps in december. And it was not only counterparty risk, it was also exchange rate risk that reflected high dollar real rates vs. short term euro and sterling rates.

  45. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    15. November 2010 at 06:34

    q,
    you should start with Milton Friedman on monetary policy (including his writings on Japan). “The Use of Knowledge in Society” by Hayek should persuade you that monetary policy is a very difficult problem. Krugman’s “The Return of Depression Economics” is a third excellent source.

    These two speeches “Some Thoughts on Monetary Policy in Japan” and “Deflation: Making Sure “It” Doesn’t Happen Here ” by Bernanke are priceless.

  46. Gravatar of Scott Sumner Scott Sumner
    15. November 2010 at 07:10

    Richard, Yes, it is dangerous to reason from a price change, and I’ve never argued that we can be sure the IOR caused lower stock prices. But auction style asset market prices can be useful in ways that other macro prices are not–we can do event studies. That reduces the risk of reasoning from a price change. Unfortunately, we don’t have good enough data to do a true event study in this case–Ideally you’d want to observe how prices respond immediately after an announcement. But given that economic theory suggest IOR is contractionary, and given the Fed publicly indicated the program’s intent was contractionary, and given stocks fell sharply after the only three applications of contraction IOR in US histroy, I think there is at least a strong circumstantial case against it.

    The coordinated rate cut of the 8th was a huge disppointment, as a far bigger rate cut was needed. They had done a bigger rate cut in January, when the need was far less acute.

    I agree that markets were also responding to news on bank bailouts–that is obvious. However it is less obvious exactly what the markets wanted in that case. When the bailout finally was passed, markets crashed, whereas they had been expected to increase. I think markets were already looking ahead and realized this was all the Fed and Treasury would do, and it wasn’t enough.

    Jon, How do you explain why the real rate on 5 year TIPS rose from 0.5% to 4.2% between July and November. There was zero counterparty risk. Isn’t that a liquidity problem?

    123, Thanks for those recommendations for q.

    I look at things from a supply and demand for base money perspective, you seem to have an interest rate perspective. I think the only way to resolve our dispute is to see what happens if the Fed does a surprise elimination in the IOR, while keeping the ff target constant. I’d expect markets to react as if the policy were expansionary. But I freely admit I may be wrong, as we are in uncharted waters.

    Do you agree that a strongly negative IOR would be inflationary?

  47. Gravatar of 123 – TheMoneyDemand Blog 123 - TheMoneyDemand Blog
    15. November 2010 at 15:25

    Scott, you said:
    “But given that economic theory suggest IOR is contractionary, and given the Fed publicly indicated the program’s intent was contractionary, and given stocks fell sharply after the only three applications of contraction IOR in US histroy, I think there is at least a strong circumstantial case against it.”
    The Fed indicated that contractionary intent is contingent on other stimulus (enormous increase of monetary base) being effective when it announced the program on October 6th.

    You said:
    “I look at things from a supply and demand for base money perspective, you seem to have an interest rate perspective.”
    No. I look at things from a supply and demand for base money perspective in terms of a three dimensional chart with thre axes labelled price level, quantity of base, interest on reserves.

    You said:
    “I think the only way to resolve our dispute is to see what happens if the Fed does a surprise elimination in the IOR, while keeping the ff target constant. I’d expect markets to react as if the policy were expansionary. But I freely admit I may be wrong, as we are in uncharted waters.”
    At this time ff target is a range, and IOR cut from 25 bps to 0 is equivalent to an expansionary ff target cut to zero.

    When ff target is a fixed value, reduced IOR is contractionary as lower quantity of monetary base is equivalent to tax on banking (as Friedman argued long ago). But this is not very important. The main problem is robustness of policy mechanism. As modern central banks prefer to operate by interest rate pegging, with low IOR and low monetary base there is a danger of interest rate undershooting or overshooting. And the risk is asymetric – excess money supply when rates drop below target is perceived as temporary and provides no monetary stimulus, while shortages of money supply when rates are above target are contractionary if credibility of interest rate pegging suffers.

    Bernanke has requested IOR in May 2008. He wanted to leave the fed funds rate at 2.0 (which was a reasonable position in May 2008), and increase the monetary base to provide more stimulus. This failed request for IOR was one of the most important wasted opportunities to prevent the Great Recession.

    You said:
    “Do you agree that a strongly negative IOR would be inflationary?”
    In current circumstances strongly negative IOR and ffr would be inflationary (if cash problem is solved like Woolsey suggested). But in general all usual caveats about interest rates apply – just like zero IOR was deflationary in Japan and inflationary in Zimbabwe, there is a risk that Plosser could engineer deflation with negative IOR.

  48. Gravatar of Jon Jon
    15. November 2010 at 21:49

    Scott writes:

    Jon, How do you explain why the real rate on 5 year TIPS rose from 0.5% to 4.2% between July and November. There was zero counterparty risk. Isn’t that a liquidity problem?

    I don’t think that’s what I was saying. My point is that the spike in LIBOR and Financial Paper was a counterparty problem. The recession stemmed from the FF target being too high through the summer and forward.

    The Fed’s statements in the September period forward discussed a diagnosis that the differential was due to limited liquidity within banking industry not generally. This is why they adopted an industrial policy approach of directing funds to the banks via TAF and sterilizing those based additions first direct with t-bill sales to the primary dealers and then by the IOR.

    As I stated before, the fact that the broader CP market tracked the FF target is indicative that the FF target itself not the banking sector was the problem.

    Ergo, IOR cannot be justified on the grounds that it accommodated the TAF as 123 argues. TAF was a program driven by a misdiagnosis (limited liquidity in the interbank market). The real problem was that the FF target was too high.

  49. Gravatar of ssumner ssumner
    17. November 2010 at 18:25

    123, I’m not a fan of three dimensional S&D. I’d rather just consider IOR a factor that shifts demand.

    Reading your answer I wonder if our debate is productive. We both see a cut in the IOR to 0% as expansionary, although we disagree about why. Maybe we just prefer to think of the same basic reason from a different perspective. For you it forces the fed funds target to zero. For me it means less demand for base money.

    Jon, I still have an open mind on the issue for two reasons. First, I simply don’t know as much about these market as either you or 123. Second, you both seem to have some good arguments. Of course my policy views are not wishy washy–I think easier money was needed, but I am not sure how best to characterize the basic problem with markets–as I see evidence both ways. But thanks for that info, and I’ll try to gradually learn more about this subject.

  50. Gravatar of 123 123
    21. November 2010 at 08:43

    Scott, you said:
    “I’m not a fan of three dimensional S&D.”
    Michael Woodford thinks four dimensional S&D is useful (the fourth dimension is composition of asset side of CB balance sheet).
    Of course nobody really likes three dimensional S&D, and the best way to collapse two dimensions into one is to consider that there are infinitely many combinations of quantity of base money and IOR that have the same expected AD. The only problem is that these combinations have different variance of expected AD. Combinations with high IOR have lower variance (especially during financial crises), and combinations with low IOR have higher variance. Many central banks have strong preference for low variance of expected AD, so they have switched to the floor system (with huge monetary base and IOR at target rate) for the duration of financial crises. Bernanke wanted to do the same in May 2008, but the law that allowed him to do this was changed only in September 2008.

    You said:
    “Reading your answer I wonder if our debate is productive. We both see a cut in the IOR to 0% as expansionary, although we disagree about why. Maybe we just prefer to think of the same basic reason from a different perspective. For you it forces the fed funds target to zero. For me it means less demand for base money.”
    Let me clarify things a little bit. The reason the fed funds rate needs to be consicered is grounded not in some theoretical differences between your and my views, but is related only to Fed’s operating procedures. Lower IOR leads to less demand for base money, but we also have to make sure that lower IOR doesn’t lead to lower supply of base money. If fed funds rate target is lowered together with IOR, this means that supply of base money does not shift, and only demand shifts.

    Here is a recent post by Mark Thoma who also said that IOR was expansionary:
    “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.”
    Source:
    http://moneywatch.bnet.com/economic-news/blog/maximum-utility/interest-on-reserves-and-inflation/1007/?tag=col1;blog-river

    I am in Brussels for the next two weeks, with infrequent access to internet, so I will post comments and replies with longer delays than usually.

  51. Gravatar of ssumner ssumner
    23. November 2010 at 13:23

    123, I certainly agree that if lower IOR leads to a lower MB, it may not be expansionary. Indeed I praised the Thoma piece in a new post.

    Enjoy Belgium.

  52. Gravatar of TheMoneyIllusion » Cameron on monetary policy in late 2008 TheMoneyIllusion » Cameron on monetary policy in late 2008
    14. January 2011 at 06:14

    […] announced it would pay interest on excess reserves (released at 8:15 am) Scott Sumner has long argued this policy greatly sharply tightened monetary policy during late 2008. So even though the […]

Leave a Reply