Response to Mankiw (and then some)

I am pleased that Greg Mankiw has recently given the interest penalty on reserves idea more visibility.  (It is also nice to be called a “prominent economist” by Mankiw.)  Unfortunately, like the rings of Saturn, this issue is much more complicated than it appears at first glance.  Many economists try to look at the entire liquidity trap issue by boiling it down to a few fundamentals; the perfect substitutibility of cash and T-bills at zero rates, or the problem of credibly promising a permanent expansion in the money supply.  Unfortunately none of those easy thought experiments are enough.  The problem must be examined from multiple perspectives at once.  Hence I plan a long post that will have readers crying “No mas, just try Sumner’s plan so I don’t have to read any more of this.”

Any examination of this issue must start with a few descriptive statistics.  These are off the top of my head and may be a bit off, but are close enough.

Normal times:

Cash = $800 billion,  Required reserves = $80 billion,  Excess reserves = $8 billion


Cash = $850 billion, Required Reserves = $90 billion,  Excess reserves  = $800 billion

T-bills are somewhere around $2 trillion.

So from a monetarist excess cash balance perspective, the problem is the hoarding of excess reserves by banks.  This is partly due to the payment of interest on reserves, which although only 1/4 percent, is slightly above the rate on T-bills.  And reserves are superior to T-bills in maturity and equal in default risk.  So it’s no surprise that lots of reserves are hoarded.  The experience of Japan in the early 2000s and the U.S. in the 1930s, however, indicates that substantial reserve hoarding can occur even without explicit interest payments.

We also know that banks hold very low levels of ERs any time the opportunity cost (in terms of the T-bill alternative) is even modestly positive.  Thus in the summer of 2008 when the target rate was only 2%, ERs were still very low.  So a penalty rate on excess reserves of say 4% should bring ERs down to extremely low levels.  In my view the Fed would then have to dramatically reduce the monetary base in order to prevent high inflation.  But Mankiw is not convinced, so let’s look at this issue more closely.

Let’s start by assuming that even with a negative 4% rate on ERs, the yield on T-bills stays near zero, or perhaps just negative enough reflect the cost of storing cash in safety deposit boxes.  (BTW, if you are wondering what would happen if T-bill yields also fell to minus 4%, one answer is that it would end the liquidity trap.  So we could stop right there, QED.)  But I think most would agree that that T-bill yields will go, at best, slightly negative.  Mankiw’s argument is that the interest penalty would have to be passed on to depositors in the form of negative rates on bank deposits, and that this would cause people to simply hoard cash in safes.

Even if Mankiw is correct, I will later make a number of arguments that this may be much less of a problem than he thinks.  But before doing that, let me question this assumption.  Suppose the government pays 4% positive interest on required reserves and negative 4% interest on excess reserves.  The penalty rate on excess reserves affects behavior at the margin, and should reduce ER holdings to the very low levels that existed before reserve and T-bill rates were nearly equalized.  In that case let’s say we go back to $90 billion in RRs and $9 billion in ERs.  Now assume that people begin to move away from T-bills because of slightly negative yields.  Do they have to go into cash?  I don’t see why.  When banks expand their balance sheets to have more deposits, they will expand everything else proportionately.  Because RRs are ten times bigger than ERs, the net flow of government cash to banks will increase, not decrease.  So if RRs are 1/10th of deposits; and if ERs are 1/100th of deposits, then the flow of interest will be 3.6% of reserves, and 0.36% of deposits.  Just enough to be slightly better than cash; and it’s FDIC-insured with no need to buy a safe!

There is the possibility that if T-bill rates go slightly negative, that cost would also have to be passed on to consumers.  But of course this means the Fed would also pay less in interest on its debt, so it would not be a problem to slightly raise the interest rate on RRs to make sure people put hoards into deposits, not safety deposit boxes (where there is no multiplier effect.)  Ten trillion in new bank deposits should be more than enough to get the monetarist excess cash balances effect working at many different margins, not just T-bills (which total around $2 trillion) but all sorts of other assets as well.  Others can check my numbers, but I think it would be much less expensive to the Treasury than fiscal stimulus, even if pursued for a year.  And I would expect it to push the U.S. out of a liquidity trap much sooner than that.

Obviously this idea might not work, but even if all the ERs that banks no longer wanted to hold went out into cash held by the public, I believe the effect would be much better than Mankiw anticipates.  Remember that the whole point of monetary expansion is to stuff more Federal Reserve notes down the publics’ throat, then what they want to hold.  (Call it a foie gras policy.)  As the public tries to get rid of excess cash balances, AD starts rising.  I think even Krugman would accept the fact that if enough money was injected into the economy, inflation expectations would rise.  The last time I made that claim he denied holding this view, so let me be more specific—I think even Krugman would accept that if the Fed printed enough cash to buy up all the assets on Earth, then inflation expectations would rise.  Of course he might reasonably object that this would be very risky.  It would either lead to hyperinflation in the long run, or would have to be later withdrawn, which could easily lead to massive capital losses by the Fed.

If the real problem with QE is that to be effective it would have to be so large as to be risky, then let’s make it as small as possible.  The Fed has already announced that they intend to do a sort of QE-type policy, but they are starting from a position of $800 billion in extra base money that is currently doing absolutely nothing for the economy.  If economists think QE is an idea worth pursuing, why not start by making things as easy for the Fed as possible, why not stop the one institution from hoarding that they are able to stop from hoarding, the commercial banking system.  The penalty rate on excess reserves would reduce bank demand for base money by roughly $800 billion, meaning that we could inject an extra $800 billion into circulation with no increase at all in the monetary base, and thus no increase in the Fed’s exposure to possible capital losses.  That’s what I’d call a good start.  Maybe it will all be put in safes (although I doubt it.)  But even if so, wouldn’t it be nice to find out now, instead of waiting until the Fed injected another $800 billion in new base money, above and beyond the current bloated monetary base?

But the preceding $800 billion head start on current policy is just the beginning of the advantages offered by an interest penalty on reserves—there are many more.  For instance, the demand for cash is probably far less elastic at zero interest rates than most academic economists suspect.  Most economists work with abstract models that totally ignore the real reasons why people hold cash.  For most of the public, cash and T-bills are not close substitutes.  People basically hold cash for two reasons:  transactions, and much more importantly, for tax evasion purposes.  Total cash in circulation is close to $3000/person.  Admittedly some is held overseas.  But even in smaller countries whose currency is not hoarded overseas, the cash ratio is fairly high.  What explains this?  It is the underground economy—cash is very anonymous and thus a good way of hiding wealth from the authorities (especially the IRS.)  For this reason, the hoarding demand for cash is not highly elastic in the short run at near zero rates.  Don’t believe me?  Then ask yourself why as interest rates have fallen almost to zero, the cash to GDP ratio has only increased slightly.

People who hoard cash will do so even at 5% nominal rates, as they can avoid paying 30% of unreported gains to the IRS.  Indeed the ratio of tax rates to interest rates correlates closely with tax hoarding in the long run, as it represents how long you can hoard cash before you would have been better off paying your taxes in the first place.  It’s true that in the very long run zero nominal rates would lead to considerably more hoarding by the public, but the other problem faced by underground economy hoarders of cash is that it is very costly to adjust their cash holdings in the short run.  They can’t go in and out of T-bills; rather they must adjust their flow of (cash) consumption purchases.  Look at the picture here, and ask yourself how easily the drug dealer could double his cash holdings, or cut them in half, and whether he even cares if T-bill yields go slightly negative.  And the short run is all important for the liquidity trap, as it is widely believed to be a short run problem (indeed this is the essence of Krugman’s “expectations trap” argument.)  The Fed needs a monetary expansion that is expected to last until the economy exits the liquidity trap.  If that exit is expected to be soon, then the problem is much easier to solve.

The other part of real world cash demand is transactions balances.  Most people walk around with very small cash holdings in their wallets.  And the amounts don’t seem particularly sensitive to interest rates in the short run.  I believe this may have to do with fear of losing their wallets and/or having them stolen.  If so, then these transactions balances still might not balloon very much at zero rates, or even slightly negative rates.  In that case Mankiw is speculating that an entirely new type of cash demand will occur at slightly negative rates, legal economy individuals holding cash as an investment.  This type of cash demand is probably very small right now, but could obviously become large if cash was seen as a desirable substitute for T-bills.

The question is how much cash is the public actually likely to hoard in the short term?  How much food can we force down the goose’s throat before it can’t take any more?  My hunch is that inflation expectations would rise significantly before anywhere near $800 billion was moved out of banks and into circulation.  I cannot prove that, but I believe I can prove that it would at least make QE more effective.  My proposal would offer the following three advantages over the current QE program:

1.  Cash hoarding would only occur at one margin (the public), not two.

2.  Penalty rates on excess reserves would eliminate cash hoarding on the margin where it currently is far more of a problem (excess reserves.)

3.  The QE program would start from a monetary base position roughly $800 billion lower than the current base, thus greatly reducing future potential capital losses by the Fed.

To those who don’t believe me, I would ask the following question:  If eliminating positive interest on excess reserves and going to interest penalties doesn’t help, then what would be the harm from raising the current positive interest rate even higher?  Why not raise it high enough so that the extra $1 trillion the Fed is contemplating injecting through QE is all hoarded by banks as excess reserves?  Most people would say that’s a terrible idea, it would insure that QE would definitely fail.  But if that is a terrible idea, insuring the failure of QE, then my interest penalty idea must be a good one, making it more likely that QE will succeed.  Notice I didn’t say guaranteeing it would succeed, just making success more likely.

Keep in mind that even people like Paul Krugman have said QE is worth a shot, and his model predicts it can work if the Fed’s inflation target is credible.  If even economists known as being particularly skeptical about monetary stimulus in a zero rate environment think it is worth a try, why not do it in the way most likely to succeed?  And again, one can’t argue that my plan wouldn’t make success more likely, unless one is willing to argue that the exact reverse policy (much higher interest rates on reserves to insure the extra QE reserve injections are hoarded) would not make the failure of QE more certain.  And I can’t imagine anyone making that argument.

Now we have exhausted all the monetarist transmission arguments that I can think of, and yet we have barely scratched the surface of the much more complex problem of liquidity traps.  So now let’s return to Krugman’s expectations trap, the argument that the real problem during a liquidity trap is a lack of policy credibility, the fact that the public might not believe the Fed’s promises to raise the price level at a particular rate over time.  Indeed, this may be the problem we face right now, as the Fed hints that it wants roughly 2% inflation, but the five year TIPS spreads show only about 0.8% expected inflation.

The good news is that no determined central bank has ever failed to inflate.  The bad news is that very few have tried at near zero rates.  Thus many people wrongly believe that near zero rates represent a barrier to inflation, when they actually represent the market correctly understanding that the central bank is engaged in deflationary policies (the U.S. in the 1930s and Japan more recently.)  In neither case did the central banks do the things (like currency depreciation) that you would expect from a central bank determined to reflate.  The only exception was the 1933 dollar depreciation program, which was successful in reflating prices.

Apart from all the advantages that I have already discussed, an interest penalty would be a strong signal from the Fed that it was serious about reflating.  So apart for the monetarist excess cash balance mechanism, you would also have the mechanism that Krugman called for—higher inflation expectations.  But the Fed should not stop there.  They must back up this action with a firm promise to follow a nominal aggregate trajectory (level targeting) of either a rising price level (100, 102, 104, 106, etc.) or rising NGDP (100, 105, 110, 115, 120, etc.)  They must also promise to try to “catch up” should they fall short.  This will further increase inflation expectations.

The preceding is almost certainly enough to get the job done, but I’m just getting started.  They have to be willing to engage in the “nuclear option,” if necessary pegging the price of a real asset such as gold or a foreign currency.  And then adjusting that peg until prices started rising.  In my view those two nominal anchors are not desirable (especially exchange rates, which could cause all sorts of diplomatic problems.)  Instead, the Fed should announce that if all else fails, if one month after their new policy is announced inflation expectations remain below 1.8%, they will start pegging the price of a 12-month CPI futures contract at 103, and then in following years 105, 107, 109, etc.  Why three percent in the first year?  To insure that you scare markets enough so that the nuclear option is not necessary.  Remember you are still doing all the other things discussed earlier (interest penalties, QE, explicit nominal targets, etc.)

As readers of this blog know, I would much prefer a NGDP growth target of 5% per year, but I am trying to reach a broader audience after the Mankiw link, and thus am using the much more generally accepted idea of an inflation target.  This futures targeting idea has been around for several decades, and no one has yet found a flaw.  As an aside, a slightly different proposal for futures targeting was criticized by Bernanke and Woodford in 1997 for having a circularity problem, but even they said the concept could work if you elicited a market forecast of the instrument setting required to equate CPI expectations with the policy goal.  This is exactly what my 1989 and 2006 proposals do.  The Fed offers to buy and sell unlimited CPI futures at the target price.  Each transaction triggers a parallel OMO, so that purchases and sales continue until the monetary base is at a level where the publics’ inflation expectations equal the policy target.  Like other “foolproof” escapes from liquidity traps, it contemplates targeting the nominal price of a real asset.  But unlike other proposals it is not subject to the problem of policy lags.

I don’t think the Fed would enjoy doing something as experimental as futures targeting right now.  But because it is a foolproof plan, just the threat of doing so, when combined with the many other steps proposed above, would mean that they would not actually have to use it.  And consider all the wild experiments they have already done.  Doesn’t it make more sense to commit to a highly experimental policy that there is less than 1 in 100 chances that you’d actually have to implement, and which will definitely work, than to continue on with other highly experimental and extremely costly policies that we already know (from TIPS spreads) are not even expected to work?


1.  Unlike other posts this was rushed out in response to the Mankiw link.  Thus I reserve the right to modify any ideas discussed here that have not been “vetted” yet, and are later shown to be foolish.

2.  Some people ask about vault cash.  That is not a problem.  It could be treated as reserves (and thus subject to the ER penalty), or exempted but capped at no more than X% of deposits.

3.  I’m not a banking expert.  If the current crisis means banks need more reserves than usual, then raise the reserve requirement to a level appropriate.  Or keep it as is, but only charge the penalty on reserves X% above the required rate.  Nothing of substance would change in my argument.

Update, 4/25/09:  JKH pointed out that the average reserve ratio is closer to 1%.  Bill Woolsey said that that reflected a 10% rate on transactions accounts, and 0% on savings accounts.  In that case I would prefer a 1% reserve ratio on all accounts, and thus an equal subsidy to all accounts.  As a practical matter that would mean actual hoarding would occur in saving accounts, where lower transactions cost would mean a higher interest rate.  If the money was hoarded by the public in savings accounts, $800 billion would support $80 trillion in M2 growth.  Where would all the bank assets come from?  Hyperinflation.

Bill has argued strongly that the interest subsidy is not needed.  I still think it refutes Mankiw’s point, but I also think Bill’s probably right.  The interest penalty on excess reserves is enough, when combined with QE and an explicit nominal target.  So I’ve come around to Bill’s view that there is no point in having an interest subsidy on RRs.



55 Responses to “Response to Mankiw (and then some)”

  1. Gravatar of Irrational Doomsday Blog Irrational Doomsday Blog
    22. April 2009 at 16:50

    Excellent piece, I especially like the point about the interest on excess reserves. I originally thought the interest was just another bailout vector to try to get some more capital into the banks.

    Let me raise your point in my own, more accusatory fashion: Why is the Fed still paying interest on excess reserves and then taking so many exceedingly risky and experimental policy moves to try to fight deflation? They make monetary injections, but they are also giving banks incentives to hoard them in excess reserves?

    The answer, if you believe that the banking system has captured the government (like in the Silent Coup article), would seem to be that it allows for incredibly huge injections, which benefit the banks, without resulting in the supposedly desired inflation. And this allows for ever bolder and larger injections.

    If they keep going, or even as it stands now, the banks will have such a preponderance of wealth as leverage, to try to oppose the banks if they threaten to start moving funds out of excess reserves risks collapsing the currency itself. It could make it difficult to tighten Fed policy or make decisions about taking over the “too big to fail” institutions if they have an inflationary suicide bomb.

  2. Gravatar of JKH JKH
    22. April 2009 at 18:31

    “In that case let’s say we go back to $90 billion in RRs and $9 billion in ERs.”

    – What would be the replacement funding for the resulting $ 800 billion hole in the Fed’s balance sheet? Is it currency or something else? If currency, how do you ensure there is demand for $ 800 billion?

    “Remember that the whole point of monetary expansion is to stuff more Federal Reserve notes down the publics’ throat.”

    – How do you do this? Doesn’t the public determine its own demand for currency?

    “The penalty rate on excess reserves would reduce bank demand for base money by roughly $800 billion, meaning that we could inject an extra $800 billion into circulation with no increase at all in the monetary base, and thus no increase in the Fed’s exposure to possible capital losses.”

    – Same question I guess.

    “The Fed offers to buy and sell unlimited CPI futures at the target price. Each transaction triggers a parallel OMO, so that purchases and sales continue until the monetary base is at a level where the publics’ inflation expectations equal the policy target.”

    – Why do you need to intervene in futures to achieve the target? Why not just do the required OMO to create the target expectations and corresponding futures prices?

  3. Gravatar of JKH JKH
    22. April 2009 at 18:33

    I wasn’t clear on the $ 2 trillion t bill reference.

    You’re not saying the Fed holds $ 2 trillion in t bills, are you?

  4. Gravatar of Devin Finbarr Devin Finbarr
    22. April 2009 at 20:18


    I love ya, but I think you’re way off the mark with this proposal. Let me try and explain why in a series of steps:

    a) Every person and corporation has a desired ratio of paper wealth to expected expenditures. For a retiree this ratio is very high, as they need to fund their retirement. For a young person, the ratio may be much lower, perhaps only enough wealth to act as a buffer in case of unexpected unemployment.

    b) The end of the bubble resulted in a dramatic fall in paper wealth. I’ve seen some estimates that total paper wealth of the world has fallen by half, an enormous amount.

    c) Let’s say before the pop the ratio of total paper wealth to total expenditures was 4 to 1. After the pop, it is now 2 to 1. But people still desire the original ratio. Retirees must restore their ratio or paper wealth to expenditures or they will run out of money too early. Young people need to restore their buffer or they will live paycheck to paycheck.

    d) There are only three ways to restore the ratio. 1) The private sector can try and blow the bubble back up again. This is a tough coordination game that is beyond the power of the private sector at this point. Nor would be a good idea if it could. 2) the government can print money and give it to people. 3) the private sector can start cutting spending.

    e) Since actors in the private sector have really good reasons for desiring the previous 4 to 1 ratio ( seniors do not want to run out of money before they die), the private sector’s choice will be to cut expenditures.

    e) When people cut expenditures, they cut durable goods and luxury spending first. Factories lie idle and millions go unemployed as a result. This is how a purely nominal problem affects the real economy. ( The famous paradox of thrift).

    f) Thus the best solution to prevent this is to print money to restore the paper wealth to expenditure ratio to the private sector’s desired level, without the private sector having to cut any expenditures. ( side note: running a deficit and issuing t-bills has the same balance sheet effect on the private sector as printing money. The large deficit the government is running is the single biggest factor keeping this harsh recession from being a depression ).

    g) The Mankiw/Sumner proposal is to force people to spend by penalizing them if they do not. If their plans work, it would maintain/boost aggregate spending by forcing people to accept the post-bust, 2 to 1, paper wealth to spending ratios. Their plans would prevent seniors from cutting their spending to reflect their lost paper wealth. As a result, seniors would run out of savings too soon. Young people would lose their desired wealth buffer, and live insecurely. In short, I think their plan would be a disaster.

    The private sector has very, very good reasons for wanting to increase its hoards of cash. Those reasons should be respected, and the printing presses should run to provide the desired cash.

  5. Gravatar of JKH JKH
    23. April 2009 at 02:33

    – Sorry, rereading, looks like you’re saying $ 2 trillion in t bills outstanding; nothing to do with the Fed?

    – And regarding my question on “replacement funding”, looks like you’re basically considering two possibilities: conversion of excess reserves to required reserves; currency issuance?

    – Regarding conversion of excess reserves to required reserves, I’ve seen a number of roughly $ 8 trillion for current banking system deposits in total (source unconfirmed, maybe dated). Using your number for required reserves, the current effective required reserve ratio would be closer to 1 per cent. Is 10 per cent a realistic reserve ratio assumption for size of the marginal change you’re considering?

    “Ten trillion in new bank deposits should be more than enough to get the monetarist excess cash balances effect working at many different margins.”

    – What types of additional assets would you foresee the banking system taking on in the associated balance sheet expansion?
    – Any additional capital requirements?
    – Is an implicit part of your proposal that you want the banking system to monetize most of the outstanding federal debt (no capital requirement?)

  6. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 02:36


    Your analysis of currency is faulty. I am not sure what is the exact problem, but I think it is too much equilibrium analysis rather than process analysis. (Or, I just don’t understand your point.)

    We are supposing that an excess supply of base money results in increased demand for T-bills. This can be because banks substitute T-bills for reserve balances to avoid the penalty on excess reserves or else the Fed buys them in standard open market operations, or that some of those with increased checkable deposits buy T-bills. Anyway, the prices of T-bills go up even more and their yields drop to zero, and then turn negative.

    I see too many people say, “zero lower bound.” But what does that mean exactly? First of all, if we imagine that FDIC insured bank deposits continue to pay interest or even have zero net interest (accounting for fees,) then as interest rates on T-bills fall (and turn negative) there will be a shift from T-bills to deposits. Some current holders of T-bills will sell them and hold onto the checkable deposits they recieve in return. The Fed and banks buy T-bills, and they are sold by current T-bill holders for deposits. The new deposit holders hold the deposits in place of T-bills because deposits are now a better store of wealth than T-bills with there assumed slightly negative yields, while deposits continue to have zero or slightly positive yields.

    If, on the other hand, we assume that the lower interest rates on T-bills and reserves (with excess reserves having a negative yield now) are “passed on” to depositors, then the interest rate on deposits falls too. And so,it is currency that becomes relevant. (I will discuss your proposal of paying higher interest on excess reserves in another post.)

    So, what are will assuming? As interest rates on T-bills and deposits fall, first to zero, and then negative, people begin to hold currency. Presumably, that would be people who already have safes (includining safety deposit boxes.) With interest on deposits negative, average people can just hold moderately more currency in the cookie jar in the house.

    The reasons why people held currency “in normal times”, including people who don’t use deposits to avoid taxes, is no longer relevant. No one is being forced to hold currency under this new circumstance. People begin to hold currency because it is a better store of wealth than T-bills and deposits, which we are assuming have negative yields.

    That drug dealers can’t change their currency holdings is irrelevant. How they respond to changes in T-bill yields doesn’t matter.

    The only way that people can start putting currency into the safes in response to lower (more negative) T-bill yields is to withdraw money from banks. As the Fed purchases T-bills, or banks buy T-bills in place of reserves, or some people buy T-bills with deposits, other people sell the T-bills that they currently own for deposits and then withdraw currency and put it in the safe. The efforts to create an excess supply of base money (both through open market operations and through penalty rates on excess reserves) flounder because the purchases of T-bills this implies are matched by people who sell T-bills for deposits, then withdraw currency and put in in a safe. These people are shifting from T-bills to currency in a way that exactly offsets the Fed’s creation of base money with open market purchases of T-bills and the banks substitution of excess reserves for T-bills to avoid the penalty rate.

    Imagining what would happen if currency just appeareed into the cash balances of current holders (drug dealers) tells us nothing. We are assuming that people are hoarding currency in place of T-bills. We have a new class of currency holders. People who would normally termporarily park their funds in zero risk, very short T-bills, waiting to see what they should do with their money later. Will the market bottom out, should we purchase that new machinery that we will need once the economy recovers, let’s wait. Because the banks (Fed and banks facing the assumed penalty rate) are buying t-bills, they are pushing the yield negative, but people who would otherwise hold the T-bills just sell them and hold currency instead. The T-bill yield only goes slightly negative because there are sellers of T-bills who match the Fed and the banks pruchases. What do they do with the money the receive? They withdraw currency and put it in the safe.

    We are making an argument that the demand for currency is rising. How does that create an excess supply of currency or base money?

    Think about the market process by which the zero bound is actually implemented. You can’t just say–well, the T-bills yield can’t be less than zero, so that is a constraint, and then what happens if currency stocks increase. Well, the traditional holders of currency can’t hold more than usual…. No. The way the zero bound works is that excess demand for T-bills at zero results in an increased demand for currency. A new sort of demand that doesn’t exist normally. Again, I will later comment on the proposal to keep the interest rate on deposits sufficiently high that there is no hoarding of currency.

    One final note. I believe that the Fed should buy up all of the T-bills, even if this results in people hoarding a matching amount of currency in safes. I don’t care about any micro loss because of excess safe purchases. But, these arguments suggest that there is no guarantee that this will raise nominal expenditure. (Not the safe purchases, but the increase in base money.) That is why I believe (and many say) that the Fed will need to purchase longer term to maturity and/or higher risk securities.

    The way I see it, and what I have always proposed, is that the Fed do as much open market operations with T-bills as possible. That is, until they have bought them all. And then, start buying other assets–more T-bonds, commercial paper and so on. Base money will rise to a point where there is an excess supply at current nominal income and to the quantity that would be the equilibrium at the target nominal income. There is no guarantee that the Fed won’t buy up all the T-bills before that point.

  7. Gravatar of Ken Ken
    23. April 2009 at 02:44

    Hey Scott,

    I like your “foie gras” metaphor. Have you ever addressed the ethical concerns of forcing people to spend their savings?

  8. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 03:03


    I know that interest rates play an key role in macroeconomic reasoning, but your error here is very fundamental.

    Suppose we raise the price of apples. This will raise the income of the apple sellers. They will buy more other goods. Yes, raising the price of apples is the key to macroeconomic prosperity.

    What is wrong with that argument? What about the apple buyers? They pay more for the apples, they have less money to spend on other things, so total spending does not rise. In fact, the entire argument in the first paragraph is offset by the consequence described in the second paragraph.

    The argument is often made using wages. The solution to the recession is to raise the minimum wage. The workers will earn more, they will buy more, etc. Ah, but the employers pay more. They hire less workers and raise the prices of the products. Those who buy the products have less to spending on other things.

    Suppose that higher interest rates increased the incomes of creditors. (Like retirees.) They will spend more…

    But the debtors must pay more. They have less money to spend on other things.

    Now, if the interest rates on short term, low risk, assets turn negative, then creditors will either pay to hold these assets (and those willing to borrow on these terms will receive that payment,) or else, they will shift to holding longer term and higher risk assets so they can continue to earn an income.

    It is an interesting intellectual exercise to imagine what would happen if all interest rates needed to be negative, but that isn’t where we are today.

    The signal that negative short term interest rates and positive long term interest rates gives is–spend now. While not everyone has opportunities where this is a reasonable response, some people do, and they should.

    It is like higher gas prices. It signals–use less. Not everyone can do this. But some people can, and they should. That fixes the problem.

    For example, suppose you are a retiree and you are going to need a new refriderator in the next five years. If you have all your wealth in “cash,” say, FDIC insured deposits, and the yield is negative, this is bad. You are digging into your capital to pay your bills. However, it is still in your interest to get that refrigerator now rather than hold onto the money (and lose a bit of it is month) and then get the refrigerator later.

    Again, what people in this situation will really need to do is move their funds to something longer term and higher risk so that they can continue to earn.

    And, of course, it is really that the people most able to do this that should. And that will solve this problem for those least able to make the change.

    Now, motivating people to buy consumer durables is part of the process, but getting firms to buy capital goods is the real goal here. Yes there is uncertainty. Sure, it is easy to understand a desire to wait and see before committing resources to capital goods that will contribute to production for the next 10 years. But, society needs the spending now. And so, the price signal is low short term interest rates now. And if .01% isn’t low enough, then -.01%, then -1% and so on.

    In my view, one of the failings of noneconomists is to see prices in a superficial manner. They are what the buyer pays and the seller receives. Changes in prices transfer funds between buyer and seller.

    Economists understand (I hope) that prices coordinate the market economic system. They need to be at levels so that buyer and seller plans match up. Focusing on how changes prices impact either buyer or seller alone, is a horrible fallacy. But even accounting for that, the “right” price, is the price that coordinates the values, preceived opportunities, and expectations of everyone.

  9. Gravatar of But What do I Know? But What do I Know?
    23. April 2009 at 03:52

    It takes a brilliant mind to come up with such a stupid idea as negative interest rates. Let’s just suppose that somehow the government could insist that banks who didn’t lend excess reserves would be penalized. Do you think for a minute that a trading vehicle that banks could “lend” to wouldn’t spring up like a mushroom after a spring rain in Bermuda or the Channel Islands? I mean, I’m no investing genius, but if all I had to do was provide a zero percent nominal return I could do it! The point is, who would sit still and allow their money to be confiscated when they could give it to someone else to hold and not lose any of it?

    If everyone is really so desperate to reduce the value of money, why not just give $10,000 to every man, woman, and child in the country–and promise there will be more next year. . .

    Or just stop collecting taxes. . .

  10. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 04:00


    Imagine there is no such thing as currency, but the interest rate on checkable deposits are zero. Bankers that hoard reserves go to prison for life, and so there are no excess reserves. Banks only hold T-bills and required reserves.

    The Fed undertakes open market operations, buying T-bills. The banks have excess reserves, but to avoid jail time, they use them to purchase T-bills. The ordinary money mulliplier process occurs.

    The increased demand for T-bills raises their price, and lowers their yield. The yield hits zero. Nominal income remains below target, so further open market purchases of T-bills occur, more excess reserves, more purchases of T-bills by banks. There is an excess demand for T-bills at zero.

    What happens? Here is the logic of the zero bound, without currency. The Fed and the banks are going to be buying T-bills. Current holders of T-bills sell them because they don’t want to hold them with a negative yield. Why? Because they can keep their money in a checking account that by assumption has a zero interest rate. The checking account is a better store of wealth than T-bills.

    And so, the Fed buys T-bills and the banks use the resulting excess reserves to buy T-bills, and so lots of T-bills are bought, but this is exactly matched by selling of T-bills so that the interest rate on T-bills doesn’t fall. But what are those selling the T-bills doing with the money? They are keeping it in their checking accounts. The new checking accounts created by the Fed in its initial purchase and by the banks when they purchased T-bills are exactly matched by the additional deposits that people want to hold as a store of wealth because T-bills would be a worse store of wealth with a negative interest rate.

    The open market operations have not created any excess supply of base money or excess supply of deposits. The increaced supply of deposits is exactly matched by an increase in demand. To avoid holding T-bills with negative yields (that would have developed if the excess demand for T-bills had been cleared with a higher price and lower yield,) people are selling the T-bills to the Fed and the banks and just choosing to hold zero interest rate deposits instead.

    Continue on…

    One problem with the process described here is that it assumes banks have a zero cost of intermediation. How can they let people “store” money in checkable deposit accounts for free, when the bank earns nothing on the assets it is using the deposits to fund? In reality, if the interest rate is zero on T-bills and that is all the banks are holding, (and zero on reserves,) then the banks are going to have to charge people for storing money.

    The market process is for interest rates to turn negative on deposits. And, of course, in reality, they turn negative before T-bills hit zero.

    And so, now, when the Fed does the open market operations and the banks get rid of excess reserves, holding deposits is always a worse option than holding T-bills. There is no zero lower bound.

    (Refer to my post above to see why the redeemability of deposits into zero interest currency creates the zero bound.)

    Now, let us suppose that for some odd reason, that negative deposit rates is consided a problem. So, the Fed decides that it is going to prevent the problem by paying the banks enough interest on their required reserves so that their earnings on their portfolio are sufficient to cover the cost of intermediation. The banks can have a portfolio of reserves and T-bills, but the interest on the reserves is enough so that banks will be willing to “store” money for free.

    So, now, we are back to where we were before. The Fed undertakes the open market operations. The banks buy up T-bills with the excess reserves. The excess demand for T-bills at a zero nominal yield would create a negative nominal interest rate. But zero interest deposits (courtesy of the Fed subsidy) is better as a store of wealth. So they just sell T-bills and hold deposits. We are exactly where we were before when we ignored the cost of intermediation. The increase in the supply of deposits is exactly matched by an increase in the demand for deposits. This demand is special, it only exists because zero interest deposits are a better store of weatlh than what would be negative yield T-bills.

    The excess cash balance approach requires _excess_ cash balances. If the demand for deposits rises an amount that matches the increase in the quantity of deposits, then there is no excess cash balance.

    If the reason people are holding deposits is because they are now an attractive store of wealth because the alterntive–T-bills, would be a poor store of wealth (due to a negative market clearing yield) then this is an increase in the demand to hold money to match the increase in supply.

    The market _automatically_ solves this problem by banks charging people to hold money in deposits when the banks can’t earn anything on the assets they hold.

    Creating a special Fed subsidy interfers with this market process.

    And, while the problem is fixed by the banks charging people for leaving money in the banks when they can’t earn anything on the assets they hold, the problem reappears because people can always withdraw zero-interest currency from banks. And so, the banks can’t charge people more for storing money than it costs depositors to just to store currency. And so, this creates a lower bound on interest for banks. Which creates a lower bound on T-bills.

    I am nearly certain that creating the subsidy for banks through paying interest on required reserves so that the currency drain is not a problem will simply create the problem I described above when there is no currency and the Fed, foolishly, created the subsidy just so that interest rates on deposits will not turn negative.

    In reality, banks don’t just hold T-bills and reserves. Even if they are constrained by capital requirements, they can replace excess reseves with long term bonds. As suggested by your post, the Fed can avoid the risk of capital loss from purchasing long term bonds by getting the banks to do so. (But this is a stupid game regarding a loophole in capital regulations.)

    The assumptions in the argument about all the excess demand for bonds at zero interest shifting to deposits is also just an assumption. Some of it can shift elsewhere, of course.

    My point is that creating a very high penalty for banks and then subsidizing deposits by paying interest on required reserves is pointless.

  11. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 04:33


    The $800 billion hole in the Fed’s balance sheet?

    You must be joking?

    Nominal expenditure has fallen and is too low because of an excess demand for base money.

    How much base money would be demanded if nominal income were at target? The current quantity of base money is less than that amount. If the quantity of base money were at that level now, there would be an excess supply of base money now.
    Nominal expenditure would recover.

    The purpose of the penalty on excess reserves is to reduce the demand for base money. That is both the actual demand and what the demand would be with nominal income on target. It creates an excess supply of base money at the current level of nominal income.

    An alternative approach is to increase the quantity of base money, of course.

    So, the answer is that nothing fills the “hole” in the Fed’s balance sheet.

    I will grant that Scott suggests that his penalty proposal will reduce the demand for base money so much that the Fed must reduce the quantity of base money to avoid inflation. And then, makes the odd claim that this will be a good point to start quantitative easing now. I wouldn’t take any of that literally. I hope that he doesn’t imagine first reducing base money and then raising it again.

    If the Fed needs to pay interest on reserves to fund all of its direct lending projects in order to avoid what would be an excess supply of base money at target nominal income, then paying banks to hold reserves may make sense. Of course, getting rid of those direct lending programs is an alternative approach. That is, more generally, give up on trying to prop up securitization.

    But this fails to be mindful of the key issue today which is the excess demand for base money at the proper level of nominal inocme.

  12. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 04:40

    What do I know:

    You don’t appear to know much.

    Zero or negative nominal interest rates doesn’t mean that people don’t have to pay back the money. If there were any reason to expect this situation to last forever, then waiting would always mean paying less. However, that is not very likely. The reality is that negative nominal interest rates would always be temporary. And so, not only will you have to pay back money, evenutally, the amount you will have to pay back will grow–like usual.

    My view is that giving everyone $10,000 or abolishing taxes would increase spending too much. However, giving everyone a smaller amount or lowing taxes some would work just fine. But the govenment would need to borrow money to pay for this. This adds to the national debt and the future interest burden.

    Penalty interest on reserves does not have that effect. The banks just have to purchase government bonds rather than hold reseserves, and balances in checkable deposis expand. If this creates too much money, then the Fed sells government bonds out of its portfolio, returning base money closer to normal.

  13. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 05:00


    The reason why people earn a return on saving is because by saving they allow fewer consumer goods to be produced for them now. That frees up resources to produce consumer goods for other people now, or else capital goods that allow for the production of consumer goods in the future. The reason people earn a return on saving is because of the value of these alternative uses of resources.

    If people want to save more now (and so, increase their ability to consume in the future) than other people want to expand their consumption beyond their current income, or else use resources to produce capital goods to expand the production consumer goods in the future, then there is nothing to pay a return on savings with.

    If people still want to save–postpone consumption–at a zero yield and no one wants to further increase consumption beyond income now, and no one wants to use more resources for capital goods, then having people pay to postpone consumption is the market process that coordinates these decisions.

    To go a bit deeper into the problem, people are trying to earn income–that is, contribute resources to current production, but rather than use current production for something, they want to get a committment for consumption out of future production. But no one else wants to provide that at even a zero rate of return. The market clearing rate of return is negative.

    Is it ethical to make people to pay to save? If that is what it takes to clear markets–yes. IF that is the intertermporal price that cooridates the decisions to consumer now or in the future with the decisons to produce consumer goods now and in the future.

    Just price theory is fundamentally mistaken. Prices are all that provide the signals and incentives needed to coordinate a market economic system. They aren’t just about a fair distribution between buyer and seller.

    Now, that is all pretty general. In reality, the near zero interest rate assets are all government guaranteed short term assets. So, the _real_ problem, is that people want to “save” by holding something with no risk–credit or interest rate risk. They want to wait. No one wants to bear that risk for them–for free– much less pay them to take their risk for them. And so, they should pay. It is more about risk than intertemporal coordination.

    If someone wants a return, they need to commit their funds for a period of time or else take credit risk. 20 year Baa corporate bonds are paying 7%. People can save and earn a return if they want. But they can’t keep there funds in a perfectly liquid form with no risk of capital loss.

  14. Gravatar of ssumner ssumner
    23. April 2009 at 05:18

    IDB, Thanks. I really have no idea why they are doing this. Central bankers seem to have an excessive fear of deflation. Earl Thompson has written on this subject. I think they should normally have a fear of inflation, but right now AD (or NGDP) is falling rapidly–we need a bit more inflation.

    JKH, How can we be sure the demand for currency is there? Well I hope it isn’t, otherwise my policy will fail.

    The Fed determines the base. The banks and public determine the reserve/cash split. My plan would sharply reduce reserve demand, so the public would almost certainly hold more. In theory deposits might soar–moving all the ER into RR, but that would be a bit far-fetched.

    There is a circularity problem if you simply try to target a an external futures market. If the policy is credible the price will be on target, and then you don’t have any information about where to set the monetary instrument. Bernanke and Woodford (JMCB, 1997) discuss this issue.

    The 2 trillion is held by the public, banks, foreigners, etc.

    Devin, I don’t want to reflate the bubble of mid-2007. I want to go back to mid-2008, when the real estate bubble had already popped, but had not overshot on the downside and also before stocks had crashed. Just a return to normalcy–a Goldilocks economy. Not 2007, not 2009.

    Money is far more powerful than fiscal policy (which is currently “not expected to work” according to TIPS spreads.)

    I am certainly not forcing people to spend. They can save all they want. Indeed I’d like to see the U.S. savings ratio soar up to Singapore’s 40% of GDP. I just don’t want to allow more saving to depress NGDP. That can be accomplished through expansionary monetary policy. All I am doing is discouraging ER holding. Nothing to do with saving. My plan doesn’t hurt seniors at all–it helps them. Their financial investments would rise.

    JKH#2, If the actual reserve ratio is much lower, so much the better for my plan–in fact that’s such great news I may update my post. Maybe instead of 10 trillion in new deposits, we’ll get 100 trillion!!

    I don’t know what assets bank would buy, and don’t care. The only goal is to reduce the demand for base money–in order to depreciate the dollar. I am certainly not trying to monetize the debt, exactly the opposite. For any given level of monetary stimulus, the lower the ER demand, the less debt is monetized. Remember that monetization of the debt is just another word for increasing the base. My plan would allow a smaller base to provide the same level of stimulus.

    Ken, I am not in any way forcing people to save. I am only discouraging banks from holding excess reserves. See my previous response to JKH. People would be just as able to hold t-bonds, stocks, t-bills, MMMFs, bank deposits, cash, whatever.

    But What do I know?, Good question! (Sorry, I couldn’t resist after you called my idea stupid.) I don’t think you understand what I am proposing, as your proposed “loopholes” such as investments in Bermuda would be fine with me. You seem to be under the illusion that many people have that I am trying to encourage lending. I do expect lending to increase somewhat, but that’s not my goal at all. It is to reduce demand for ERs, in order to reflate the price level. So let them use all the loopholes they want (except vault cash–long prison terms for banks officials who hold too much vault cash and don’t report it.)

    Bill, I don’t follow your worry about people substituting T-bills for deposits. I said deposits would go up by 10 trillion, but JKH points out it’s more like 100 trillion. So people reduce T-bill holdings by 2 trillion but deposits go up by 100 trillion. Isn’t that monetary expansion? My point was that with that much monetary expansion you’d start affecting other asset prices, not just T-bills. BTW, I have no objection to your idea of the government buying up all the T-bills, I just don’t see that as necessary.

    You said my comments on current cash hoarding were not relevant. I do understand why you make this point, but I am making a pragmatic argument that hoarding may be less than people imagine. In a simple model where the substitutibility of T-bills for cash is all that matters, my comment would be off topic, but it is not all that matters, lots of other things like expected inflation also matter. And the speed at which people adjust their cash holdings and build safes also matters. My point was that many economists probably think that as interest rates approach zero the public will hoard lots of cash. Not so, interest rates are already near zero, and cash hoarding has only gone up slightly. I did concede that if T-bill yields went negative people might hoard lots more cash. But we are talking not about legal cash hoarding doubling, as one might naively think by picturing the current $800 billion cash held by the public doubling, rather we are looking at legal cash hoards rising 10 fold or 100 fold, as legal cash hoards are now almost certainly very small. Right now almost all cash is either transactions balances or illegal hoarding (which doesn’t care about T-bill yields.) My point was that to suddenly get a public not used to legal hoarding to scale up their legal hoards so dramatically would be very disruptive. It might happen, but it’s just as likely that the very act of forcing that much cash into the economy would raise inflation expectations. That was my point. Technically, you are correct if you just look at the issue in a static, substitutibility fashion. And you may be right even in a broader sense, which is why I ended with several fallback options.

    I’m not sure I addressed all of your issues, but I think we need to first resolve some key issues:

    1. Could my subsidy plan make FDIC-insured bank deposits more attractive than legal cash hoarding?

    2. If so, would deposits soar by 10 or 100 trillion?

    3. If so, would this do much more than simply substitute for T-bills (which total 2 trillion?)

    4. We are try to move money out of ERs. It can go into cash or RRs. Why isn’t RRs better than cash, given the 10 or 100 fold multiplier effect on the money supply from RRs?

    When I can better pin down exactly where we disagree, I will be able to better address your questions, as some seemed to simply assume the bank subsidy plan would fail, which is an assumption I am not ready for yet, but will certainly accept if I can’t convince others.

  15. Gravatar of ssumner ssumner
    23. April 2009 at 05:32

    Bill, Rereading your second post, I really think the issue is magnitudes. You seem to assume banks could increase their deposits to 10 trillion by merely putting matching T-bills on the asset side. But there aren’t nearly enough T-bills in the entire world. That is exactly my point. Getting people to hoard in banks means that a given injection of base money can do far, far more. So if deposits rise by 10 trillion, banks will be forced to put non-T-bill assets on their assets side. And this then starts the monetarist excess cash balance mechanism working.

  16. Gravatar of Alex Golubev Alex Golubev
    23. April 2009 at 05:50

    ARE banks hoarding EXCESS reserves? are we talking about loss reserves? Sorry i’m not entirely sure how their BS works. balance sheet that is. if these are loss reserves, then i doubt we want to incentivise the banks not to save. You’ll help the velocity, but run the banks straight into a wall, no?

  17. Gravatar of Alex Alex
    23. April 2009 at 06:35


    Ok Scott you convinced me please don’t write more long posts (joking I actually enjoy long posts that reflect some original thinking from the part of the author). By paying interest on reserves the Fed just created a new short term government bond which is actually more attractive to t-bill at the current yields. Hence the 800 billion in excess reserves is not what we usually call the Monetary base and the true series for the monetary base should be adjusted just like the adjust it for changes on reserve requirements. Now the question is why did the Fed did this. Did they really thought that they were injecting liquidity into the whole economy or did they just wanted to inject liquidity to the banking sector but not to spill to the rest of the economy (they might want to do this to make sure that there are no runs against banks, to show in advance that they are the lender of last resort)

    Now about the cash hoarding issue by individuals. I’m your typical cash hoarder. I’m no drug dealer, I don’t have unreported sources of income, I just grew up from a country in which money in banks tends to vanish from time to time. What affects how much cash I hoard? Well a lot of it is related to the cost of banking. I went got some travel reimbursements in cash (like $,1000) and they went straight to my stash. Why? Well I can’t mail them to the bank as I do with checks which means I would have to go in person to the bank to make the deposit which I hate. Now I could decide to pay less with my credit card and use more cash for a couple of months and spend it (with out increasing my total expenditure). True but given that my money is one of this big banks which thanks to the Fed didn’t go broke I’m happy to keep it in cash. Even then I get 2% cash back when I pay with plastic and the interest rate I get on my deposits is less than 1% (before tax) and as you mentioned it is not safe to be going around with cash in your wallet so in the end the cash is still there in my drawer unspent (I also think that if robbers come to your house it is safer if you actually have something to give them so that they can leave as soon as possible). In any case, nominal interest rates, banking costs, rewards, risks, etc, would only affect whether my assets are in the form of cash or a CD, savings account, money market or T-bills, safe deposits, etc. and how I carry out my transactions, but it does not affect how much I spend (except maybe through wealth effects).

    The other way in which I can reduce my cash holdings is by spending more, i.e. I get $1,000 in cash and I go out and by a flat-screen TV which I wouldn’t have bought otherwise. Now why would I do this? Why would I consume more today. I would consume more today if the real interest rate drops. Giving me a wad of cash will not make me go out and spend if I think that goods are going to be cheaper tomorrow. Now I come from Argentina and I know that if you give everybody a wad of cash and don’t take it back later then prices will increase, so if I get the wad of cash I’ll go out and spend as soon as it takes me to drive to the nearest Best Buy. But that’s me. Some people might not see it that way, some people might not go out and spend it right away but given that American’s behavior in the last decade I have no doubt that they will go out and spend spend spend (I don’t care if initially they pay back credit card debt since that will just give the purchasing power to share holders in credit card companies in the end). So the only issue that remains is the commitment of the Fed to keep the money there. Now here I actually see a big move as counterproductive. Suppose that the Fed really increases monetary base by 800 billion. Do we really believe that the Fed will leave it there and let prices double? I don’t. If this is the case then I know the Fed will pull it out and then I have to guess how much will they pull out in the end. So it might be better to just aim at a more modest increase of 10% which is easier to live with afterward. (The one full proof commitment devise involves the Treasury increasing nominal debt by so much that we know that the Fed will have no other option tan to monetize it but I know you don’t like this one).


  18. Gravatar of David Pearson David Pearson
    23. April 2009 at 06:40


    Great post. After reading it, one can only wonder why the Fed keeps sticking to its ER-promoting policy. What the heck are they thinking?

    But then, the answer is right in front of us, and I’ve cited it before. The Fed is not, in their mind, engaged in QE. They are targeting credit spreads by intermediating the securities markets. To do this, they need a big balance sheet — they are in effect replacing the balance sheet capacities of countless hedge funds, Bear Stearns, Lehman, and the shrinking Citi’s and BofA’s.

    Your interest-penalty plan would have the same indirect effect as the Fed wants to achieve: higher nominal asset prices. The Fed just thinks it can be more surgical, targeting only site of the tumor. The tumor is the securities market, and the radiation is coming from the Fed’s expanding balance sheet.

    I’d be interested in a post from you on why Credit Easing will not work. I think that precise criticism of the Fed would help others also argue for an end to the payment of interest on reserves.

  19. Gravatar of JKH JKH
    23. April 2009 at 07:01


    “The Fed determines the base. The banks and public determine the reserve/cash split.”

    We’ve been through this before. This is not true. The logic is wrong.

    The Fed and only the Fed determines the CURRENT level of excess reserves. I’ll show this below. Because of this, the banks and the public do not determine the reserve/cash base split.

    In specifying this, it’s important to acknowledge the distinction between the accepted accounting definition of excess reserves, as defined by the Fed, and the economic concepts of excess supply and excess demand. I think you are almost certainly making this distinction, but the language of the discussion can sometimes be confusing in that sense.

    All Fed transactions affect excess reserves at the margin:

    E.g. 1: the Fed buys an asset and credit’s the seller’s bank’s clearing account. This increases excess reserves.

    E.g. 2: the Fed accommodates the “conversion’ of excess reserves to required reserves as a result of bank deposit expansion. This obviously reduces excess reserves.

    E.g. 3: the Fed accommodates the public’s demand for additional currency. This obviously reduces excess reserves.

    E.g. 4: taking into account all other transactions that will affect the system excess reserve position for a given day, the Fed decides it wants to make a final discretionary adjustment to the aggregate system excess reserve level otherwise pre-determined by all other transactions expected to be settled on that day. It does this by OMO. In that way, the Fed has the final determination as to the ultimate level of system excess reserves for that business day.

    The banks and the public don’t determine that Fed decision, so they don’t determine the reserve/cash base split.

  20. Gravatar of JKH JKH
    23. April 2009 at 07:11

    “You must be joking?”

    The general problem here is that much of the economic theorizing that is intended to support the proposal ends up being incompatible with the actual operation of the central bank balance sheet. My question about what happens to the “hole” in the Fed’s balance sheet is an essential one in that context. If you don’t understand the importance of the question, and the logical requirement for an answer, it’s because you don’t understand the operation of the central bank balance sheet, and in particular the way in which double entry accounting entries are required as a reflection of the actual process by which various forms of money are created and destroyed. My various questions here are rhetorical in that context, in an attempt to engage. But I see as before meaningful engagement is obstructed by defensive economic theorizing. Unfortunately, that chasm is symptomatic of why the proposal will not be influential.

  21. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 08:45


    Don’t forget the issue of capital requirements.

    Capital requirements aren’t a problem if the banks buy government bonds. It doesn’t have to be T-bills, but they have to sell new stock or else retain earnings (and not use them to write off past losses) if they are going to make commercial loans, home mortgates, buy investment grade corporate bonds, etc.

    I don’t understand the $100 trillion.

    Excess reserves are 800b. The marginal reserve requirement is 10%. That is $8 trillion in transactions deposits.

    Part of it will come ending sweep accounts. These are already transactions accounts from the point of view of the depositors. It is just that the banks tell the Fed they are something else (they are swept into something else right before it is time to report transactions accounts to the Fed.) No one bothers to keep track of this, but it is probably at least equal to current official transactions accounts. But that is small. Now, if the banks began to make savings accounts into transactions accounts, then they could come up with 1/2 the amount of transactions accounts (that is, 1/2 of the 8 trillion) pretty easily. It would mean that they would have to stand ready to cover checks from all the savings accounts and that would be costly.

    Anyway, unless the banks raise more capital, they can’t hold “risky” assets. They can only accumulate government bonds.

    Increasing the quantity of money (even to 10 trillion) will do no good if the demand rises an equal amount. Subsidizing the banks so that they can continue to pay interest on deposits is obviously aimed at increasing the demand for money.

    As I explained before, if the banks purchase other kinds of government bonds, this will be expansionary. I hate to say this, but it looks to me like the banks are driving down the interest rates on those bonds.

    My argument only applies to the degree that the Fed pays extra interest rate on required reserves to banks so that they can pay enough interest on deposits to avoid a currency drain in as the near zero nominal bound is approached. If the banks were only allowed to own T-bills, and all of them were like the 4 week ones a few months ago (really close to zero) then my arguments would apply. But in reality, there are all sorts of T-bills and their yields are not all .01% these days, and there are notes and bonds too, with interset rates much higher.

    Why 4%? why not -.25% on excess reserves?

  22. Gravatar of ssumner ssumner
    23. April 2009 at 09:54

    Alex, ERs have nothing to do with loss reserves.

    Alex#2, Let’s say you are a “typical cash hoarder.” Then when interest rates fall to zero you don’t hoard much more cash. That is exactly my point (the legality issue is irrelevant.) Now if you tell me you would hoard a lot more cash at zero rates, than I have a very simple response–you obviously cannot be a typical cash hoarder, because we know for sure that most people don’t behave that way.

    David, What the Fed doesn’t understand is the “tumor” is that all of NGDP is falling fast. When that happens almost all assets (except T-securities) go down. The Fed is treating the symptom not the cause.

    JKH, The problem with your analysis is at the end, when you refer to “for that business day.” The problem is that banks can very quickly get rid of excess cash balances they don’t want, just pay out more cash then they take in. Of course all banks and all of the public cannot simultaneously get rid of base money all at once. There’s a word for their attempt to do so; rising AD. Otherwise, prior to last September there would be no explanation for their ability to keep ER levels very low.

    Bill, JKH says there are about 8 trillion in deposits and I know there are about 80 billion in RRs, so why isn’t the deposit multiplier close to 100, or at least much more than 10? But let’s say it is 10. I don’t see why banks couldn’t expand their balance sheets. Regarding the bad banks, suppose they bought the entire national debt, and reduced long term rates. Isn’t that what QE is supposed to do? And banks could do it with a multiplier of at least 10, maybe 100. And there are 1000s of sound banks all over the country which could add lots of deposits and a wide variety of assets (if they weren’t being paid to hoard reserves.) Worst case the money goes out into cash, which is where it’d be otherwise anyway. The way I look at it when you want more AD there are three types of base money:

    ER — very bad (4% penalty)
    Cash — a little bad (0%)
    RR — very good (4% subsidy)

    We both agree on reducing ER down close to zero. I think base money in deposits is ten times better than cash. I see your point that my plan would slightly increase M1 and M2 demand. But I think it would increase M1 and M2 supply by far more–especially if JKH figure are right. I doubt that a slight interest subsidy would make real M2 demand double, or increase 10-fold.

    Of course this is all hypothetical, and none of this would happen or we’d end up with hyperinflation. I hope readers realize that Bill and I are debating this as a thought experiment, which plan would produce the most hyperinflation if carried through. In practice, if the entire plan was adopted the Fed would have to pull most of the extra $800 billion out of circulation to keep AD from exploding.

  23. Gravatar of JKH JKH
    23. April 2009 at 11:07


    “The problem with your analysis is at the end, when you refer to “for that business day.” The problem is that banks can very quickly get rid of excess cash balances they don’t want, just pay out more cash then they take in.”

    What I described is how a central bank actually sets its target excess reserve position for the day. I don’t know what you mean when you say “banks can very quickly get rid of excess cash balances they don’t want, just pay out more cash then they take in.” This is simply not true, and not the way in which the reserve system works.

    In the context of my single day model, which is a reasonable representation of how the system operates, the planned net delivery or draw of currency between each bank and the central bank has been notified in advance, so the central bank knows what it’s dealing with in terms of the gross effect of that day’s currency flows on the excess reserve position. It has similar knowledge with respect to updating of any changes to required reserves. It therefore knows how to adjust for these expected reserve effects in determining the full net effect of its actions when deciding on where to set the excess reserve position for that day. The system excess reserve position is the Fed’s autonomous decision.

    Beyond that, given that the day’s net currency exchange with the commercial banks and any required reserve changes are already accounted for through pre-notification, there is no way that commercial banks can change the amount of excess reserves in the system, other than dealing with directly with the Fed, which if it happens is at the Fed’s discretion, in the OMO market, or at the discount window. Other than at the Fed’s discretion, there is no way that the banking system as a whole can “get rid” of its excess reserve position. Individual banks can attempt to change their own positions, essentially by offloading their excesses to their competitors, which is the essence of the competitive interbank funds pricing process under the umbrella of a supply constrained reserve system as operated by the central bank.

    This “get rid of” thing seems to be a common idea running through a lot of the economic description here. It is completely wrong. As I said earlier, it contradicts the constraint of double entry bookkeeping in accounting for the source and destination of financial flows throughout a system of bank balance sheets.

    The central bank determines the creation and destruction of excess reserves. It allows for changes to required reserves and currency flows in doing so. And that’s all there is to it, in term of explaining excess reserve levels prior to last September, today, or at any time.

  24. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 13:05


    There may be thousand’s of sound banks, but you are proposing that they raise capital at a rapid rate. Sell new stock? While they may be making profits now (I guess,) how fast will that increase retained earnings?

    Perhaps you should be more careful in your terminology. Checkable deposits (called “transactions deposits”) have a 10% marginal reserve requirement. Savings accounts have no reserve requirement and Certificates of Deposit have no reserve requirements.

    Checkable deposits are a small fraction of total deposits. Last I looked, it was about 10%. The 10% reserve requirement against the 10% of deposits that are checkable is about 1% of total deposits.

    Excess reserves is relative to checkable deposits. Only an expansion of checkable deposits will turn the excess reserves into required reserves. Expanding savings accounts or Certificates of Deposit will not have that effect.

    If anything, banks will be motivated to shift their funding from savings accounts and CD’s to checking accounts. In the case of the shift from CD’s, I think that is a monetary expansion. With the shift in savings accounts–not so much. Reporting what are really checking accouts as checking accounts rather than hiding them using sweep accounts will reduce excess reserves without making any real change.

    The notion that banks will expand their $2 trillion in CDs because their is a penatly on excess reserve makes no sense. I also don’t see them expanding their $4 trillion in savings accounts. And, I don’t see them passively adjusting savings accounts if people want to shift fund to savings accounts. I expect lower interest on savings accounts to stop that.

    If reserves stays the same, then it is about an 8 trillion expansion in checkable deposits. Usually, there is some currency leakage, but who knows?

    JKH has said many times before that there is nothing to the money multliplier process. Hey, the Fed always chooses the level of excess reserves. (In his mind, the process that you are describing of banks buying assets with newly created checkable deposits would result in the Fed undertaking OMOs to keep excess reserves on target. He refuses to get beyond talking about the Fed’s current operating procedures. Heck, we know perfectly well that they will would do open market sales like crazy to keep the Federal Funds rate from falling too low.)

  25. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 13:37


    I oppose paying interest on required reserves. I oppose paying a subsidy so that banks will pay higher interest on deposits. Let the interest rates on deposits adjust according to market forces. If banks earn less, then let them pay less on deposits. If there is a currency drain, then create more base money.

    When you create a penalty rate on excess reserves, the reality is that banks won’t have any. There is not need to make it very high.

    I think the approach of paying banks interst on their required reserves so that they will keep interest rates on deposits high enough (not too negative) so that people will leave their money in banks rather than hold currency is a mistake. It is just a too clever financial manipulation.

    If this sort of subsidy would be necessary to avoid a currency drain, then you are talking about whatever assets it is that the banks are buying hitting the zero bound.

    And if they are hitting the zero bound, then your are making deposits superior to those assets. You are motivating people to hold deposits rather than those assets.

    Anyway, I don’t really think that is a problem. Pay the banks more interest on their excess reserves. I don’t think that the banks need to buy up the entire national debt and the subsidy that you want to pay for funding through checkable deposits is slightly deflationary, but… it just menas that the necessary amount of base money is slightly higher than it would be if there was no subsidy.

    If the entire national debt is purchased and interest rates on 30 year govt. bonds are at the zero bound, and the banks are still paying interest because of the Fed subsidy… then we can reevaulate.

  26. Gravatar of Frank Frank
    23. April 2009 at 15:57

    you state that banks are hoarding reserves; but there is an industry of securities analysts and other industry critical economists (Roubini/Krugman) that say the banks lack sufficient capital and need to be nationalized. Are they hoarding reserves in anticipation of more asset write downs?

  27. Gravatar of ssumner ssumner
    23. April 2009 at 16:14

    JKH, I meant a few days later the banks will get rid of any excess reserves they don’t want to hold. You gave an example when the Fed suddenly added reserves to the banking system. I agree that by the end of the day the banks may not have succeeded in getting rid of those excess reserves. But it wouldn’t take long, a few days at most. They could lend out any reserves they didn’t want to other banks. The system as a whole gets rid of excess reserves by not replacing cash that is withdrawn. The banks went from 8 to 800 billion voluntarily. The Fed didn’t force them to. They are quite capable of getting rid of those excess reserves. If there was a penalty rate and I was a bank president I would not hold a lot of excess reserves. I’d buy something like T-bills. And please, please don’t say that if I bought T-bills the money would go to another bank. There is something called velocity, isn’t there?

    Bill, I assumed that if banks bought up another 8 billion in assets, above and beyond T-bills, it would raise asset prices and encourage recovery. I’m not used to using this sort of monetarist approach, so maybe I am mistaken. But I had thought that monetarists argued that deposits were substitutable for not just T-bills, but all sorts of other assets. My argument was based on there being only 2 trillion in T-bills.

    I don’t know enough about banking to understand why they couldn’t expand their balance sheets proportionately. Obviously during hyperinflation banks balance sheets get much bigger, very fast, how do they raise the necessary capital? We are basically talking about a hyperinflation scenario here.

    The subsidy would just have to be big enough to prevent cash hoarding. That is it would have to cover the negative interest rate on bank assets, which would be tiny because cash is always a zero rate alternative. Essentially it is the cost of storing cash in safety deposit boxes. In addition there is the marginal cost of adding deposits, which is really small because most people already have bank deposits, so it’s not all that much more than the marginal cost of adding a zero to each person’s account. I don’t care about the checking/savings distinction, so you could make the reserve requirement simply 1% on all deposits. Then you’d have your 100 trillion. That would have to boost AD, wouldn’t it? That must be close to the total wealth of the country. How could asset prices not rise sharply? Of course all these thought experiments are silly, but they do show how easy it would be for the Fed to create hyperinflation if they wanted to. I think I might back off a bit, and just leave the subsidy as a thought experiment to show Mankiw was wrong, but recommend no subsidy, as you argue. I’d sort of like to leave the idea out there for comment, but not make it my main point. I’ll think about it overnight.

  28. Gravatar of ssumner ssumner
    23. April 2009 at 16:16

    Frank, No, capital and reserves are completely unrelated. If there was a penalty rate on ERs they would hoard T-bills or some other safe asset that can be used to meet their capital requirements at 100% of market value.

  29. Gravatar of Bill Woolsey Bill Woolsey
    23. April 2009 at 17:04

    What have banks done during hyperinflation about capital?

    I don’t know.

    Perhaps they didn’t have capital requirements at all. Or perhaps they were a fixed nominal amount. Or perhaps a bribe to the right public official means that the capital requirements don’t really matter.

    Or, perhaps banks are permitted it count appreciation of real assets that belong to the bank, like their buildings. This would increase their net worth and so capital.

    Or perhaps they do issue new stock. Banks may have strong nominal profit prospects during a hyperinflation.

    I suspect it all depends.

    Under current conditions, the U.S. Treasury could buy stock. They have. Or the capital requirements could be relaxed. Cut them in half, and the banking system could double in size.

    A bank’s net worth must be a certain fraction of its assets. However, these assets are risk adjusted. Reserves and government bonds have a zero weight. So, reserves can be shifted to government bonds without any impact on capital requirements. But shifting either government bonds or reserves to commercial loans requires that the bank obtain funds from stockholders. Either it must retain earnings to add to its net worth or sell more stock.

    Under normal conditions, an expansion of base money leads to an expansion of reserves, which leads to an expansion of transactions deposits. If people with excess transactions deposits shift some of them to savings accounts, this is good for the bank, because it reduces the banks reserve requirement.

    Now, the banks are holding excess reseves. The proposal is to charge them for storing money at the Fed so they don’t want to hold them. The individual bank purchases assets (like government bonds) and this reduces their reserves and excess reserves. However, what really happens is that whoever they purchased these bonds from receives a balance in a checkable deposit at some other bank, increasing the required reserves of that other bank and so reducing the excess reserves in the system. In general, the excesss reserves disappear as checkable deposits expand. However, if people shift their checkable deposits to savings accounts, this is no longer good for the bank. It reduces the banks required reserves, but no longer is this a good thing which gives the bank more reserves to lend and earn money. It instead increases the payments they must make to the Fed. The logic of increaed checkable deposits spilling over to savings accounts no longer works as usual. Rather than banks providing a better yield on savings account, at least partly because there is no reserve requirement, the opposite logic occurs. I would not expect M2 to grow in proportion to M1.

  30. Gravatar of ssumner ssumner
    23. April 2009 at 18:04

    Bill, I probably shouldn’t even be speculating about banks, as that is not my area. But I was under the impression that S&Ls did poorly in the late 1970s and early 1980s because they lent long and borrowed short. But maybe that’s not true of banks. And I recall reading that banks in places like Brazil learned how to profit from hyperinflation. After hearing your response I would definitely equalize the required reserve ratio between DDs and TDs. Since checking accounts are more costly in terms of paperwork, I would envision higher rates on TDs. But everything relative. I’d like to see just enough subsidy for rates of no more than .10% on TDs, and presumably less on DDs. And I anticipate the program only lasting a few weeks or months. And I’d be perfectly happy with dropping the whole subsidy idea and just going with the penalty rate on ERs.

  31. Gravatar of Devin Finbarr Devin Finbarr
    24. April 2009 at 01:10


    Imagine a gang of thieves slowly slipped counterfeit bills into the money supply for a long period of time. Eventually, half the money supply consists of counterfeit bills. The thieves had long traded their counterfeit bills for wine and women. The bills were now evenly distributed across the economy, in every home and business. Suddenly, the crime is discovered. Now no one will except the counterfeit bills. Paper wealth plummets, aggregate demand plummets. Factories site idle.

    What is the proper policy response to this scenario?

  32. Gravatar of Devin Finbarr Devin Finbarr
    24. April 2009 at 02:09

    Obviously during hyperinflation banks balance sheets get much bigger, very fast, how do they raise the necessary capital?

    In recent years, during inflationary times, most of the new money has been created in the shadow banking sector ( the money market funds). These funds have no reserves nor capital requirements. The limiting factor for money market funds is making borrowers fit their models for liquidity risk and default risk. Those models have all blown up, hence the credit freeze and the deflation.

    Bill and JHK are right, the banking system is constrained by capital requirements and the lack of borrowers with high enough credit ratings.

    BTW, Scott, even if adding reserves did allow the banks to lend more, how would this money actual end up stimulating aggregate demanding? Is your general idea, that the Fed easing should result in banks lending out money at lower and lower interest rates ( 30 year home mortgages for 2% … 1% )? Eventually the interest rates will be so low that people will borrow again, spurring aggregate demand? I still don’t understand the exact cause and effect that leads from the banks wanting to get rid of excess reserves to aggregate demand rising.

    Devin, I don’t want to reflate the bubble of mid-2007. I want to go back to mid-2008, when the real estate bubble had already popped, but had not overshot on the downside and also before stocks had crashed.

    Let’s say you were running an annuity company. And let’s say that some law had just passed making stocks held by annuity companies non-transferable. Once you buy, you hold it forever, and you have to live off dividends. What dividend yield would you demand in order to invest?

  33. Gravatar of Devin Finbarr Devin Finbarr
    24. April 2009 at 02:49


    In your response to my first post, you seem to me missing the bubble/bury the corpse effect.

    Imagine at point A all equities are owned by annuity companies that hold the equities for ever. Every year, equities dilute by 1% overall (option grants, raising capital, IPO’s, etc ) and the annuities spend $100 billion buying up these shares. There is no other trading. By definition, the market cap of all stocks is $10 trillion.

    Let’s say that over time young adults start investing a portion of their retirement savings in stocks. Let’s say that individuals invest $10 billion a year, then $20 billion the next year, finally up to a total of $100 billion a year. There is now $200 billion chasing 1% of the market, so the total market cap of all stocks is now $20 trillion.

    The oldest cohort of investors now starts reaching retirement. Recently, they have assumed that the $20 trillion price is a new plateu, and thus their paper wealth to expenditures ratio reflects the high price.

    But when this cohort reaches retirement, they must start selling stocks. Let’s say that instead of the amount of sellers each year being equivalent to 1% of the market, it’s now 2% of the market. The same $200 billion is buying stocks. The total market capitalization now drops in half. Now that it has dropped in half, these retirees must drastically reduce their expenditures. Aggregate demand plummets.

    No real wealth has been destroyed. But there is tremendous asset price deflation, and if this asset price deflation is not checked by a government policy of balance sheet repair, then it will result lower aggregate demand, unemployment, and idle factories.

    The mistake the retirees made was that they based their spending decisions on the current “mark to market” price of their assets. They should have based their spending decisions on the dividend yield of the stocks. Then they would be immune to the illusions of the bubble effect.

    ( BTW, the second part of your response is all correct, but it does not address the argument I am making. There are some people who say that because there was a bubble in paper wealth, that our prosperity was “phony”. I am not one of those people. The houses built were real houses, the cars real cars, etc. The collapse in paper wealth is all nominal. It only affects the real economy via sticky wages, sticky debt, and the non-neutrality of deflation. )

  34. Gravatar of JKH JKH
    24. April 2009 at 03:30


    “The banks went from 8 to 800 billion voluntarily. The Fed didn’t force them to.”

    This is absolutely staggering. It is no wonder that I’m having such a difficult time communicating.

    The system went to $ 800 billion in excess reserves because the Fed created them as a result of its own asset and balance sheet expansion. The Fed forced the reserves into the system. The banks had nothing to do with it. This misunderstanding is also why your demand oriented framework for the analysis of excess reserves is on quite shaky ground.

    I do know I’ve seen this type of fundamental error made by some other economists. (One of the Mankiw links does it as well.) Others get it right. That makes it all the more frustrating. But it’s really unfortunate that some economist colleague somewhere hasn’t pointed this out to you, as it is a misconception that seems to run through your work. I’d be a bit surprised if Krugman doesn’t understand this. The actual dynamics probably support his position on fiscal easing. In fact, I’m surprised he hasn’t highlighted it yet. But somebody somewhere that you would listen to with some respect must be able to tell you that the banking system didn’t accumulate this excess reserve position voluntarily. (Although I guess you’d listen to Krugman less and less now.)

    This is why I’ve been focusing so much on the way the system works now, much to Bill Woolsey’s constant irritation. The error in terms of this particular example is beyond fundamental. The only good news is that I now have an iconic example of the error. Perhaps I’ll attempt to write up a broader summary, although I feel that I’m knocking my head against a wall by now. (BTW you seem to have interpreted the facts of my example above backwards from what I intended. I thought I stated the assumptions and the sequence of events and decisions fairly carefully.)

    BTW, you’re making the same sort of error on potential currency withdrawal that you made with potential deposit expansion – the scenarios you pose fail the reality test in terms of order of magnitude and plausible proportion. The current excess reserve position would require somewhere between a doubling and a more than 10 fold increase in the size of the banking system, in order to “convert” excess reserves to required reserves, depending on which ratio you use at the margin. Similarly, the conversion of the current excess reserves to currency would require a doubling of currency. These are not realistic assumptions, although the currency conversion is marginally more conceivable in the event of a run on the private sector banking system. I get the impression you don’t follow order of magnitude in the actual flow of funds all that closely; if you did, you might put your theory to the test in a different way.

    Maybe you should consider a survey among economists – did the Fed or the banks create the $ 800 billion excess reserve position that developed in the banking system from September 2008 to December 2008? That’s a fairly straightforward question. A lot may get it wrong, but a few who understand the banking system will get it right.

  35. Gravatar of ssumner ssumner
    24. April 2009 at 03:53

    JKH, I’ve been a monetary economist for 30 years, I’ve used many money textbooks, and I have never seen any other monetary economist who looks at monetary policy the way you do. (Of course lots of non-economists do.) If the Fed had not been paying interest on reserves, a massive base injection might well have caused hyperinflation. I’ll say it again. Banks do not have to hold excess reserves. Please find me an example of a money textbook that says ER is not a choice variable by commercial banks. I have never seen one.

    And your final question is not clear at all. Of course the Fed created the $800 that banks are currently using as ERs and of course the Fed encouraged them to hold those ERs, and of course banks might hold those funds as ERs even without the Fed’s encouragement. The question is whether banks are forced to. There is no law forcing them to hold ERs that they don’t want to hold, just as there is no law forcing you to carry cash in your wallet that you don’t want to hold.

    All of monetary economics revolves around a simple illusion called the fallacy of composition; what’s true for the individual is not true for the group. That’s the fundamental principle that underlies all of monetary theory. And you are telling me that it is wrong.

    You might look at Mishkin’s textbook, which is the most popular on the market. Mishkin was on the Board of Governors and is a brilliant economist, so I think he knows how the Fed conducts OMOs.

  36. Gravatar of JKH JKH
    24. April 2009 at 04:17

    “All of monetary economics revolves around a simple illusion called the fallacy of composition; what’s true for the individual is not true for the group. That’s the fundamental principle that underlies all of monetary theory. And you are telling me that it is wrong.”

    This is extraordinarily difficult. That’s my point. What an individual bank can do or attempts to do with excess reserves is not what the system can do, given the fact that the Fed always has the final say on the amount of excess reserves in the system. Given that the Fed is the final arbiter on the level of excess reserves, absolutely the banking system is forced to hold them.

    Let me try a more macro example:

    Suppose the Fed tries your penalty system for a while.

    Suppose required reserves go up by $ 200 billion, say as a result of a $ 4 trillion increase in banking system deposits (I’ve used a midpoint 5 per cent effective reserve ratio at the margin, just for fun.).

    Suppose at the same time the Fed has done an additional $ 1 trillion of its own balance sheet expansion via additional credit easing.

    At the margin, the Fed funds the $ 1 trillion asset expansion with a $ 1 trillion in excess reserves.

    The net increase in excess reserves is $ 800 billion.

    The banking system has certainly caused a gross effect on excess reserves. I’ve never denied this sort of effect. This gross effect is non-discretionary from the Fed’s perspective.

    But the Fed has the final say on the supply of excess reserves, in this case as a result of its discretionary asset expansion and decision to let further excess reserve expansion fund those assets.

  37. Gravatar of JKH JKH
    24. April 2009 at 04:22

    The textbooks are a problem.

    They’ve always tended to explain the multiplier as a dynamic process, which is wrong.

    The multiplier is a static equilibrium constraint. It’s comparable to the kind of identify whereby the current account deficit must equal capital inflows. It’s a constraint on economic modeling, but it doesn’t explain the process.

  38. Gravatar of JKH JKH
    24. April 2009 at 05:02


    “I have never seen any other monetary economist who looks at monetary policy the way you do.”

    Here it is from the Fed’s own mouth, explaining the irrelevance of the multiplier in normal times:

    “In practice, the connection between reserve requirements and money creation is not nearly as strong as the exercise above would suggest. Reserve requirements apply only to transaction accounts, which are components of M1, a narrowly defined measure of money. Deposits that are components of M2 and M3 (but not M1), such as savings accounts and time deposits, have no reserve requirements and therefore can expand without regard to reserve levels. Furthermore, the Federal Reserve operates in a way that permits banks to acquire the reserves they need to meet their requirements from the money market, so long as they are willing to pay the prevailing price (the federal funds rate) for borrowed reserves. Consequently, reserve requirements currently play a relatively limited role in money creation in the United States.”

    This is consistent with what I’ve said in general. It applies as described up until the extraordinary excess reserve easing that started in September of 2008. The page link has been removed from the Fed’s web site, replaced with a notice that the Fed is updating the section on reserves. This is understandable, given the complexity of the easing transition, and new environment. But when it is finally updated, I fully expect it will support my general interpretation. At that time, I shall return to remind you.

  39. Gravatar of JKH JKH
    24. April 2009 at 05:03

    And to confirm my model of daily excess reserve management:

    “Reserve forecasters at the New York Fed and at the Board of Governors in Washington, D.C., compile data on bank reserves for the previous day and make projections of factors that could affect reserves for future days. The staff also receives information from the Treasury about its balance at the Federal Reserve and assists the Treasury in managing this balance and Treasury accounts at commercial banks.

    Following the discussion with the Treasury, forecasts of reserves are completed. Then, after reviewing all of the information gathered from the various sources, Desk staff develop a plan of action for the day.

    That plan is reviewed with interested parties around the system during a conference call held each morning. Conditions in financial markets, including domestic securities and money markets and foreign exchange markets also are reviewed at this time.

    When the conference call is complete, the Desk conducts any agreed-upon open market operations. The Desk initiates this process by announcing the OMO through an electronic auction system called FedTrade, inviting dealers to submit bids or offers as appropriate.”

  40. Gravatar of Alex Alex
    24. April 2009 at 05:32


    I think that JKH view is that a monopoly controls the quantity (and the market decides the price) while you argue that it controls the price (and the market decides the quantity) you can argue all you want about which one is correct but in the end who cares they are equivalent. You think that the Fed controls the supply of Base Money and the markets determines the composition (given other variables like reserve requirements, interest/penalties on reserves and other market conditions) while JKH argues that the Fed chooses the amount of Reserves and the markets chooses the amount of Base Money (given all the other variables). The problem is that I think JKH is mixing other concepts. Accounting identities and double entry book keeping, static or dynamic models, are not related to the issue.


  41. Gravatar of Bill Woolsey Bill Woolsey
    24. April 2009 at 06:39

    Suppose the banks create $8000 trillion in checkable deposits, and so, the current $800b are now required. But, now, the Fed makes another $800b open market purchase all from banks. Gee, excess reserves are again $800b. Therefore banks don’t choose excess reserves?

    What is the period of analysis?

    Implicit in this “argument” is that there are two periods. In period one, banks adjust their checkable deposits and so use up excess reseves. In period two, the Fed takes that as given and then creates whatever excess reserves it wants. In period 2, banks aren’t allowed to make any changes in checkable deposits.

    And now, we ignore period one and say that the Fed controls excess reserves because in period 2, the banks can’t do anything about their excess reserves by assumption.

    Suppose, instead, we use “month” as period. The banks can purchase assets as fast as the Fed can.

    Anyway, Scott is absolutely right that each bank and the banking system (not including the Fed) choose to hold excess reserves. Scott undestands exactly how it works. The amount of checkable deposts change increasing the excess reserves for the system as the whole.

    No one denies that the Fed can create more or less base money at will. And, that it is possible that the Fed could create base money faster than banks purchase assets, expand checkable deposits, and be subjected to greater reserve requirements and so have less excess reserves.

    As far as I can tell, your quotations from Fed publications were nothing new to me, and have no impact of the ability of penalty rates on excess reseserves to cause a larger expansion of checkable deposits. In no way does it contradtion the process that Scott described.


    You know, we could explain exactly what the Fed was doing in the Great Depression and how all of it made perfect sense given the tissue of fallacies behind their real bills thinking. So?

  42. Gravatar of Bill Woolsey Bill Woolsey
    24. April 2009 at 06:56

    JKH, do you have a copy of your vita online?

    Do you have any papers on banking theory?

    Is your training in economics? finance? accounting?

    Just curious.

    To try one more time, the Fed, of course, creating the excess reserves initially. The banks chose to hold them rather than purchase other assets, which would have expanded checkable deposits to the point where they would have been required. This is all so elementary.

  43. Gravatar of JKH JKH
    24. April 2009 at 07:45

    Bill Woolsey,

    Your latest explanation is getting closer to something I can agree with at least in terms of the operation of the current system. (I said “closer”.) That would be progress at least in terms of me not being forced to come back and keep correctly blatantly false logic otherwise. A proposal for change is typically a proposal to change the existing state of affairs, and to the degree that the existing state is badly explained or poorly understood, it must be corrected, with the result that the proposal to which it serves as an input loses credibility. So that’s a step forward in terms of moving me forward from having to correct evident errors in the explanation of the existing state.

    As to the proposal itself, I see a number of fundamental problems. It is not enough to propose something where the sponsor claims he simply doesn’t care what assets the banks buy, and then expects the proposal to have credibility in the real world. And it is not enough to ignore more or less the potential capital considerations. And it doesn’t add credibility to be oblivious to the issue of realistic proportionate adjustment and orders of magnitude. The numbers we are talking about in terms of using up existing excess reserves for deposit expansion are absolutely colossal. The actual effective reserve ratio for the banking system would suggest they require at least a 10 times expansion of the size of the entire banking system. Who can treat an idea like this seriously? And a somehow engineered result that directs all of those deposits into checkable form would still require a doubling of banking system size. And where are the assets going to come to generate a deposit expansion that is blithely imagined at somewhere in the range of $ 10 trillion to $ 100 trillion? When treasury bills are limited to $ 2 trillion? Where are they going to come from in any event, before even considering their risk effect and capital consequences? And what are the capital consequences going to be, unless banks purchase nothing but treasury debt? (which I referred to as the monetization of treasury debt by the commercial banking system, not by the central bank, as Scott mistook it to be earlier). And even then, remember that regulators require banks to put up capital against interest rate risk. What sort of interest rate risk position and capital requirement would be created by putting on just $ 5 trillion in treasury bonds against short term deposits? Do you really think that banks don’t think about details like this? Do you really think the Fed doesn’t think about this, in terms of exactly what it DOESN’T expect banks to do? These are all issues of real world constraint and proportion, while you are talking about a reverse engineered deposit expansion simply based on a bunch of excess reserves that the Fed has forced fed the system in order to fund its credit easing. The Fed did not put these reserves into the system to “unleash” the mighty work of “the multiplier”. It put them in as a source of financing for its own balance sheet. And that’s why it’s paying interest on reserves – because it has no intention of allowing or suggesting the feasibility of such a reckless notion as multiplier expansion working overtime on an excess reserve position that it has suddenly expanded by 40,000 per cent for the first time in 100 years, for entirely different reasons.

  44. Gravatar of JKH JKH
    24. April 2009 at 08:00

    Bill Woolsey,

    I have a degree in pure mathematics, MBA in finance, and a CFA. I have (had) 30 years experience working in the Treasury Division of a major Canadian Bank. I managed its Bank of Canada account in the Volcker era, was responsible for global money market operations after that, and for liquidity and interest rate positioning for the entire bank after that. I wrote a paper in 1978 suggesting the possibility that the Bank of Canada might monetize the value of Baffin Island on its balance sheet, at such point when it ran out of other ideas. I’m a generally more agreeable human being than might appear otherwise from my writing style on this topic.


  45. Gravatar of ssumner ssumner
    24. April 2009 at 16:38

    JKH, Imagine an economy with two people, Jack and Jill. The Fed injects money in the economy. Do they also determine how much of the money is held by Jack, and how much by Jill? Obviously not. Now let’s call Jack the banking system, and Jill is the public. Sure the Fed can somewhat influence the distribution of base money through reserve requirements and interest on reserves. But once those are set, the banks and public determine how much will be in the banking system and how much will be held by the public. As a practical matter most reserves will probably be held by the public unless you are in a liquidity trap, and have no penalty rate on excess reserves. So if they add a lot of excess reserves, it will start in the ER category, but quickly flow to where it is most wanted. Imagine a small lake right next to Lake Huron, connected by an open channel of water, with no rapids. It has 1% of the surface area of Lake Huron. The Canadian government pours 1 million gallons of water into the small lake. Ten seconds later all of that water is in the small lake. In the short run the Canadian government controls the amount of water in that small lake. But one day later only 1% of the water that was added is still in the small lake. 99% is in lake Huron. That’s how ERs worked before the liquidity trap.

    JKH#2, I also don’t care for the multiplier approach

    JKH#3, I totally agree that the fed usually supplies reserves passively to target the Fed funds rate, and I agree the required reserves ratio is usually not used as a tool of monetary policy.

    Regarding the question of where all these assets will come from, consider Zimbabwe. Their banking system increased trillions-fold. I don’t recall them having much trouble finding enough assets for the banks to hold. Germany during 1923 is another example. It is easy for the Fed to massively expand bank balance sheets if they want to, of course I hope everyone knows I think this would be a terrible idea. I onlly do these thought experiments to show how absurd the “liquidity trap” excuse for Fed inaction is.” If they have trouble boosting AD, just let me at the controls for a while. (Yes, I’m just kidding here.)

    Alex, I am taking the quantity of money approach in this example.

    JKH, Don’t worry about being disagreeable. I think everyone sounds more frustrated communicating by email than in real life. That is also true of me. I often sound exasperated on the internet, whereas I am mild-mannered in person. It’s the nature of communication where you can’t see the other person.

  46. Gravatar of Nick Rowe Nick Rowe
    25. April 2009 at 02:30

    Totally off-topic (sorry): This blog could benefit from having a “Recent Comments” section on the right hand side. So we can see which posts have new comments. (I don’t know if it’s easy to set up; if not, don’t sweat it. Content is what really matters, and it’s good.)

  47. Gravatar of ssumner ssumner
    25. April 2009 at 05:38

    Thanks Nick, I will look into it. BTW, that would also be a big help to me. At least once a week I go through all 100 or so posts looking for new comments.

  48. Gravatar of Dilip Dilip
    25. April 2009 at 08:16

    If you have someone who knows their way around WordPress (or if you can it yourself), try installing this plugin:

    It should start displaying recent comments on the sidebar.

  49. Gravatar of Dilip Dilip
    25. April 2009 at 08:18

    Or you could try this one too:

    if you are not interested in displaying avatars — which I don’t is supported by your current commenting system anyway.

  50. Gravatar of ssumner ssumner
    25. April 2009 at 12:29

    Dilip, I’m probably showing incredible ignorance, but I assume avatars are those little pictures attached to each commenter? I’ll check this out at school next week. Those links will be a big help.

  51. Gravatar of Dilip Dilip
    25. April 2009 at 20:33

    Yeah — thats what avatars are. I think you should try the second link I sent and then try the first if that doesn’t work out.

  52. Gravatar of Random Links XXXXIV « Random Musings of a Deranged Mind Random Links XXXXIV « Random Musings of a Deranged Mind
    18. May 2009 at 17:09

    […] Scott Sumner: […]

  53. Gravatar of Understanding the Excess Reserve Tax Understanding the Excess Reserve Tax
    14. July 2009 at 19:31

    […] original proposal, as he notes in the comments, can be found here. The basics of the proposal are that the Fed would set a modestly negative rate to be paid on bank […]

  54. Gravatar of TheMoneyIllusion » Zero Rate Bound, RIP TheMoneyIllusion » Zero Rate Bound, RIP
    26. January 2010 at 12:04

    […] public holding currency.  I won’t get into all the minutia of this issue; I discussed them in another post.  The bottom line is that negative rates on reserves don’t completely eliminate the need for […]

  55. Gravatar of TheMoneyIllusion » Guess who discussed negative interest rates on money in 1998? TheMoneyIllusion » Guess who discussed negative interest rates on money in 1998?
    28. January 2010 at 11:27

    […] A few days ago I did what I thought was one of my more important posts, pointing out that the Fed’s likely new operating target will eliminate the problem of the zero rate bound.  There are always some misunderstandings in the comment section when I bring up negative rates on reserves.  What about vault cash?  What about the impact on bank profits?  What if firms don’t want to borrow even at very low rates?  The simple answers are vault cash can be dealt with in many ways–it isn’t a problem.  And the system can be set up in a way that bank profits aren’t hurt.  And the purpose of the plan is not to get firms to borrow more, but to reduce bank demand for the medium of account.  This is all discussed in earlier posts like this one. […]

Leave a Reply