So much for yet another legacy.

From Richard Koo’s latest report:

But nightmare scenario awaits when private loan demand recovers. The problem is what happens when private loan demand recovers. Loan books could grow more than tenfold in the US and five fold in Japan and Europe if bank reserves remain at current levels, triggering inflation rates of 500% to over 1,000%.

To avoid this outcome, central banks will have to mop up excessive reserves by raising the statutory reserve ratio, raising the interest rate paid on reserves, and selling government bonds. All of these measures will serve to lift interest rates, sending bond yields sharply higher and triggering a possible crash in the bond markets.

A sharp increase in government bond yields could lead to fiscal collapse in countries with a large national debt. For Japan, where the national debt amounts to 240% of GDP, the results would be catastrophic.

Expanding quantitative easing because it appears to be doing no harm is grievous error. Mr. Abe and his advisors may believe that all they have to do once their anti-deflationary policies succeed and JGB yields start to rise is have the BOJ buy more bonds. However, bank reserves under quantitative easing have risen to a level capable of fueling a 500% inflation rate, in which case the BOJ would have to sell, not buy, JGBs.

Nomura | JPN

BOJ purchases of JGBs in that situation could cause the potential inflation rate to rise from 500% to 600% to 700% and trigger an economic collapse.

I do not know whether the German finance official who was opposed to reckless quantitative easing based his view on this kind of scenario. Nevertheless, it is extremely dangerous to assume that since quantitative easing does no harm in a balance sheet recession, it can be continuously expanded without concern. The real danger posed by this policy will become apparent only after private-sector balance sheets are repaired, and then it will happen suddenly.

BOJs excess reserves could become a time bomb. I would now like to bring some actual numbers into the discussion so that readers may appreciate the implications of this scenario.

Only 7.7 trillion in bank reserves are required to maintain Japans money supply. With the Japanese government now running annual fiscal deficits in excess of 40 trillion, BOJ financing of the entire deficit would require the Bank to supply reserves equal to more than five times the amount needed to maintain the money supply. Over a two-year period, it would have to supply reserves equal to more than ten times the required amount.

In other words, the purchase of one years worth of newly issued government debt by the BOJ has the potential to generate a 500% inflation rate. I suspect few Japanese are willing to accept such a trade-off.

Moreover, the BOJ has already engaged in substantial quantitative easing under heavy pressure from politicians, pushing excess reserves to 29.8 trillion. In my view this represents a time bomb.

155 Responses

    1. @AP Lerner, I love how the interviewer goes over Mosler’s points and almost all of his responses are “Well, yeah what he’s saying is true… BUT”.

      True is true… there is not “but” to the truth.

  1. Being a novice at MMT, are we saying Koo gets it wrong because banks don’t lend reserves? I’ve read that banks are constrained by capital (or should it be called equity?) not reserves. Do I have that right?

  2. Can someone maybe give a brief translation for the uninitiated among us? What’s wrong about this, the assumption that QE is going to cause inflation?

    1. @jerry,

      Banks don’t need reserves in order to lend. Bank create reserves (deposit) when loan is created. If/when deposit leaves that particular bank (or any other reason bank may be low on reserves) then that bank can always obtain necessary reserves from other banks holding excess reserves. If entire system low on required reserves then the cost of obtaining those reserves (interest rate) will increase, but Fed always accomodates need for banking system reserves (it must, or system would collapse). Ultimately, banks not reserved constrained in their lending, reserves merely influence % rate of loans, not quantity. Koo’s concern (should) exists regardless of amount of reserves in system.

      1. @Don,

        When a loan (Asset) is created the commercial bank creates a corresponding deposit (liability).

        But it does not create reserves (an asset for the bank).

        Reserves are deposits at the central bank, that can’t be created when the commercial bank grants a loan.

      2. actually in the first instance loans do create deposits and any corresponding required reserves.
        that’s because a required reserve, when there is a 0 balance in the reserve account, is, functionally, an overdraft in the reserve account.
        and at the end of the statement week the fed actually books any said overdrafts as loans from the fed

    2. When the Fed buys these assets they are doing so in the realm of reserves. They buy a bond from a bank and do so by marking up (with computer keystrokes) he banks excess reserve account. This increases total banking system excess reserves which are part of the Fed’s balance sheet and monetary base. But private sector banks cannot lend excess reserves to anyone but another bank who may need more required reserves which does not change the banking systems total reserves. Excess reserves have nothing to do with commercial loans although the mainstream incorrectly believes banks loan excess reserves. unlike what most economic textbooks say, there is no correlation between the monetary base and the spendable money supply. The M2 money supply is increased when banks make loans which make deposits (or if the Fed makes a short term loan). Therefore, the Fed’s asset buying will not cause inflation because they are not adding financial assets to the private sector. By the way, before October 2008, the Fed could not do what it is doing because all the excess reserves would drive the Fed Funds rate to zero, not allowing the Fed to control is target rate. Since 2008, they started paying interest on excess reserves allowing the fed to easily control a positive interest target rate. The other commenters are shocked that Dr. Koo doesn’t seem to understand that the central banks are not adding money to the money supply when they do QE.

      1. @John Wilkins, Ah okay yes that all makes sense. The language of this type of writing is always cloaked in the obfuscation of high finance, not sure why they can’t just speak plain english.

        This seems to be the same sort of thing espoused by the Peter Schiff/ Ron Paul acolytes over at Zero Hedge, endless doomsday warnings over non-existent inflation that is “just around the corner”..

      2. @jerry, I think if you read some of this stuff repeatedly, you come to realize that the people defending those views have really crude mental models of how the world works. It’s just:

        a. Fed creates lots of money.
        b. There is then too much money.
        c. Too much money causes higher inflation.
        d. So there is then higher inflation.

        There are no institutions or operational details in the model at all – just this mysterious thing “the Fed” and this other mysterious think “money”.

        It’s like saying that the umpire has rubbed up too many baseballs before the game, and that’s going to cause an increase in passed balls and wild pitches.

  3. to some degree you’re on to something, adam.

    i don’t believe for one moment koo is clueless. he has an agenda. you just can’t take what these commentators say at face-value.

    i’ve been following this little spat at the diaoyu/tiaoyu/senkaku islands and taiwan sent ships over there to “attack” japan BEFORE the chinese did.

    and if you notice, japan is being attacked from 3 sides–china, taiwan and KOREA!! korea is grumbling about some small, insignificant islands west of japan.

    but i think this apparent militarism is really about japanese monetary/fiscal policy and so, i think that adam has the “causality” reversed. if china/taiwan and korea really cared about the islands, they woulda done something a lot more serious then firing water cannons at japanese ships. it’s not about the islands or theoretical oil reserves or some fantasy about nascent japanese militarism.

  4. The disaster portrayed by Koo doesn’t have to be because raising interest rates is not the only way of damping down excess demand. Assuming reasonable cooperation between government and central bank, much of the damping can be done with restrictive fiscal policy.

    But that’s not to argue that the gyrations involved in a big QE operation followed by QE reversal is a clever way of regulating an economy.

  5. Banks got bailed out with a few trillion of printed fiat currency so now all that is left is to raise taxes and drive inflation so the rest of us can pay for it.

    1. @Alicia,

      Raising taxes is unlikely to drive inflation. If higher taxes coincide with a supply shock, it might look that way, as it did in the 1970s. But there’s no sound economic model I know of that supports higher taxes causing inflation.

      1. please read more carefully 😉
        it’s about offering a $10/hr transition job to anyone willing and able to work to facilitate the transition to private sector employment

        i never used the expression ‘making people work’

  6. Japan is going to Blow!

    Centrally planned economies always end in the same way – crash and burn.

    The USA is the next largest centrally planned economy and will also crash and burn – I’d say 5-20 years before the empire crumbles.

    The USA was a very short lived empire in the global sense ~100 years – the primary reason is a greed driven capitalist society founded on fascism (the joke being fascism was replaced by meaningless work “freedom”) – prone to boom, busts, turmoil, and constant wars to protect oil priced in USA currency units.

    After the use empire collapses, there will be void, I believe south america, russia, india, and china will drive the 21st century forward as long as the move directly to democratic socialism and not American style fascist capitalism

  7. The world is in desperate need of a Central Banker that understands how it all works. Will someone please resurrect Marriner Eccles ASAP!

    1. @Adam1, If we did have a central banker who understood how it all worked, his first speech would be entitled,

      “Grow Up, Do Some More Damn Fiscal, and Stop Looking at Central Bankers to Solve Problems They Can’t Solve”

      Seriously, the mainstream conventional wisdom on central bankers these days is a kind of wild-eyed superstitious primitivism: central banker as Merlin.

      1. “except that would be factually incorrect.”

        Expect that it wouldn’t. Particularly not in the rest of the world outside your tiny part of it.

  8. I followed the link from FT Alphaville.

    I agree that banks are not constrained from lending by a lack of reserves, but that is not the same as saying that they won’t lend an excess. In the event that the economy picked up, the banks would have a large stock of a liquid asset that they no longer need. From the point of view of an individual bank, it can spend these reserves on loan assets (aka lending reserves) or something else, like an acquisition or a shiny new head office, but the point is, it is all spending that will drive up prices. What would MMT disciples expect the banks to do with their excess reserves?

    Koo seems right to me.

    1. @RebelEconomist,

      Reserves are assets to a bank. You can’t spend them on anything you can only exchange them with other assets held by the few other entities allowed to hold Bank Reserves.

      “In the event that the economy picked up, the banks would have a large stock of a liquid asset that they no longer need. ”

      They never need them anyway, and they can’t get rid of them to anybody else because nobody wants them who is allowed to hold them. And they don’t constrain the bank from doing anything else.

      So why would they be bothered doing anything with them?

      1. @Neil Wilson, most potential bank counterparties have bank accounts, but even if they don’t, the bank can always exchange its reserves for currency at the central bank and spend that. I do appreciate that “spend” is a bit of a misleading word in that the reserves are not destroyed in the process, but unless the central bank takes back or immobilises the reserves in some way, they “hot potato” around until the rising general price level validates the existing stock of reserves – ie the inflation that Koo has in mind.

        Just as banks are not constrained from lending by a lack of reserves, they do not HAVE to get rid of an excess, but if reserves no long yield as much return or safety as other assets, a bank will WANT to exchange them for something better.

      2. @RebelEconomist,

        ” but unless the central bank takes back or immobilises the reserves in some way, they “hot potato” around until the rising general price level validates the existing stock of reserves – ie the inflation that Koo has in mind.”

        Then why aren’t they doing that now? There’s plenty of things to buy and the assets are cheap. So where are the Securicor wagons?

        It’s nonsense.

        Banks are not going to spend currency – because banks would have to receive that currency back to go into bank accounts – which forces them to save the currency again.

        Currency costs money – it’s expensive for banks to process and store. Why would banks bother when they can lend for a better return – particularly in an improving market.

        Moreover if a bank spends any cash it owns, then it has to reduce the liabilities *it owns* by the same amount. Not deposit accounts but its own capital buffer. The liabilities it receives back for whatever asset it as purchased are very likely to be of the wrong type to use as a capital buffer. How does it fulfil its regulatory capital requirements then?

        You’re not thinking the process through – piles of paper are precisely the same as piles of Bank reserves. They are both receipts for liabilities at the central bank and both pay essentially nothing. They offset liabilities on the right hand side of the balance sheet – which have to be disposed of at the same time.

        The spending you believe will happen isn’t happening and doesn’t happen. And if it did you’d tighten the capital requirements and stop it stone dead.

      3. @Neil Wilson, “There’s plenty of things to buy and the assets are cheap.” Really? Name some, and don’t forget their riskiness.

        “Banks are not going to spend currency – because banks would have to receive that currency back”. Collectively, yes; individually (on which basis they decide what to do), no.

        “Moreover if a bank spends any cash it owns, then it has to reduce the liabilities *it owns* by the same amount”. No it doesn’t; it acquires another asset in return for the cash.

        The only sensible point you make is that banks will need more capital if they expand their balance sheets. But like reserves, capital does not constrain bank lending. It is just a cost of business, and if lending is profitable, banks will have no problem raising more capital.

      4. @RebelEconomist,

        “Really? Name some, and don’t forget their riskiness.”

        In which case we’re at indifference level already.

        “No it doesn’t; it acquires another asset in return for the cash”

        It does. You spend liabilities, not assets. That’s how you buy things. you swap the debt you owe for a debt somebody else owes you and step out the middle.

        If you acquire another asset – say property – then the liabilities you acquire with that asset will not be the same class of liabilities as they were before. Therefore the bank will be short of capital.

        And of course there will likely be tax to pay on the purchase

        ” It is just a cost of business, and if lending is profitable, banks will have no problem raising more capital.”

        But lending can’t be profitable, because the bank is spending all its effort trying to buy other assets, rather than just creating them out of thin air.

        All you have described is a bank buying assets. Cash is irrelevant to the discussion.

        The limit of a bank purchasing assets is the spare liabilities it owns in its own right that are not employed leveraging loans.

        The level of central bank reserves owned is completely beside the point.

      5. @Neil Wilson,

        “In which case we’re at indifference level already.” We are going round in circles here. The point Koo is making is that we are at indifference levels now, hence QE has not caused inflation, but if the economy picks up, we won’t be, and then the banks will try to exchange their reserves for (then) more attractive assets.

        “You spend liabilities, not assets.” I appreciate that base money represents a central bank liability (arguably; I think people like Chris Cook have a point here), but it is an asset to its owner. You can argue that they hand over this asset in settlement of a debt that they owe for their new asset, but I don’t think that that adds much clarity, especially in a DVP transaction. Every financial asset is someone else’s liability.

        As for the rest of what you write, it is just confusing. Base money is basically a bank loan to the central bank. If an improving economy makes other assets, including loan assets, more attractive, then the banks can be expected to try to swap their base money for some other asset. “Cash” means various different things in finance, so better not to use the term I find, but cash surely includes base money, so cash is hardly “irrelevant”.

      6. banks always want the most attractive assets.

        and ‘the improving economy’ you reference means ‘increased aggregate demand’ / more sales precedes the supposed reserves effects in your defense of Koo

        so you/koo at best have your causation confused

        (and in any case the level of reserves isn’t causal in this regard)

      7. @RebelEconomist, Good lord, if the econmony picks up such that a real positive interest policy is needed, the Fed will drain reserves and the banks will get back those toxic assets the Fed purchased with those reserves. Said toxic assets now having more value as the economy has picked up.

        The reserves exist for two reasons: 1) liquidity for the payments system, 2) interest rate maintenance.

      8. @Neil Wilson,

        Neil: So why would they be bothered doing anything with them?

        Asset-liability management and interest rate risk management. It is very counter-productive to say that banks should not be bothered.

        However Koo is still wrong.

      9. “I appreciate that base money represents a central bank liability”

        But you don’t appear to appreciate that it represents a liability in the private banks own books – and a particular type of liability, regulatory capital.

        Run through it again.

        The private bank has Reserve Assets at the central bank, which it swaps for, say, Printed Cash. Now when the bank tries to buy that property from Joe Builder what happens is that the Printed Cash asset goes down *but so does the bank’s profit and loss account*.

        The property purchased goes to Fixed Assets and the liability goes to the property equity reserve.

        And the property equity reserve isn’t regulatory capital, whereas the profit and loss account was.

        So the bank has immediately reduced its ability to make loans.

        So it will then have to pay the price to switch some deposit liabilities to regulatory capital to recover its loan making capacity.

        Strip away the meaningless middle transaction, match the bond sale to the property and what you have left is a operation running a business that is selling equity bonds to buy property.

        So its running a sideline as a property investment trust. Big deal.

      10. @Sergei_new,

        “Asset-liability management and interest rate risk management. It is very counter-productive to say that banks should not be bothered.”

        There’s a limit to what you can do with Bank Reserves because few entities can hold the assets and they’ve probably got enough already.

      11. @Sergei_new,

        Neil, bank reserves are just one asset. It is very dangerous to draw conclusions about individual bank strategies from system-wide statements about just one asset type. The fact that banks in aggregate cannot get rid of reserves does not mean that individual banks cannot and will not try. Moreover, some banks might be even *forced* to try.

      12. @Sergei_new,

        We’re talking about Bank Reserves here and the effect of lots of them in the system – because that is the vertical transaction that apparently is going to bring about the end of the world.

        And try as they might unless some other entity wants the Bank Reserves an individual bank *can’t* get rid of a systemic excess of them.

      13. @Sergei_new,

        Neil, you do not listen to what I say and keep on preaching the same old story which I know pretty well. After you learned about MMT you still need to keep your mind open to new ideas and new questions.

        You are absolutely wrong to say that interest rate risk of individual banks does matter since systemic excess reserves override it. Because it does matter and there are regulatory limits for interest rate risk even in the USA. And individual banks have to comply on their individual level. No arguments about excessive *systemic* reserves will bring you forward in your understanding of the banking system and its dynamics. Something you claim you mastered fully.

      14. @Sergei_new,

        Neil 1: “I don’t say that – so why are you saying I am?”

        Neil 2: “There’s a limit to what you can do with Bank Reserves because few entities can hold the assets and they’ve probably got enough already.”

        There is no limit if you are forced to do it. Systemic considerations are irrelevant. It is also irrelevant whether any other entity wants to hold additional reserves. Because the problem shifts away from reserves into another dimension. You simply might get into a very unstable situation with all the excess reserves you can have in the system. Because no amount of excess reserves will save banks from losses on their interest rate risk positions.

      15. @Sergei_new,

        ” Because the problem shifts away from reserves into another dimension.”

        Off we go into another tangent completely at odds with the matter under discussion.

        So bank reserves don’t cause anything. Thank you for confirming that.

      16. Warren,

        “some properties do qualify as assets for capital purposes, such as the bank’s executive offices, branches, etc.”

        Yes, but there aren’t enough of those to ‘soak up’ all those excess reserves that are supposedly about to bring about the end of credit as we know it (after being converted into cash of course).

        Most properties, or other assets, don’t qualify – which means that a bank is going to be limited in its options.

        In other words the capital restrictions do actually slow things down. Banks have to raise extra capital to continue bidding up prices.

      17. nor do said properties ‘soak up reserves’ 😉

        and US banks can’t buy non qualifying properties or any non qualifying assets in any case.

        capital restrictions do slow things down temporarily as it takes a bit of time for banks to raise more capital

      18. “nor do said properties ‘soak up reserves’ ;)”

        Absolutely. Better make that crystal clear before words are twisted again…

        “and US banks can’t buy non qualifying properties or any non qualifying assets in any case.”

        That’s not global though is it? Deregulation seems to allow banks to do all sorts of things they shouldn’t really be doing.

        ISTM that the confusion here is because people aren’t taking into account that those that would have held government bonds now have to hold private bank liabilities – which are then matched by the central bank reserve asset held by the private bank.

        So the private bank is playing middleman, which means that their balance sheet is bigger.

      19. @Sergei_new,

        Neil: Off we go into another tangent completely at odds with the matter under discussion.

        At odds with 500%? Yes. Tangent? No. Time bomb? Who knows what can explode. You do not need to defend your “reserves-loan growth-inflation channel”. Only Koo is attacking it here. But there are plenty of other by-side effects with potentially very tricky consequences.

      20. @Sergei_new,

        Neil: So the private bank is playing middleman, which means that their balance sheet is bigger.

        As middlemen banks are heavily regulated. Bigger balance sheet generally correlates with higher risk. Risk aversion can trigger some fancy stuff once it reaches some boiling temperature. Even if your assets are CB reserves.

      21. @Sergei_new,

        Warren: hence the ‘interest rate determination process’ as banks trade fed funds with each other

        That is just one channel, interest rate channel, which is valid only up to one year. There are also other channels and also with longer maturities.

      22. @Sergei_new,

        Warren: what other ‘by-side effects with potentially very tricky consequences’

        Interest rate risk in banks is a decently complex and tricky topic. Each bank might have its own approach and solution driven by its own balance sheet structure. It is one thing to set regulatory limits. It is completely another thing to intervene directly into such matters. You cannot deny that such matters exist. And also that banks cannot hoard in their strategies since many of them face similar problems from similar balance sheet structures. So instead of having a diversified system you might end up with a very homogeneous one only waiting to explode. And it all got started with apparently harmless risk-free assets.

      23. yes, there is interest rate risk, but that’s easier to manage if all you have is reserve balances at the BOJ.

        the tricky part is hedging long term fixed rate assets.

      24. @Sergei_new,

        Warren: for fed funds it’s only relevant overnight/within the two week statement period etc. what are you referring to?

        Reserves are only good for cash interbank market. Kind of short-term and therefore pointless approach. But to try to get rid of reserves banks can also do things other than trading with each other. Which might push price indifference levels on longer assets.

      25. @Sergei_new,

        Warren: yes, there is interest rate risk, but that’s easier to manage if all you have is reserve balances

        Easier? Manage? What interest rate risk do you manage with reserves? Hm.

      26. @Sergei_new,

        Warren: what else can banks do to ‘get rid of reserves’? and that they can’t do anyway?

        All the way down here I keep on saying that it is not about can or cannot. It is about have to and chose to. And yet noone here wants to listen. Why is that?

      27. @Sergei_new,

        Warren: Didn’t you say interest rate risk is a problem with lots of reserve balances?

        Yes I did but the risk is NOT in reserve balances since they carry no interest rate risk at all. What they do is change the amount of interest rate risk on the balance sheet because all liabilities do carry interest rate risk.

      28. @Sergei_new,

        Unless you explain exactly what you are frightened of – with balance sheet transactions to show what the problem is then there is nothing to debate.

        All we see here is somebody shouting that we should be frightened of the dark, but never explaining precisely why.

        There are three changes with a reserves approach

        (i) non-banks hold their financial savings as savings bonds with banks, not the government sector – essentially privatising that function.
        (ii) banks therefore operate as a bigger middleman in the saving function than they do now.
        (iii) the banks balance sheets are bigger.

        Quite how that causes a problem and people moving from having cash under the mattress to savings in bank accounts doesn’t is clearly a mystery – since the latter is similarly people moving their ‘government bonds’ to savings accounts at banks.

      29. @Sergei_new,

        Neil: Unless you explain exactly what you are frightened of

        I am not in a position or interest to lecture you on basics of ALM and IRR in banks. I am surprised that Warren does not get it but maybe he is not familiar with ALM and IRR either. Saying that balance sheet transactions can resolve IRR problem is a very strange approach. You can of course keep your eyes shut. But please do not forget that the IRR management in banks was basically invented after the S&L crisis. Your comment says that shouting about it was a waste of time until it exploded the whole banking system.

      30. @Sergei_new,

        Warren: huh? with reserves 0 duration and paying interest at the fed’s target rate where’s the interest rate risk?

        Sorry, this is not how banks balance sheets work. Not even close. Your view is a bond trader view who managed to squeeze exceptionally favorable funding conditions from the treasury/ALM.

      31. my view is as owner and director of my bank.

        to your question, assume all my assets happened to be in balances in my fed account, which
        pays interest at the fed’s target rate.
        where’s my ‘interest rate risk?’

      32. @Sergei_new,

        “Sorry, this is not how banks balance sheets work.”

        Then explain, in simple terms, how they do work and why we have a problem.

        You can hardly complain that nobody listens and then refuse to explain when invited.

        If you have some great insight share it with us. If not then you are the victim of some fear monger who hasn’t explained it to you properly.

      33. @Sergei_new,

        Neil, banks have a regulatory limit on interest rate risk in banking book of 20% of equity. The IRR is calculated as a standard interest rate shock on the balance sheet. The delta between the base case and the shocked scenario shall not exceed those 20%. This is ALM 101.

        The valuation of the balance sheet is a very tricky part and full of assumptions. But when this IRR limit is exceeded the regulator calls management in and asks to explain why and expects actions to reduce the risk.

        But what you want from me is impossible. I.e. it is impossible to tell whether there is a problem or not. It is impossible to tell what positions and assumptions individual banks have, what positions they would like to have and how in aggregate it will play on the total banking system level.

        But what I am trying to say is that excess reserves can force banks into a corner. Because excess reserves have a very clear interest rate profile which is forcefully injected into banks. But if enough banks with sufficiently large balance sheets hoard into the same strategy and collectively try to get rid of reserves which they can not which only re-enforces their desire it is very possible that banking system gets into a very unstable situation.

        For the greater insight you need to be a regulator and do lots of number crunching on continuous basis.

      34. @Sergei_new,

        It is the regulators that are injecting the reserves! To suggest that they will sit there and not take that into account in regulation is a little absurd.

        So it would not be excess reserves that would be forcing banks into a corner. It would be bad regulation.

        Reserves are functionally loans to the central bank. You don’t get anything with less risk than that.

      1. @WARREN MOSLER,

        “excess reserves are held as assets and they size in aggregate is determined by the fed, not the banks”

        Not really. Participation in OMOs is largely voluntary, so the central bank can only determine the aggregate stock of reserves if it is willing to offer terms that will be accepted by its OMO counterparties. And that is the problem. To unwind QE, the central bank may have to offer terms that will cost it and its government shareholder dearly, perhaps to the point that they may prefer inflation to paying that cost.

        But no doubt you will respond with some other argument that I reject, so, in the absence of any support here, I will give up trying to persuade you on this occasion. It has been many months since I last looked at this blog, and it is noticeable how little the MMT narrative has moved on. Even if you still think you are right, I would have at least expected some change of arguments to cover the type of doubts that are commonly raised. If you have not done so already, it might help to have a meeting with some of MMT’s more informed critics, especially those with practical experience, to try to resolve differences of at least descriptions of operational procedures. As it is, MMT does not seem to be moving on from being a belief-based, self-referential sect.

      2. yes really.

        the cb also sets the cost of funds for the banking system, and the overnight repo rate is a function of that

        and what does ‘cost govt. shareholders dearly’ mean?

        The point of a cb selling secs would be to support long rates at higher levels.
        if it doesn’t want to do that it shouldn’t be selling them.

        for the cb, it’s always about price, not quantity.

        a+b=b+a is the theory of addition, and it hasn’t move on either.

        sticks and stones…


      3. @Adam1,

        That does not make much difference; it is analgous to exchanging non-interest bearing reserves for a (perpetual) floating rate note, compared with exchanging reserves for the fixed interest assets that the central bank bought under QE. It still generates (opportunity) cost for the central bank and its government shareholder.

        Actually, this is the kind of dialogue I had in mind. I am sure that MMT must have encountered this issue before, and by now either ought to have accepted the point that I make or have come up with an argument that moves the debate forward.

      4. @RebelEconomist, “In the United States, a primary dealer is a bank or securities broker-dealer that is permitted to trade directly with the Federal Reserve System (“the Fed”).[2] Such firms are REQUIRED to make bids or offers when the Fed conducts open market operations, provide information to the Fed’s open market trading desk, and to participate actively in U.S. Treasury securities auctions.” – Wikipedia via FRBNY website

      5. @WARREN MOSLER,

        “The point of a cb selling secs would be to support long rates at higher levels.” No it would not. The point would be to mop up the excess of base money. If a central bank could sell the securities it bought under QE for the price it paid, I am sure it would be delighted.

        “what does ‘cost govt. shareholders dearly’ mean?”
        Let’s say that the Fed have bought a trillion dollars of bonds more than they need to supply the quantity of money required in the post-crisis economy, and that the DV01 of the Fed’s portfolio is 0.05. That means a loss of $50bn to the US taxpayer for every 1% increase in the yield of the bond portfolio from purchase to sale. Not crippling perhaps alongside other government expenditures, but not small change either.

        “for the cb, it’s always about price, not quantity.” That is because the nearest substitute for base money is short-term debt, so central banks use the price of that debt to gauge the adequacy of the money supply. The relationship between the stock of base money and other macroeconomic variables like inflation is implicit in the macroeconomic models central banks use to set their operating interest rate.

      6. @Jerry,

        That’s why I said “LARGELY voluntary”. Imagine a three month treasury bill auction. I bid 1% discount. I have made a bid as required, but it is not going to be successful in the present market. And, if I cannot foresee making any money out of being a primary dealer, I can always give it up, as many banks have done in the past (and some, such as Nomura IIRC, have subsequently been allowed to return).

      7. @Warren Mosler,

        “you obviously haven’t read page one of the 7dif or the mandatory readings?” Oh here we go; the typical MMT response: if I only spent more time reading, I would see the light!

        Actually I have read some of them, and they have helped me to develop my understanding. In fact I even put LR Wray’s book on a UK university reading list in about 2005 which I believe contributed to Edward Elgar reprinting it.

        If you know your stuff, and your understanding is sound, you should be able to answer my arguments with a short factual or logical response, as I am doing here. But you can’t, because I suspect (I remain open to be convinced) your understanding is not sound. And the reason, to go back to my original comment, is that MMT is too insular. Over and out.

      8. @RebelEconomist,

        It is particularly difficult for me to present my arguments to you because I was trained as an engineer not economist and I use different mental models.

        First of all you have stated that the quantity of base money has a profound impact on the economy. I disagree.

        The key issue is whether commercial banks will lend more if they hold excess reserves, assuming constant overnight deposit rate supported by the Central Bank paying interests on these reserves. I say “no”.

        “An important channel through which this approach to monetary policy works is by influencing the demand for credit. The supply of credit is assumed to adjust to demand, which in a world of deregulated financial systems and stable, well functioning markets, is a reasonable assumption.”

        One of the goals of QE as stated by central bankers was exactly to force commercial banks to increase lending.

        “Whereas routine operations are working through interest rates to influence the demand for credit, quantitative easing seeks to influence the supply of credit. The aim is to provide more reserves than banks wish to hold, with the intention that they will try to dispose of these excess reserves by increasing their lending.
        This type of activity is usually undertaken only when interest rates have fallen to zero or near-zero levels. While, theoretically, a central bank could undertake quantitative easing at any level of interest rates, in practice it is difficult to do so, as large-scale provision of excess reserves forces the interest rate to zero. The central bank could stop this from happening by paying a market-based interest rate on reserve holdings, but this would undermine the purpose of the exercise by reducing the incentive for banks to lend their reserves.”

        So Ric Battelino implicitly confirms that the transmission channel of the QE has something to do with interests rates but later partially contradicts himself.

        We cannot say that QE “worked” by forcing more credit.

        The view that maintaining overnight interest can be implemented without draining excess reserves to zero has been confirmed (with some reservations) in the following paper by FRB in the context of monetary policy tightening.

        “These examples provide evidence that a central bank may successfully tighten monetary policy without draining reserve balances (or their equivalent) if it can increase the rate of interest paid on those balances. However, the examples do not guarantee that that capacity is available in
        all circumstances. A key question is whether raising the rate of interest paid on central bank balances would be just as effective in boosting market rates when balances are extraordinarily large. In models of market interest rate determination such as the one described in appendix 2,
        the scale of balances outstanding need not damp the effectiveness of tightening using the interest rate on reserves as a policy tool. ”

        Nothing confirms the hypothesis that the quantity of excess reserves significantly impacts the willingness of commercial banks to extend loans. If they lend more, the competition will push lending rate lower. Assuming creditworthiness of customers, the quantity of credit created over a period of time by private banks depends on the interest rate only. This is the case because a market for loans exists and the usual law of supply and demand applies there, but the supply of funds is perfectly elastic if the CB targets an interest rate. (Please notice that I am ignoring all the feedback loops, I am just analysing that component in isolation). We need to take into account bank spreads too. This is one of the “frictions” in the transmission mechanism during a financial crisis. But looking at the lower limit, spreads cannot fall too low so lending rate will never be lower than overnight deposit rate (or banks will start making loses).

        The second issue you raised is that “To unwind QE, the central bank may have to offer terms that will cost it and its government shareholder dearly, perhaps to the point that they may prefer inflation to paying that cost.”

        If I understand this, you meant that either the CB would need to sell bonds at the price (discount) consistent with the new overnight interest rate – to drain excess reserves and prevent the overnight rate staying at close to zero or to keep paying interests on excess reserves.

        (The issue of “government sector financial capital losses” is insignificant in the reference frame of MMT).

        Well I agree with that point – if you read my previous comment I was saying the same. The problem is wider – if the interest rates are to go higher, either the CB or the Treasury has to pay higher interests or make the non-govt sector benefit from capital gains on the existing stock of govt liabilities.

        But so what? MMT advocates keeping interest rates close to zero and using fiscal policy to adjust the aggregate demand. If you read more of Warren, he said many times that high interest rates may be stimulatory precisely because of the interest payments on the existing stock of debt. So new credit creation may be suppressed but the income channel influence may override the monetary channel. Better to avoid pulling that lever…

      9. @WARREN MOSLER,

        Warren: excess reserves are held as assets and they size in aggregate is determined by the fed, not the banks

        Correct, but banks react individually and competitively.

  9. I cannot get past my fear that direct monetization of government debt will eventually result in substantial inflation. Regardless of the nuances of reserve accounting, is that Koo’s point ?

    Also, can you explain what went wrong in the Weimar Republic ?

    1. It’s somewhat like worrying that putting too much oil in the engine sump will make the car go too fast.

      Weimar ‘overspent’ to a fault, like 50% of gdp annually, mainly on buying fx/selling marks to pay war reparations

      1. I understand that non-trivial inflation is no joke to the vast majority of us, ordinary people, wage earners and all… But under that sudden “monetization of government debt” scenario, can we say, there is so much more to tax now, and the govt can even afford to run budget surpluses for some time?

        Also, I’d expect war reparations to be “payments in real terms.” Is that why you put ‘overspent’ in quotes? And their buying fx/selling marks was for propelling exports, sort of like mandatory transfer of output for reparations?

      2. @Nihat,

        inflation is of little consequences to ordinary individuals. Under any sort of competitive system, businesses absorb it and put it into their prices, individuals absorb it in wage increases and pensions follow it via index linking. Real assets just change price accordingly. Shareholders get an increase in dividends from increased business profits.

        In fact it can help enormously as it rots debt and masks business mistakes.

        It’s those with the debt assets that lose out. But if the entity holding the debt assets happens to be the currency issuer, then even that isn’t a problem. If money lenders are reluctant to lend for investment purposes, then the state can inject for investment purposes.

        So there is a wide range of inflation rates that are likely ‘stable’ and the clarion call for perennially low inflation, regardless of the social consequences, can almost certainly be traced back to banks and money lenders.

        I’m very sceptical of the standard inflation story. It feels like a myth to me.

      3. @Nihat,

        “Monetization” of the debt really doesn’t mean anything. The following two scenarios are equivalent:

        1) Treasury issues bonds for an amount equal to the deficit and central bank buys some back until they hit the central bank’s target yield.

        2) Treasury “prints” the deficit and the central bank issues CDs at the central bank’s determined (target) yield.

      4. Neil,

        I see what you’re saying, and it’s theoretically so very nice and credible. An automatic, orderly response to a trivial problem…

        What you might be missing is, not every segment in a society wields the same power or has the same influence and leverage when it matters, and substantial and sustained inflationary disturbance is likely to hit the less powerful and more numerous segments the hardest. People on fixed income (wage earners, pensioners, even salary earners…) are particularly vulnerable especially if unions are non-existent or castrated. Yeah, they’ve gotta eat, too, but they can eat less… Let’s see on how much less they can live lest we profit less than we could…

        I am afraid those that have the power to not give tend to not give.

        But, I’d agree with you, an MMT govt with proper appreciation of public purpose could deal with it effectively.

      5. everyone thinks we’re always on the razor’s edge of hyperinflation.
        one false move by the cb and the currency goes up in smoke.

        the bank of japan has been trying to inflate as hard as it can for going on two decades and failed
        the fed has tried same going on 5 years and failed.
        maybe it’s not as easy as people think for a cb to inflate?
        exactly as mmt has been saying?

      6. Warren & Brian, I thought, from the spirit of the question, “monetization” had to be understood as an attempt to spend for real goods and services. I stand corrected.

      7. Inflation is of consequence to everybody holding the devalued currency.

        Inflation hurts and can destroy the function of the currency as a store of value, and when inflation gets high enough or volatile enough, it can eventually threaten the role of the currency as the unit of account of the economy.

        And if inflation can benefit debtors that’s only at the expense of savers.

      8. ‘store of value’ is more a ‘characteristic’ than a ‘function’

        the burden of proof is on you to show how ‘inflation’ however defined reduces the gov’s ability to provision itself and reduces potential real output.

        not that I favor inflation. just saying

      9. It doesn’t really matter whether inflation reduces the ability of government to provision itself or not, but when inflation hurts the exchange rate, and the government needs imports, then it will.

        High inflation diverts real resources into speculation on exchange rates, inventories, etc. which reduces real output.

      10. sorry, there are no internal contradictions.

        now the burden of proof is on you to show how inflation increases the ability of government to provision itself and how it increases real output.

      11. @Mamoth, Mam, if the OPEC decided to raise the price of crude tomorrow morning to $200/bbl how could the US govt’s ability to just simply pay that price “be reduced”, if it still wants to provision the Military and/or the SPR with oil at what ever price the OPEC people post?

        this ability cannot be “reduced” for the US govt if the OPEC people prefer to be paid in USDs for oil imo…


      12. @Matt,

        When we talk about government, I think about any government.

        What if the OPEC countries decide to accept only gold or bitcoins in payment?

      13. that would be the same as if they decided to buy those with their dollars.
        and if that drove the price up ‘too high’ they’d have to save in something else or not sell oil

      14. @Mam,

        “When we talk about government, I think about any government.”

        You and Ramanan! 😉

        “What if the OPEC countries decide to accept only gold or bitcoins in payment?”

        Then looks like: “Houston we have a problem…”

        But for now, they seem to more readily accept the USD balance than their own currency balances and the US govt ability is not “reduced”, if North America soon becomes petro independent, these arrangements with OPEC will probably be upset greatly …. rsp,

  10. You mean to say that there is no connection between monetization and inflation ? An answer withoug a metaphor would be helpful.

    1. @gerry,

      Even neoclassical economists accept that Treasuries and USDs are just interest-bearing and non-interest-bearing liabilities of the USG, respectively (with some qualifications, given institutional arrangements re the Fed).

      A primary effect of swapping one for another is a change in interest income to the non-sovereign sector. All else equal, what folks call ‘monetization’ these days is deflationary as it undermines interest and credit channels. See Japan.

      To be inflationary, the Fed would have to pay well above the intrinsic value of those interest-bearing liabilities without any offsetting ‘sterilization’.

    1. @gerry,

      I think the idea is that when wealth is already liquid and fungible it hardly makes a difference. Even the Fed uses MMT principles because housing equity is supposed to make people feel wealthy. So if home equity is considered a source of liquid wealth that impacts behavior why would “only having 50k in treasuries” instead of cash make a difference in buying a car?

  11. Let me say that what Neil wrote is correct. The quantity of money lent and interest rates are inter-dependant at any given point of time and the demand for credit depends on the price of money (interest rate).

    The physical analogy could be the state of ideal gas

    Since nR = const only p, V and T can change. One of these parameters can be an independent variable (let’s say V). We can either control T or p but never both.

    In the modern era central banks can either control (with a reasonable accuracy) the quantity of bank money or interest rates. Again – never both.

    Let’s dissect the following argument:

    “To avoid this outcome, central banks will have to mop up excessive reserves by raising the statutory reserve ratio, raising the interest rate paid on reserves, and selling government bonds. All of these measures will serve to lift interest rates, sending bond yields sharply higher and triggering a possible crash in the bond markets.”

    Excessive reserves are always mopped up overnight to maintain the overnight rate so the quantity of reserves is not an issue. Credit expansion is not limited by the quantity of those excessive reserves. It is limited by the demand for credit. There is a state of dynamic equilibrium on the credit markets. If for a whatever reason (“irrational expectations”) demand curve shifts then the supply curve has to be shifted (by changing interest rates) or the rate of new credit creation will adjust not necessarily in the way central bankers want. Sadly I haven’t seen a proper probabilistic model of credit market equilibrium. But I hope you know what I mean.

    Too much new credit creation will lead to inflation as new demand financed by dM/dt (credit creation) may push aggregate demand above the productive capacities level. This is how we actually think about the macroeconomic processes – this is the “true” IS/LM.

    The essence of the “inflationist” arguments raised by “deficit doves” about the danger of allowing for the stock of public debt to grow is that if there is a strong recovery in the economy the central bank would need to jack up the real (not nominal) interest rates to let’s say 10%. This would inevitably shift upwards the whole bond yield curve (its short-term end is anchored at the overnight money rate). The long end of the bond yield curve is actually the interest rate paid on 10 or 30 years bonds. Now if the public debt is let’s say 200% of the GDP and the government has to pay 11% interest on bonds this leads to pumping 22% of GDP to the bondholders. (this is a monetary flow dM/dt) This state is clearly not sustainable as spending propensity of these people is not equal to zero and because the stock of public debt may grow exponentially – there is a positive feedback loop there. It is like putting off fire by pouring more petrol.

    This situation is I believe caused by what Warren calls the inflationary effects of higher interest rates coming via an “income channel”.

    Real interest rates stayed very low in Japan after the bubble burst in the mid 1980-s. Is it likely that the economy “recovers” or rather enters another debt-fuelled real-estate boom phase? Unlikely, but what if?

    Reference for short-term real interest rates:

    Does the whole analysis mean that “deficit doves” are right? I am not talking about the “run-away money multiplier” story which is clearly incorrect.

    I think that “deficit doves” are wrong. There is absolutely no need t jack up real interest rates to let’s say the 10% range to halt inflation if another real estate boom appears or there is a sudden boom of investment. Just look at the Chinese, how they have recently halted the housing bubble growth. A tax applied to all bank loans minus roll-over (net expansion of credit) would have exactly the same effect as raising credit rates. In the case of a housing boom administrative limitations on new mortgage loans and putting some “red tape” or “green tape” would do. Also – “automatic stabilisers” – actually putting the budget in surplus – would do all the heavy lifting.

    Real bond rates can stay at 1%. The car doesn’t need to be slowed down with a handbrake if the normal braking system functions well. But first we need to stop denying the existence of the brake pedal.

    It is the sacrosaint dogma of not interfering with the “market” what would suffer. But this is a story from another fairy-tale.

    1. @Adam (ak),
      “In the modern era central banks can either control (with a reasonable accuracy) the quantity of bank money or interest rates. Again – never both.”

      In theory the CB can control the supply of money, but operationally, its almost impossible to control the quantity of money within its defined charter role. A CB’s primary chartered role is to protect the payment system that underpins the financial system. Reserves are primarily use to clear and settle payments between banks. If the CB tries to control the supply of reserves directly it quickly finds itself denying reserves to solvent banks just wishing to clear and settle payments – and those payments may just as likely being caused by depositors as lenders. Banks rarely fail for insolvency, the usually fail because of illiquidity and the inability to keep the promises they’ve made to their creditors – including depositors which was to clear and settle the checks they wrote. So, the CB becomes in violation of its charter if it starts to deny the system reserves when it needs them. Volker tried in the early ’80’s and always ended up adding reserves to the system and eventually abandoned this policy and reverted back to targeting interest rates.

      As for everything else you write – love your analysis!

      1. it’s stronger than that.

        loans create both deposits and required reserves, as a required reserve in the first instance is an overdraft at the fed which is functionally a loan from the fed

      2. @WARREN MOSLER,
        Agreed, I was just trying to illustrate that if the CB dried to deny overdraft to a bank it was effectively denying depositors access to their funds because it can’t tell exactly what caused anyone bank to overdraft if it’s denied the whole system the necessary funds to clear and settle payments.

      3. and more so. if a bank makes a loan which creates a deposit and carries a reserve requirement, and the bank does nothing,
        the overdraft is there. the reserve requirement ’caused’ the overdraft. the cb would have to somehow impede the loan itself
        in real time for the overdraft not to be there, right? am i missing something?

      4. @WARREN MOSLER,
        Agreed, a loan typically creates an increase in reserve demand because it causing an increase in deposits and a deposit typically causes an increase in necessary reserves OR the increased transaction volume causes the system to demand reserves overall to facilitate interbank clearing.

        What I was trying to point out is that changes in liquidity preferences by depositors also impacts the demand for reserves. If depositors move significant sums of money from non-reserve requiring deposit accounts to reserve requiring deposit accounts the system is going to generate demand for more reserves. Or if depositors decide to hold more cash it too will cause a higher demand for cash (which has to first be bought with reserves).

        So, just as you were pointing out the CB would have to impede the creation of the loan they would have to proactively prevent depositors from accessing their money as they choose to prevent any change in the demand for reserves. Neither is ever going to happen.

      5. @Adam1, I don’t even think in theory the CB can control the money supply, or I just haven’t heard the theory. BTW the money multiplier is not a theory.

      1. @WARREN MOSLER,

        Government can control both. The Central Bank determines bank loans or interest rates. The OCC, or FDIC, determines what is the appropriate level of leverage on equity capital for banks to be allowed to issue government guaranteed deposits. This varies depending on circumstances, but it is a safe bet that the two units can work together.
        What I see happening is that if there is ever excess demand for bank loans, it would rather promptly meet up with the capital constraints placed on them through Dodd-Frank. There is also a much diminished shadow banking system compared to 2007 which reinforces market discipline. Loan spreads would expand and bank credit would become more expensive in an economic upturn.
        That would be a natural brake against inflation occurring.

      2. good point, but capital grows ‘endogenously’ and is infinitely available at a price.

        but yes, the govt could simply issue bank charters that limit total loans for that bank.
        call it rationing.

        but then it would lose control of rates to ‘end borrower’,
        as (even the most credit worthy) borrowers would bid up the available credit.
        IBM and others trying to finance inventory and sales to clients, for example.

        so the cb could control the marginal cost of funds to the banking system, which could be 0,
        but not the cost to borrowers. What would gm pay for funds to finance inventory vs dumping part of the inventory
        to get down to it’s borrowing limit?

        And if govt limited what banks could charge,
        there would then be a non bank market for funds that would have a market determined rate.

  12. If eccessive bank lending is what they fear I wonder why don’t any mainstream economist propose imposing lending limits to the banks.

    This problem of government supposedly running out of money is so dumb.

    1. @PZ,

      Yeah the whole inflation business is what they will say but that happens between the moment when people will gladly take the government’s money but refuse to pay it in taxes. In other words its a lot easier to let people in the bar than it is to bounce them. So far the only people able to say no to taxation is the 1% and they already did most of their buying.

    1. that was a fraud problem that won’t likely happen again. same happened originally with credit cards.
      but other fraud problems will likely surface in the normal course of human history

  13. QE4 does not cause an increase in inflation? QE$ is being used as free money to de-leverage the big banks rather then letter them collapse as you would under a capitalist system.

    You are not seeing inflation because NET POSITIVE DEFLATION = Inflation – Deflation.

    The de-leveraging of a quadrillion or so of pure speculative bets created by the “banks” cause far more deflation than inflation.

    QE4 is the continued banking bailout at the expense of Americans. Period.

    what are you people smoking!

Leave a Reply

Your email address will not be published. Required fields are marked *