Wednesday, April 17, 2013

The Three Places Reserves Can Go

This post is very similar to my previous list of ways in which reserves can leave the banking system, but here I'm not worrying about what's inside or outside the banking system, which greatly simplifies matters. Also keep in mind that the titles to both this post and the prior one are technically incorrect and misleading: by definition reserves are base money INSIDE the banking system, so they can't go anywhere outside of it. Basically there's only three places reserves (really the base money that makes up the reserves) can go (not making a distinction between required and excess reserves here):

  1. To an entity with a Fed deposit account: Treasury, GSEs, banks (foreign and domestic), foreign central banks, IMF, etc. Note that private individuals and non-bank businesses and organizations don't have Fed deposit accounts, and thus reserves cannot go to them*. Also note that not all banks (for instance) have reserve accounts either.
  2. Withdrawn from commercial bank customer deposits as paper bills or coins, in which case they cease to be reserves (vault cash). They return to vault cash reserve status once they're redeposited at a bank.
  3. Back to the Fed (central bank), where they are annihilated**. This is because reserves are liabilities of the Fed***. Electronic reserves are literally annihilated. Paper bills and coins, of course, might not physically be "annihilated" but instead might be sold again (i.e. exchanged for electronic reserves) to another bank. Of course they might also be literally annihilated (i.e. taken out of circulation permanently due to wear and tear).
Please see my other post for more details and an explanation of terms used here!

*Reserves can be credited to the bank where a private individual or non-bank business or organization holds its account with the bank instructed to in turn credit the private non-bank entity's deposit. This happens, for instance, when the Treasury pays a non-bank individual or business for services to the government from its Treasury General Account (TGA) Fed deposit (Note that funds held in the TGA and other non-bank Fed deposits are not technically "reserves" unless they are transferred to banks). It also happens when banks pay for goods or services or when they pay salaries or dividends to private non-banks when the non-bank recipients of the funds hold their deposit accounts at other banks. Note that in both cases, however, Fed deposits don't disappear: they simply move from one Fed deposit to another. Also note that this doesn't mean that banks can "loan out reserves" to non-banks. That's not how commercial bank lending works. I suppose you could claim that reserves could be loaned out indirectly in this manner, but that's always the case, even when there are no excess reserves in the system: when there are no excess reserves in the banking system, commercial banks will simply borrow the needed reserves from the Fed... and the bulk of this borrowing is on a very temporary basis (the exception being required reserves needed, by regulation, to support expanding commercial bank demand deposits). This is because the Fed (central bank) ALWAYS provides the reserves (if needed) to transfer deposits, clear payments, or expand the deposit base. That's one of the main things it's chartered to do! Also, if it didn't, it couldn't control the overnight rate (one of the other main things it does). Thus having excess reserves in the banking system really doesn't make commercial bank loans more likely. Non-bank private entities literally have no use for Fed deposits, which is why they don't need Fed deposit accounts. Non-bank private entities are really only concerned with their bank deposits and physical cash, and that physical cash (most of it anyway) originates from bank deposits (e.g. ATMs).

**Note that both coins and paper bills (notes) are minted by Treasury, but the Fed only pays Treasury the cost of production for the paper bills whereas they pay face value for the coins.

***Coins (as opposed to paper notes) are actually assets of the Fed after being purchased from Treasury at face value (but before being sold to banks): kind of a historical oddity, and I think the basis of the Trillion Dollar Coin idea for skirting the debt limit. The vast bulk of bank reserves (paper notes and electronic reserves) don't show up on the Fed's balance sheet until they are sold to banks, and then they appear as a liability to the Fed (in the case of electronic reserves they of course don't even exist until they are credited to banks!). Of course when coins are sold to banks, they are removed  from the Fed's balance sheet (i.e. they are erased as Fed assets).

51 comments:

  1. Tom,

    "Reserves" (i.e. reserve balances) are bank deposits. They're just bank deposits created by the central bank - the Fed.

    Generally deposits held at the Fed by commercial banks are called "reserves" or "reserve balances", whereas the Treasury's deposits are simply referred to as "deposits" or the Treasury's "balance". This differentiates the two and reflects the fact that Treasury deposits at the Fed are not counted as part of the money supply, whereas bank reserves are (they are part of the 'monetary base').

    ReplyDelete
    Replies
    1. Yes, reserves are bank deposit assets at the Fed. That's why I like to append "-asset" or "-liability" sometimes... to keep this concept clear. So in that case reserves are private bank deposit-assets, whereas private deposit accounts at the bank are bank deposit-liabilities.

      True, the TGA is outside the banking system. If what you say is true, then technically the reserve deposits lose their reserve status when transferred to the TGA but regain it again when transferred back to private banks. This still represents s place where reserve deposits can go however. They are not annihilated until they go back to the Fed.

      Delete
    2. sure, they're balances with different names. As you say, if "reserve balances" were referred to as "bank deposits" then there would be a lot of confusion.

      Delete
  2. Paper bills and coins, of course, might not physically be "annihilated"

    If they're not annihilated, does this mean they represent "funds" which the Fed can then "use"... ?

    ReplyDelete
    Replies
    1. If you have more info, let me know. But I think when the banks take possession they are a Fed liability. When they come back to the Fed, I'd assume that this liability is erased. So in that sense they can be used to exchange for other Fed liabilities.

      Delete
    2. "When they come back to the Fed, I'd assume that this liability is erased."

      That's my point.. when govt liabilities come back to the govt, they are erased.

      However this erasure is represented as a positive balance in the Treasury's account.

      The numbers (or notes, or coins) are not physically annihilated. They continue to exist as an accounting entry on both the Treasury and Fed balance sheets.

      Delete
    3. OK, show me how that works. Lets start with just paper bills first. Lets say the Treasury can create those for a negligible amount of money (i.e. no cost). So the Treasury mints a $100 bill, the bill goes to the Fed. Is anybody's balance sheet affected so far? Now a bank wants to buy the bill, so the Fed sells it. Don't we have something like this (I'll write PE for "pre-existing"):

      Treasury: A: PE, L: PE, E: PE
      Fed: A: PE, L: PE + $100 bill - $100 reserves, E: PE
      Banks: A: PE + $100 bill - $100 reserves, L: PE, E: PE

      Now the bank sells it back, don't we have "PE" for everything?

      Delete
  3. When the bank buys notes from the Fed (with reserve balances), the Fed's liabilities switch from reserve balances to notes but the quantity of its liabilities doesn't change. Is that what you mean?

    ReplyDelete
    Replies
    1. Yes, that's what I mean. And prior to that, when the Fed holds the notes, I don't see how they'd appear on the Fed's BS. Thus when the notes return to the Fed when the banks sell them, I don't see how they appear on the Fed's BS. I assume the Fed holds a record of them (so their employees don't walk off with them), but as far as the BS goes, I don't see how they appear when the Fed holds them. I assume they don't appear there.

      Delete
    2. I'm not sure. I know the regional Federal Reserve banks are required to post collateral with Federal Reserve agents (who represent the Board of Governors within the banks) in exchange for the notes.

      It might be worth calling or emailing one of the Fed banks to find out how they account for notes held in their vaults.

      Delete
    3. Calling or emailing: Sure, but I'll leave that up to you! ;)

      I'm all for having accurate information on this blog of course, and I'd be very happy to see more of the details of the mechanics, but I don't want it to detract from the simplicity of my main point: that reserves can only go to so many places and they don't flow out into the general economy and cause hyper-inflation, like Peter Schiff seems to think. If I've made a major blunder... I'd be more than happy to hear about it, and I'll change the post accordingly! I've been editing the "List of ways..." post... adding more items to the list when needed, which is what partly inspired this one... to take a step back and simplify it a bit. (BTW, I've already changed this post slightly in response to one of your comments... could you tell?). The main point is to answer a common question I see on pragcap "Where do all those ERs go and why don't they cause inflation?"

      The obvious analogy is if I write out an IOU to you. Before I hand it to you it's essentially worthless (only the negligible value of the paper and ink should appear on my BS, not the face value of the note). When you take possession it becomes a face-value liability on my BS. When you hand it back, that liability goes away.

      So I can see the paper notes appearing as depreciating assets on the Fed BS (consolidated) regarding the production costs the Fed paid for them, but I'd be very surprised to see them at face value on the consolidated Fed BS as long as the Fed is holding them!

      Regarding your two statements above:

      "That's my point.. when govt liabilities come back to the govt, they are erased."

      "However this erasure is represented as a positive balance in the Treasury's account."

      In this 1st sentence, you again consolidate the Treas and Fed together under "govt." That's fine, but not what I prefer to do (as you know), so that's why I've essentially stopped using "govt" in these descriptions. I'd say instead that reserves coming back to the Fed cause them to be erased as Fed liabilities.

      The 2nd sentence I don't see as generally true... even in a consolidated govt BS. It depends on why the paper notes return to the Fed. When the banks sell paper notes back to the Fed, the liabilities are erased from the Fed BS, but they are replaced w/ other liabilities: electronic reserves "keystroked" into existence to purchase the paper notes. I don't see that affecting either the size or equity of the Fed, Treasury, or consolidated Fed/Treas BSs. However, if paper notes (or electronic reserves!) return to the Fed as interest payments or fees or fines levied by the Fed on the banks (assuming paper notes are accepted for these payments), then yes, they affect the Fed and consolidated Fed/Treas BSs in term of equity and size. And if funds are remitted from the Fed to Treasury, then they also affect the Treasury TGA balance in isolation (and likewise stop affecting the Fed BS). Thus I'd say it depends on how the paper notes return to the Fed: in exchange for electronic reserves, or as payments to the Fed.

      Now perhaps there are some interesting details about how the various Fed regional banks maintain their own separate BSs w/in the system (as you allude to), and I honestly didn't even think of that, but I don't think it will change anything in terms of a consolidated Fed BS.

      OK, thanks for keeping me honest phil!

      BTW, if you want to see why I want to keep this post as simple as possible, check out the following months old John Tamny article on Forbes (Tamny now writes for the WSJ as well!):

      http://www.forbes.com/sites/johntamny/2012/07/29/ron-paul-fractional-reserve-banking-and-the-money-multiplier-myth/

      Pay special attention to his reasoning about why the money multiplier is a myth (he gives an example)! Ha!

      Delete
  4. Tom,

    "In this 1st sentence, you again consolidate the Treas and Fed together under "govt."

    Sorry for the abbreviation. Actually I wasn't consolidating the Treasury and Fed, I was referring to "the government of the United States of America". This is a specific legal entity, not a hypothetical consolidated entity:

    "The government of the United States of America is the federal government of the constitutional republic of fifty states that constitute the United States, as well as one capitol district, and several other territories. The federal government is composed of three distinct branches: legislative, executive and judicial, which powers are vested by the U.S. Constitution in the Congress, the President, and the federal courts, including the Supreme Court, respectively; the powers and duties of these branches are further defined by acts of Congress, including the creation of executive departments and courts inferior to the Supreme Court.

    The full name of the republic is "The United States of America". No other name appears in the Constitution, and this is the name that appears on money, in treaties, and in legal cases to which it is a party."

    http://en.wikipedia.org/wiki/Federal_government_of_the_United_States

    The Treasury is a department of the United States government, and the Fed is the United States' central bank. Both are parts of "the government", though the Fed describes itself as "independent within the government":

    http://www.federalreserve.gov/pf/pdf/pf_complete.pdf

    The Treasury's liabilities (bills, bonds etc) are liabilities of the United States government, and the Fed's liabilities (notes, deposits etc) are also liabilities of the United States government.

    When the Treasury has a positive balance in its account at the Fed, it counts this as an asset. But to the United States government that deposit is also a liability, meaning that the US government is holding its own liability. As such, that liability has been 'extinguished'.

    ReplyDelete
    Replies
    1. Brad Delong, former assistant deputy secretary of the US Treasury, put it like this:

      “The way (Warren Mosler) sees it:

      - The federal government spends: it writes a check to somebody…
      - That somebody then takes that check and deposits it in their bank…
      - The bank credits them with a balance equal to the government check and presents the check to the Treasury's fiscal agent, the Federal Reserve…
      - The Federal Reserve then credits the bank with the amount of the check, crediting it to the bank's reserve deposit balance…
      - The Treasury and the Federal Reserve can debit the bank's reserve deposit balance by open-market operations--by selling Treasury bonds--and thus transform the liquid non interest-paying debt of the government that is high-powered money into interest-paying debt…
      - The Treasury can debit the bank's reserve deposit balance by imposing taxes and debiting the bank's reserve deposit by presenting check the taxpayer writes to the bank for payment…

      Briefly: (a) federal government spending creates fiat money--liquid debt of the federal government. This money can then be uncreated by (b) levying taxes or by (c) open-market operations that transform non-interest paying reserve deposits into interest-paying federal debt. Whether you do (b) or (c), and how much you do of (b) or (c), are technical financing decisions, but there is no such thing as "undermining the financial integrity of the nation". Federal spending is always funded by money creation, and the government then adjusts the money supply by levying taxes and conducting open-market operations.

      From my perspective, Warren's way of putting it is both (i) completely, obviously, and tautologically correct; and yet (ii) somehow obtuse…

      The problem is that Warren's claim that "the deficit can present no financial risk" is both right and wrong. it is right in that the government can never find itself in a situation in which it is forced to default on its interest-paying debt. How could it be? It amortizes its interest-paying debt by simply crediting the payee's reserve deposit account. The deficit and debt can present no nominal financial risk. But it is wrong in that the government can find itself unable to keep whatever commitments it has made on what the real return on government debt--both interest-paying and non-interest paying--will be. The inability to keep such commitments is also a kind of "financial risk", and deficit and debt accumulation can create it.

      Are our deficits and debt accumulation currently creating such real financial risks? No”.

      http://delong.typepad.com/sdj/2013/03/bill-black-is-justifiably-irate-monday-hoisted-from-comments-weblogging.html

      JKH actually agreed with the statement that “taxes extinguish money” in the comments section, but qualified it:

      “Taxes only extinguish money tautologically if there is a cumulative deficit in excess of the tax”

      http://delong.typepad.com/sdj/2013/03/bill-black-is-justifiably-irate-monday-hoisted-from-comments-weblogging.html#comment-6a00e551f080038834017c37fac341970b

      (see the bottom of the page)

      Delete
    2. I thought we were discussing the Fed buying back paper notes from the banks with electronic reserves? When those "liabilities" come back there's no net "erasure" on the consolidated govt balance sheet since notes come in and electronic reserves go out. Of course those notes could come back in other ways.

      Delete
    3. Were we? Sorry, my mistake!

      Sure, if the Fed swaps one liability for another there is no net erasure. One sort of liability has been 'erased' and another sort created - a direct swap.

      Delete
  5. "physical cash originates from bank deposits (e.g. ATMs)"

    Actually you can cash a Treasury check at a bank (or elsewhere) without having a bank account. There's also a "Direct Express" debit card for people who don't have bank accounts:

    http://www.fms.treas.gov/directexpresscard/index.html

    ReplyDelete
  6. True... I qualified the statement. Thanks.

    ReplyDelete
  7. Tom, I am still reading as much as I can on this because I think have the same question as you (can excess reserves cause inflation). I think it *is* possible for the reserves to cause inflation because Bernanke has referenced inflation and reserves before....I just can't figure out the mechanism by which inflaiton would be created.

    Anyway, I kind of think I'm on the cusp of having an understanding. I feel like I might be close to an epiphany, haha. I read two Fed papers on the myth of the money multiplier today which made a few more concepts fall into place in my head. You've probably seen them, but I'll link them here just in case.

    http://www.federalreserve.gov/pubs/feds/2010/201041/201041pap.pdf

    http://www.newyorkfed.org/research/staff_reports/sr497.pdf

    I have to say, the Fed puts out some really great papers (for nerds like me who love to read about the monetary system, anyway).

    ReplyDelete
    Replies
    1. pupcakes, if you can figure that out you're a better man than me! It seems to me I've seen Cullen argue it both ways. And I should state now, that this isn't necessarily CPI type inflation, but asset price inflation I'm talking about. I've asked him recently in response to his "The Short Supply of Assets" article if he's changed his thinking on this. I just put it in the forum, so perhaps he'll cover it. I hate to pester the man, but that one I'm really curious about too. Here's my original question:

      http://pragcap.com/the-short-supply-of-assets/comment-page-1#comment-143642

      I've edited my own post on this multiple times... but I don't really know, so I left it ambiguous (more so each time!). See my Example 4 discussion at the bottom.

      I don't think that a static level of ER necessarily can cause asset price inflation. I think they essentially just sit there... waiting to be sucked up by the Fed again once they unwind their bond holdings. They do tend to force theo overnight rate to sit right at the IOR, but that's about it. So in that sense I don't see ER "escaping" and causing inflation, however, as long as the Fed continues to buy T-bonds, I can imagine a mechanism. The increasing levels of ERs are more just another symptom of that continued buying rather than an actual cause of the inflation. But again, I eagerly await Cullen's response to see what his take is and if it's changed.

      I'm also a fan of David Glasner... I think he's more of a neo-classical, and generally a proponent of Market Monetarist views as well, but he's a bit more thoughtful and detail oriented that your average MMer (In other words I agree w/ him more than your average MMer! Ha!). He also shares a lot of views w/ MR. I actually bounced a lot of the ideas that formed the basis of my "List of Ways..." post (which led to this one... see the link at the top) off of him long before I created this post. Glasner touches on the idea here:

      http://uneasymoney.com/2013/04/29/they-come-not-to-praise-market-monetarism-but-to-bury-it/

      Being an MM sympathizer, it's no surprise that he affirms that asset price inflation can occur. That's kind of the point to a MMer.

      Glasner also has on multiple occasions tackled the "money multiplier myth" and tried to put it to rest. I.e., he's not a "loanable funds" guy.

      Delete
  8. Here is my feeling, tell me what you think:

    Banks can create loans essentially out of thin air, i.e. they are not capital or reserve constrained (let's assume for now that loans are a function of demand based on price and other economic factors). Banks have to ensure they meet the reserve requirement after the fact but they can just create loans, right? If they don't have the reserves, they need to go to the interbank market to cover the requirement (let's forget about IOR for a moment).

    The more loans they create, the more likely it is banks will need to access the interbank market to meet the reserve requirement which will drive rates up as banks compete for money. As the rates that banks pay for money go up, banks will increase the interest rate they charge on loans. Could we say the price we pay for loans could be asset price inflation? Or if the price we pay for loans goes up, it will follow that the price of goods/services/assets would also rise?

    What I'm kind of thinking is it's the velocity of money in the interbank system which drives inflation. Which is why the Fed pays IOR, so the banks have no incentive to lend to each other and drive rates up (though they probably don't really need to right now anyway as reserves are at record levels).

    In your opinion (you know more about this than I do), is this line of thinking on the right track at all or am I missing something basic?

    ReplyDelete
    Replies
    1. pupcakes, that's pretty good, but I don't think you've got it quite right. Keep in mind that I'm a layman too!... but here's my read. First of all, banks ARE capital constrained in their lending, by regulatory capital requirements. See my Examples 3, 3.1, 3.2 (probably 3.2 is the best, if you just look at one), and "Calculating Capital vs Equity" posts.

      Now I'm good with everything you write until this "...which will drive rates up as banks compete for money." Actually, I don't think that will ever happen as long as the Fed is targeting the overnight rate, either through a combination of ERs and IOR or through their more traditional method of OMOs. The Fed (and other CBs, such as the BoC in Canada) are able to achieve their target pretty precisely because they can always step in if need be. Let's take the traditional OMO way... if the Fed is targeting 2% say, and the rate in the market starts to fall below this, they perform OMS (open market sales), thus removing reserves from the banking system and driving the rate back up. Likewise if the rate gets too high, they perform OMPs (you guessed it, purchases), thus injecting reserves and driving rates back down. Also the discount window is there as a last resort too. Now if we have permanent ER and no IOR, the overnight rate literally goes to zero! IOR is set above that to make sure that doesn't happen (one of the reasons anyway).

      So keep in mind that the Fed is in some ways like any other bank. It can create reserves out of thin air, but it is even less constrained than a regular bank in this regard! It literally has a bottomless "pocket" with which to purchase anything it wants. That's why you don't want to get into a fight with the Fed!

      I don't know if you call the price you pay for loans asset price inflation. Maybe, but I don't think that's what people usually mean. I think they usually mean financial asset prices: the price of stocks, bonds, and other financial assets on the secondary market.

      Here's the mechanism I see, somewhat influenced by that article at pragcap and also by the MMers: The Fed buys so many bonds that they are creating a shortage, even though Treasury continues to auction them. Investors (or "savers" as Cullen would say) who are interested in buying bonds are interested in safe assets. Prices up on bonds means yields go down. So people are paying more for lower yielding, but safe assets. The MMers claim this tends to raise the price of ALL financial assets, ... I'm not sure, but this might be referred to as a "portfolio rebalancing effect." One MMer named "dtoh" (on Sumner's site) took a lot of time to explain this too me, but it's still a bit fuzzy to me. I think he said what was important was the "risk adjusted rate of return" and that the formula for the price was that it was proportional to 1/(risk adjusted rate of return)... the price of financial assets that is.

      Now at the margin, this is going to cause *some* investors to turn their attention to something other than financial assets because they figure they can get a better risk adjusted rate of return by doing something else with their money, like building a new factory, etc. dhoh referred to this as investing in "real goods and services" as opposed to "financial assets." If you read that Glasner piece, I think he at least hints at this mechanism.

      OK, but I'm getting a bit off track here... regardless if this marginal effect occurs, the financial asset price rise in theory precedes it.

      OK, that's pretty much everything I "know" about it! Ha!

      So, to get back to your "meeting reserve requirements" line of attack... the Fed basically ensures that can ALWAYS be done, ER or no ER, and furthermore that it can be done at the rate they target! So I don't see "competition" over reserves being a factor.

      Delete
  9. BTW, in case you're interested in the MM perspective, here's my thread w/ dtoh on Sumner's site:

    http://www.themoneyillusion.com/?p=19141#comment-225110

    I thought dtoh was a lot better to communicate with than Sumner himself, and like Glasner, shared a lot of the same ideas about MR. I bounced an early version of my "List of Ways Reserves Leave the Banking System" off of dtoh too just to make sure we were on the same page regarding some of the basics... and as far as I could tell we were (this isn't as clear w/ Sumner himself I've found). Nor Nick Rowe for that matter. Although when pressed, none of those three (Rowe, dtoh or Glasner) said they thought that "physical cash" actually mattered much. You get a very different impression from reading Sumner... in that he often refers to "$100 bills" and such as if the presence of physical cash somehow makes a difference for his "hot potato" idea (an idea, BTW, that dtoh says is unnecessary in explaining MM).

    ReplyDelete
  10. Thanks for that great explanation Tom. All your points are taken. Quick question. When you say this:

    Now I'm good with everything you write until this "...which will drive rates up as banks compete for money." Actually, I don't think that will ever happen as long as the Fed is targeting the overnight rate, either through a combination of ERs and IOR or through their more traditional method of OMOs. The Fed (and other CBs, such as the BoC in Canada) are able to achieve their target pretty precisely because they can always step in if need be.

    I think we're on the same wavelength, if I'm reading it correctly. If banks actually start to compete for interbank loans (potentially driving interest rates up), you say the fed will step in to buy assets (keeping rates down), which is......targeting inflation.

    So what I'm saying is if banks increase their loans to the point where they need to seek out interbank loans frequently enough to start to drive up interest rates to the point where the fed steps in, then.....that's inflation. Does that start to answer my original question as to how inflation starts to happen? (Although I haven't figured out how excess reserves from QE fit into this, but I still need to think it through quite a bit).

    Thanks for chatting with me about this and I'll stop bugging you for now.

    ReplyDelete
    Replies
    1. Maybe to put it a different way...there's no competition between banks because the Fed always will step in to stop it, because the competition between banks for reserves is an inflationary thing. Am I making sense?

      Delete
    2. Yeah, I think you've basically got it. Although I'm a little uncomfortable still with the use of the word "inflation" to describe interest rates going up. I think those are two different things. During the so-called "Great Moderation" ... between approximately 1982 and 2006/2007, you had the so-called "traditional monetarists" (as opposed to "Market Monetarists"... going back to before MM was invented I think), having discovered what they thought was the secret to success in managing the Fed: Target a rate of inflation. I think they picked the Milton Friedman recommended 2% as kind of a baseline. What mechanism did they use to target it? Basically the overnight rate. The "formula" they eventually came up with was called the "Taylor Rule" developed by John Taylor. Traditional monetarists still exist (Taylor for one!), Greenspan, and John Cochran, to name some others (they are often associated with the "fresh water" or Chicago school... of neo-classicals. Friedman taught there). This is how I see it: The overnight interest rate can be controlled on a short term time scale by the Fed. Day to day they can hit their target, "defending" the targeted overnight rate w/ OMOs. Now on a longer time scale, their *real* goal may be to hit that 2% CPI rate. How do they do it? Well one method is to use the Taylor rule to make periodic adjustments (say every six weeks in Canada... about the same here) to the targeted overnight rate. Now I think the basic idea is if inflation is too high, you adjust the overnight target upwards, and visa versa if it's too low. So I tend to think of inflation working opposite to the overnight rate. When rates are too low, you risk rising inflation. When they're too high, you risk too low of an inflation rate. Does that make sense? At least that's how it USED to work back when we weren't at the zero lower bound (ZLB) (i.e. a 0% overnight rate), which is where we've been since late 2008. Before 1982, from about 1978 or so, we had another unusual set of circumstances: "stagflation" which was inflation combined with high interest rates and recession. Back then I think the Fed tried a different scheme: target the actual M2 "stock" of money (bank deposits + reserves, etc.). That was Friedman's original idea, but it was abandoned quickly. Recall that we'd only recently converted to fiat money at that point... between about 1971 (when Nixon "closed the gold window".. you can still see his speech on youtube, BTW... kind of interesting!) and about 1975 when we were "fully fiat" w/ floating exchange rates, etc. Friedman had a lot to do with all that as I understand it.

      So like I say, the inflation targeting scheme worked surprisingly well when it was effective. I think the current circumstances have a lot of neo-classical economists, be they monetarists (of whatever stripe) or neo-Keynesians scratching their heads a bit since things are a little different at the ZLB. But my takeaway message is "don't conflate inflation rates with interest rates": they're two different things.





      Delete
    3. pupcakes, perhaps some interesting references for you:

      Nixon closing the gold window in 1971:
      https://www.youtube.com/watch?v=iRzr1QU6K1o

      Milton Friedman praising Greenspan (in a WSJ editorial) for his role in the "Great Moderation" as chairman of the Fed the year of Greenspan's retirement, and month's before Friedman's death in 2006:
      http://online.wsj.com/article/SB113867954176960734.html

      Traditional monetarist John Cochrane commenting on another traditional monetarist's (John Taylor's) thoughts about monetary policy and why the new-fangled "Market Monetarists" are wrong about targeting NGDP:
      http://johnhcochrane.blogspot.com/2013/04/taylor-on-monetary-policy.html

      MMer Scott Sumner's short response to Cochrane (although he has whole posts on Cochrane if you search):
      http://www.themoneyillusion.com/?p=20895#comment-245316

      Austrian thinker David Stockman on "Milton Friedman's contraption" (he's not a fan... nor are most Austrian economists):
      https://www.youtube.com/watch?feature=player_detailpage&v=CAkdB-2qFHY#t=1106s

      MM sympathizing neo-classical economist Nick Rowe on the "Great Debate" between post-Keynesian Steve Keen and Paul Krugman (neo-Keynesian neo-classical):

      http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/03/banking-mysticism-and-the-hot-potato.html

      http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/04/the-supply-of-money-is-demand-determined.html

      A great summary (w/ links!) about the "Great Debate" between Keen and Krugman:

      http://unlearningeconomics.wordpress.com/2012/04/03/the-keenkrugman-debate-a-summary/

      and finally, MMTer Scott Fullwiler (chiming in on the K vs K debate) but also posting some banking basics (which partly inspired this blog!) and discussing how the overnight rate is "defended" by the Fed:

      http://www.nakedcapitalism.com/2012/04/scott-fullwiler-krugmans-flashing-neon-sign.html

      Enjoy!

      Delete
  11. pupcakes, last bit on this... in case you're interested in the ZLB situation and how interest rates are affected there, etc., I thought this was a rare "meeting of the minds" across a wide variety of schools of though. I first saw it on pragcap with this post:

    http://pragcap.com/all-your-dorks-are-belong-to-this

    Which referrenced this Fullwiler piece:

    http://neweconomicperspectives.org/2013/01/understanding-the-permanent-floor-an-important-inconsistency-in-neoclassical-monetary-economics.html

    and here's Steve Randy Waldman, taking on the role of Unlearnedeconomics in the Keen vs Krugman debate, by acting as a central clearing house for all the traffic on this particular topic... both summarizing the debate and providing links to articles by all the various players:

    http://www.interfluidity.com/v2/3763.html

    ReplyDelete
  12. Hi Tom, sorry, I've only just seen your last post here. I'll have a look at the links over the weekend. I'm still reading a ton of stuff on this and some more concepts are falling into place but I'm still trying to find the link between excess reserves and possible inflation. I think the possibility is there because Bernanke has referred to it, but I can't figure out the mechanism. I'll report back if I learn anything new. :)

    ReplyDelete
    Replies
    1. No worries. Sure, let me know if you find anything. You can try posting a question to Scott Sumner, David Glasner, or Nick Rowe too... they're all economists, ...generally on the MM side of things, but they all answer questions. Perhaps it's the expectations game that matters more than the excess reserves themselves. If the central bank declares that it will do what it takes to hit a specific inflation target... perhaps the other players in the market decide it's useless to try and fight the central bank, and fall into line.

      Delete
  13. I'll let you know if I come across anything for sure.

    And thanks for pointing me to Scott Fullwiler, his writing style works well to my style of thinking and learning and makes esoteric concepts easy to understand (which a lot of very smart finance writers are not as good at). I'm still trying to figure out the difference between all the schools of thought (MMT, MR etc) so that I can determine for myself the bias that each writer has. I can't tell that now so it's hard for me find the line between objective truth (that all people will agree on) and opinions which come from a certain school of thought.

    ReplyDelete
    Replies
    1. Yeah, it's funny, part of the reason I started this blog was because of Fullwiler. I don't know the whole history, but Cullen has hinted at it... I think when Cullen was first starting to question some of the MMT ideas like the Job Guaranteee (JG), etc., he got slammed by the MMT folks, including Fullwiler... I think they would leave long nasty comments on pragcap (I've found some old old comments by Scott -- but didn't really read them in detail so I can't confirm if they were nasty or not). I came along after all that so Scott's article "Krugman's Flashing..." was a real eye opener for me! Not knowing the history I'd ask Cullen about the article. He was always game, and even wrote something to the effect of "Scott really knows that stuff well." But I didn't realize that there was some real ugly history there!... so I'd often post links to Fullwiler's article on pragcap to help people... but after learning a bit about the history I started to feel a little bad about that. I feel a lot more comfortable posting links to my own versions! (I still credit Scott in a link in Example #1). Same goes for that econviz tool... I've emailed the guy that runs that... he's an amateur like me and has nothing but respect for Cullen, but if you look on his site all the info came from MMT. It's a great tool... but I always feel a little awkward posting links to it on pragcap!

      Good luck on your quest! I'd be happy to add my two cents if I'm qualified to do so, so don't hesitate to ask. I've only been digging into this kind of thing since 2010... and it's NOT my profession... but perhaps I've been down a similar road myself. I feel a bit more comfortable now with my knowledge of some of the main schools, and can at least summarize some simple principles from each I think. I know a lot more about some than others, however. I consider pragcap to be a "home base" now, but at various times I was interested in William K. Black, Michael Hudson, David Stockman, Chris Whalen, Nick Rowe, David Glasner, Scott Sumner, Steve Keen, Scott Fullwiler, etc. I would listen to some of these guys talk on youtube (especially Stockman, Whalen, Black, Keen, and Hudson) and would be mesmerized... trying desperately to fit what each of them was talking about with the other guys. It's a *little* more clear to me now.

      Delete
    2. "I think when Cullen was first starting to question some of the MMT ideas like the Job Guaranteee (JG), etc., he got slammed by the MMT folks, including Fullwiler... I think they would leave long nasty comments on pragcap"

      That's what Cullen likes his followers to think. Why not look at the evidence? Here is the comment in which Cullen was supposedly "attacked" by Fullwiler as an "ideologue". You know, the "attack" Cullen always talks about?:

      http://pragcap.com/the-politics-of-mmt/comment-page-1#comment-90652

      As you can see Fullwiler is polite. Cullen's comments about the JG are inane and self-contradictory. He then goes off on a daft rant in which he praises unemployment, and contradicts himself yet again.

      http://pragcap.com/the-politics-of-mmt/comment-page-1#comment-90703

      Delete
    3. Scott apologized to me both in that comment and later personally. Obviously, he felt like he had crossed a line.

      Either way, it doesn't matter. I was never against the JG and I said so explicitly in the comments you cite. Yet here you are again trying to frame me as some anti full employment person. No, I am just not fully convinced that the JG and permanent "zero involuntary unemployment" is the most efficient way to operate the economy. MMT changes the definition of FE in order to make a moral argument. That's all. It's highly misleading and you should portray it as a moral position rather than trying to make it some sort of economic truth (like the silly 9 bones, 10 dogs analogy which fails when you understand who really issues the bones).

      Delete
  14. Yeah, the Krugman Flashing piece was so good. It got me all fired up because it made it so easy for me to understand. I am also reading a short book on how the Fed works, which is available for free as a pdf on the Fed's website, and iirc, most of what Scott Fullweiler says is verified by what I am reading in Fed papers - and I of course consider them the primary and most trustworthy source, so I'm using Fed papers as my measuring stick at the moment until I learn more.

    A few other things - I've been reading all night and have come across some interesting pieces that I'm wondering if you've seen:

    This Fed paper about QE and inflation seems to be in agreement with a thought I've been having in the back of my head recently, which is that inflation is not, and never needs to be an issue as long as the Fed can pay IOR. Theoretically they could keep raising the rate and do it forever and remove any incentive for banks to loan (apart from the fact that it would get expensive and other considerations):

    http://www.frbsf.org/education/activities/drecon/2010/0510.html

    This one is my next one to read:

    http://www.clevelandfed.org/research/commentary/2010/2010-4.cfm

    Also, I have a few other pages open on my browser that I am slowly making my way through which go into more detail on MMT, MR and other schools of thought. If any of these seem fishy to you and you think I should disregard them, let me know - you've been learning about this longer than I have (some of these links might actually be ones that you provided):

    http://www.nakedcapitalism.com/2013/03/what-is-modern-monetary-theory-or-mmt.html

    http://themodernprogressive.org/thehumaneconomy/index.php/understanding-the-modern-monetary-system/

    http://www.creditwritedowns.com/2012/04/progress-on-the-monetary-policy-and-banking-debate.html

    Also read your long convo with dtoh on how reserves leave the economy. At first you seem to be talking past each other a little...it seemed like you were both defining "cash" differently (he was using it to mean currency + electronic and you were taking it to mean hard currency only). and he was saying "cash" drives ngdp growth, which causes inflation. I think you and I both agree that hard currency held by the public has negligible impact on inflation, but I'm not sure if this was ever resolved between you and dtoh.

    Anyway, it's good to be able to bounce ideas off someone. I work in the industry but the finer points of central banking and monetary policy are not things that people discuss a lot so it's nice to be able to talk about it to other people. My holy grail is to find a super expert who can explain things to be in terms of what are undisputed facts, what is an opinion and what the counterargument to every opinion is. For the time being, I'm slowly cobbling together this information on my own and you, Cullen and a few others have been a great help. :)

    Also, do you ever read Frances Coppola? She has some pretty interesting things on her blog as well.

    ReplyDelete
    Replies
    1. Coppola, yes, I'm not a regular on her site, but she sometimes writes guest articles for Cullen, and sometimes leaves some comments too. She seems very knowledgeable ... and she's an amateur too I think, right?

      Regarding your other links... no, I'm not that familiar. You will soon leave me in the dust! Please come back and straighten me out when I go off the rails!

      Regarding convo w/ dtoh: Yes, he admitted that cash is not crucial to make MM work, but he said it's more a symptom in his view (i.e. it's correlated w/ MM working). I basically asked the same question of Nick Rowe and David Glasner: their attitude is similar: physical cash is NOT necessary for MM to work. Nick likes the "hot potato" idea whereas Glasner and dtoh don't really use that concept. I think Sumner like "hot potato" as well as "Chuck Norris Effect"... Ha! Just recently I got a little clarity from Sumner on his views regarding cash. I *think* he agrees w/ my list of places reserves can go (I was scolding him for a radio interview in which he mentioned "lending out reserves")... and he agreed! So I was pleased... HOWEVER... I do think he (more than the others) thinks physical cash really does have an effect in the economy, especially in regard to the idea of lowering IOR to 0 or below (i.e. he thinks that will cause people to hold more cash, which will me a mechanism to increase AD).

      Delete
    2. Also, Sumner said lowering IOR to 0 or below is NOT his preferred method. NGDPLT is better. But he and Glasner agree that IOR > 0 is not a good idea.

      Here's dtoh on physical cash being necessary:

      http://www.themoneyillusion.com/?p=19141#comment-225414

      Although he does have (earlier) a list of side effects should we get rid of it:

      http://www.themoneyillusion.com/?p=19141#comment-225305

      Here's Sumner's final word on me scolding him for "lending out reserves":

      http://www.themoneyillusion.com/?p=21126#comment-247583

      The actual radio interview (and a few other exchanges between us) is actually in his next post:

      http://www.themoneyillusion.com/?p=21135

      I was happy to see Sumner reply here several times (even to dtoh) simply "Tom's right." :)

      I don't comment a lot at Sumner's site, but we are more likely to clash a bit. Overall I'm happy that Sumner seems to agree on what can happen to reserves. I think he's a bit more knowledgeable (perahps!) than Cullen gives him credit for regarding banking basics... it's just that he has a very neo-classical bias that the should not be part of macro theory (in fact he has a post entitled that "Keep Banks Out of Macro"). I think Sumner had some real arguments with MMTers about the time the "Krugman's Flashing" article came out, and I have the impression he was somewhat taken aback by it... he engaged w/ MMTers at first, but now he basically thinks they are crazy and wants nothing to do with them. I don't think MR is even on his radar screen. Rowe and Glasner are a bit more charitable towards MMTers!

      Delete
    3. Also, regarding IOR and lending, you might take a look at this comment by JKH at monetaryrealism.com (3rd paragraph down or so):

      http://monetaryrealism.com/krugman-on-says-law/#comment-15559

      He's responding to Morgan Warstler (Morgan is a character! ... he comes off as being super obnoxious!)

      Delete
  15. Nice blog and the content seems very useful about Commercial Banking Services and it helped me a lot.
    Thanks for sharing the post.

    ReplyDelete
  16. What about the reserves held by the China through SAFE? Are they held in an account at the fed or maybe as treasuries?

    ReplyDelete
    Replies
    1. dannyb2b: I don't know! I've never heard of SAFE. Tell me about it.

      But in general foreign central banks do have Fed deposit accounts at the Fed. The balance in these accounts may not technically be called "reserves" because only Fed deposits held by commercial banks are called that (e.g. the Tsy has a Fed deposit (the TGA) but the balance in that deposit is not considered to be "reserves."). Similar for other Fed deposit holders (IMF, GSEs, etc.).

      By "treasuries" you mean Tsy debt (T-bonds, bills, etc), correct? That's very different than reserves. China holds those as well I think (also perhaps primarily though their central bank).

      Delete
    2. http://www.swfinstitute.org/swfs/safe-investment-company/

      Though it doesn't buy the reserves you were talking about ;)

      Delete
  17. Tom,
    I like your summary of go-out list a lot from this post. I have few related questions.
    Q1: Do you have a "go-in" list of situations that money will be going into reserves?
    Q2: Does repo/inverse repo between Fed and bond dealers change bank reverses? How does the accounting work here?

    Thanks for further information

    ReplyDelete
    Replies
    1. Peiya Liu,

      I don't have a go-in list, but you can imagine that they are the reverse of what I list above, namely:

      1. From an entity with a Fed deposit: e.g. Treasury pays someone (sends them a tax rebate check say), and they deposit the check: the bank credits the person's bank deposit, and the Fed credits the bank's Fed deposit.

      2. A person deposits cash into their deposit at the bank. Remember that vault cash can be counted as reserves, so the bank doesn't even have to convert it into electronic form by sending it back to the Fed in exchange of electronic deposits, but they can.

      3. The Fed purchases as asset: the result is more reserves (see my example 4.1) in the banking system.

      Repo and reverse repos do change the reserve balances. Again, see my example 4.1 for the accounting. Repos add reserves, reverse repos remove them:

      http://en.wikipedia.org/wiki/Repurchase_agreement#United_States_Federal_Reserve_use_of_repos

      Hope that helped!

      Delete
    2. Tom,

      I still have one question. Do you know where FED put IOER/IOR interests into bank accounts? Are those interests still in the bank's reserve or excess reserve, or outside bank's reserve account?

      Thanks

      Delete
    3. Peiya Liu, I think they simply credit the bank's Fed deposit, but I don't know for sure.

      Delete