George Selgin on monetary reform

George Selgin has a fascinating new paper that discusses various monetary reform ideas:

To improve the Fed’s current operating framework and reduce the chances for another financial crisis, I offer the five following prescriptions, all of which embody a Bagehotian perspective: (1) abolish the primary dealer system, (2) limit or abolish repos, (3) abandon “Treasuries only,” (4) revive the Term Auction Facility, and (5) stop last-resort discount window lending.

I like the way his proposals simplify and open up the monetary policy process.  I won’t go into all the details here, but the thrust of the argument is that we should move from a bank-oriented regime to a market-oriented regime:

Conducting open market operations in a variety of securities, and not just in Treasuries, would increase the ability of such operations to take the place of both discount-window lending and emergency credit facilities during financial crises. It would therefore allow the Fed to perform its last-resort lending duties during such crises without departing substantially from “business as usual,” and especially without allowing the performance of those duties to interfere with the conduct of ordinary monetary policy. An expanded list of securities would also allow the Fed to spread its tri-party repo settlement risk across more than two clearing institutions (Board of Governors 2002, section 2: 3–4). Finally, security diversification would be a natural complement to counterparty diversification: taken together, the two innovations would allow the Fed to satisfy in a straightforward manner Bagehot’s requirement that central banks supply liquid funds freely, on any good collateral—a requirement which (as we have seen) isn’t necessarily satisfied by channeling funds through a handful of privileged firms only, and only in exchange for Treasuries.

I’m no expert on banking, so I’d be interested in what others think of his proposals.   George provides detailed arguments for all five of the proposals listed in the quotation on top.

BTW, elsewhere I’ve called for increasing the size of the FOMC from 12 to 7,000,000,000, and also changing the one-man-one-vote decision-making system to one-dollar-one-vote.  In other words, let the market implement monetary policy.  It makes no sense at all for the very same institution to set the policy target, and also try to implement it, especially as they’ve never provided an explicit policy goal.  The current system makes it almost impossible to hold the FOMC accountable.  (I can already see the comments I’ll get—”that’s a feature, not a bug.”)


Tags:

 
 
 

60 Responses to “George Selgin on monetary reform”

  1. Gravatar of 123 123
    6. April 2012 at 12:32

    Excellent paper.

    Good points:
    - finally a deserved praise for the ECB framework
    - “by dealing with Treasuries only, the Fed plays favorites with the U.S. Treasury.”

    Weak points:
    - I like LTROs better than Selgin’s auctions. It is easier to make LTROs work.
    - totally wrong about September 2008 “Because the Fed sterilized most of its subprime asset purchases, by reducing its Treasury holdings by over $250 billion and by having the Treasury increase its deposits at the Fed by about $300 billion, the purchases actually reduced the availability of liquid funds to solvent banks.” Stephen Williamson has proved this was the right thing to do.
    - ignores the clearinghouse risk (see Craig Pirrong)

  2. Gravatar of 123 123
    6. April 2012 at 12:38

    Craig Pirrong has posted this yesterday re clearinghouses:
    http://streetwiseprofessor.com/?p=6203

  3. Gravatar of Lars Christensen Lars Christensen
    6. April 2012 at 12:54

    Scott, I love George’s paper. His analysis is totally correct. The Fed’s primary dealer set-up had very negative implications in 2008 and his ideas for a “market maker of last resort” in my view makes a lot of sense.

  4. Gravatar of L Street – Selgin’s prescription for Money Market reform « The Market Monetarist L Street – Selgin’s prescription for Money Market reform « The Market Monetarist
    6. April 2012 at 13:02

    [...] (April 6 2012): Scott Sumner also comments on George’s paper here. Share this:EmailLike this:LikeBe the first to like this post. 5 Comments by Lars Christensen [...]

  5. Gravatar of 123 123
    6. April 2012 at 13:03

    “Because the Fed sterilized most of its subprime asset purchases, by reducing its Treasury holdings by over $250 billion . . . . the purchases actually reduced the availability of liquid funds to solvent banks”

    Here is a key quote from Williamson:
    http://newmonetarism.blogspot.com/2011/08/liquidity-traps-money-inflation-and.html
    “The interest rate on reserves (IROR), which is currently 0.25%, is higher than the interest rate on 3-month T-bills, which is currently 0%. Thus, T-bills command a higher liquidity premium than do reserves, as they are actually more useful in financial exchange (inside and outside the US). ”

    So basically the Fed has increased the stock of liquid assets available when it sold treasuries.

  6. Gravatar of dtoh dtoh
    6. April 2012 at 13:33

    Selgin’s thesis is based on the assumption that we can and should eliminate TBTF.

    Whether we should is moot because politically we can not. Selgin’s whole thesis collapses because its underlying assumption is false.

    Instead we should a) make the TBTF guarantee explicit and put it in the form of a put/call on bank equity at a price set at Fed discretion, b) give the Fed power to flexibly set minimum and maximum asset/equity ratios for banks, and c) require all compensation above t certain level to bank employees be in the form of common stock which vests over a period following termination of employment. This largely eliminates moral risk.

    Because of the explicit guarantee, bank demand deposits become a near perfect substitute for cash (excluding for tax evasion, etc.).

    Banks can get cash at will by selling Treasuries to the Fed.

    The Fed controls NGDP by controlling the price of financial assets by adjusting the required asset/equity ratios for banks. To increase NGDP, the Fed raises the asset/equity ratio, banks make more loans (i.e. they short financial assets), which raises prices (i.e. lowers returns), and therefore market participants exchange financial assets for real goods and services (i.e. NGDP goes up.)

    It’s very simple. Everything else is just details.

  7. Gravatar of Major_Freedom Major_Freedom
    6. April 2012 at 13:34

    ssumner:

    BTW, elsewhere I’ve called for increasing the size of the FOMC from 12 to 7,000,000,000, and also changing the one-man-one-vote decision-making system to one-dollar-one-vote. In other words, let the market implement monetary policy. It makes no sense at all for the very same institution to set the policy target, and also try to implement it, especially as they’ve never provided an explicit policy goal.

    What is this? A belated April fool’s joke?

    Every other blog post contains a direct or indirect advocacy for the Fed to set a policy target of 5% NGDP.

    Now you’re saying let the market decide monetary policy, which not only completely nullifies all Fed targeting policies, but also nullifies central banking itself?

    George Selgin:

    Finally, security diversification would be a natural complement to counterparty diversification: taken together, the two innovations would allow the Fed to satisfy in a straightforward manner Bagehot’s requirement that central banks supply liquid funds freely, on any good collateral—a requirement which (as we have seen) isn’t necessarily satisfied by channeling funds through a handful of privileged firms only, and only in exchange for Treasuries.

    Unless Selgin is advocating that the Fed buy “stuff” from every last individual who deals in dollars, such that every individual gets an equal nominal dollar boost to their sales, and not just a “privileged” few who happen to own and be in a position to sell “a diverse set of securities” to the Fed, then Selgin cannot be considered to be someone who is really solving the problem of funds being channeled “through a handful of privileged firms only.”

    He’ll only be calling to replace one privileged system for another, albeit less, privileged system.

    Of course, if the Fed did send checks to every individual, it would completely nullify the whole reason the Fed was designed in the first place: To enable a privileged class of bankers and statesmen to gain at the expense of everyone else by way of the Cantillon effect.

  8. Gravatar of M.R. M.R.
    6. April 2012 at 14:05

    Selgin writes:

    “One argument against open market operations using private securities is that such purchases are risky. … But the argument isn’t entirely compelling, because … the risk can be kept negligible by means of sufficient “haircuts,” and because, if last-resort lending is desirable at all—if it is a genuine public good—there’s no reason for not having taxpayers shoulder some of the potential cost of providing it, just as they shoulder the cost of supplying emergency assistance to victims of natural disasters.”

    How about this slight modification:

    “One argument against [deposit insurance] is that such [commitments] are risky. … But the argument isn’t entirely compelling, because … the risk can be kept negligible by means of sufficient “[capital],” and because, if [deposit insurance] is desirable at all—if it is a genuine public good—there’s no reason for not having taxpayers shoulder some of the potential cost of providing it, just as they shoulder the cost of supplying emergency assistance to victims of natural disasters.”

  9. Gravatar of ssumner ssumner
    6. April 2012 at 14:13

    123, Don’t the LTROs expose the ECB to default risk? I keep reading that it’s a backdoor bailout of the heavily indebted governments.

    Lars, I just added an update with your post on the subject.

    123, What’s the intuition on why T-bills are more liquid than reserves?

    dtoh, If we can’t get rid of TBTF then we should just nationalize the liabilities of the banking system. We’ve effectively done so already, why not do it formally? But I’m not willing to give up on eliminating TBTF. Policymakers don’t like doing these bailouts (as we saw with Lehman.) If we can put into place a system that eliminates the need for TBTF, then we can eliminate it.

    If we can’t then yes, we should have much higher capital requirements, even if it drives the entire US banking system out of business, and we had to rely on Canadian banks for loans.

    If we target NGDP at 5% per year growth, wouldn’t the asset/equity ratio eventually approach infinity?

    MF, No joke, you just haven’t been paying attention. I’ve always favored having the market implement monetary policy–I published a paper on a NGDP futures targeting system back in 1989, where the market determines the money supply and interest rates.

  10. Gravatar of ssumner ssumner
    6. April 2012 at 14:17

    MR, Actually, deposit insurance is already financed by taxpayers.

  11. Gravatar of AFG AFG
    6. April 2012 at 14:23

    This seems like a reason to get rid of the FOMC, not have it change the target.

    With the mandate at “keep inflation stable and unemployment low,” any personnel change on the FOMC can cause a shift in target, inherently making it less credible. Why not just have congress set the NGDP target?

  12. Gravatar of Bonnie Bonnie
    6. April 2012 at 14:30

    I think you just like being the skunk at the garden party of elitists. What do we need accountability for when Bernanke made the cover of “The Atlantic” being praised as the monetary savior of the western world? Without him, we would have all been doomed to eating out of trashcans, err, um, well instead of 15% of us, anyway. (that is sarcasm, of course).

    Really, when I saw that magazine on the rack, I didn’t know whether to die laughing or sobbing uncontrollably.

  13. Gravatar of Major_Freedom Major_Freedom
    6. April 2012 at 14:30

    ssumner:

    MF, No joke, you just haven’t been paying attention. I’ve always favored having the market implement monetary policy–I published a paper on a NGDP futures targeting system back in 1989, where the market determines the money supply and interest rates.

    Seriously, has the idea of a free market been completely lost in the halls of academia these days?

    With the Fed targeting NGDP, that is not a market driven monetary system! A market driven monetary system would be the market determining not only money supply and interest rates, but NGDP as well.

    By targeting 5% NGDP, the Fed is determining how much money is in the economy, and they are determining interest rates, not those in the market. The Fed is creating the money, not those in the market. The Fed is changing interest rates by inflating the money supply, not those in the market.

    It would be like me having a counterfeiting operation out of my basement, in which the paper I print is enforced as legal tender, and I print at such a rate that results in NGDP of 5% growth each year, and then I say

    “The monetary system is market driven, because the market determines how much money I have to print by individuals spending a particular amount out of their income, and if they want to hold more money, it means they want me to print more to maintain 5% growth NDGP.”

    Tell me, where in your “free market” analysis is the free market process actually deciding on 5% growth in NGDP? Oh that’s right, nowhere. The Fed decides it and constantly overrules the market’s determination of NGDP.

    As long as the Fed exists and prints money to generate a 5% growth of NGDP, over and above what would result from individual private property exchanges, then that is not a free market monetary system!

    It’s so sad that the actual nature of a free market is lost on those who ironically call themselves “market” monetarists.

    Just like statist progressives hijacked the 19th century term “liberal”, so too it seems that statist monetarists are hijacking the term “market.”

    The free market is not what you think it means. The free market means government has its hands off. That means no government intervention in money production. That means the market decides what money is, what money is produced, how much spending of money takes place, and therefore what NGDP happens to be at any given time.

    It doesn’t mean the Fed imposes an artificial, non-market NGDP, and then sit back and say “OK market, do your stuff and determine how much money I should print and therefore how much I will distort interest rates, by reacting to your spending habits and printing when you hoard, and not printing when you spend.”

    This is indeed a joke.

  14. Gravatar of 123 123
    6. April 2012 at 14:34

    Scott,

    LTRO’s have several good features:
    - hairucts and overcollateralization
    - mark to market. If the value of a collateral drops, borrowers have to post additional collateral. So moral hazard of lending to PIGS is mitigated.
    - full recourse. this also mitigates the moral hazard
    - no interest rate risk, as the LTRO has a variable interest rate

    While there is some risk, it mainly comes from the possible errors in setting the haircut levels. But mostly LTROs conform to Selgin’s ideal of adding liquidity without making a subsidy, and they conform better than Selgin’s own auction scheme. There is no bailout. Bundesbank has voted for the LTROs. SMP (Trichet’s TWIST-PIGS program), Fed’s operation Twist is a partial bailout.

    T-bills are more liquid because everyone can hold them, while only the risky banks can hold reserves.

  15. Gravatar of Major_Freedom Major_Freedom
    6. April 2012 at 14:39

    To reiterate:

    A free market monetary system would have all money production, and all money spending, carried out by private property owners, where individual market participants collectively determine what NGDP happens to be via the collective outcome of individual money production and individual money spending decisions.

    By having a central bank imposing a 5% NGDP growth, it means the market process is not determining NGDP. What if the market process decided on an NGDP growth of 2%? Or -2%? Shouldn’t NGDP then be 2% or -2% if we are to say that monetary policy is market driven?

    Your readers are being seriously mislead by this empty claim that the Fed targeting NGDP is somehow consistent with market driven monetary policy.

  16. Gravatar of Bonnie Bonnie
    6. April 2012 at 15:04

    Major_Freedom:

    I am not being mislead. I completely understand the difference between free banking and the Fed. The problem is what we have now is as seriously dysfunctional as it is deeply embedded in our markets, and we need to be able to get from here to there with the least possible chaos, or we’ll never make it from a political point of view. It might be your preference to just flush the entire thing over night, but the people who get caught in the middle might end up with an entirely different point of view than you’re doing it for their own good. When you lose their support, you lose the entire battle; and in that sense, it doesn’t matter who is technically right or who is wrong.

    The kind of stuff you’re talking about would take years to fully implement, if we decided to do it, and this NGDP plan is a good start in the right direction, regardless of how the political stuff ultimately works out.

  17. Gravatar of Benjamin Cole Benjamin Cole
    6. April 2012 at 16:06

    Interesting. One truism of QE that only buys Treasuries is that it ratchets down public debt.

    I like monetizing the debt, deleveraging all taxpayers (rich people should actually like this idea–who do they think will pay taxes to pay off the debt—and remember it is income taxes that pay down the debt, not payroll taxes).

    If the Fed buys just any old debt, it may be pressured to buy cruddy debt from private issuers due to political pressure. The benefits of the Fed purchases may fall on the sellers (maybe vulture funds who bought low?). Buying only federal debt seems to benefit everyone more or less fairly.

    Monetizing debt will also help assure bondholders that sovereign debt will never go bad. See Greece. This is important.

    Re free banking: Before the recent real estate bust, house prices had not busted in decades and decades. That’s why home mortgage were so easily securitized (in addition to greed etc). Thousands of mortgages on very safe assets were pooled–seemed all but risk free. It actually made sense at the time.

    In free fractional gold-standard banking, I could see people reasonably leveraging up on houses as in 2006, both as debtor and lender. After all, house prices never go down, and never all at once for years and years—so everyone thought in 2006-7. As fractional gold-standard banks competed for deposits ,and depositors and lenders wanted better yield, the banks would lend on homes. The competitive forces on banking would be the same as out banks faced in 2006.

    Okay, so we get the same bust. Except there is no central bank to print more money.

    See Japan for the results when a central bank decides not to print money following recession real estate bust. With free banking, the depression that followed would wipe out generations. Except for defending human rights and freedom, I do not think it is worth to spend generations in penance. Much less, just to worship gold.

    Sovereign central banks are an immensely useful and powerful tool to have. You gotta have the A-bomb in your monetary arsenal for ultimate deterrence.

  18. Gravatar of Don Geddis Don Geddis
    6. April 2012 at 16:28

    M_F: you’re just not reading very carefully. Scott has never claimed to advocate completely free banking. Scott wrote: “let the market implement monetary policy”. You replied: “Fed to set a policy target of 5% NGDP. Now you’re saying let the market decide monetary policy?”

    You obviously missed the distinction that Scott wrote, where he is suggesting that the Fed set the target, but the market do the implementation. He is only advocating taking the IMPLEMENTATION of monetary policy away from the discretion bureaucrats, not the policy/target itself.

    We all know that you would prefer free banking, with no government involvement. You ought to start your own blog to advocate your ideas, instead of constantly polluting this one with your off-topic rants.

  19. Gravatar of Major_Freedom Major_Freedom
    6. April 2012 at 16:30

    Bonnie:

    There is no inherent logic contained in NGDP targeting that would peacefully transform our monetary system from Federal Reserve led to free market banking.

    There is however an inherent logic that would violently transform our monetary system from Federal Reserve led to free market banking, and that is eventual destruction of the currency.

    I argue that targeting a constant NGDP, through thick and thin, requires an exponentially increasing money supply to confuse people into spending more money on an increasingly distorted economy that doesn’t align with individual preferences.

    As such, the inherent logic of NGDP targeting resulting in an eventual collapse of central banking, is actually WORSE than peacefully transitioning to a free market banking system NOW, while the system of money is still intact.

    You and Sumner and all the other “reverse Fabians” who believe targeting NGDP is “a step in the right direction towards free banking” are catastrophically wrong. You are in fact being misled. You’re being misled and the signal for it is the feeling you are probably feeling right now, at some level, of morally and intellectually capitulating to some degree. You are accepting SOME bad. It is this accepting of bad that contains the seeds of long term destruction. It will come back and bite you, and if not you, then your children, or your children’s children.

    Your claim that what I am suggesting is “impossible”, betrays the fact that humans make the world what they want it to be. It is not predestined. If everyone thought the way you did, then no progress would ever be made. It is precisely the “dangerous” people you chastise who are the only reason you even know which direction is the right direction, and yet you criticize them for not capitulating the way you are.

    Forcing a constant growth in “spending” in a world of individual humans who do not want to engage in constant growth of spending, is doomed to fail, the same way that forcing communism on people who do not want to be economically controlled is doomed to fail.

    Your mindset is far too myopic. You’re only thinking in the immediate moment, and you’re not taking into account the long term effects of forcing a constant growth in NGDP, which I argue is destruction of the currency. I argue that destruction of the currency is FAR worse than retaining a functioning currency system and radically reforming it.

    Just look at the aggregate money stocks of currencies around the world. Look at Australia, the NGDP model of the world, and its M3. It’s exponentially rising. This is expected because inflation distorts the real economy, and as more and more people try to fix more and more of the errors, the more the central banking system has to increase aggregate money to mislead people into spending enough money to sustain NGDP.

    At some point, the whole money printing counterfeiting scheme is going to come crashing down, and I guarantee you that you will interpret the skyrocketing NGDP to be the central bank “failing” to control it, when in reality they can’t control it forever.

    It’s like a boiler tank pressure “gradually” rising over time, where for a long time, the pressure just keeps rising, and the tank is “stable” over time. But then, suddenly, and “without warning”, the tank will explode.

    That’s how you have to thin of NGDP targeting. Sure, you can see for many, many, many years a gradual increase in NGDP, and everything seems fine, but at some point, people will not accept any more money to sustain NGDP, and suddenly and without warning, the currency will be abandoned and people will fly into real values. It typically happens very quickly.

    Humans aren’t by nature an entity built to collectively spend 5% more money each year every year. Humans are by nature built to collectively spend an indefinite amount of money, based on their knowledge and on the capital structure at the time. Economic calculation requires individuals to be able to hoard cash and decrease their spending. NGDP targeting prevents people from doing that on their own.

  20. Gravatar of Major_Freedom Major_Freedom
    6. April 2012 at 16:37

    Don Geddis:

    M_F: you’re just not reading very carefully. Scott has never claimed to advocate completely free banking. Scott wrote: “let the market implement monetary policy”. You replied: “Fed to set a policy target of 5% NGDP. Now you’re saying let the market decide monetary policy?”

    Speaking of not reading very carefully, I did not claim that Scott did advocate for free banking.

    You obviously missed the distinction that Scott wrote, where he is suggesting that the Fed set the target, but the market do the implementation.

    This is just word butchering. The market cannot be considered “implementing” anything if the Fed is printing money and setting a target that overrules the market’s determination of NGDP.

    He is only advocating taking the IMPLEMENTATION of monetary policy away from the discretion bureaucrats, not the policy/target itself.

    The market CAN’T “implement” monetary policy if market participants are constantly overruled regarding their cash holding versus spending ratios, where the money printers will print and spend more and then less whenever market participants want to implement a different money holding and spending ratio than “5% NGDP”.

    It’s a poisoning of the meaning of markets.

    We all know that you would prefer free banking, with no government involvement. You ought to start your own blog to advocate your ideas, instead of constantly polluting this one with your off-topic rants.

    Cures to poison always seem to be “pollution” to those who advocate for poison.

  21. Gravatar of dtoh dtoh
    6. April 2012 at 17:22

    Scott,

    “If we can’t get rid of TBTF then we should just nationalize the liabilities of the banking system.”

    Why nationalize. Just a put a formal guarantee in place. Government is awful at operating things, but only bad if all they are doing is the guarantee.

    “We’ve effectively done so already, why not do it formally?”

    That’s exactly what I’m suggesting.

    “But I’m not willing to give up on eliminating TBTF.”

    Totally unrealistic IMHO.

    “Policymakers don’t like doing these bailouts (as we saw with Lehman.)”

    But they still do them.

    “If we can put into place a system that eliminates the need for TBTF, then we can eliminate it.”

    We can’t

    “If we can’t then yes, we should have much higher capital requirements, even if it drives the entire US banking system out of business, and we had to rely on Canadian banks for loans.”

    Why higher? Sometimes it needs to be lower. You need to adjust depending on what level of NDGP growth you need.

    “If we target NGDP at 5% per year growth, wouldn’t the asset/equity ratio eventually approach infinity?”

    Why would it? Only if there are no additional additions to overall bank equity. The Fed would control leverage, but profitability would be determined by normal competitive factors. If leverage is low (by Fed regulation), then spreads will increase to generate a market level of profitability, which in turn would attract new capital.

  22. Gravatar of Morgan Warstler Morgan Warstler
    6. April 2012 at 17:46

    MF,

    “I argue that targeting a constant NGDP, through thick and thin, requires an exponentially increasing money supply to confuse people into spending more money on an increasingly distorted economy that doesn’t align with individual preferences.”

    OK, what if the NGDPLT was 2%?

    For the next 20 years, a total of 2%.

    Which means even in a turn of good fortune the Fed is pissing on the booms year to keep everything going at 2%.

    Please explain how this is exponential increase in money supply.

  23. Gravatar of Major_Freedom Major_Freedom
    6. April 2012 at 18:53

    OK, what if the NGDPLT was 2%?

    Same thing, except the exponential growth in money will be slightly less convex.

    Which means even in a turn of good fortune the Fed is pissing on the booms year to keep everything going at 2%.

    Please explain how this is exponential increase in money supply.

    You have to understand that a constant growth in nominal spending via inflation does not imply constant growth in money supply.

    You have to understand that inflation distorts the real economy into an unsustainable configuration, and the only way that inflation can get people to spend enough money in the various malinvested projects, to sustain them, would be to create more money than what the necessary spending requires, which of course itself distorts the real economy even further, thus requiring yet another increase in the amount of money relative to spending, and so on.

    It’s not about picking a “right” NGDP target. It’s the targeting of NGDP via inflation in the first place that is the problem.

  24. Gravatar of W. Peden W. Peden
    6. April 2012 at 19:06

    MF,

    Why would an NGDP target that was below trend real output be inflationary?

  25. Gravatar of Major_Freedom Major_Freedom
    6. April 2012 at 19:09

    And one more thing Morgan:

    Your hypothetical scenario of a “boom” taking place and the Fed forcing 2% price inflation anyway, as if 2% inflation requires monetary tightening, is rather perplexing, since 2% price inflation means the Fed would be loosening, not tightening. In a free market, prices tend to fall. It requires inflation to get prices to rise 2% per year.

  26. Gravatar of Major_Freedom Major_Freedom
    6. April 2012 at 19:10

    W. Peden:

    Why would an NGDP target that was below trend real output be inflationary?

    When I say inflationary, I mean monetary inflation, not price inflation.

  27. Gravatar of Negation of Ideology Negation of Ideology
    6. April 2012 at 19:16

    Some interesting ideas. I basically favor all of them except (3). The justification he uses for abandoning Treasuries only is utterly bizarre:

    “by dealing with Treasuries only, the Fed plays favorites with the U.S. Treasury.”

    The government is favoring itself? It’s bailing itself out? Are we going to complain that the IRS “plays favorites with the U.S. Treasury” as well?

    A dollar, whether the old US Note, a Federal Reserve Note, or Federal Reserve Deposit is a security and obligation of the US government. Just like a Treasury Bond. Exchanging one government security for another government security is not the government playing favorites. If a corporation issues stock and buys back some bonds is it playing favorites with itself?

    Anyway, I know I sound very critical, but besides that I liked the paper.

  28. Gravatar of Major_Freedom Major_Freedom
    6. April 2012 at 19:28

    Propagation of Ideology:

    I think Selgin is taking for granted the notion that central banks should “help the economy” and not just the Treasury.

    It is playing favorites because everyone is forced by legal tender laws and taxation into the US dollar standard. It would be like everyone being compelled by law into owning Apply stock, and then watching Apple issue more shares only to its own board. Everyone else’s shares are diluted. This doesn’t “help the economy.” It benefits only certain people. Selgin is essentially saying “If everyone is going to own Apple stock, then Apple should at least issue new shares to people other than just those in the board.”

  29. Gravatar of Negation of Ideology Negation of Ideology
    6. April 2012 at 19:45

    Major –

    “It is playing favorites because everyone is forced by legal tender laws and taxation into the US dollar standard.”

    The Fed has nothing to do passing legal tender laws or taxes. Congress does that.

    “It would be like everyone being compelled by law into owning Apply stock, and then watching Apple issue more shares only to its own board.”

    No, it is nothing like Apple issuing shares only to its board – unless you’re claiming the Fed only issues dollars to members of Congress. It would be like Apple issuing more shares to buy Apple corporate bonds back from its creditors. Would that mean that Apple’s board of directors are playing favorites with Apple’s corporate Treasury?

  30. Gravatar of George Selgin George Selgin
    7. April 2012 at 04:29

    In quoting the phrase “help the economy” repeatedly, MF clearly implies that the phrase is mine. In fact it isn’t. That’s a good example of the wrong way to use quotation marks.

  31. Gravatar of Becky Hargrove Becky Hargrove
    7. April 2012 at 04:50

    George Selgin,
    Your paper was quite helpful, even non-economists such as myself can understand the basic points you make. Plus, I agree that it’s best to do what is actually possible in the here and now, then work from there.

  32. Gravatar of George Selgin George Selgin
    7. April 2012 at 05:06

    Thanks, Becky. And thanks for the post, Scott.

  33. Gravatar of W. Peden W. Peden
    7. April 2012 at 06:35

    Major Freedom,

    “When I say inflationary, I mean monetary inflation, not price inflation.”

    Then what did you mean when you said-

    “You have to understand that a constant growth in nominal spending via inflation does not imply constant growth in money supply.”

    - ?

    If inflation is a constant increase in the money supply, then the above sentence is false, since “inflation” and “growth in the money supply” are synonyms.

  34. Gravatar of W. Peden W. Peden
    7. April 2012 at 07:21

    Wait, I see: by “constant” you mean at the same rate of growth, rather than there constantly BEING growth.

  35. Gravatar of ssumner ssumner
    7. April 2012 at 11:17

    AFG, Yes, I’ve called on Congress to set a NGDP target path.

    Bonnie, That article was originally called “the villain” then the name was changed.

    MF, You completely misunderstood–check out Don Geddis.

    123, So let’s say Italy and Spain default. The banks go bust. Are you saying the banks still repay the ECB?

    Ben, Yes, with fiscal stimulus the national debt gets bigger, with monetary stimulus it gets smaller.

    dtoh, You said;

    “But they still do them.”

    No, they didn’t bail out Lehman, which shows they’d prefer not to. With NGDP targeting they don’t need to.

    I’m afraid your capital req. model won’t work without a numeraire, without something to anchor the price level and NGDP. You are trying to achieve a nominal target with a real model of the economy, it just won’t work. Basically you have to raise velocity higher and higher, and eventually the model would collapse. You need to raise the monetary base.

    If you don’t believe me, imagine trying to peg NGDP growth at 5% a year in an economy without banks.

    Negation of Ideology, I’m also comfortable with a Treasuries only policy–I don’t see much distortion.

    George, MF has been known to misrepresent peoples’ views once or twice . . . or a billion times.

  36. Gravatar of Major_Freedom Major_Freedom
    7. April 2012 at 11:46

    W. Peden:

    “When I say inflationary, I mean monetary inflation, not price inflation.”

    Then what did you mean when you said-

    “You have to understand that a constant growth in nominal spending via inflation does not imply constant growth in money supply.”

    I meant monetary inflation.

    Wait, I see: by “constant” you mean at the same rate of growth, rather than there constantly BEING growth.

    Correct.

    George Selgin:

    In quoting the phrase “help the economy” repeatedly, MF clearly implies that the phrase is mine. In fact it isn’t. That’s a good example of the wrong way to use quotation marks.

    You’re right, that is implied, but it wasn’t my intention. I used the quotes to signify it being a dubious idea, not to attribute it to you. My apologies. When I actually quote people, my personal method is to italicize their comments, or explicitly state “This is what he said:”. Sorry for the confusion.

    ssumner

    George, MF has been known to misrepresent peoples’ views once or twice . . . or a billion times.

    You just made that up to kick an intellectual antagonist person when he’s down. I wasn’t quoting Selgin, although it definitely looked that way. I thought the “I think Selgin is taking for granted the notion that…” made it clear that the quotes means it represents a notion I am addressing. But it does seem the wrong way to use quotes.

    MF, You completely misunderstood–check out Don Geddis.

    Both you and Geddis completely misunderstood the point of my response. My point is that it is a butchering of the English language to say “the market” is “implementing” monetary policy what with the Fed targeting a rate of NGDP.

    It would be like saying the market “implements” the color of Model T, with Henry Ford deciding to paint them all black.

    An actual market “implementation” of NGDP would be for the market process to decide NGDP.

  37. Gravatar of Major_Freedom Major_Freedom
    7. April 2012 at 12:10

    Propagation of Ideology:

    “It is playing favorites because everyone is forced by legal tender laws and taxation into the US dollar standard.”

    The Fed has nothing to do passing legal tender laws or taxes. Congress does that.

    Irrelevant. The Fed-Treasury relationship is taking advantage of it. It would be like a white plantation owner in the 18th century taking advantage of slave laws, and then saying “I didn’t make the laws.”

    At any rate, the Fed does get involved with laws. Remember the audit the Fed bill? Also, in the 1960s, the Rockefellers financed women’s lib to double the taxation base, so that the banks can double their risk free lending to the state.

    “It would be like everyone being compelled by law into owning Apply stock, and then watching Apple issue more shares only to its own board.”

    No, it is nothing like Apple issuing shares only to its board – unless you’re claiming the Fed only issues dollars to members of Congress.

    That’s not required. It is everything like Apple issuing shares only its board. I am saying when the Fed buys Treasury debt, that favors those at the Treasury and those who receive Treasury money. “Traditional monetary policy” sees the Fed ONLY buying Treasury debt, which means those at the Treasury and those who receive Treasury money are favored by traditional monetary policy.

    It would be like Apple issuing more shares to buy Apple corporate bonds back from its creditors.

    In order for this scenario to serve as an adequate analogy, it must be the case that everyone in the country would have to own Apple debt, and then Apple issues shares to buy back debt only from a select favored group of creditors all the time, while every other creditor instantly becomes subordinated and their value diluted.

    Would that mean that Apple’s board of directors are playing favorites with Apple’s corporate Treasury?

    If everyone owned Apple debt the way everyone owns US dollars, then yes they would be playing favorites.

  38. Gravatar of 123 123
    7. April 2012 at 12:44

    Scott: “So let’s say Italy and Spain default. The banks go bust. Are you saying the banks still repay the ECB?”

    According to the logic of your question, the only asset central banks should hold is gold. :)

    But seriously, what I’m saying is that LTRO’s are carefully constructed to provide liquidity only while avoiding subsidy. All the assets in the world have an element of risk, but the ECB is more than compensated for the risk it takes on. LTRO’s are a good business. Mark to market means that the additional collateral is posted when the risk of default grows. Full recourse means that in the event of default, the ECB has an additional claim on other assets of the defaulted bank. Bagehot would be happy, Selgin should be happy too.

  39. Gravatar of Saturos Saturos
    7. April 2012 at 13:54

    “…increasing the size of the FOMC from 12 to 7,000,000,000, and also changing the one-man-one-vote decision-making system to one-dollar-one-vote. ”

    And what do you think are the chances of the FOMC agreeing to this proposal? As for Congress implementing a monetary reform over their heads, that seems about as likely as Ron Paul’s agenda – indeed, right now it seems more likely that they will “End the Fed” than allow “markets to play havoc with the money supply”.

  40. Gravatar of Major_Freedom Major_Freedom
    7. April 2012 at 13:59

    Saturos:

    And what do you think are the chances of the FOMC agreeing to this proposal? As for Congress implementing a monetary reform over their heads, that seems about as likely as Ron Paul’s agenda – indeed, right now it seems more likely that they will “End the Fed” than allow “markets to play havoc with the money supply.”

    It’s funny isn’t it? The typical defense market monetarists give for their apologizing for the Fed, is something like “A free market isn’t going to happen soon, so given we have a Fed, they should do X.”

    Then they give these whoppers of “There can be 7 billion FOMC members rather than 12, and one dollar one vote rather than one man one vote”, like it’s something that those at the Fed will consider any more seriously than shutting its doors tomorrow.

    It’s like market monetarists are allowed to call for “idealistic” reform, but they won’t allow others to do it.

  41. Gravatar of StatsGuy StatsGuy
    7. April 2012 at 16:33

    ” the thrust of the argument is that we should move from a bank-oriented regime to a market-oriented regime”

    This is one to think through. My only concern with bagehot is what happens if there is not enough low risk collateral to purchase… Essentially, you can drive rates to zero further and further out along the time curve, but this gets back to the idea of liquidity crisis vs. solvency crisis. This is a gift to holders of low risk assets, but eventually, the CB will need to buy higher risk assets (equity, or preferred equity) unless it can actually create the expectation of higher future AD.

    This would create a much more active CB – at least, one that is more VISIBLY active. Selgin’s rules, which remove discretion from the CB, would absolutely be necessary, but I doubt Selgin has any expectation this could happen, given the immense political power of the banks.

  42. Gravatar of dtoh dtoh
    7. April 2012 at 17:39

    Scott,
    You said;
    “No, they didn’t bail out Lehman, which shows they’d prefer not to.”

    Lehman was the exception. Next time around, the bankers will all say look what happened when you didn’t bail out Lehman. Of course they prefer not to…. but they will.

    “If you don’t believe me, imagine trying to peg NGDP growth at 5% a year in an economy without banks.”

    Imagine there’s no bankers… and no money too. It’s easy if you try. The Fed just operates a credit card company. You can have a debit or credit balance with the Fed. You get interest on a debit balance pay on a credit balance. The Fed has an algorithm that determines whether or not you have hit your credit limit. All transactions in the economy take place via credit card. The Fed wants more spending (NGDP), they just cut rates and loosen credit criteria.

    You can do the same thing with one year non-interest bearing promissory notes where the Fed just raises and lowers the price of the notes.

    You can also eliminate the Fed as the intermediary so all they do is approve/disapprove the credit and set the rate.

    Very easy.

  43. Gravatar of Some Links Some Links
    8. April 2012 at 12:48

    [...] Scott Sumner discusses George Selgin’s outstanding recent paper “L Street: Bagehotian Prescriptions for a 21st-Century Money Market.” [...]

  44. Gravatar of ssumner ssumner
    8. April 2012 at 14:11

    123, I’ll let George respond to that, I don’t have strong views on LTRO one way or another.

    Saturos. The Fed has no say in the issue of the FOMC’s composition, that’s up to Congress.

    I don’t expect my proposals to be implemented in the short run, but perhaps someday. When Irving Fisher advocated price level targeting I don’t doubt everyone told him the Fed would never agree to it. Admittedly inflation targeting is slightly different, but close enough that he’d consider it a win.

    Statsguy, The last thing you want to do is drive rates to zero way out the yield curve. If you’ve done that, then money is way too tight.

    You said;

    “This would create a much more active CB – at least, one that is more VISIBLY active.”

    I disagree, the CBs that seem very active are those that let NGDP growth fluctuate sharply, creating the need for unconventional policies like QE. The Australian central bank seems much less active, and is doing a better job of controlling NGDP.

    dtoh, There’s still no nominal anchor in your system.

  45. Gravatar of dtoh dtoh
    8. April 2012 at 17:01

    Scott,
    You said;
    “There’s still no nominal anchor in your system.”

    The nominal anchor can be anything as a long as people accept it for exchange. Call them credits. You could kick start the system by making one credit exchangeable into one $1 Fed Reserve Note. Then simply stop exchangeability, make Fed (approved) credits legal tender. As long as the Fed is controlling the supply it works fine.

    Fed follows economist consensus views. Economists are not dumb. Economist aren’t following you argument. Ergo, if you’re right, your argument is deficient.

    You can simplify the argument by starting from a credit only model. Or… if you insist on having money, use a model where money expires at midnight on the day it is issued. Then V becomes constant, and daily NGDP equals OMP for that day. The Fed just bids up the price it will pay for financial assets to the level of NGDP it needs.

    With your hot potato argument, you get the counter-argument, yes…but V will just drop. If you present your argument as an increase in nominal spending caused by a change in the relative price of financial assets to real assets, it’s much clearer.

  46. Gravatar of Doc Merlin Doc Merlin
    8. April 2012 at 22:28

    “(1) abolish the primary dealer system,
    +1

    (2) limit or abolish repos,
    +1

    (3) abandon “Treasuries only,”
    Treasuries only was actually the purpose of the Fed. It was created to allow the US government to monetize debt more effectively.

    (4) revive the Term Auction Facility, and
    I don’t really think this will have any effect. Right now there is so much liquidity at the top of the capital markets that its kind of absurd. The lack of liquidity is that the smaller firm and the consumer level.

    (5) stop last-resort discount window lending.”
    +1

  47. Gravatar of George Selgin George Selgin
    9. April 2012 at 09:31

    Doc, you are wrong about (3): although I’m the last to deny the fiscal rationale behind the establishment of many early currency monopolies, it played hardly any role in the case of the Fed; indeed, as I explain in my paper, the Fed’s founders did not wish to see it purchase much government paper precisely owing to fear of it’s becoming a slave to the Treasury. Of course it did eventually become precisely that (indeed, already during WWI it appears to have set low discount rates in order to enhance war bond sales to the general public). But I don’t think the record shows this to have been one of the motivating purposes of its establishment.

    As for (4), you must bear in mind that, without it or something like it, prospects for (5) become especially slim.

  48. Gravatar of StatsGuy StatsGuy
    9. April 2012 at 10:40

    “Statsguy, The last thing you want to do is drive rates to zero way out the yield curve. If you’ve done that, then money is way too tight.”

    Remember, there are two questions about the chuck norris effect – first, will chuck punch you, and second, will chuck’s fists actually hurt you if he did.

    If people think that – EVEN IF the Fed will do everything it can, some people may need to be convinced that the Fed’s policies will actually hurt them. This may not happen often, but it may happen – so we need a credible plan on what to do if we got there, while recognizing that in building a credible plan we are reducing are chance of actually getting there.

    “This would create a much more active CB – at least, one that is more VISIBLY active.”

    Scott –
    “I disagree, the CBs that seem very active are those that let NGDP growth fluctuate sharply…”

    I’m not sure how this works if you squeeze most of the endogenous money out of the system, although it’s likely that liquidity crises would happen less often to begin with.

  49. Gravatar of Major_Freedom Major_Freedom
    9. April 2012 at 11:21

    George Selgin:

    although I’m the last to deny the fiscal rationale behind the establishment of many early currency monopolies, it played hardly any role in the case of the Fed; indeed, as I explain in my paper, the Fed’s founders did not wish to see it purchase much government paper precisely owing to fear of it’s becoming a slave to the Treasury. Of course it did eventually become precisely that (indeed, already during WWI it appears to have set low discount rates in order to enhance war bond sales to the general public). But I don’t think the record shows this to have been one of the motivating purposes of its establishment.

    The Fed’s founders were not just bankers. It initially included two Senators, Nelson Aldrich and A. Piatt Andrew, the latter of whom was assistant Secretary at the Treasury. Later on, Democrats Glass and Owen got involved. The newer Glass-Owen version of the bill contained a change from the Aldrich bill, namely, instead of the new currency being an obligation of the private banks, it was to be the obligation of the US Treasury, among some other changes.

    The Aldrich bill that saw almost total banker control led to Democrat worries that the Fed would just be used as a tool of the rich (known as the “Money Trust”). So the Glass-Owen version of the bill gave more power to the government.

    Since the Fed’s banker founders needed to convince both Democrat and Republican politicians to go along with the cartelization plan, they agreed to act as fiscal agent of the Treasury, which of course included buying US Treasuries.

    It was a tiny price the bankers were very much willing to pay in order to get the cartelization bill passed by the newly Democrat controlled House. That was their main goal. It’s why the final Glass-Owen version of the bill that contained more government control was almost universally supported by the bankers. The publicized worries of the bankers that they would become slaves of the Treasury if they bought too many Treasuries, were more crocodile tears than anything else.

  50. Gravatar of George Selgin George Selgin
    9. April 2012 at 12:34

    “which of course included buying US Treasuries.”

    In fact there was no substantial buying of such by the Fed until the mid-1920s, when the Fed took advantage of a minor provision of the law (designed to help it smooth its interest earnings) to engage in its first major OMOs. Only in the 1930s were such operations officially sanctioned by the revised FRA. To reiterate, the original plan was for the Fed to stick to discounting commercial paper, and to have only such paper and gold backing its notes. The Fed didn’t stick to the plan, of course; but it is not correct to suggest that it was intended from the start that it should be a major purchaser of U.S. Treasury securities.

  51. Gravatar of Major_Freedom Major_Freedom
    9. April 2012 at 13:06

    George Selgin:

    In fact there was no substantial buying of such by the Fed until the mid-1920s, when the Fed took advantage of a minor provision of the law (designed to help it smooth its interest earnings) to engage in its first major OMOs. Only in the 1930s were such operations officially sanctioned by the revised FRA. To reiterate, the original plan was for the Fed to stick to discounting commercial paper, and to have only such paper and gold backing its notes. The Fed didn’t stick to the plan, of course; but it is not correct to suggest that it was intended from the start that it should be a major purchaser of U.S. Treasury securities.

    I hope you don’t believe that your switching of goal posts went unnoticed. You just introduced the adjectives “substantial” and “major”, when neither you nor myself even considered the extent of the purchases before, only whether or not purchases per se were originally intended.

    You seem to be conceding that Treasury purchases were originally intended, but now you’re saying they weren’t “substantial” or “major.” Fine, they weren’t “substantial” or “major” at first, but my initial point was that Treasury purchases were in fact tied into the Glass-Owen bill, which had almost universal support from the bankers.

    After all, the Fed’s founders were composed of not just bankers, but politicians as well. It shouldn’t be so surprising that the Fed would be politicized from the beginning.

    Sure, the rate at which the Fed bought Treasuries substantially increased in the 1920s, and then again in the 1930s, but that is besides the point. The point is that they were buying Treasuries since the start.

    In any event, an increase in the amount bought could not have happened if the Fed were not already buying Treasuries. That’s how an initial intentional buying of Treasuries turned into “substantial” and “major” buying of Treasuries.

    The 1930s FRA amendment didn’t legally introduce the Fed buying Treasuries. That legality already existed in the original FRA, under Section 14:

    Section 14. Open Market Operations

    Purchase and Sale of Cable Transfers, Bank Acceptances and Bills of Exchange

    Any Federal reserve bank may, under rules and regulations prescribed by the Board of Governors of the Federal Reserve System, purchase and sell in the open market, at home or abroad, either from or to domestic or foreign banks, firms, corporations, or individuals, cable transfers and bankers’ acceptances and bills of exchange of the kinds and maturities by this Act made eligible for rediscount, with or without the indorsement of a member bank.”

    The 1930s FRA amendment just added the FOMC to do what the Fed was already allowed to do. While the original intent of Treasury purchases was to enable the Treasury to finance basic things like roads and bridges, it spawned into quite the interventionist agency later on.

    Now one can quibble over just how much the Fed purchased over time, or in the 1920s and 1930s compared to initially, but it is not correct to claim that Treasury purchases were not intended from the start. They were intended, just not to a substantial degree as you later stated.

  52. Gravatar of George Selgin George Selgin
    9. April 2012 at 16:03

    By insubstantial I mean trivial. As Casey Stengel might say, “You can look it up.”

  53. Gravatar of ssumner ssumner
    9. April 2012 at 16:36

    dtoh, You said;

    “As long as the Fed is controlling the supply it works fine.”

    But if they control the supply, then they control NGDP, and there’s nothing for the capital regs to do.

    You said;

    “With your hot potato argument, you get the counter-argument, yes…but V will just drop.”

    I’m trying to persuade economists who understand the fundamentals of monetary economics, not the crazies, so I don’t worry about that argument.

    You said;

    “Fed follows economist consensus views. Economists are not dumb. Economist aren’t following you argument. Ergo, if you’re right, your argument is deficient.”

    Let’s see. And obscure economist from Bentley starts a blog (along with a few other obscure economists like Beckworth.) Absolutely no one is talking about NGDP targeting. NO ONE!!

    Three years later you get Goldman Sachs, Christina Romer, Paul Krugman, Tyler Cowen, Brad DeLong, The National Review, Matt Yglesias, etc, etc, talking about NGDP targeting. And you want me to give up because I have no hope of convincing other economists? With failure like that, who needs success? :)

    Statsguy, Obviously there’s a grain of truth in what you say, but I keep insisting on pushing back against the view that this is what easy money looks like, and even easier money would be even lower rates for longer, and an even more bloated base. At a minimum I’m more likely to be right on average, in the long run. But yes, Chuck Norris must be willing to do whatever it takes.

  54. Gravatar of Major_Freedom Major_Freedom
    9. April 2012 at 16:59

    ssumner:

    Three years later you get Goldman Sachs, Christina Romer, Paul Krugman, Tyler Cowen, Brad DeLong, The National Review, Matt Yglesias, etc, etc, talking about NGDP targeting. And you want me to give up because I have no hope of convincing other economists? With failure like that, who needs success?

    I think your sarcasm inadvertently and amusingly revealed a stark truth.

    Question:

    In an NGDP targeting monetary system, what’s to stop the major banks from hoarding the newly created money the Fed sends to them, in order to get the Fed to continue to give them more money in an attempt to boost NGDP that isn’t happening?

    What I mean is, NGDP is ultimately up to the people who own money. If they don’t spend “enough”, then the Fed will keep sending more and more checks to the banks until the banks start increasing their loans or dividend financed consumption or whatever, to get NGDP to rise to the target rate.

    If you’re a major bank that owns substantial quantities of US treasuries, what’s to stop you from hoarding the money you got from selling treasuries to the Fed, in order to coax the Fed into creating even more money? In an NGDP targeting world, the Fed will have to depend on non-Fed entities to spend money. If the non-Fed entities don’t spend enough money, then the Fed will just create more money and send it to the banks.

    The major banks can easily game the system by simply abstaining from investing the money the Fed sends them, which the Fed has no choice but to continue to send money to the banks until the banks do spend.

    The major banks can stockpile trillions of dollars in this way, and then they can buy up huge portions of the real economy.

    How will an NGDP targeting Federal Reserve system get around this moral hazard?

  55. Gravatar of dwb dwb
    9. April 2012 at 17:09

    excellent paper, i am finally getting around to reading it.

    thoughts:
    - “extending” (lets not say “abolish” just yet) the dealer network is obviously an important idea. Some of the Fed’s own research on the “monetary transmission mechanism” (i dont have the papers in front of me) emphasize that the reserves-mechanism is more important for smaller banks and thrifts than for large well-capitalized banks (that are rated and can issue short term paper). But: extending the network is not free and has some operational costs (risk monitoring, operational and IT infrastructure etc). There is anti-money laundering, security, and so on, its not a small deal to do it. The Fed’s bureaucracy and payroll will get much bigger (maybe at the regional banks as well).

    -The fed replies too much on repos, agreed. But the repo market in general is VERY LARGE and the Fed only participates in a small bit. Yet there is such a thing as “General Collateral” repo on non-treasury securities that mm funds and banks and corporates lend each other to manage liquidity, but i doubt the Fed would be competitive in this market. The haircuts are market determined, and for political or optical reasons the Fed would probably want a deeper than market haircut on non-treasury collateral to insure against the consequences of any Congressional apoplexy. Thats probably a non-starter. Even then, in the size the Fed needs to do, Treasuries are the most liquid thing out there {been a while since i worked on the repo desk, but i thought they also accepted Fannie/Freddie/GNMA securities as collateral too}.

    Probably the Fed itself would agree it relies too much on short term repos (maybe short term TAF liquidity facilities are better) but if the fed *insists* on targeting the effective fed funds 6-8 weeks at a time, on a daily basis, I don’t see much choice.

    -i like the TAF. maybe its semantics, or the Fed did not know what it wanted the TAF to be. But i’d like to see more use of TAF and less of repos. In fact, if a bank borrows from TAF for 6 weeks at the fed funds rate, let them go lend it on repo, just like a TBill.

    of couse, TAF is when they want to ADD liquidity. whats the anti-TAF?

  56. Gravatar of dtoh dtoh
    9. April 2012 at 21:34

    Scott,

    As long as the Fed is controlling the supply it works fine. > “But if they control the supply, then they control NGDP, and there’s nothing for the capital regs to do”.

    I’m not saying you can’t control NGDP with expectations and OMO. I just saying you can do it better with capital regs (or to describe it better… minimum asset regs.) And.. in the case I described, there is a massive 100% opportunity cost (currency expires worthless at mid-night every night) so OMO works perfectly. When the opportunity cost gets close to zero, OMO doesn’t work so well (V drops). The point I was trying to make in my comments in this thread however was not on the relative merits of OMO versus asset/equity regs, but rather that it’s easier to explain either approach through a model that compares the relative price of financial assets to the price of real goods and services. In the real world if I have excess cash, I don’t spend it; I buy financial assets. What causes me to spend more on real good and services is when the value of those financial assets increase relative to real goods and services.

    With your hot potato argument, you get the counter-argument, yes…but V will just drop. > “I’m trying to persuade economists who understand the fundamentals of monetary economics, not the crazies, so I don’t worry about that argument.”

    But that is the only argument. If you say, that OMP will not cause an increase NGDP, definitionaly you have to be saying that V will drop.

    “And you want me to give up because I have no hope of convincing other economists? With failure like that, who needs success?”

    I think you have been wildly successful, but…. you haven’t convinced the FOMC yet. Just think how much more successful you would be if your argument was even better. You used to talk a lot more about the impact of asset pricing. Once you cut through the details, asset prices (together with expectations) is the mechanism which moves NGDP

    BTW – I finally watched the video discussion with you, Cowan, Delong and Smith. Very impressive. Thought you did a much better job than any of the other panelists.

  57. Gravatar of ssumner ssumner
    10. April 2012 at 11:37

    MR, The banks can’t “stockpile” because they must give up assets to get those reserves.

    dtoh, Thanks for the comments on the video. You said;

    “I think you have been wildly successful, but…. you haven’t convinced the FOMC yet. Just think how much more successful you would be if your argument was even better.”

    So you are saying a Sumner plus dtoh blog would have been even more wildly successful, and would have convinced the FOMC by now? Possibly, but I find that a stretch.

    All I can do is reiterate:

    1. OMOs can to it alone.

    2. It’s theoretically possible that capital regs could be adjusted to smooth V at the zero bound, but I’m not convinced. And even if this is true, I doubt we’d ever be at the zero bound with 5% NGDPLT.

    I don’t feel I understand bank regs well enough to say more. I’m certainly not saying you are definitely wrong, just that it seems a side issue to me.

  58. Gravatar of Major_Freedom Major_Freedom
    10. April 2012 at 12:42

    ssumner:

    MR, The banks can’t “stockpile” because they must give up assets to get those reserves.

    You mean government debt, right? It can be a slush fund mechanism.

    OK, let me spell out the process that I have in mind:

    Treasury issues $100 billion in debt to the major banks, after which the Treasury spends the money. The money and debt ends up in US banks. The bank hoards all the money except that which is needed to satisfy withdraw and transfer requests. Say they net out $90 billion.

    Meanwhile, the Fed is trying to target NGDP by purchasing government debt from the major banks. So the Fed buys the $100 billion of debt from the banks. The banks then hoard all of this money, knowing that the Fed will continue to “ease” the more the banks hoard cash. The banks now have $190 billion.

    The Treasury then issues another $100 billion in debt to the major banks. The banks use the $100 billion they got from the Fed prior, to buy the latest debt issue. The Treasury then spends the money. The money and debt once again end up in the major banks. Again the banks hoard all the money except that which is needed for withdrawal and transfer requests. Say the banks net out another $90 billion from the next $100 billion. Now the banks have a total of $180 billion cash ($90 billion net from the first wave of inflation plus $90 billion from the second wave of inflation) plus $100 billion worth of government debt.

    Meanwhile, the Fed is still trying to target NGDP by purchasing government debt from the major banks, because the last round saw the banks hoarding the money, so it didn’t work. So again the Fed buys another $100 billion of debt from the banks. By the same process of hoarding $90 billion, the banks now have a total of $270 billion in cash.

    This process can repeat ad infinitum, with the Treasury issuing new debt to the major banks, who can buy the debt using the cash they got from past monetization of debt by the Fed.

    Is this not possible?

  59. Gravatar of dwb dwb
    10. April 2012 at 13:01

    It’s theoretically possible that capital regs could be adjusted to smooth V at the zero bound, but I’m not convinced.

    BASEL has some proposals out for countercyclical capital rules but i am deeply skeptical. As i’ve said, i view ngdp targeting as demand side policy, and capital regs as supply side policy.

    now, its fine to cut taxes during a recession but capital works a little differently:

    1. Its not the stock of capital but the return on capital that drives loan/investment decisions, which is governed by expected future growth. scarcity of capital is not usually an issue for a healthy bank with plenty of investment opportunities. there are tons of investors willing to invest capital (Buffett invested billions in Goldman at the trough).

    2. You are asking banks to increase leverage at a time when loans are defaulting (increasing leverage BECAUSE loans are defaulting). Feels like absolving them of their excesses and bad loan practices.

    3. Recessions are a time when safe assets are scarce, and investors pull money from risky assets (hence bank runs like Lehman). Allowing banks to have less cushion is counterproductive.

    4. We sort-of had a system like this pre-90s because banks did not have to mark their loans / assets to market. Part of the reluctance by FASB to implement mark to market accounting was exactly to avoid cyclical capital. So banks were in fact more highly leveraged at downturns. It did not provide a cushion or create loans, i would say i just postponed the inevitable. zombie banking did not do much for Japan.

  60. Gravatar of dtoh dtoh
    10. April 2012 at 18:34

    @ dwb

    Its not the stock of capital but the return on capital that drives loan/investment decisions, which is governed by expected future growth. scarcity of capital is not usually an issue for a healthy bank with plenty of investment opportunities. there are tons of investors willing to invest capital (Buffett invested billions in Goldman at the trough).

    It’s not a question of loan/investments, but rather a question of getting people (consumers and businesses) to spend more be it consumption of investment. The way to do this is to raise the price on financial assets relative to real goods and services. Then people exchange financial assets for real goods and services, i.e. increase spending (NGDP).

    You are asking banks to increase leverage at a time when loans are defaulting (increasing leverage BECAUSE loans are defaulting). Feels like absolving them of their excesses and bad loan practices.

    Not absolving, but rather offseting. Reduce leverage by lowering asset prices (higher rates) when then economy is over-heated (above the NGDP target) and increase leverage when NGDP growth is below target.

    Recessions are a time when safe assets are scarce, and investors pull money from risky assets (hence bank runs like Lehman). Allowing banks to have less cushion is counterproductive.

    Yes, but this is what you need to do if you want to smooth out the cycle. And… I have commented before that I think the Fed should provide an explicit backstop for the banks, but do it in the form of an equity put/call rather than through loans.

    We sort-of had a system like this pre-90s because banks did not have to mark their loans / assets to market. Part of the reluctance by FASB to implement mark to market accounting was exactly to avoid cyclical capital. So banks were in fact more highly leveraged at downturns. It did not provide a cushion or create loans.

    The real problem was that you had a lot of financial insitutions which were not subject to asset/equity regulation. Also a lot of the problem with the mark to market was that the banks have to mark to bid. When you get a big drop in liquidity (i.e. wider bid offer spread), banks are marking down assets simply because a lower liquidity rather than any fundamental change in the value of the asset.

    BTW – I thought your comments earlier in this thread on the primary dealers and the repo market were spot on. I was thinking about writing a comment to the same effect, but you did it much better than I could have.

Leave a Reply