Wednesday, August 7, 2013

Market Monetarists: Please Explain!

In recent discussions with David Beckworth about a recent post of his, I've become confused as to what the Market Monetarist (MMist) position really is on some key issues. Commentator "Jared" brought David's post to my attention when he commented about it at pragcap. Since then, the discussion has moved from David's site to pragcap and back again (under a different post). Jared's latest on this has been very clarifying.

Basically the discussion has boiled down to the following: David talks about leaving the interest on reserves (IOR) rate at 0.25% as the rate on short term Treasury debt rises, as a way to increase inflation and nominal GDP (NGDP) and all the things that MMist love. He claims this can be done while committing to a permanent increase in the stock of "base money" meaning that the Fed keeps the stock of excess reserves (ERs) above zero (and doesn't let them decrease). In fact not only CAN it be done, he claims that's the way to do it. Or at least I thought that's what he was claiming until his latest comment. He also brings up cash, but I think Jared effectively dismissed that pretty quickly. Plus David responded to me with one of his previous posts (wherein he claims he did a better job of explaining himself) where he explicitly considers the case of a "Cashless Society." Actually, when pressed, I've found that all the MMist I regularly read claim that cash isn't really important. That includes Nick Rowe, David Glasner, Scott Sumner, and now David Beckworth. Even some hyperinflationists have told me that cash isn't important, even though their arguments seemed to imply that it was. (I'll try to provide a few more links here when I get the chance: I think I saved links to all the occurrences where MMist told me cash wasn't important). Although one hyperinflationist ("bart") took a slightly different tact when he explained that my argument was "made null" by not taking into account "rehypothecation" and "collateralization" (presumably of the ER). He lost me there.

Meanwhile, the discussion with David has now gotten into the realm of the common MMist claim that the endogeneity of inside money really only holds over a short term (six week) basis between Fed meetings (or BoC in Nick Rowe's case). Rowe also touched on that here, and so has Scott Sumner and David Glasner. So have I for that matter, though I was less endorsing the view than explaining it. Jared admits some difficulty with this concept, but ultimately concludes that inside money is ALWAYS endogenous. I don't know if I can follow him there yet. I'm still undecided. However, I'd love to get back to the original issue: how can short term rates rise above the IOR rate when ER > 0? And if they can't, then what's the point of increasing the "base money" stock through central bank (CB) asset purchases (i.e. Quantitative Easing (QE))? Would it have to do with the rest of the yield curve? Is it just for psychological purposes?

Also, I suppose we could go the Japanese route and veer off into "non-traditional" asset purchases by the CB, but then is that really still "monetary policy" or have we entered the realm of "fiscal policy?" I think I know where each side comes down on that question, but it seems a little beside the point, since folks like Sumner have said that it's highly unlikely that these kinds of asset purchases would ever be necessary, while Glasner laid out a specific scheme where the kinds of assets purchased didn't really matter. I don't doubt that if the Fed starting buying non-traditional assets (Cullen Roche often mentions "bags of dirt") they could definitely raise the inflation rate. (I know... I owe you more links).

By all means, read Jared's argument(s). They are very good. I perhaps have muddled the issue by building on Beckworth's cashless society concept to conceptually simplify my argument. My assumptions are:

1. No cash (paper reserve notes or coins)

2. No reserve requirements (reserve requirement = 0%)

3. Only a single commercial bank (or just think of all commercial banks aggregated together)

In such a system "base money" = electronic Fed deposits at the bank (i.e. reserves, or excess reserves, since none are required). Since reserves can only go three places there's limited items our one commercial bank can buy with them. Essentially it boils down to Tsy debt from Tsy auctions. If the Fed were selling (which it's not... according to David's earlier comments, it's committed to keeping stocks of base money elevated: i.e. ER > 0) they could buy Tsy debt or mortgage backed securities (MBS) from them too (in which case the base money would be destroyed). Now when the bank spends on Tsy debt (or when private sector entities pay taxes) the reserves sent to the Treasury General Account (TGA: Treasury's Fed deposit, from which it can spend) end up in the bank again when Treasury spends into the real economy. So our bank can spend base money on Tsy debt at Tsy auctions, but that base money will end up right back on the bank's balance sheet shortly thereafter. It's only a temporary way to get rid of it. That is, unless Tsy runs a surplus (which they're not).

Everything else that the bank buys it does so by crediting the bank deposits of entities in the private sector. This includes any goods or services the bank uses (electric bill, office supplies, etc), employee salaries, shareholder dividends, interest to depositors, Tsy debt held in the private sector, or any other financial assets (bonds, stocks, etc.), or new loans (you can view new loans as the bank buying loan agreements from the borrowers). Crediting bank deposits means creating inside money ex-nihilo.

I think it's pretty clear that in such a world, whenever short term Tsy debt comes up for auction, and the amount that's being auctioned is less than the stock of ER, that the bank(s) will bid the price up and thus the yield down to the IOR rate. If there's any differential, the banks won't hesitate to buy it all. If I'm wrong about that, what am I missing? Could it be that expectations of higher rates are so strong that the bank is not willing to risk a decline in principal even on very short term debt? Could that be possible?

Now perhaps my simplifying assumptions go to far. If so, which ones, and why? I'm asking for help here in understanding. My simplifying assumptions are not essential (Jared does an excellent job arguing this same position w/o resorting to them). I only made them in an attempt to make the problem more clear w/o significantly affecting the macro picture. But perhaps they are not justified. Either case, I think it would be instructive (to me anyway) to know exactly where I've gone wrong here and why, if indeed I have.

Getting back to Beckworth's position, I thought that Jared clearly summed up the dilemma of deciphering exactly where David stands, by pointing this out:

"...our discussion began with this quote from Friedman, ““Oh well, we’ve got the interest rate down to zero; what more can we do?” It’s very simple. They can buy long-term government securities, and they can keep buying them and providing high-powered money until the high powered money starts getting the economy in an expansion.” This certainly sounds like a causal (and temporal) story running from CB asset purchases (injections of high-power money) to economic expansion. But it seems as if you’re now saying that the central bank does not really need to conduct asset purchases to increase the monetary base, it just needs to credibly commit to a permanent increase in the monetary base, which could FOLLOW the inside money creation. Is your view different from Friedman’s" 

Update: Jared and JP Koning chime in on the thread. I especially liked this from Jared:

"If we cut through the jargon, by "different degree of policy accommodation to changes in demand for bank reserves," you mean the Fed will raise or lower the rate at which it provides reserves to private banks, right?" 

Update 2: Sumner makes an amazing statement:

"In a sensible system the base money is endogenous. You set the NGDP target, and the public tells you how much base money they want to hold. I’m all for that."


  1. After our discussion with Beckworth, I’m even more convinced that short-term rates are bound by IOR (given ER > 0). But I can now see a scenario where even under this condition, QE-style asset purchases can be inflationary.

    Imagine this: the Fed pumps in $2 trillion of ER, with IOR = 0.25%. Congress signs legislation forbidding the Fed from raising IOR, or draining the ER in any other way, until deposits have grown to the point where all of the $2 trillion of ER have been converted to required reserves. That would lock-in short-term rates at 0.25% for a very long time (and the larger the monetary base, the longer that would take). Now I think most market participants believe the economy will recover within the next few years, or at least well before deposit growth turns all the ER into RR. And since rates will still be locked really low, then we can rightfully expect inflation sometime in the future. But because nobody knows exactly when the economy will reach full employment and become inflationary, people will spend more now to make sure they purchase goods and services before the inevitable inflation occurs. Increasing the size of the monetary base early on through QE signals how long and severe the inflation might be. The larger the base, the longer and more severe the inflation will be, providing more incentive to spend more now and get it while it’s cheap.

    What do you think?

    1. I like it! I took a slightly different tact here:

      but I came to a similar conclusion: that QE (or the effects of QE, namely ER > 0) could be inflationary. I looked at it by revisiting the "portfolio rebalancing" that David brought up.

      You're more explicitly thinking in terms of an expectations channel perhaps... but I don't think the two views are incompatible.

    2. I like this guy "dtoh"'s explanation:

      (and DOB in the comment below)

      Here's another:

      and another: (nice and short, this one!... and Sumner's response to me below it)

  2. Thanks for these links and the update. I'll take a look.

    Here's another piece of Sumnerian obfuscation for you.

    1. Is dtoh just using different words for the HPE?

      Thanks for your link, BTW, that was a good discussion between the three of you (Franky, you, and Scott).

      "obfuscation" ... Ha!.. it's far from the 1st time I've seen that word associated w/ Scott:

      That's the been the top 1 or 2 hits that comes up for Mike's blog when you Google it for the better part of a year now.

  3. I've been traveling and, thus, fallen behind the constant progression of the econoblogosphere. It looks like I've missed some intense stuff between Cullen, Krugman, and Sumner. Attempting to catch up, I spotted Sumner being quite unlike himself in clarity and brevity:

    "The Fed does monetary policy by setting expectations, and then providing the amount of base money for the monetary base market to be in equilibrium when NGDP expectations are on target. It’s that simple. And that’s fundamentally what they are doing even if they target nominal rates."

    That actually makes sense to me, and it's consistent with Beckworth's claim that the timing of the increase in the monetary base is not important. AND it's consistent with Friedman's (and MMers) support for QE. The QE is simply a method of "setting expectations." It's not necessary because there are alternative methods. If QE works, it works as a communication strategy. The claim that QE is merely an asset swap misses the communicative point.

    What I still don't understand, or just don't agree with, is the idea of the demand for base money. It seems to me that non-banks demand more income, mainly in deposits, not cash; and bank's don't want excess reserves; they're simply stuck with them.

    1. Hey Jared! Welcome back from your travels... wherever they took you. Thanks for the Sumner quote... I think I must have read that one, since it was one of the two articles Cullen commented on (that I'm aware of).

      Nick Rowe has been over at pragcap making comments recently... did you notice?

      He's got a bunch in there... I haven't read them all very carefully yet... he and Cullen have a good thread or two going. (It's creepy ... that's not the 1st time I posted that link to one of my favorite Nick Rowe articles (I did it once before on another blog)... and then within minutes he's there correcting me! ... he says he has "spies all over the internet." LoL!

      Plus, Nick paid me a visit here today:

      I also ask him (towards to bottom) to evaluate my recent discussion w/ Sumner wherein he appears to be saying "Yeah, QE was a mess... ideally we'd just go back to providing just enough base money as the economy called for... pretty much exactly like we did prior to 2008... no need for excess reserves and all that 'mess'... all we need to add is to make NGDPLT explicit! The HPE is 'very weak' at the ZLB. That's what I've been trying to tell people now for 5 years!"

      Hahaha!... well you read it and tell me what you think. Cullen I think is still left a little confused by Scott after his interchange (which was days after mine: the one I paraphrased above):

      ... and I can see why!

      ... if you see there, after Scott moved on to that 2nd post (the one you linked to here), Cullen (after noting my frustrations trying to understand Scott) told me "Tom, you're going to love this one!"

      I have to say... I'm still having a hard time squaring all that.

      You write:

      "What I still don't understand, or just don't agree with, is the idea of the demand for base money. It seems to me that non-banks demand more income, mainly in deposits, not cash; and bank's don't want excess reserves; they're simply stuck with them."

      I'm with you on that 1st part about the non-banks. In terms of the banks however... reserves at not a terrible thing for them actually: 1st off they just get them by accident as a consequence of the Fed buying assets from non-banks during QE... then they swell the banks' balance sheet (not equity, of course) with risk free interest paying (not a lot) assets. Better than a sharp poke in the eye! So I wouldn't say they "don't want them." It's better than NOT having them... and they didn't have to do anything to get them either.

    2. Jared, forgot to mention that Nick has a new one up regarding Steve Keen (coming after his foray onto pragcap to "argue with" Cullen):

      And then this was up previously, but it looked interesting:

      That's on my too-read list: it may lie at the heart of that favorite article I keep posting of Nicks... the one in which I apparently wasn't interpreting "exogenous" or "endogenous" correctly (back to my discussion w/Nick on my "Links" pg):

      (he claims Glasner's use of "endogenous" is different than the one he means... and perhaps for Sumner too)

      OK, more reading for me, and less typing! :)

    3. After reading your duscussion with Sumner about QE being a mess, I think your decsription is right on. I think what he means by "sensible system" is simply one where the Fed communicates a clear and consistent message; and for him, that means announcing an NGDPLT. He think it's currently a mess because of the taper talk, and even before that, Fed Presidents publicly worrying about inflation and bubbles signaling that the FOMC would not let inflation rise above 2%. If the Fed would just commit to a 5% NGDPLT, then markets would know that the Fed would not be tightening anytime soon. Now they're not so sure; in fact, they're betting the Fed will tighten sooner rather than later with the taper talk. But Sumner thinks QE could work (as a communication strategy) if the message of "easy money" wasn't undermined.

    4. Great Jared. Thanks for reading and for your excellent feedback!

  4. I was seeking this convinced info for a long time. Thank you for offering such great information and good luck. updates and articles