Home > Macro/Monetary Theory > Sproul and Sumner on Backing and Quantity Theory of Money

Sproul and Sumner on Backing and Quantity Theory of Money

Mike Sproul recently did a post on J.P. Koning’s blog contrasting his version of the real bills doctrine with Scott Sumner’s (perceived) view of the quantity theory of money.  Sumner’s reply his here.  I think both are partly correct and partly off base.  In short, I think that money is “backed” by real assets, though I think Sproul is missing the true nature of this backing.  But I also think that the price level depends on the quantity of money in circulation.

The big, glaring problem on Sproul’s side is this:

The quantity of convertible FRN’s can be increased by any amount without affecting their value, as long as they are fully backed.

The obvious question this raises, which Sumner points out, is “then how have we had all this inflation?”  Sproul offers this explanation:

Two explanations:
1) The quantity of FRN’s outruns the Fed’s assets by 2%/year on average. This might happen because the Fed issued $100 while getting a bond that is only worth $98 (after transaction costs), or maybe the Fed’s printing and handling costs exceed its interest earnings by 2%/year.
2) The Fed has enough assets to fully cover all its FRN’s with zero inflation, but through habit, ignorance, or painful experience of deflation, the Fed chooses to maintain inflation at 2%, in effect, defaulting on 2% of its obligations per year.

As far as I can tell this is only one explanation.  Regardless of the mental process involved, according to the theory, in order to create secular inflation, the CB has to somehow cause it’s liabilities outrun it’s assets.  The trouble is that this explanation appears to be purely hypothetical with no evidence that this has been the case.  And in fact, this does not seem to be the case.  If this were the case, it seems like it should be relatively easy to show the Fed’s balance sheet leaking assets (relative to liabilities) at a rate roughly equal to the inflation rate at any point in time.  I don’t think that is possible because I suspect that all of the money issued by the Fed has in fact been “fully backed” by Sproul’s definition.  Assuming this is the case, the claim fails the empirical test in a very simple way that is not subject to any kind of difficult identification problem.

On the other hand, I take issue with this statement by Sumner.

We all agree that gold need not be “backed” to have value.  The QTM implicitly thinks of cash as a sort of paper gold.  It’s a real asset that has value because it provides liquidity services.

I guess I agree that gold doesn’t need to be “backed” to have value but I think Sumner is implying something that I don’t agree with which is that gold is not valuable for any reason other than the fact that it provides liquidity.  God has value independent of it’s use as a medium of exchange or store of value and that is what allows it to be used for those purposes.  I think this is evidenced by the fact that it is still valuable even though its use as a medium of exchange is very limited.

If you want, you could argue that this is because of modern industrial applications that didn’t exist thousands of years ago or perhaps that it is because of some lingering emotional attachment to the metal that is functionally obsolete but is the remnant of thousands of years of ingrained social habit.  But let us just conduct a kind of thought experiment regarding the development of gold as a medium of exchange.

Here is one possible scenario:

Imagine we are in a primitive society with some knowledge of metal working.  Some guy discovers a shiny gold metal that is pretty rare.  He thinks it looks cool.  He digs some of it out of the ground and takes it to the blacksmith and says “can you make a cool looking design in the shape a dragon on my shield with this?”  The blacksmith says “beats me, let’s find out.”  He tries it and discovers that he can and it looks pretty cool.  Then this guy goes around with his cool looking gold dragon shield and everybody is like “whoa that guy has the coolest looking shield I’ve ever seen how did he do that?”  They all want shields and swords and plates and cups and what-have-you with the shiny gold metal because it looks cooler than these things look without it.  So they start looking for the metal.  They discover it is fairly hard to find.

Also, these people have quite a bit of other goods like grain and goats and stuff like that and some of them have so much that they are willing to trade some for seemingly frivolous items like a gold-plated shield or candelabra.  So the price of gold becomes fairly high.  Now, in light of this, people eventually realize that gold is very durable, uniform, easily divisible and has a high value to weight/volume.  So these people figure it is more convenient to accumulate some of their wealth in the form of gold than additional grain, goats, etc.  And then, having some of their wealth in that form, they discover that it is easier to carry out transactions using the gold than using the goats.

In this scenario, gold becomes valuable because people think it looks cool and then it becomes a medium of exchange.  It’s ability to provide liquidity services no doubt adds a liquidity premium above what it’s price would be otherwise, but it does not account for the entire value.

Alternative scenario:

On the other hand, imagine a primitive society where somebody discovers a shiny yellow metal and nobody cares, nobody has any particular use for it and it has no value.  But then that guy figures out “hey, this stuff is durable, uniform, easily divisible, and has a high….er, zero value to weight/volume–but if I could convince everyone to use it to trade because it would be easier than using other goods, it could be a high value because of its ability to provide liquidity services, so I just have to do that.”  And then this guy set about convincing everyone to trade stuff for the otherwise worthless metal that nobody cared about and somehow managed to do that and it became valuable purely because of its utility in exchange.

I think it is fairly obvious which one of these is more likely.  But if you are still not convinced that the first scenario is totally plausible just look around at how much of our wealth we currently devote to stuff that looks cool.  Is a Gucci or Prada handbag really that much more functional than a $20 bag from Sears?  Or just imagine that there was a particular kind of cool looking pebble that could only be gathered in small quantities at the tops of very tall mountains.  Do you suppose that people would go to a lot of trouble to climb those mountains and get them and make little necklaces and bracelets and paperweights out of them to impress their friends?  And do you think people who had never climbed a mountain might go on ebay and pay significant sums of money for them….to impress their friends?

Maybe you think that these pebbles would only be valuable because they would be rare.  Well there are pebbles on the tops of mountains.  There are only so many pebbles on Mt. Everest.  Why aren’t they valuable?  Answer: because they don’t look cool!  People value things that look cool.  When something looks cool and is rare, the value tends to be high.  Is it “rational” for people to value things just because they look cool?  Sorry, that’s a fundamentally non-economic question, you’ll have to ask someone else.

This might seem like a pointless debate but it matters a lot.  If you believe that gold became “money” because it had some independent value to begin with, then Sumner’s “paper gold” argument doesn’t make sense.  You are left with a fundamental discrepancy between the nature of “hard” currencies like gold and the nature of “fiat” currencies.  You can’t help but wonder if there might be some kind of backing involved in the latter.

And of course, I think there is backing by real goods.  The problem is that Sproul is missing it (at least most of it) because he is thinking of “money” in terms of base money and looking for the backing on the central bank’s balance sheet.  This is difficult because the assets on the CB’s balance sheet are mainly nominally denominated not “real” assets.  So the CB may stand ready to redeem treasury securities for cash if the value of cash (in terms of treasury securities) falls too much.  But those securities only represent future dollars which they are trading for current dollars.  This would do little to prop up the dollar if, say, people decided they would rather use pesos, or gold, or some other medium of exchange.  The value of dollars and T-bills could both fall together in terms of other goods.

Now, the RBD folks can get around this but only by essentially turning their theory into chartalism/MMT.  In other words, they assume that they can basically pull in their outstanding liabilities (dollars) by liquidating their assets (treasury debt) until the outstanding liabilities fall to the point where they can be further redeemed by some smaller amount of hard assets like gold or silver which the CB is presumed to posess.

In order for this to work out, the government would have to go along by paying off the debt.  To see why, imagine that all money is made up of federal reserve notes (base money) which has been issued in exchange for treasury securities.  Then one day you wake up and notice on the news that everyone has suddenly “lost confidence” in the dollar.  People are switching to pesos and the value of a dollar in terms of real goods is plummeting.  Now let’s see if we can figure out some relationship between money and real goods which might prevent this decline in value.

Start by imagining that, shortly after you turn on the news and just before you run out to trade your life savings for whatever real goods you can get your hands on, someone from the central bank calls.  They say “hey, we’ve got a deal for you, don’t go trade your life savings in for a pack of gum, we’ve got these treasury bills we will sell you at an attractive rate!”

Trouble is, your problem isn’t that you want to hold less money today and more in the future, it is that you are trying to get out of the dollar game all together.  So this isn’t likely to seem very attractive, no matter how low the yield.  So you say “sorry, right now I can still get a pack of gum, in three months when those T-bills mature, even if the yield is 100%, I might not even be able to get that.”

“But wait” he says “there’s more!  Now listen, friend, we’re just trying to claw back enough dollars that we can redeem the rest of them for this silver we’ve got, they’re not worthless, don’t trade them for gum.  You see, we have all this debt from the government.  Now we are going to call that debt, in order to reduce our dollars outstanding.”

“Great” you say “good luck with that, I’m gonna go get some gum now.”

“I don’t think you’re following me, friend.  See, you know how you aren’t really interested in those bills?  Well, it seems not many people are, which means that when the government has to pay off those bills we are holding, guess what they’re gonna have to do.”

“Tax somebody I guess….”

“That’s right friend.  And what are they going to demand in payment of those taxes?  Dollars.  How much do you figure your house is worth right now?  Million bucks?  What about your stock portfolio?  Gonna have any capital gains to report this year?  Well you’re gonna need some dollars then.  And if you don’t, trust me, somebody is gonna need those dollars around mid April.  How many dollars will they need?  Well just about enough to pay off all of this debt that we have on our books, which happens to be just about the number of dollars that are out there–except for the silver of course.  So you might want to hold onto those dollars and not trade them for a pack of gum.”

So there you go, we’ve made it to real goods–wait, almost.

“Oh and just so we’re clear, what do you think is going to happen to you, friend, if you don’t happen to have the dollars when the taxman comes around looking for the money to pay off the government debt that we, the central bank, have on our balance sheet?  Probably not shrug and go away….”

“By the way, taxes are due in about three months.  I’ve got some three-month treasury bills here with very attractive yields.”

Okay, now we have made it to real goods.  The CB’s balance sheet has a bunch of debt offsetting the dollars.  The debt is from the government.  In order to claw back enough dollars to be able to redeem the remainder with the small amount of “real” goods on their balance sheet, they just need the government to tax those dollars out of circulation and use them to pay off those debts.  The connection to real goods comes from the government’s threat of confiscation if taxes are not paid in dollars.  But, as I said, this essentially boils down to chartalism.

And actually, I think taxation does partially explain the value of money.  But MMTers and RBDers are both missing the bigger picture. To see this, notice that we can accomplish basically the same thing by cutting out the central-government middle man.

Imagine that the government collects taxes in gold or some other real good so that this obligation has no bearing on the value of money.  And further imagine that the government has a balanced budget so that there is no treasury debt to make up the assets of the central bank.  So the central bank and the central government are totally separate.  But now imagine that the central bank creates dollars by making loans to consumers and carries those loans as an asset on the books and carries the dollars as an offsetting liability.  By the way, let me point out that “liability” is a perfectly appropriate classification for these dollars because the bank is contractually obligated to settle their accounts receivable (an asset) at a fixed rate when presented with these dollars.

See where this is going yet?  The bank creates money, the money circulates, it is carried as a liability on the bank’s balance sheet and the loans that create the money are carried as an offsetting asset.  And the dollars are backed by those loans.  That is, they are redeemable for assets (the loans) at a fixed rate.  There can’t be any more clear definition of “backing” than that.  But wait!  There are no real goods involved yet.  But that is an easy fix.

Imagine that the bank requires you to put up some real goods as collateral on these loans.  Problem solved.  Now you have money created through the process of bank lending along with a corresponding debt.  The dollars are carried on the bank’s balance sheet as a liability and the debt carried as an asset.  The dollars can be exchanged for the debt at a fixed rate and paying off the debt allows you to keep your collateral.  So basically, your dollars are redeemable for your house at a fixed rate.  The dollar is backed by your house.  And since everybody has houses, cars, boats, businesses, TVs, etc. that are “redeemable” in this way for dollars, you will find that these dollars make a pretty good medium of exchange.

All I have done here is separate the central bank from the central government and combine it with the commercial banks.  Now, once we have noticed that money is backed in this way, there is no reason we must believe that the price level is independent of the quantity in circulation as long as it is thusly “backed.”  The explanation of this can get complicated but notice two things.

1.  Though money is backed by real goods in the sense that it is redeemable for them at fixed rates.  Those goods and rates vary from person to person and time to time.  It’s not like a dollar is always backed by a certain number of oz. of silver and so as long as there are no CB balance-sheet shenanigans, it should always be worth the same as that many oz. of silver.

2.  The quantity of money in circulation at any point in time depends on people’s willingness to hold debt.  This depends on interest rates.  Since this amounts to promising to pay dollars in the future in exchange for dollars today, it is possible to push more dollars into the system today that people then use to buy stuff and push up prices.

This is where it starts to get tricky and expectations about the future stance of XYZ become important.  I don’t want to start down that road here but just let me reiterate that money doesn’t have to be totally unbacked bits of paper with no actual connection to real goods in order for some relationship to exist between the quantity of money and the price level.

 

 

Advertisements
Categories: Macro/Monetary Theory
  1. June 6, 2014 at 10:37 pm

    Dueling thought experiments:
    Suppose a landlord collects rent of 50 oz. of silver per year, in perpetuity. At a 5% interest rate, this makes his land worth 1000 oz. (=50/.05). The landlord’s balance sheet would show 1000 oz. net worth on the liability side, while the asset side would show either 1000 oz. of land, or alternatively, 1000 oz. of “rents receivable”. This is shown in line 1 of the T account in Table 1.

    Table 1
    ……ASSETS…………………………………..LIABILITIES
    1) 1000 oz. rents receivable…………1000 oz. net worth
    2) ………………………………………………+200 oz. bank notes (dollars) spent on groceries
    3)……………………………………………….-200 oz. net worth

    In line 2, the landlord buys 200 oz. worth of groceries, not with silver, but by writing out 200 paper IOU’s (bank notes) that each say “Good for 1 oz. rent on my property”. Assuming the grocer rents from the landlord, or knows someone who does, he would accept the landlord’s bank notes at par. Note that we could change “landlord” to “government”, and “rents receivable” to “taxes receivable”, and this story would be a good description of the way that early paper moneys were issued in 1690 in the American colonies.

    Table 2
    ……ASSETS…………………………………..LIABILITIES
    4) ………………………………………………+400 oz. bank notes (dollars) spent on groceries
    5)……………………………………………….-400 oz. net worth (Net worth is now 400 oz.)

    In Table 2, we suppose that the landlord issues another $400 of bank notes and spends them on groceries. This reduces his net worth to zero, but note that the $600 of bank notes is still adequately backed by 1000 oz. worth of taxes receivable, so each bank note is still worth 1 oz., even though the quantity of money has tripled. The backing theory is right and the quantity theory is wrong. If the landlord wanted, he could sell his land for 1000 oz., redeem every bank note for 1 oz. each, and still have 400 oz. left over. Or he could directly sell 600 oz. worth of his land for the $600 of bank notes, and then retire the bank notes.

    Table 3
    ……ASSETS…………………………………..LIABILITIES
    6) +2000 oz. of land……………..……….+$2000 bond issued to pay for the land

    In Table 3, the landlord buys another 2000 oz. of land by issuing a $2000 bond. Note that the bond is denominated in dollars, but since $1=1 oz., nothing important would change if the bond were for 2000 ounces instead of $2000. His issuance of this bond might shake peoples’ confidence in him, and his bank notes might lose value as a result, but to keep things simple I will assume that this effect is negligible.

    Table 4
    ……ASSETS…………………………………..LIABILITIES
    7) ………………………………………………+600 oz. bank notes (dollars) spent on bonds
    8)……………………………………………….-$600 of bonds re-purchased.

    In Table 4, the landlord prints 600 new bank notes and uses them to conduct an open-market purchase of his own bonds. This doubles the quantity of bank notes from 600 to 1200, but each bank note is still fully backed by the landlord’s 3000 oz. worth of land, so $1 is still worth 1 oz. To make this open-market purchase resemble the Fed’s open-market purchases, the landlord would have to conduct this purchase through a “central bank” of his own creation. All he would have to do is designate his kitchen as his central bank. Then his kitchen prints $600 of bank notes, and uses them to buy $600 of the bonds that the landlord had previously issued to the public. This does not change anything that matters, just as it did not matter when the US government created the Fed as a separate entity.

    Table 5
    ……ASSETS…………………………………..LIABILITIES
    6) +300 oz. of silver……………..……….+$300 of bank notes spent on silver.
    7) -300 oz. of silver……………..……….-$300 of bank notes redeemed for silver.

    Line 6 of Table 5 shows the landlord conducting an open-market purchase of silver. Once again the new bank notes are fully backed, so $1 is still worth 1 oz., but the landlord’s stock of silver allows him to offer silver convertibility to the public. This might continue for years, until one day all 300 oz. of silver are paid out, and the bank suspends silver convertibility (line 7).

    The suspension of silver convertibility has practically no effect on the community, but all Hell will break loose among academic economists. They will start saying things like
    “In what sense is cash a liability of the (landlord)? I thought once we left the (silver) standard the (landlord) was no longer required to redeem dollars?”

    “Dollar bills are not debt. The (landlord) is not required to redeem them for anything but themselves. That’s not debt.”

    “They’ll spend the new cash almost immediately after being paid for the bonds. That’s the famous “hot potato effect” that underlies the quantity theory.”

    Eventually, things will get really crazy and they’ll start saying things like “MV=PT”.

    • Free Radical
      June 7, 2014 at 1:45 am

      I’m not sure what this is meant to prove. You are showing that the accounting can work out when there is a silver standard. I agree. As far as I know, pretty much everyone agrees. But you have done nothing to show how the value of a dollar is maintained when convertibility is suspended, you just seem to assume that everything carries on as though convertibility were still in effect even though it no longer is. The fact is that once you leave the silver standard the landlord is no longer required to redeem dollars for silver. That is a meaningful distinction.

      What you might call the “pure QTM” types seem to believe that some magical force kicks in to keep the value of the currency positive and relatively stable even though there is nothing backing it.

      What you seem to believe is that we act as though there is a contract backing it even though there isn’t one.

      What I am saying is that there IS a contract backing it! Or rather, there are many contracts. You don’t have to imagine either of these things, you just have to notice that there are real (not imagined) contractual obligations to redeem dollars for real goods that form the foundation of the entire banking/money creation mechanism. It just isn’t one single universal standard.

      Also when you say “so each bank note is still worth 1 oz., even though the quantity of money has tripled. The backing theory is right and the quantity theory is wrong” you haven’t proven anything wrong, you assumed that the money supply tripled and the price level didn’t change but you did that in the context of a model with a commodity standard so the conclusion that the value of the dollar remained at the level of that standard is not very persuasive. I suspect you could do something similar without a commodity standard but when you do it will amount essentially to assuming that the value doesn’t change not proving it doesn’t change. The fact of the matter is that the price level has changed with changes in the quantity of money in ways that are broadly consistent with the quantity theory but directly contradict your claim that the price level doesn’t change unless the backing is impaired. This doesn’t mean money isn’t “backed” but your theory of the price level seems to fail any attempt at empirical scrutiny. This is because you are missing the true nature of the backing.

      Finally, MV=PT (or PY as I tend to write it) is a very helpful equation for thinking about these issues, there’s nothing crazy about it, you just have take it for what it is and not try to make it anything more or less.

      • June 7, 2014 at 2:37 am

        ” The fact is that once you leave the silver standard the landlord is no longer required to redeem dollars for silver. ”

        But the landlord still redeems dollars for 1 oz worth of rent. That’s what ties the dollar to 1 oz of silver, even when explicit silver convertibility is suspended. In the same way, the Federal government still redeems dollars for taxes due.

        ” You don’t have to imagine either of these things, you just have to notice that there are real (not imagined) contractual obligations to redeem dollars for real goods that form the foundation of the entire banking/money creation mechanism.”

        That puts the cart before the horse. I know you don’t like my stock/money analogy, but the value of GM shares is determined by GM’s balance sheet, and not by a set of private trading contracts written by people who did not issue the shares in the first place. Same for money.

        “you assumed that the money supply tripled and the price level didn’t change but you did that in the context of a model with a commodity standard so the conclusion that the value of the dollar remained at the level of that standard is not very persuasive.”

        All so-called fiat money is actually on a commodity standard, in the sense that there are physical goods underlying the assets of the money issuer. It’s just that people don’t see the implicit convertibility. For example, people who see that the landlord’s bank notes are convertible into physical silver will say “that’s a commodity standard, so your conclusion isn’t persuasive.” But then it’s easy to forget that even with silver convertibility suspended, the “rent convertibility” means we are still on a commodity standard, so the QT still won’t work.

        There are lots of empirical studies that explicitly tested the QT vs BT during lots of historical episodes, not just the 1970’s in the US. I list them in my paper entitled “There’s No Such Thing as Fiat Money”. They include studies by Sargent, B. Smith, Calomiris, Siklos, Bomberger and Makinen, etc.

        MV=Py is more misleading than useful.

      • Free Radical
        June 7, 2014 at 3:47 am

        “But the landlord still redeems dollars for 1 oz worth of rent. That’s what ties the dollar to 1 oz of silver, even when explicit silver convertibility is suspended. In the same way, the Federal government still redeems dollars for taxes due.”

        Yes and that is why this is not a model of fiat money, it is a model of fixed convertibility. Changing it from one oz. of silver to one oz. of silver worth of rent does not end the convertibility, it just changes the form in which it is convertible.

        “That puts the cart before the horse. I know you don’t like my stock/money analogy, but the value of GM shares is determined by GM’s balance sheet, and not by a set of private trading contracts written by people who did not issue the shares in the first place. Same for money.”

        No that is wrong. There is a contractual relationship that entitles the holders of stock to a given share of the assets of the company. This is not the case for holders of dollars. You are just imagining that it is the case. But the fact is that that obligation is written down and well recognized in the first case and not the second and this is an important distinction.

        “All so-called fiat money is actually on a commodity standard, in the sense that there are physical goods underlying the assets of the money issuer. It’s just that people don’t see the implicit convertibility. For example, people who see that the landlord’s bank notes are convertible into physical silver will say “that’s a commodity standard, so your conclusion isn’t persuasive.” But then it’s easy to forget that even with silver convertibility suspended, the “rent convertibility” means we are still on a commodity standard, so the QT still won’t work.”

        Yes, you created a hypothetical situation in which there was still a commodity standard after silver conversion was suspended because you assumed it was replaced with a rent standard. But that doesn’t prove the quantity theory wrong in any way. The irony is that what you are describing is very close to my theory where your “rent” is my mortgages/consumer credit/etc. (thought this does not maintain the link to the value of silver as you claim) but I don’t think you are seeing it because you are determined to end up with the value of a dollar in terms of silver staying constant even though that clearly hasn’t been the case.

        I don’t think there is anything misleading about MV=PY unless you try to apply it to something other than money (like stocks) but this is probably not worth arguing about.

      • June 7, 2014 at 12:34 pm

        Mike,

        “..the value of GM shares is determined by GM’s balance sheet, and not by a set of private trading contracts written by people who did not issue the shares in the first place. Same for money.”

        Actually, it’s pretty essential to this view of money that it involves a tripartite arrangement with the money issuer and not just the debtor and money holder. Circuit theory is a useful model here. The value of money in the hands of the workers (and why they will accept in payment) derives from the fact that the firm is contractually obliged to deliver money back to the bank. You need all three parties to be involved to arrive at the value.

    • August 12, 2014 at 12:29 am

      I guess finding useful, reliable inriamftoon on the internet isn’t hopeless after all.

  2. June 7, 2014 at 1:30 am

    Errata: “This reduces his net worth to zero” should say “to 400”

  3. Tom Brown
    June 7, 2014 at 2:44 am

    Have either of you ever seen this paper?

    http://www.minneapolisfed.org/research/sr/sr218.pdf

    On a related note, Jason Smith is adapting information theory to construct a theory of the price level. I’m curious of either of you have any comments on that approach. Here’s a good example:

    http://informationtransfereconomics.blogspot.com/2014/05/models-matter.html

    and here

    http://informationtransfereconomics.blogspot.com/2014/05/limits-of-information-transfer-model.html

    I find it intriguing that some standard macro results seem to fall out of the information transfer model as special cases. Thoughts?

    • Free Radical
      June 7, 2014 at 3:48 am

      Tom,

      Let me look into it.

    • Free Radical
      June 7, 2014 at 6:23 am

      Tom,

      I took at look at the kocherlakota paper, I basically skimmed it so I didn’t get all the ins and outs but I think I have a pretty good handle on what is going on there so I will try to comment and if I see a need to look at it more closely later I will. Some of these points are pretty subtle but I will see what I can do.

      The basic point of the paper, summed up in the shortest and simplest possible way, seems to be that the only purpose of money is to transfer information about past trading. In a sense I can get behind that thesis but I would put a big asterisk after the word “only.” I think it is right to think of money as a sort of placeholder representing services rendered (or good produced/traded) in the past and redeemable for goods/services in the future. I have two gripes about the approach to making that point which is taken here.

      1. It is unnecessarily complicated to make that point. This is just sort of how economics works these days though, so that’s just me complaining off in my corner.

      2. It doesn’t explain how or why money works for that purpose. Furthermore, it assumes something which bugs the hell out of me right from the beginning, namely that “of course, money is intrinsically useless.”

      I would argue (as I did in this post) that originally, money was made up of things that were “useful” (or at least intrinsically valuable) and that this is what makes it possible for them to carry out the functions we normally associate with money. This means that paper money has a different character than “hard” currency and “fiat” money different from convertible paper money and the meaning of each and the reasons they work as money are important. But there seems to be this notion that fiat money is obviously nothing other than intrinsically worthless pieces of paper that we have somehow arbitrarily decided to assign value to and so through a process of backward induction, we have arrived at the conclusion that this must be the explanation for all monies. I think that explanation fails for both.

      However, what we have here is a model constructed around the assumption that money has no intrinsic value and that means that it can’t perform any function in the model other than convey information. The only way it can possibly matter is by allowing us to construct strategies that are contingent on it. So it shouldn’t be surprising to find out that if we just knew all of information, and could make strategies contingent on that information, there would be no further role for money.

      But all of these equilibria (at least as far as I can tell) are still constructed using beliefs that I consider suspect like: as long as every other young person so far has traded goods for dollars, assume that all will in the future. It’s possible to construct a Nash Equilibrium that way but there is a kind of circular reasoning involved that I find less than compelling and which can’t be avoided when you start by assuming that money is intrinsically worthless.

      Now for the record, I think we are moving toward a system of “money” which is purely informational (as opposed to physical), where your bank just keeps track of how much money you have and transfers it from one person to another as goods and services are exchanged. So in that respect, we are sort of moving from a world of physical currency into one where money truly is essentially just memory. But make no mistake, there will be contractual obligations attached to that money and connected to real goods, i.e. if you can’t accumulate enough of these dollars in your digital bank account to pay off your mortgage, the bank will take your house. It won’t just be people saying “well we see here that you mowed the Smiths’ lawn so go ahead and have ten gallons of gas, you deserve it.”

      • Tom Brown
        June 7, 2014 at 1:33 pm

        Mike, thanks so much for your review of the Kocherlakota Fed paper! Those are some helpful insights. One of the interesting things you’ve stated here: “So in that respect, we are sort of moving from a world of physical currency into one where money truly is essentially just memory.” is also interesting to me in light of Jason’s information theoretic approach. I think it’s fair to say that in searching for the best measure of money as information in his model, he picked one that empirically fit the data best, and in that regard he found that the physical currency portion of the monetary base was the best fit. Somewhat ironic, no? The ratio of the currency portion of the monetary base to NGDP he found to be a descent determinant of how small changes in the money supply would effect the price level across several different countries and within the same country at different periods of their history. I.e., this ratio best fit the information transfer model curve. For example:

        http://informationtransfereconomics.blogspot.com/2014/05/switzerland-and-sweden.html

        So in his estimation Sweden is in a position in which the price level should be responsive to changes in the money supply (which, if true, perhaps puts even more blame on their CB for their recent deflation). Switzerland, in contrast, should have a price level which is unresponsive to changes in their money supply (If I’m reading this correctly). This one shows the underlying theoretical surface a bit better:

        http://informationtransfereconomics.blogspot.com/2014/01/it-really-does-seem-to-be-about-size-of.html

        (sometimes when Jason writes “MB” he means currency… he’s a little sloppy about that)

      • Free Radical
        June 8, 2014 at 2:12 am

        I can’t really tell what Jason is doing there. It seems like he is just saying that the price level is correlated with the size of the money base. That is what I would expect but I don’t see how it fits into an “information transfer model” it just seems like it boils down to MV=PY to me. I have recently tried to show that something like the “flattening out” he observes though. I characterize this in terms of the demand for loans and essentially what it boils down to is that when you hit the ZLB (or the interest on reserves lower bound) and keep expanding the base, you can’t move any farther out along the demand for loans and so it just ends up accumulating as bank reserves rather than being loaned (“multiplied”). You could see if this were the case by looking at broader measures of money.

        Of course, it’s also possible that as the money supply increases, velocity decreases which I would also expect to be the case. Real money balances should expand to the point where the marginal liquidity premium on money is equal to the nominal interest rate. When the nominal rate is lower, real money balances should be higher (velocity lower). I believe this result has been obtained empirically by others.

      • Tom Brown
        June 9, 2014 at 6:10 pm

        Mike F., thanks for taking a look at Jason’s page, and for your thoughts. I pointed him to your post as well and asked him what his thoughts were on the backing theory. This was his response:

        http://informationtransfereconomics.blogspot.com/2014/06/money-unit-of-information-and-medium-of.html?showComment=1402164679857#c5292041805245727671

        I’m definitely not an expert on Jason’s theory or the ITM, but I know a tiny bit about basic information theory, and I think that’s where all the logarithms creep into Jason’s formulas. For example, if you know that a particular event has three possible outcomes with probabilities P1, P2 and P3, such that P1 + P2 + P3 = 1, and P[123] all must be on the interval [0,1], then the maximum information carrying capacity of this event is

        -(P1*log(P1) + P2*log(P2) + P3*log(P3))

        If you take the logs to be be base 2, then this measures information in bits. For example, flipping a fair coin is:

        -(0.5*log(0.5) + 0.5*log(0.5)) = 1 bit

        If the coin is unfair and comes up heads 100% of the time, then you have

        -((1-epsion)*log(1-epsilon) + epsilon*log(epsilon)) as epsilon approaches 0, which is 0 bits

        A roll of an 8 sided fair die is -8*(2^-3)*(log(2^-3)) = 3 bits.

        This is equivalent to three flips of a fair coin, etc.

        His basic argument rests on the idea that spending money transfers information from demand to supply, and we can use information theory to analyze this process (e.g. calculate the maximum possible information capacity of this process, etc).

        What I find interesting, is that this approach has some concrete things to say about a few subjects which I find very mysterious in econ: e.g. expectations, representative agents, utility, cause and effect, etc. At this point it’s an open question if this approach is valuable or not, but to me it’s at least intriguing.

      • Free Radical
        June 9, 2014 at 11:06 pm

        So part of what I’m trying to do in this post is point out that “backing” and QTM are not exactly mutually exclusive. It seems like people see it as either Mike Sproul’s backing (where a dollar is equal to a constant quantity of some asset) or the standard money-is-backed-by-nothing-and-is-only-valuable-because-of-use-in-exchange theory. This dichotomy seems to be evidenced in Jason’s response:

        “Regarding the backing theory, I am pretty sure that 100% backing is wrong — mostly because the quantity theory of money is at least partially correct.”

        My point is that the value of the dollar is not anchored to the gold/silver in the vaults of the central bank, but it is anchored to a wide range of assets throughout the economy through various debt contracts. This means that you can still pump more money in and it circulates in the way imagined by the QTM and pushes down prices but the value stays positive and (relatively) stable because it is backed by those real assets. For instance take this:

        “When a monetary system is established (an maybe after inflationary episodes or in the case of monetary “phase transitions” e.g. after WWII), there really does need to be something to establish the scale of the money and lock in the coefficients of the model.”

        And this:

        “It seems like backing might become important when the information-carrying capability of money becomes ineffective or questioned.”

        Putting aside the fact that I’m not exactly sure what he means by the information-carrying capability, one must wonder why this wouldn’t be constantly in question. If the theory is that it works because nobody questions it then I find that theory unconvincing. I say that if you question the value of money, you find that it doesn’t evaporate because it is secured by a vast network of contractual obligations to pay specific amounts of money at specific points in the future or else forfeit real goods. It is clear that that network of contracts exists, there is no debating that. Yet we choose to ignore them and opt for a theory of magically levitating money instead because it is more elegant on paper. (Or maybe it is because we can’t separate this type of backing from that proposed by people who insist on predictions about the price level that are clearly violated….)

      • Tom Brown
        June 10, 2014 at 6:36 pm

        Mike F., thanks again for your thoughts on Jason’s comment. I think I see what you’re getting at, regarding backing theory not necessarily being mutually exclusive to QTM.

        In case you are curious what is meant by “information carrying capability” these two posts cover the basics of the concept (building on my super simple information theory examples above):

        http://informationtransfereconomics.blogspot.com/2014/03/apples-bananas-and-information-transfer.html

        http://informationtransfereconomics.blogspot.com/2014/03/how-money-transfers-information.html

        Equation 6 in that latter post, and the fact that it reduces to the QTM when kappa = 0.5 is the key takeaway for me. (kappa > 0.5 in most cases, and is closer to 1 in Japan, though it can change over time).

      • Free Radical
        June 10, 2014 at 8:35 pm

        Tom,

        No problem, I’m here to help =). I will try to take a look at those posts but might be a while as I am now discovering a whole new world of monetary circuit theory (as indicated in my latest post). Thanks for the interesting links though.

      • Tom Brown
        June 10, 2014 at 10:22 pm

        Mike, you might find this interesting too:

        http://www.themoneyillusion.com/?p=26884#comment-351878

        Part of an exchange between Smith and Sumner in the comments to Scott’s response to Sproul’s article. He discusses what functional form gives the best empirical fit. the log P ~ k * log M form falls out of the ITM. The whole thread between them is interesting.

  4. June 7, 2014 at 5:32 pm

    Mike F. and Nick:
    “this is not a model of fiat money, it is a model of fixed convertibility. Changing it from one oz. of silver to one oz. of silver worth of rent does not end the convertibility, it just changes the form in which it is convertible. “
    My point is that so-called fiat money is actually backed and convertible. Quantity theorists make the mistake of seeing that explicit gold convertibility has been suspended, and they don’t see the other channels (tax, bond, rent, loans) through which the money is convertible, and they wrongly conclude that the money has no backing.
    “No that is wrong. There is a contractual relationship that entitles the holders of stock to a given share of the assets of the company. This is not the case for holders of dollars. You are just imagining that it is the case. But the fact is that that obligation is written down and well recognized in the first case and not the second and this is an important distinction.”
    Suppose a bank issues 100 paper dollars against which it holds 100 oz worth of assets. Those dollars will be worth 1 oz each. Then people start trading and making contracts that next week, 20 pounds of corn will be delivered for $20. If the 100 oz of assets is lost, then the corn contract is not enough to keep up the value of the dollar, especially when you consider that people can pay with derivative money, that is, with money that is not one of the original dollars, but is just a claim to $1 (e.g., checking account dollars, credit card dollars, gift card dollars, etc)
    “Yes, you created a hypothetical situation in which there was still a commodity standard after silver conversion was suspended because you assumed it was replaced with a rent standard. But that doesn’t prove the quantity theory wrong in any way. “
    If the money is backed and convertible, then it is described by the BT, not the QT.
    “The irony is that what you are describing is very close to my theory where your “rent” is my mortgages/consumer credit/etc. (thought this does not maintain the link to the value of silver as you claim) but I don’t think you are seeing it because you are determined to end up with the value of a dollar in terms of silver staying constant even though that clearly hasn’t been the case.”
    My rent example properly designates rents receivable as the asset of the money-issuer on the liability side of the issuer’s balance sheet, and it properly places the dollars issued. Given that rent payments are a form of silver convertibility, then it does maintain $1=1 oz.

    • June 8, 2014 at 9:11 am

      “Suppose a bank issues 100 paper dollars against which it holds 100 oz worth of assets. Those dollars will be worth 1 oz each. Then people start trading and making contracts that next week..”

      Forget the corn. Suppose those private contracts are loans of the original dollars. And we end up with private banks holding $10,000 of loans against $9,000 of deposits and $1,000 of equity. So now there are debtors with total debts of $10,000 chasing total money of $9,100. Even if the original bank loses its 100 oz, those debtors still need to get hold of more money than is available. As long as the original bank’s money can be used to repay debt created by the other banks, it will retain a value.

      • June 8, 2014 at 3:02 pm

        Nick:
        I addressed this in my “Short squeezes..” post on JP Koning’s blog a few weeks back. All the banks have to do to avoid being stuck in a situation of needing more money than is available, is to put a suspension/alternate settlement clause on the money they issue. For example, each paper dollar could say “Bank will pay either 1 oz of silver OR bonds worth 1 oz, with a 30-day delay if a bank run is happening”

      • June 8, 2014 at 3:14 pm

        OK, but it’s not the banks being squeezed in my example – it’s the non-bank borrowers. If anything the banks are benefiting from the squeeze, because it’s what creates demand for their paper.

      • Free Radical
        June 8, 2014 at 6:24 pm

        YES! Thanks Nick!

      • June 10, 2014 at 7:42 pm

        Nick:

        The usual meaning of short squeeze refers to what happens to the bank when everyone wants their money at once. It doesn’t refer to the money demand of borrowers, and anyway, the borrowers don’t have to pay their loans in base money. They can pay it in various derivative moneys.

        Some problems with this “contractual obligation” theory of money:
        1) All value theories are forward looking, but this is backward looking. We find the value of apples, houses, stocks, and bonds by looking at their future services, not past contracts
        2) Stocks and bonds (like money) are valued according to the assets and liabilities of their ISSUER. This obligation theory makes valuation depend on the assets and liabilities of anyone who makes a contract. It certainly wouldn’t pass muster with accountants.
        3) Past contracts can be paid in derivative money, in lieu of base money.

      • Free Radical
        June 10, 2014 at 8:40 pm

        1. No it is not backward looking. Money has value because people know that people are obligated to use the money to pay off debt in the future. The fact that this obligation took place in the past does not make it backward looking.

        2. Money is not a bond, this point is based on a false premise.

        3. This is not a theory of why base money has value (although that is included in the theory) it is a theory of why “a dollar” (or what have you) has valuable. This applies to everything that can be considered a dollar which can most easily be defined as whatever can be used to pay off a bank loan without conversion into something else, or essentially base money and derivative money.

  5. June 7, 2014 at 6:25 pm

    Errata:
    “on the liability side of the issuer’s balance sheet, and it properly places the dollars issued.”

    should say:

    and it properly places the dollars issued on the liability side of the issuer’s balance sheet.

  6. Max
    June 11, 2014 at 11:22 pm

    Good post. I like how you peeled back the layers of backing: base-money -> t-bills -> taxes -> private property.

    • Free Radical
      June 11, 2014 at 11:31 pm

      Thanks Max!

  7. Tom Brown
  8. June 12, 2014 at 2:47 am

    Mike F.:
    If the value of money is determined by the fact that past obligations create a demand for money, then the central bank’s assets become unnecessary. If that were true, we should see lots of central banks that hold no assets, while their currency maintains its value due to past obligations. There are no such banks. All central banks (and private banks) hold assets, because it is those assets that back the money issued by those banks.

    • Free Radical
      June 12, 2014 at 4:50 am

      I don’t think this makes any sense. Most of the assets central banks hold are exactly the type of debt I am talking about. The central bank serves a role in the banking system. Essentially that role is to regulate the extent to which credit can expand, exactly how they do that is sort of involved (I have done some posts on it) but I don’t really feel like you are following me at all if you are suggesting that my theory that money is backed by debt implies that the assets of the central bank are irrelevant when most of those assets are debt. Perhaps you are confusing my theory with a different one….

  9. July 26, 2014 at 2:32 pm

    Wonderful blog! I found it while searching on Yahoo News.
    Do you have any tips on how to get listed in Yahoo News?
    I’ve been trying for a while but I never seem to get there!
    Appreciate it

  10. August 30, 2014 at 12:46 am

    I read a lot of interesting articles here. Probably you spend a lot of time writing, i know how to save you a lot
    of work, there is an online tool that creates readable, google friendly posts in minutes, just type
    in google – laranitas free content source

  11. January 7, 2015 at 3:51 am

    Interesting articles you post here, i have shared this post on my fb

  1. April 10, 2015 at 1:44 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: