Tuesday, March 7, 2017

Real and Nominal Layers

JP Koning has written a thought-provoking post on the monetary system ("Money as layers"). (Based on the dozen or so posts I've read from him earlier, JP strikes me as a thoughtful, balanced writer who comes up with interesting, and often out-of-the-ordinary, topics. I've learned a lot from him. Johan, thanks for notifying me of this latest post of JP's!)

JP uses the layered structure of dreams from the film Inception as a metaphor for our monetary system. He says:

Like Inception, our monetary system is a layer upon a layer upon a layer. Anyone who withdraws cash at an ATM is 'kicking' back into the underlying central bank layer from the banking layer; depositing cash is like sedating oneself back into the overlying banking layer. 
Monetary history a story of how these layers have evolved over time. The original bottom layer was comprised of gold and silver coins. On top this base, banks erected the banknote layer; bits of paper which could be redeemed with gold coin. The next layer to develop was the deposit layer; non-tangible book entries that could be transferred by order from one person to another. Bank customers could "kick" out of their deposits and back into banknotes, and then kick out of banknotes into coin. Conversely, they could sedate themselves from coin into notes and finally deposits. 

 (Note the terms 'sedate' and 'kick', as I'm going to use those when I build on JP's idea. 'Sedating' is moving further away from the bottom layer, while 'kicking' is moving towards the bottom layer.)

If we take reality to be the bottom layer in Inception, then it would make sense to have the bottom layer in our economy comprise of real goods and services. We can call it the real layer. JP only talks about the bottom, or foundation, layer of the monetary system  not the economy. To him that bottom layer used to be the precious metals (to me this is different from gold and/or silver coins, although JP seems to mix these together?) and is nowadays banknotes issued by the central bank.

I would place precious metals in bullion form on the real layer. Gold and silver coins, with a face value higher, or potentially higher, than the price of the metal itself (intrinsic value), used to form what I'd like to call a nominal layer ("credit layer" wouldn't probably be a bad name, either). Banknotes[1], deposits, etc, are all nominal layers.

So, we have the real layer and we have nominal layers. Just like dreams are connected to reality, so are the nominal layers connected to the real layer. As an anonymous reader said under JP's post: "People consider these subordinated assets to be claims on real commodity wealth". What provides this connection is the (nominal) price we set on goods and services when we trade them. It is the unit of account which forms a link between the real layer and any nominal layer (see my first post for how I view the unit of account).

Just like dreams can feel very real, so can nominal wealth feel very much like real wealth. And for a good reason: most of the time, an individual can convert nominal wealth into real wealth  that is, kick himself out of a nominal layer into the real layer. This can happen either directly, or indirectly via other nominal layers closer to the real layer than the starting layer.

Here the layer, or hierarchy, metaphor doesn't work that well: it is possible for an individual to kick out from many of the nominal layers directly into the real layer, without visiting any possible layers in-between. This, often but not always, means that the counterparty to the trade, the one who sells the good or the service, sedates from the real layer directly into that specific nominal layer, say, a commercial bank deposit. (An example of when this is not true: the credit entry on the seller's account brings the account balance to zero from a previous negative/debit balance (an overdraft). In that case it is hard for me to see how we could say that the seller ended up on that layer.)

Where JP isn't too clear is what is the quality, or qualities, that separates one layer from another. Does it have to do with the (perceived) riskiness of the layer, the institution behind the layer, or even with the chronological order in which the layers appeared, or seem to have appeared? As Johan Meriluoto points out (and JP confirms), this idea of layers is similar to Perry Mehrling's idea that "the system is hierarchical in character".

When it comes to what makes one layer different from another, Mehrling seems to focus on the institutions (although risk as a factor lurks always in the background). He gives an example of a simple hierarchy of a central bank, commercial banks and security dealers. Mehrling's hierarchy fluctuates: one hierarchical level, or layer, can look (qualitatively) much like another in good times, while under market stress the hierarchical character gets amplified. Thus, in what way and to what extent the layers (are perceived to) differ from each other varies with time.

It would be hard to argue that the layers are only about risk. A commercial bank deposit which is covered by a public deposit guarantee can arguably be viewed as being, at all times, on par with currency. And when JP suggests that a (private) deposit is somehow a subordinated layer compared to a (private) banknote, riskiness as a differentiating factor disappears entirely from the picture  after all, the risk of theft or misplacement makes a banknote less secure than a deposit.

When it comes to this question of priority between banknotes and what he calls "non-tangible book entries", I find myself at odds with JP. What he says makes some sense if we consider the recent history of banking as it applies to the general public (say, from the 18th century onward), but I'm not at all convinced it is true about the early history of monetary systems (which is, unfortunately, not known). What makes it untrue even if we only consider the recent history is that the book entries have existed, all of the time, side by side with banknotes. They were not the "next layer to develop" after banknotes.

That might very well be a minor detail for JP. What makes it somewhat important for me is that as I'm trying to build a monetary system from scratch (as it might have happened thousands of years ago, although I'm not trying to make a historical argument; this is a thought-experiment which I believe can help us understand the current system better), I find it makes sense to view/describe/understand banknotes in terms of the non-tangible book entries. The latter should be logically prior to the former. (You might get a better idea of what I mean by this if you read first Part 3 and Part 4, and then Part 7 of my series.)


[1] Banknotes, like gold coins, might on some occasions have an intrinsic value higher than face value. When this is so, they kind of "kick" themselves back on the real layer.


  1. Great post.

    "Where JP isn't too clear is what is the quality, or qualities, that separates one layer from another. Does it have to do with the (perceived) riskiness of the layer, the institution behind the layer, or even with the chronological order in which the layers appeared, or seem to have appeared? "

    I was thinking of the medium, i.e. banknotes vs deposits as separates layers.

    "What he says makes some sense if we consider the recent history of banking as it applies to the general public (say, from the 18th century onward), but I'm not at all convinced it is true about the early history of monetary systems (which is, unfortunately, not known)."

    I was thinking about the British system, where if I remember correctly banknotes were the dominant exchange medium issued by banks after the founding of the BoE in the 17th century, but by the 19th century deposits had begun to dominate.

  2. Thanks, JP!

    Yes, those are separate layers. I wasn't too clear about it, but what I was after when I talked about the qualities is what gives one layer a place higher or lower in the "hierarchy of layers"?

    I actually thought you might have had the British system in mind -- even the famous London goldsmiths :-) My point about banknotes vs. deposits relates to my point about the qualities. In this case you seem to imply that the dominant (i.e. more widespread?) medium should be higher up in the hierarchy/closer to the foundation layer? In your blog text, on the other hand, it sounded almost like deposits were "invented" after banknotes, and that gave banknotes a priority.

    Perhaps your point is more about there being different layers than there being a hierarchy of layers? At least in the case of banknotes vs. deposits? In Inception, there seems to be a clear hierarchy. Likewise, Mehrling uses a simple example where there seems to be a clear hierarchy.

    I don't see this necessarily as a problem. I just thought it interesting to consider what it is exactly that makes the structure of layers hierarchical.

  3. The description of a hierarchy could be based on observation, i.e., who's liabilities tend to serve as settlement instruments for someone else. Moreover, financial crises often illuminate which instruments people flee towards and which are left behind... often indicating some institutional power structure behind the implied hierarchical structure.

  4. I think the question of whose liabilities serve as settlement instruments is a matter of logic, not 'merely' of observation. Staying within the confinements of finance for the moment, I can settle my personal debts with bank money or central bank money. Banks, otoh cannot settle their debts to one another with their own liabilities. They must resort to central bank / inter bank money to do so. Likewise, two central banks cannot settle debts among themselves with their own liabilities. In most cases there is no institution further up in the hierarchy, the ECB being the notable exception. So, in general there are 2 layers to out modern monetary systems, plus whatever else serves as means of payment that isn't attached to the banking system. By that definition, bank deposits and bank notes are on the same layer. So, I suppose I don't agree with JP Koning in that regard.

  5. Sorry for the silence, but I caught a nasty flu one week ago. Finally recovering.

    Yes, layers imply a hierarchy, and as Oliver says, there's no clear hierarchy between (private) banknotes and deposits. But I don't think we need to get more stuck in that.

    As you know, I don't like to talk about banks' liabilities. CB liabilities are not really CB liabilities -- or, are best viewed not as such. Commercial banks are liable to let an accountholder "change recordkeeper" (from one private bank to another, or to the CB), but that is different from having a $100 liability towards someone whose checking account balance is positive $100. But we don’t necessarily need to go there, either ;-)

    I think the most interesting intersection is between currency and commercial bank deposits. Is there a hierarchy here? Ask Greeks, and they probably tell you there is. After all, the classic “bank run” is about deposit-holders demanding currency. Yet, the authorities have usually done everything they can (“… and believe me, it will be enough”) to secure that currency and deposits are perfect substitutes; are valued at par. In the absence of e-currency, or CB checking account for everyone, this “promise” is becoming more and more important, as we use less and less cash.

    Any thoughts?

  6. Sorry about the flu. If it was like the stuff going around here, it was tough.

    In this morning's comment, you seem to be talking about money, not the gift economy. [In the money economy, someone creates money when they make a promise to pay; then the promise (or it's recorded equivalent) is used as trade item.]

    Now to the subject of my comment. You write "I think the most interesting intersection is between currency and commercial bank deposits." It certainly is an interesting intersection.

    It is often written that "banks create money when they make loans." That statement does not agree with my opening definition--banks are actually making money available when they make a loan, not making a new promise to pay. Or are they?

    In my opinion, commercial banks can be viewed in both ways: They are making money immediately available AND they are creating new money by making a promise to pay in the future.

    Confusing? Yes, but, if we can discover ways to create new money WITHOUT ambiguity, we can establish a reference method(s) that make commercial bank money creation understandable.

    My favorite reference method comes from a government-with-money-creating-powers. This government has three possible methods of obtaining money-to-spend: tax-collection, borrowing previously-printed money, or actual new-money creation by 'printing'.

    Notice that both tax-collection and borrowing involves money previously-created. This means that borrowing (as an isolated action) by itself does not create money. This leaves only 'printing' as a method of creating money.

    How can government 'print' money and not be an obvious counterfeiter? Government can 'borrow' from itself. The Central Bank can create notes (for circulation) and exchange these notes for Government Treasury Bonds. The CB ends up with valuable bonds and Government ends up with money to spend.

    Now we have the background to understand commercial banks. Commercial banks can only lend money-first-created-by-government. This creates a limit of how much money is actually available to distribute. Against this limit, banks lend but consequentially (as a system whole) receive the loaned money immediately back into the banking system. This reversal of the loan proceeds results in reuse of the original money at the same time that the number (amount) of recorded deposits at the commercial banks increases.

    This dual divergence accounts for the illusion of money creation at the same time that money supply remains constant.

    Is this making sense Antti? Does this make the intersection of commercial banking and currency more understandable?


  7. Pardon the late reply. I've been very busy. You ask: I think the most interesting intersection is between currency and commercial bank deposits. Is there a hierarchy here?

    I remember some economist, it may have been Mehrling, saying that hierarchies only become apparent when there is a crisis. I think that may apply in the Greek example, too, although I'm not acquainted with the specifics. Is Greek NCB cash treated differently than deposits in banks registered in Greece? In Cyprus, at least, I think they tried to fence off a whole country. That would have included the NCB, putting the wedge not between Cypriot cash and Cypriot deposits but between anything Cypriot and the rest of the eurozone.

  8. I'm busy as well: I capitulated and got myself a day job :-)

    Roger: What you explain sounds like the "money multiplier" story? I don't buy into it. And instead of saying that (commercial) banks create money, I say that they do bookkeeping. It is a very natural thing for a company to do. Then I try to figure out what that bookkeeping is about.

    Oliver: Yes, I think Mehrling talks about that in the paper Johan has linked to (perhaps in Koning's blog?). I think Greece and Cyprus are special cases, due to euro. The "whatever it takes" guy sits in Frankfurt, not in Athens.

    I was thinking about a more general case, for instance the "whatever it takes" people in Washington. In 2008/2009, Treasury, the Fed and FDIC together did whatever it took. And it took full guarantees for practically ALL bank debt. In the modern system especially, one cannot anymore think that everything else than currency (in the absence of e-currency or CB checking accounts) is risky. It cannot be. Nearly all checking accounts are expected to be 100 % safe, because it is those accounts we use, instead of currency. People who have checking accounts are not "lending money" to private businesses, banks. They are using a public service/utility. Currency alone cannot efficiently serve that role. (This is an important distinction when we talk about possible CB e-currencies; see for instance Koning's recent post on FedCoin.)

    Makes some sense?

  9. Antti: I think we have a common interest in building a logical monetary (or gift) system, beginning from scratch.

    We seem to differ in the physical meaning of the numbers entered by "bookkeeping". I confine the numbers to physical origins.

    I hope you are feeling better after your flu bout. I still have my flu symptoms after three months but there is real improvement now. It is nice to feel better! :-}

  10. Good to hear you're better,Roger! I'm not yet symptom-free, but getting there.

    All: I suggest you read this extensive post by Carolyn Sissoko: https://syntheticassets.wordpress.com/2017/03/21/banking-theory-a-monetary-theory-thats-more-heterodox-than-heterodoxy/

    Mehrling's hierarchy mentioned and questioned, etc. I find myself often agreeing with Carolyn. It's not that surprising, considering that her PhD thesis advisor was Joseph Ostroy (whom I quote in my early posts)!

    I'm planning to write my next post on taxation. I'll try to give a new angle on it, different from the cartalist view.

  11. I'm about half way through the linked post and must say, I agree with most everything she says. I also remember her discussing with Nick on his blog and having linked to a paper I very much agreed with. I'm not sure it's necessary to throw the term hierarchy out the window, though. I think it suffices to accept that the hierarchy is not causal in nature, i.e. there is no top down causality. Let'y see whether my children will let me read through the rest...

  12. Apparently, I was nearly all the way through already. In addition to the above, I particularly liked that she mentions the production cycle and the need by producers to have access to credit lines. That's very much the monetary theory of production that I'm always on about.
    I'm not sure why she feels the need to contrast her writing with Chartalism, as there is much writing within related traditions that is much closer to her's than Chartalism. Maybe because it's more well known.
    I was also when Nick commented that he found it a difficult read. I find it much more intuitive than many thing he writes. But I suppose that just goes to show economic schools are a bit like languages.

  13. Sissoko writes in her conclusion that: "In short, banks provide the liquidity that makes it possible for markets to approach the neo-classical ideal."

    ... What exactly does she mean by the neo-classical ideal?

  14. Oliver: Yes. Again, you and I agree on nearly everything. I guess she picked chartalism as her target because it's popular, and also because she wanted to show -- by comparison with monetarism -- that chartalists don't go far enough.

    Johan: Could it be that she means that banks make any cash-in-advance constraints non-binding, so that agents can trade with each other rather freely as long as they respect their lifetime budget constraint? That's how I interpreted it.

  15. Hmm, yeah, maybe? I was just curious since, for example, in Keynes the finance gap between desired investments and existing savings is funded by bank credit even though the money supply is said to be exogenously determined by the state. I.e., bank credit is the (implicit) "heart" of the multiplier.

    The neoclassical model is, of course, a different beast altogether. No banks in the standard model, so (all) profits are distributed to households as interest payments from firms' bond issuance… which is a problem because aggregate real profits generally differ from aggregate real interest payments. So the model itself is inconsistent. Rectifying the inconsistency will, on the other hand, break the dichotomy between real and nominal variables, so monetarism's version of the QTM breaks down as well.

  16. Hi Antti, are you in hiatus? Canoeing with Nick Rowe? Has all been said?