« Why I=S is a bad place to start doing macro, again | Main | Financial Assets > Liabilities »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

So Bitcoin is subject to monetary policy by money quantity and LETS is subject to monetary policy by interest rates?

JKH: well, if Bitcoin were managed it could use quantity and/or interest rates, if it wanted to. But LETS needs to be managed in a sense that Bitcoin doesn't. (Though whether Bitcoin would work *well* without being managed is another question. Bitcoin does have a nominal anchor, though it might not be a very good nominal anchor, but it's better than having none.)

> The New Keynesian (Neo-Wicksellian) answer to this question is that LETS lacks an automatic equilibrating mechanism that would create a nominal anchor, and would need to be actively managed. And the way to actively manage it would be to adjust the rate of interest on LETS balances. The correct response to those two mischievous fairies would be for the managers of LETS to cut the rate of interest paid on positive balances (or charged on negative balances), to encourage each individual to want to buy more goods than he sells. Whether this would work is another question.

I think this just means that the LETS system lacks a liquidity premium in its original specification, since R$1 and G$1 can be created together from a zero balance. Imposing an interest rate and hence an interest rate spread means there is a benefit to holding a net Green balance.

Even though the aggregate net money stock is zero, individuals want to hold a net positive personal stock. With bitcoin, the aggregate net money stock is positive, so individuals must hold a positive personal stock to make purchases.

With this framework, the 'nominal anchor' of Bitcoin is precisely the marginal (or is it average?) value of liquidity. This also means even green pieces of paper (that is, our real currency) have a short-term nominal anchor, and this is supported by behavioural finance advice along the lines of "to reduce your personal spending, pay for things with cash and not credit cards."

Majro: for some reason your comment went into spam.

We need to distinguish two different spreads:

1. The spread between the interest rate charged on red money minus the interest rate paid on green money (that's like the 0.50% the Bank of Canada has between the Bank Rate and the Deposit Rate).

2. The spread between interest rates paid on bonds and the interest rate paid on green money (simple textbook macro models like ISLM normally assume money pays 0% interest - like currency - so that spread is equivalent to "the" rate of interest in those models).

I think you are talking about spread 2, because that is the liquidity premium.

> Majro: for some reason your comment went into spam.

Thank you for the rescue.

> We need to distinguish two different spreads:

I'm not yet thinking about a world with bonds for the LETS liquidity premium, I am indeed thinking about the Bank/Deposit rate spread.

The liquidity value (perhaps I shouldn't call it a premium?) is the answer for "why should I, with certain expectations of future income with stochastic timing [to avoid income/solvency effects], hold a money balance rather than make my next-intended purchase?"

In a Green-only world of which Bitcoin is one of them, the answer is 'if I make a lower-value purchase now and my income is delayed, I won't have the money to make a higher-value purchase later'. The liquidity value is built in to the limited gross stock of money, so there is implied value in holding an average positive balance of money.

In a LETS world with no interest paid on G$ or charged on R$, there is never any reason to hold (on average) a net money stock. To effect a desired positive average balance, the monetary authority would need to impose an interest rate spread like the Bank/Deposit rate.

Majro: OK. I (think I) understand you now.

BITCOIN: There are some whose experience with the currency would have them, rather, call it BITCHCOIN. :-)

LETS as described would be a bit like giving everyone a credit card with no credit limit, no interest rate, and no repayment schedule. If I had a credit card like that I would never work again.

To make LETS work you would need at least credit limits and repayment schedules. You could for example say that everyone can have as much credit as they like but it can never exceed their net-worth and when they die their stuff is sold off to pay down any outstanding debt. To fully anchor the system the value of the currency unit would have to be pegged (by convention?) to something real , for example a bundle of goods, or (if its not too Marxists) an certain amount of standard labor time. With these rules then people's desire to take on debt would be bounded by the needs of trade, plus any genuine reasons why they may want to buy more or less than they spend in the present period compared to future periods.

If a fairy came along and changed the thing that anchors the price-level then this would be like changing the real value of people's negative or positive money balances and the nominal value of their other assets. In the long term the gross money supply would adapt to the new price level.

I am now trying to see what effect the default zero inter4est rate would have. This might make people want to hold negative money balances in any given period - but that would be impossible in the aggregate. I am sure this would be sub-optimal in some way - but can't work out how.

"In Bitcoin, everyone starts with a balance that is either positive ("green money") or zero. There are no negative balances ("red money"). In aggregate, the Bitcoin account holders have positive net wealth in their accounts.

Bitcoin is like a fixed quantity of green bits of paper with positive value that fell from the sky into people's pockets. (Yes, I know about mining, and I'm ignoring it.) The buyer gives green paper to the seller in exchange for goods received."

There are no bitcoin bonds "backing" bitcoin, right?

Well, my reference model is not one with "perfectly synchronised payments and receipts of money", but instead one where "individuals' payments and receipts [are] less synchronised."

Given such an environment, it seems to me that the value of B$ in equilibrium comes from the fact that it does not meet the needs of trade (unless all agents are identical and endowed with an identical just-right B$ supply -- a case I consider uninteresting). Given heterogeneous agents and uncertainty over when B$ will be needed, if you give everybody enough money to definitely meet the needs of trades, then the money supply is inconsistent with the price level (i.e. equilibrium requires somebody to carry positive money balances forever and is therefore not equilibrium). And if you don't give everybody enough money to meet the needs of trade, then in equilibrium the needs of trade are not met.

So to me the tension between these two environments is that in one it is possible to meet the needs of trade -- but the money supply needs to be managed -- and in the other there is no hope of ever meeting the needs of trade.

csissoko: Thanks for your Tweet!

That imperfect synchronisation is my reference model too. (But note how it is logically incompatible with a representative agent model, since the representative agent has expenditures always equal to receipts of money, in every minute of the year, just like they must always be equal in aggregate.)

"Given such an environment, it seems to me that the value of B$ in equilibrium comes from the fact that it does not meet the needs of trade..."

I think that's right.

"So to me the tension between these two environments is that in one it is possible to meet the needs of trade -- but the money supply needs to be managed -- and in the other there is no hope of ever meeting the needs of trade."

To me too. But maybe that tension can be made smaller rather than bigger, if the "elasticity" is somewhere between zero and infinity, and it's managed "right" (whatever that means).

Yes Bitcoin definitely imposes the cash-in-advance constraint that you write about, csissoko, while LETS does not. I like Nick Rowe's occasional reminders that in addition to elasticity, we also need a nominal anchor for our money, and this makes LETS dependent on an external numeraire. But I think you are pointing out that Bitcoin is similarly dependent because we cannot have an equilibrium if people try to use Bitcoin as the numeraire.

MF: "I am now trying to see what effect the default zero inter4est rate would have. This might make people want to hold negative money balances in any given period - but that would be impossible in the aggregate. I am sure this would be sub-optimal in some way - but can't work out how."

If everyone wanted to hold negative balances (impossible in aggregate as you say), they would try to buy more goods than they sell, which would mean an excess demand for goods, and rising prices.

@ Nick: You're welcome. Enjoying the discussion.

@Deepwatrcreatur I would say instead that the equilibrium outcomes that can be attained with Bitcoin will be a subset of those that can be attained with LETS. But this is assuming the full range of policy outcomes with LETS (including "perfect" policy). Once one adds a policy rule to LETS, it may no longer be true that it permits a broader set of outcomes.

'If everyone wanted to hold negative balances (impossible in aggregate as you say), they would try to buy more goods than they sell, which would mean an excess demand for goods, and rising prices.'

And I suppose if zero was too high an interest rates then prices would have to fall, and if they couldn't fall because they were sticky you would get less trades and a recession ?

Deepwatrcreature: "Bitcoin is similarly dependent because we cannot have an equilibrium if people try to use Bitcoin as the numeraire."

I think we *can* have a (unique) equilibrium if Bitcoin is the numeraire. It's pretty much straight out of Patinkin's Money Interest and Prices. (Whether we get to it, in a reasonable time, is another question.)

csissoko: "Once one adds a policy rule to LETS, it may no longer be true that it permits a broader set of outcomes."

Think I agree.

MF: Yep.

Nick, I think you're right: Bitcoin can be a numeraire because it is a base money. But LETS is a system of credit and thus relies on a base, outside money.

Deepwatr: We nearly agree. But I would say that LETS is money too, in the sense of being a medium of exchange. If I'm an apple producer who wants bananas, I first sell my apples for LETS, then buy bananas with LETS. (And the banana producer buys carrots with LETS, and the carrot producer buys my apples with LETS.)

Bitcoin supports arbitrary side annotations on any transaction, thus it is perfectly feasible to write an IOU note into the ledger and thus create "red" money.

For example, I could grab a handful of satoshi and annotate "I promise to pay 1 bitcoin to whoever owns this parcel of satoshi at the start of New Year's Day 2018, at midnight GMT". Those handful of satoshi can now circulate exactly like an IOU note might circulate, anyone who voluntarily chooses to accept this IOU note might value it at whatever price comes to mind on the day. At some stage, the critical payment date comes up and either the promissory note pays up, or it defaults.

People forget that Bitcoin is NOT intrinsically fungible (as compared with gold for example, which IS fully fungible), instead Bitcoin appears fungible merely by social convention. The owner of a Bitcoin can decide to break with the fungible convention and create a unique annotation (typically promise of some future delivery) which then operates and circulates via the same blockchain ledger. This is a well known process, it is goes by the name "colouring". Think of it like taking an Federal Reserve Note and then colouring in part of the note to indicate something special about that note.

At the moment there is not automated mechanism for scraping out the coloured bitcoins from the entire blockchain ledger and then producing a red money tally, nor is it obvious what such a tally would used for in practice. Coloured coins may promise delivery of various objects and services at a range of future dates; some of which only have value to particular people. However, unless the Bitcoin concept dies for some unrelated reason, it seems very likely that common conventions for handling coloured coins will emerge, the annotations will become formalized, enforcement bureaus will setup shop, possibly default insurance will be available. Use your imagination.

Hmmm, that's strange my comments are not posting. Have I offended the wrong people?

Yes, in terms in csissoko's "New Monetarism" LETS do help reduce search costs. But I think you and some others are also asking how the over-issuance of LETS is prevented, so that they are only used to meet the needs of trade, by qualified borrowers. I believe csissoko tackles this general question in her paper on Real Bills and trade credit, and I wonder how much of that analysis carries over to the LETS system.

Tel: you must have offended the Typepad demon, because it put your comments in spam! Sorry about that. I fished them out.

"Bitcoin supports arbitrary side annotations on any transaction, thus it is perfectly feasible to write an IOU note into the ledger and thus create "red" money."

I'm not 100% sure I understand this. But I *think* what you have created there is a simple bitcoin bond, and not red money. That bond is a promise by you to pay Bitcoins, so the value of that bond depends on whether people trust you. Like any bond, it can be traded, but it is unlikely to be used as a medium of exchange, simply because it is less liquid. It may be traded for Bitcoin, but it is unlikely to be traded for apples and bananas.

Could one helpfully rewrite this thought experiment with the old Cambridge cash balance idea? Let money be solely a store of value and the only store of value (island of stone money, big heavy gold coins, etc.). The only demand for money is the demand for savings, so let the fraction of income saved be an exogenous constant. The equilibrium condition is then

M/P = k Y

Which is just the quantity equation, of course. Now, is M the net or gross? It seems that it has to be the net, therefore monetary equilibrium cannot be obtained in a balanced red-green money world, unless k = 0, P = 0 or Y = 0 (I'll be piously silent about the behavior at infinity). Y going to 0 would be Bad. So let's listen to Marshall and leave it alone. If I just take k != 0 by fiat, then P must go to zero, but your argument in the third to last paragraph shows that convergence is unrealistic.

Old Man Rowe was able to live with red money because he didn't use money as a store of value - he used his land as a store of value. So he isn't a disproof since in this world money is the only store of value.

Am I getting this asymmetry right or is there some way to conceive of M in gross terms (in this approach)? In a green or unbalanced red-green show, money can be a beautiful store of value, but in a balanced red-green show money can't be handsome only handy.

Ah! Big mistake: in the *real* Cambridge approach, k was not an exogenous constant. k was derived from the utility of savings. *I'm* taking it to be an exogenous constant for simplicity's sake! This is important because one _could_ conceive of a stable balanced red-green world where P, Y & k move until the utility of savings goes to zero.

Thanks for taking the trouble to rescue my comment!

Financial instruments are necessarily zero sum... creating a bond MUST always generate equal amounts of "red" and "green".

I guess you could argue that with a coloured Bitcoin, the red and the green do not independently circulate so you have only one single entity (the coloured coin, or bond if you prefer) which is circulating. This could be seen as "green" I suppose, but the red remains also attached to it in as much as the bond issuer has a liability. Personally, I've always been squeamish about the idea of red money independently circulating to begin with, so I'm not bothered with the concept of a coloured bitcoin circulating as a single entity.

If you want to get technical about it, suppose I buy back my own bond, then I should be able to extinguish that bond somehow, or at the very least toss it onto the shelf and pay myself on the due date (extinguish it the slow way). Now further suppose that I find someone else willing to issue an identical bond to what I have already issued (i.e. pays the same amount on the same day) and I buy a newly issued bond from that person, and then I use that bond to swap over with the current holder of the bond that I issued. In a way I have circulated the red side of the bond over to a different party, although it's a cumbersome process and requires three parties to all agree (plus some transaction costs).

Anyhow, I can see what you are getting at that the Bitcoin system is designed to make the ready circulation of green money (and the green side of bond issue) easy to do, and it poorly facilitates the circulation of the red side of any financial instrument. This doesn't really matter in terms of the key component of your argument:

LETS expands and contracts automatically to meet the Needs of Trade. Bitcoin does not. That is both an advantage and a disadvantage.

If Bitcoin can issue bonds into itself then it can also expand the aggregate money volume. Merely circulating the green side of the bond is quite sufficient, because there's no upper limit on the number of bonds that can be generated (other than credulity of the people accepting the bonds).

Like any bond, it can be traded, but it is unlikely to be used as a medium of exchange, simply because it is less liquid. It may be traded for Bitcoin, but it is unlikely to be traded for apples and bananas.

Hmmm fair point, I agree that coloured bitcoins (or bonds) would be less liquid, but then again probably not really important when it comes to expansion and contraction of the money supply.

Here's my argument: Money is both a medium of exchange and a store of value, it's impossible to separate these functions. The main pressure on the monetary supply is not coming from everyone suddenly requiring more medium of exchange. What happens is the total "store of value" component is larger than the readily exchanging component... and we see this with Bitcoin too. Speculators simply buy bitcoins and sit on them, for various reasons, partly because they think the technology is still growing, and partly because they want protection from inflation if government chooses to print more fiat.

Now if everyone were to buy bitcoins for long term speculation and just sit on them (and no one exchanges), then the value of a bitcoin would be very low. The system is a ledger and it creates value only because it does get used as a medium of exchange. The spectators would defeat themselves. However, at the same time, it is important for people to have that "store of value" option because confidence is the driver that attracts people to choose some medium of exchange over others. With Bitcoin bonds circulating in the system, this satisfies the people who want to just hold something... so yeah probably they will circulate more slowly and be less liquid, but that's fine because they fulfil a purpose by doing that. In turn this releases other non-coloured bitcoins which then go and circulate with high liquidity.

The ultimate limits on expansion will be the limit of credulity, in terms of being confident that those bonds get paid on time. As I said to begin with, all financial instruments create both a red side and a green side, so there's inevitably an element of confidence involved. If your system appears to have the capacity to expand without some kind of pressure coming from confidence then I would argue your system is unstable and will quickly be exploited. Why work when you have a credit card that goes to infinity?

One can imagine a world where people buy goods using IOUs. These IOUs can then be used a money. Any given trade can be completed with either an existing IOU held by the buyer or a new on they issue. Keeping track of the credit worthiness of any IOU issuer and the fact that they are denominated in lots of different goods (whatever the traders agree on) opens a business opportunity for banks who could standardize the units used for IOU (using for example a fixed bundle of commodities) and have a better view of credit worthiness. Individuals rather than issuing their own personal IOUs go the bank , post some collateral, and in return get (for a fee) the banks IOUs to buy stuff with. As the bank is denominating its IOU in terms of a fixed bundle of commodities it is important for its operations that it keeps their value stable. To do this it has to control the level of gross money to maintain the price level, which it will do via interest rate policy (either on red money, green money or both).

I therefore. believe that something like LETS could (and probably has) arisen in the free market.

Nick said: "I'm not 100% sure I understand this. But I *think* what you have created there is a simple bitcoin bond, and not red money. That bond is a promise by you to pay Bitcoins, so the value of that bond depends on whether people trust you. Like any bond, it can be traded, but it is unlikely to be used as a medium of exchange, simply because it is less liquid. It may be traded for Bitcoin, but it is unlikely to be traded for apples and bananas."

Tel, maybe you can help me.

What is the difference between a simple bitcoin bond and "red money"?

To add to what Tel say: There doesn't seem to be any obvious reason why you couldn't have credit money in a bitcoin world. You could for example have fractional reserve banking based on bitcoin in theory, couldn't you ?

Too Much Fed, any financial instrument such as a bond MUST necessarily have both an asset side ("green money") and also an exactly equal liability side ("red money"). You add up the total of all financial assets in the entire world, total assets equals total liabilities. It's the fundamental accounting tautology.

The implication is that if the Bitcoin system can trade any financial instruments at all, then we must conclude that Bitcoin also supports red / green money.

The Bitcoin bond can be traded, and so it could circulate, but Nick points out above that effectively only the "green" side of the bond does circulate, the "red" side is stuck to the issuer (the "red money" exists in the system, it just doesn't move around). I would argue that's exactly how you want it to work, but anyway for purposes of monetary expansion who cares whether it trades only on the asset side?

In other words, the Bitcoin bond is just one item (red and green tied together), not two items (red and green apart). This idea of "red and green money" implies that the "red" can go off into the world as an independent entity not tied to anything. The holder of the "red" would then have a general liability, but not a specific liability TO anyone in particular. Personally I'm highly suspicious of that idea.

Tel, let me try this.

Betty does not qualify for an "overdraft" of her checking account. Andy does. Andy gets an "overdraft" of his checking account. In Nick's scenario, Andy sends goods and "red money" to Betty. Betty defaults. What happens?

It appears Nick is trying to do liability swaps instead of asset swaps?

"The Bitcoin bond can be traded, and so it could circulate, but Nick points out above that effectively only the "green" side of the bond does circulate, the "red" side is stuck to the issuer (the "red money" exists in the system, it just doesn't move around)."

Should bitcoin be considered a bond or some kind of "currency"?

"I would argue that's exactly how you want it to work, but anyway for purposes of monetary expansion who cares whether it trades only on the asset side?"

I agree with the "red" side being stuck to the issuer and that is how it should work. See the liability swap question above. The monetary expansion purposes part could be interesting to examine in more detail.

Betty does not qualify for an "overdraft" of her checking account. Andy does. Andy gets an "overdraft" of his checking account. In Nick's scenario, Andy sends goods and "red money" to Betty. Betty defaults. What happens?

Well the existing system already handles this. When Betty goes into a shop and swipes her credit card it will check if the card has reached its limit, if so the transaction will fail and Andy will refuse service. Now if Betty does have some available credit what happens is her bank sends an electronic promise over to Andy's bank and allows the transaction to continue. Betty's account goes further into overdraft and Andy's overdraft is reduced. Logically we can divide this into two steps:

Step 1: Betty does a transaction with her bank whereby new money is created, red into Betty's account and green available as liquidity. This step may fail depending on how much the bank trusts Betty, her previous payment record, how deep her current overdraft is, etc.

Step 2: Green money moves from Betty's bank over to Andy's bank (actually this doesn't properly happen until settlement at the end of the day in our current system but that's an implementation detail). This adds to Andy's balance.

Thus, in the current system, there's no way for Andy to actually send red money to Betty, but anyway we get the same result in effect by sending green money the other way.

This would suggest that despite no operation being available to transact in red money, it makes no difference to the outcome so therefore we don't need that operation. Forget about it, Bitcoin can do the same job as LETS anyhow (just like our current system can do). The current limitation of Bitcoin is that "Step 1" from above is not available, because no one will offer overdraft services, in effect there's no banking, and as I mentioned above there's no accepted convention for writing IOU notes, nor does the average Bitcoin user trust an IOU note written by a pseudonymous entity. But this could easily change.

Oh Christ. It's gone off the rails, again. How many times do I have to explain:

"(Green) money" is not just another name for any financial asset.

"(Red) money" is not just another name for any financial liability.

"Money" is a ***medium of exchange***, people. And a "medium of exchange" is a good (if green, and bad if red) that all other goods are bought and sold for. And a medium of exchange doesn't even have to be a *financial* asset, it might be a real asset, like cigarettes (green) or garbage (red).

MF: "There doesn't seem to be any obvious reason why you couldn't have credit money in a bitcoin world. You could for example have fractional reserve banking based on bitcoin in theory, couldn't you ?"

Yes, you could have fractional reserve banks issuing their own money, convertible at a fixed exchange rate into Bitcoin, just as commercial banks today issue their own money, convertible at a fixed exchange rate into Bank of Canada money. Presumably though there would have to be some advantages, for some people, to prefer using the money issued by those banks to using Bitcoin (or Bank of Canada money). More convenient for some reason.

But calling it "credit" money does not clarify, it obfuscates. When I hear the word "credit" (used as a noun), I reach for my shovel.

And this post is not about that stuff anyway.

TMF: stop.

Tel: "Financial instruments are necessarily zero sum... creating a bond MUST always generate equal amounts of "red" and "green"."

Bitcoin is an asset to the owner. To whom is it a liability? Some might reply: "It is a liability to the whole community", but that is just a desperate attempt to salvage the "assets must equal liabilities" dogma.

And remember, if I sell my car to you, that presumably means that you value my car more than I do, so we *create* value when I sell it to you. Similarly, if I issue a financial asset, and sell it to you, that presumably means that you value your asset more than I value my liability. So in subjective terms (Austrian, right?) assets > liabilities.

Which will be the subject of a whole nother post, one of these days.

I'm trying to shoehorn bitcoin into my limited, fundamentalist idea of what money is and also trying to figure out whether you are using the dichotomy of LETs and bitcoin as an analogy for the existing banking system in which commercial bank credit and central bank fiat coexist.

Starting with the latter question, I think there is a fundamental difference between credit money, fiat money (i.e. the product of QE) and what I understand bitcoin to be.

Wikipedia writes:
Bitcoins are created as a reward in a competition in which users offer their computing power to verify and record bitcoin transactions into the blockchain. This activity is referred to as mining and successful miners are rewarded with transaction fees and newly created bitcoins.

That, to me, is the description of a type of income based money emission. Miners are awarded income in new bitcoins as a reward for investing in mining equipment and providing its computing power for the community. Since there is no central authority, the creation of this type of money cannot rely on credit risk asessment, imo. I'd be interestd to know how Tel sees this. Nor is there any central authority that will buy existing IOUs in exchange for new money, as with central bank fiat. In that sense I strongly protest the fact that you ignore mining and just act as though bitcoins fall from the sky. I think it's quite fundamental. Bitcoin can only be said to be purely green because it is created in a real exchange for the service of mining. Otherwise it would have to be red/green just like any other money.

So one ends up with three types of money: Money that is created in exchange for a service / good (albeit at a predetermined rate, that's a bit of a funny one), money that is created through an act of a central bank (fiat) and money that is created according to the needs of borrowers (credit). I don't see a nominal anchor problem with any of them, but that may be due to my limited understanding.

Personally, I think bitcoin is a superfluous hype the surplus value of which stems from the fact that it caters to the fears of of silicon valley libertarians and the like. The more pertinant question to me is whether a pure credit economy, assuming overdrafts are not unlimited and there is a central management that takes the assessment of credit risk serisously (i.e. it is managed), is good enough as a stand alone system or whether it is imperative that a(central) bank also engages in the business of buying and selling non-bank securities. Or indeed, although I very much doubt it, whether a pure fiat money system is viable to stand alone.

Or, to condense that into one sentence:
To get your hands on money (and for money to have any value) you either have to give up something real (goods / services) or at least promise to do so in future (sell IOUs or take on credit).

There is no such thing as free money. That also solves the nominal anchor problem, imo.

Deepwtr said: "Bitcoin can be a numeraire because it is a base money. But LETS is a system of credit and thus relies on a base, outside money."

What do you mean by LETS relying on a base, outside money? That it necessarily needs outside money for the system to work (in real life)? Or that if we have a system like LETS, the (very much money-like) credits cannot act as a numeraire, and thus something else, outside money, must act as a numeraire for there to be an equilibrium (in the model)?

I'm asking because I think we don't need any specific numeraire (in real life). We can have an abstract unit of account. I've written a post about this earlier: "Unit-of-account and Numeraire".

Bitcoin is an asset to the owner. To whom is it a liability? Some might reply: "It is a liability to the whole community", but that is just a desperate attempt to salvage the "assets must equal liabilities" dogma.
I would say each basic bitcoin is not a financial instrument. It's a token to demonstrate proof of computational work (i.e. "mining") and that generally also implies expenditure of physical energy (see Landauer's principle, which I accept on faith because it is beyond my capacity to prove or disprove). No one owes you anything just because you have a bitcoin... it is what it is, so have fun with it. The price could drop to zero at any time (or not). That's conceptually different to an IOU note, or a bond, or even a betting slip.

Do you see the distinction I'm making here? If I have an IOU note for three chickens that means there's a contractual agreement between me and some other person (the holder of the note and the issuer of the note). However Bitcoin doesn't have an issuer, no one promised you that it was worth anything, and it's not backed by anything either (and it never was backed by anything, not even for a little while).

The Bitcoin platform could be used as a system to create financial instruments. It could be used as a platform for various things, and I think that's probably where bitcoins get their value from... the intrinsics and useful properties of the platform. I should also point out that no one anywhere has any remotely plausibly theory for long term pricing of bitcoins, so holding onto them is 100% speculative. If you analyse Bitcoin in terms of a financial instrument then the value can only be zero, but that's clearly wrong.

Similarly, if I issue a financial asset, and sell it to you, that presumably means that you value your asset more than I value my liability.
Yeah, there's subjective individual value which is unknowable to anyone other than the individual (and I would argue also indescribable, although others may disagree) and there's the face value of the financial instrument (e.g. the bond value) which is nominal of course, and there's also the market value of that same instrument.

Accounting standards cannot document individual subjective value, if they could do then transactions would not balance they would be positive sum. I dare you to go to an accounting school and spend a few days explaining to them how this new system is going to work. Really, I double dare you... and write a blog about it.

Oliver: I think I agree and disagree with you.

If Bitcoins are created as a reward for "mining", then we can view them as some super-cheap, plastic "medals" which the owner, The Great Miner, can carry around, feeling proud. I don't see how the mining effort would make the Bitcoins money. Say, you found a riverbed where you could easily pick up some lumps of gold (a finite, smallish amount), without any real effort. Let's further assume that you live in an economy where gold bullion is commodity money (not talking about a gold standard). The others have mined their gold, but you got yours without any real effort. Your gold would still command the same exchange value as others', right? The same should be true for a limited number of Bitcoins that just dropped from the sky.

I think the "mining effort" is just a convincing enough way to make the Bitcoins seem valuable. You have to have some story that convinces people that Bitcoins are not created "out of nothing" (isn't this the defining idea of Bitcoin, to be unlike "fiat money"?). They are not created "out of nothing". They are super-cheap, plastic medals awarded to people, and somehow they command a considerable value on the secondary market.

To me, Bitcoins are pure "commodity money" with no intrinsic value (weird?). If we call them money, we should never call "fiat money" money. If we want to call the latter 'money', then Bitcoin is no money at all. (Nick says it is a medium of exchange, but I think it would be unwise to equate MoE with 'money' in this case because it leads nowhere.)

Where do I agree with you? I agree with you in that Bitcoin cannot be money. You might think it can be money because of the mining effort, but if the mining effort doesn't make Bitcoin valuable in the way you suggest, then you end up agreeing with me. Right? :-)


(I know very little about Bitcoin, so let me know if I've misunderstood something.)

Tel: Funny how we both expressed similar thoughts almost simultaneously!

On assets > liabilities, see my new post. Liquidity of assets is key.

Nick: I still haven't fully understood what is the defining feature of an overdraft (like in LETS) which makes it "red money" to you? Or perhaps I have, but in that case I don't agree with you.

The following quote is from my own blog, from a comment I made to Oliver. In it I try to show how a "traditional loan" can be converted to an overdraft. I'd like to hear if you see that the mortgage in question becomes "red money"?


-----------------------------------------------------------------------------------------------------------

My point is that we can take the traditional loan and convert it into an overdraft, without touching the credit contract.

You have a 20-year mortgage you took five years ago, $250,000 of which is outstanding, and you have a $5,000 balance on your checking account? OK. Let's convert it into an overdraft.

Now you have a total overdraft limit of $250,000. You have an unused overdraft of $5,000. Your checking account balance is negative (debit) $245,000. Your overdraft limit will be reduced monthly by the same sum as was your "traditional loan's" monthly amortization. Interest charges will be debited on your account just like they have been debited until now.

The interest charges in these two cases can be made equivalent by setting one interest rate on used overdraft and one interest rate on full overdraft limit, reflecting the interest charged on the traditional loan and interest earned on a credit balance on checking account:

If

X is the interest charged on the traditional loan, and
Y is the interest earned on the credit balance on checking account, then

Y is the interest charged on the used overdraft, and
(X-Y) is the interest charged on the full overdraft limit. (I hope I got it right.)


So, I'm saying that this overdraft has now replaced your traditional mortgage and still nothing essential has changed. Other than that now we say that you repay some of your debt every time there is a credit to your account and incur some new debt every time there is debit to your account. We don't say that you repay your debt, or your loan, monthly when the overdraft limit is reduced.


----------------------------------------------------------------------

Antti: I don't see how the mining effort would make the Bitcoins money.

I've always been inclined to not call bitcoin money. I was trying to play along with Nick to see where it leads me. Suffice to say, I'm not at all happy with where it has lead me...

I think I would call it a type of equity in the bitcoin Project that has the added benefit that it can be used to purchase things from anyone else who has joined the clib. It seems clear to me that the value of one bitcoin is equal to the effort it takes to mine it. The fact that it follows a strict emission rule, just makes it become more sarce over time which, following the paradox of thrift, is not necessarily a good thing. And whether the value of mining bitcoins then translates into an exhange value for anything else, is another question altogether. It's self-referential, really. So that's why I would agree with you that it's not money.

Oliver: "I've always been inclined to not call bitcoin money."

I sensed this, and that's why I said I agree with you. I had a feeling we are more or less in full agreement, but as your in-house contrarian, I had to find something to disagree on.

'But calling it "credit" money does not clarify, it obfuscates.'

I think I see what you mean (the term has been hijacked by people who use it to obfuscate). However under any monetary system there will people who defer spending and those who bring it forward - and the IOUs that underlie this clearly represent credit and if they become the basis for exchange then "credit money" seems a reasonable term.

In an FRB world all the money "multiplied-up" from the base is based (more-or-less) upon credit. Of course the ability of credit money to function as money depends upon people's desire to hold it rather than the real thing. This was clearly the case under the gold standard as gold was costly to store and inconvenient to cart around, but may not be true for BC.

I could explain something about the mechanics of Bitcoin but I don't think it will help this discussion, so I won't unless invited.

Jim: those mechanics can be interesting, but yes, it wouldn't really be part of this discussion, which is only using Bitcoin as an example to illustrate a general point.

Nick: “In Bitcoin, everyone starts with a balance that is either positive ("green money") or zero. There are no negative balances ("red money"). In aggregate, the Bitcoin account holders have positive net wealth in their accounts”

Nick,

I have only a general knowledge of bitcoin, so feel free to see this comment as me thinking out loud, but, as you say, your post is not really about bitcoin.

Many of these discussions are about terminology. Words and their meaning are much more important than we tend to assume. I once worked on a project that investigated why the UK government was so bad at managing outbreaks of foot and mouth disease in cattle. The prime root cause turned out to be that the UK agriculture department did not have a consistent or coherent definition of the word ”farm”! Unbelievable but true.

Start by defining some terms which are used in these discussions but which clearly suffer from “farm”-like problems.

A “real asset” is an asset which has an innate value. For example, gold.

A “financial asset” is an asset which has value only because someone trustworthy guarantees that it has value. For example, as I type this sentence, I am holding a worthless piece of paper in my hand so that I can type the exact wording “I promise to pay the bearer on demand the sum of twenty pounds”. This promise is signed by Chris Salmon, the chief cashier of the Bank of England, “for the governor and company of the Bank of England”.

The definition of a financial asset requires that it is created as an asset / liability pair where the precise nature of the “liability” depends on type of financial asset but, at minimum, it represents the obligation of the issuer to provide some sort of guarantee for the asset, as per the chief cashier’s words.

Note that a "liability" does NOT mean that the holder of the liability “owes” anyone anything, apart from fulfilling an agreed obligation such as acting as a guarantor or paying optional interest on commercial bank reserves.

The “medium of exchange” is an asset which is used as one side of virtually every exchange involving any other asset. At any point in time and in any one country, there is normally a unique medium of exchange, specified by the government. The role of medium of exchange almost requires that it is unique.

I want to add two further definitions.

A “natural asset” is an asset which exists in nature.

An “artificial asset” is a man-made asset.

Note that there is a distinction between a “real asset” and a “natural asset”. Note also, that it makes sense to talk in green and red terms only for financial assets as that is the only asset class which is created as asset / liability pairs.

Using my terms:

Gold is real, natural and NOT the medium of exchange.
Money is financial, artificial and the medium of exchange.
A loan is financial, artificial and NOT the medium of exchange.

Using these terms and your quote, I speculate that:

Bitcoin is real, artificial and NOT the medium of exchange.

That would suggest that we should think of bitcoin more as an artificial equivalent of gold than as an equivalent of money. I don’t know what “mining” means in the context of bitcoin. It may, or may not, be a co-incidence that both gold and bitcoin are mined.

Even more speculatively, who are the users of bitcoin and why do they prefer it to the government specified medium of exchange? Are bitcoiners the equivalent of gold bugs for the technologically savvy? I read somewhere that bitcoin is popular amongst the libertarian right. I am happy to be corrected if that is wrong.

@Jamie you are not wrong about the libertarian right, well, a particular subset of it known as techno-libertarians.

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad