|
Post by oliver on Sept 15, 2017 7:56:13 GMT
This is actually getting interesting. Think of Free Banking era in the US, with private bank notes. Was there only one UoA, the US$, or were there as many UoAs as there were banks? A $5 note issued by a bank in Town A could buy, say, 50 eggs in Town A. The same note might buy only 40 eggs in Town B, even if the sticker price of the eggs was the same as in Town A. Vice versa for a $5 note issued by a bank in Town B. If we agreed that during Free Banking there was only one UoA, the US$, would this imply that credit balances in the ledger of the central bank of Zimbabwe are denominated in US$, even if merchants in Harare would ask for their goods either a $5 Fed note or a $10 bond note? Just a thought. I'm starting to think that it doesn't matter that much if you think that there could be two UoAs, 'dollars' and 'LETS dollars', in our LETS world. Is it that different from imagining that a merchant in the US in 2017 would say that the price of a certain good is $9 in "year 2010 dollars" but would only accept $10 worth of Fed notes for the goods? Only the latter price would have any practical meaning. I think you're right. It's a matter of degree whether one thinks of it as being a world with one UoA + some minor discounts per issuing agency / bank or of various separate UoAs. This is what I argued on your blog once, when I claimed that the smallest unit for an independent UoA is a single, commercial bank. It's also confusing, of course, to think of various UoAs all with the same name.
I guess in the free banking period, it makes sense to talk of one UoA because they all referenced a common, commodity standard / UoA and all carried the same name (dollar).
Zimbabwe is obviously attempting to do the same thing. It is like a free bank vs. the more or less homogenous, domestic USD currency area. But, owing to its history of not being particularly trustworthy, it isn't doing a very good job. The trust that the USD conveys cannot just be transferred to the reserve bank of Zimbabwe by fiat.
To do a better job, Zimbabwe would have to set up a real currency regime like Hong Kong, with a giant treasure trove of forex reserves that it could throw at anyone who doubted its commitment. It probably hasn't got the current account surplus to pull that off, but issuing promises that it will do its best just isn't going to cut it given its history and the state of its economy.
|
|
|
Post by JP (admin) on Sept 15, 2017 16:04:09 GMT
I'm starting to think that it doesn't matter that much if you think that there could be two UoAs, 'dollars' and 'LETS dollars', in our LETS world. Is it that different from imagining that a merchant in the US in 2017 would say that the price of a certain good is $9 in "year 2010 dollars" but would only accept $10 worth of Fed notes for the goods? Only the latter price would have any practical meaning. We don't have to get bogged down in the unit of account stuff. I only brought up this whole distinction because I wanted to know more from you about the relationship with the abstract skilo/USD unit of account, the purchasing power of LETS balances, and the economy wide price level. [On the topic of the U.S. free banking period, I believe that the dollar coin was used to define the $ unit of account. Notes were denominated in dollars, but they traded at to a discount to face value depending on location. Retailers would set prices in terms of the coin, but put surcharges on banknotes the size of which was determined by the cost of transporting them back home. i.e. distance from home could not cause inflation.] Let's quickly review where we agree/disagree. In a LETS world we have--as you say--one UoA, skilo or imaginary dollar, and one system with records denominated in that UoA. I have no problems with this. Do we agree that some mechanism has to keep the value of the balances in-line with the skilo unit of account? They don't just bind to each other automatically, right? Look at bond notes--they don't bind automatically to the US$ unit of account. I used an admittedly extreme example of all the debtors skipping to Mars to show this. Oliver gave three ways the LETs administrator might adjust for this departure. I think that the first two would keep the value of LETS balance in lockstep with the unit of account, but the third would cause the value of LETS balances to fall to a discount to the unit of account, right? [OK, going back I see Oliver disagrees: "There are no 2 values that can trade at a discount to one another." Maybe the Zimbabwe example will change his mind?] A classic mechanism for binding the two is to make LETS balances convertible into units of account, but since skilos don't actually exist, this isn't an option. There probably has to be another day-to-day mechanism for ensuring that there is no wavering between the purchasing power of balances and the unit of account? Open market operations?
|
|
|
Post by Antti Jokinen on Sept 15, 2017 18:29:31 GMT
They don't just bind to each other automatically, right? Look at bond notes--they don't bind automatically to the US$ unit of account. As you said, one cannot convert LETS credit balances into skilos, because the latter don't exist. In Zimbabwe, RTGS credits used to be convertible into Fed notes at par, but aren't anymore. As Fed notes still circulate, and are preferable to bond notes, it is clear that bond notes trade at a discount. If we go back to Free Banking, there the Reserve Bank of Zimbabwe could be compared to a State Bank few people trusted, but nearly all people in that state needed to do business with. Because dollar coins and any notes issued by a sound bank were in short supply in that particular state, people's everyday business required they use the State Bank's notes, unless they wanted to engage in bilateral barter. The state government was also banking with the State Bank. All this understandably led to a two-price system, where the price of goods offered for sale varied depending on whether the buyer was offering dollar coins or State Bank notes. This is all about alternatives. To answer your question, LETS records and 'skilo' bind to each other more or less automatically if people have to choose between LETS and bilateral barter. In bilateral barter the nominal/absolute price is not relevant; real/relative price is what matters. And bilateral barter, even if it's delayed (i.e. there are bilateral debts), is arguably less effective than trade through LETS -- even if inflation runs high (there are plenty of studies on why people continue to use currency even during hyperinflation). Oliver talked earlier (perhaps somewhat cryptically?) about communal decisions. If sellers continue to demand a skilo-denominated credit entry on their LETS account when they sell something, the sticker price will mean just that -- it will state the skilo amount of the credit entry they expect. No matter the circumstances around the LETS system. No matter the inflation rate. Once people get used to the system, they might not even realize that 'skilo' (SK) is not defined by a LETS SK1 credit balance -- that 'skilo' is logically prior to the records of trades of goods priced in skilos. And this might not only be true of ordinary people, but also of monetary economists. Once we reach this point, skilo and a LETS SK1 credit balance are more or less inseparable. In people's minds, a SK1 credit balance has become a unit of account (and probably a "medium of exchange" and a "store of value", too). I don't want to leave any stones unturned, so if you're OK with going back and forth on things that are difficult to understand, let's do it. This conversation has been very useful for me so far.
|
|
|
Post by Antti Jokinen on Sept 16, 2017 9:00:29 GMT
Makes sense. Though I still think it's more correct to view these as cases of one UoA. Here's why: Each bank would like its credits/notes to trade at par with dollar coins, or more generally, with strongest credits out there. Whether it is so, is decided by the public. So we should view it as a proposition by the bank, saying "Here's our ledger with records denominated in the one and only UoA, US$ -- feel free to use it when you trade". I could write a personal IOU for $10. If someone didn't accept it at face value for the goods he sells, would that mean I was referring to a personal UoA when I wrote the IOU? I wish people would accept it at face value, and the same goes for the TOMs a bank writes. Again, this is probably not important. We must remember that the modern banking system is built on the principle of each credit trading at par, and authorities have shown that they are willing to go to great lengths to ensure that it is so.
|
|
|
Post by Antti Jokinen on Sept 17, 2017 16:46:45 GMT
I continue my answer to JP above.
He said:
Yes. That "mechanism" is people's commitment to using the LETS system when they trade. If a seller insists on using the LETS, and the purpose of LETS is to record the skilo value of the goods he sells, then the sticker price must match the amount he expects to be credited to his account. (Thus, if a potential buyer has a SK100 credit balance, he can find out what it's worth by observing what the seller offers for sale at SK100.)
Do we agree?
|
|
|
Post by JP (admin) on Sept 18, 2017 1:36:03 GMT
This is all about alternatives. To answer your question, LETS records and 'skilo' bind to each other more or less automatically if people have to choose between LETS and bilateral barter. My brain must not be working this weekend---I'm really not following you on what binds LETS balance to the unit of account. Maybe we can take a different tack. You said earlier that the unit of account, say the $, is defined by some vague set of memories. It is very sticky. Merchants set their price in terms of this unit. Now assume that a bunch of LETS debtors suddenly skip for Mars. (Let's follow Oliver's option #3 here). The purchasing power of credits has to fall by, say, half. What happens to the price level? The logic of the situation requires that balances immediately lose half their purchasing power but the unit of account is pretty much frozen in time. This seems like a contradiction.
|
|
|
Post by oliver on Sept 18, 2017 9:20:59 GMT
This is all about alternatives. To answer your question, LETS records and 'skilo' bind to each other more or less automatically if people have to choose between LETS and bilateral barter. My brain must not be working this weekend---I'm really not following you on what binds LETS balance to the unit of account. Maybe we can take a different tack. You said earlier that the unit of account, say the $, is defined by some vague set of memories. It is very sticky. Merchants set their price in terms of this unit. Now assume that a bunch of LETS debtors suddenly skip for Mars. (Let's follow Oliver's option #3 here). The purchasing power of credits has to fall by, say, half. What happens to the price level? The logic of the situation requires that balances immediately lose half their purchasing power but the unit of account is pretty much frozen in time. This seems like a contradiction. The price level = unit of account. They are inseparable if there is only one system other than barter to choose from. So, if there is an event that changes supply and demand in such a way that market participants realise this but nominal entries in their accounts are not altered, there will be an adjustment of the price level. The unit of account will have moved. For all existing entries, this means an adjustment of purchasing power vs. before. For all future entries, this means they will reflect the movement of the UoA accordingly. The vector of causation runs from real market events (via accounting rules / culture) to entries, not vice versa.
The price stickiness does not apply to the LETS entries vs. the UoA, it applies to the UoA vs. changes in real market activity. It's saying that it takes a lot of cumulative Events in the market to change the price level in a significant way in the first place. Not that actual changes in the price level won't find their way into the accounting. A whole bunch of people / debtors moving to Mars is a significant, albeit unrealistic event. Also, one as to see the debt they leave behind in relation to the demand for goods that also goes missing. The effect of people leaving and joining the system on effective supply and demand is not as clear cut as checking whether they are net debtors or net creditors.
The institutional setup, that is, the way the accuntants are mandated to record losses and gains (my points 1- 4), will determine whether such events have an impact on the price level = UoA in the first place. A 'lax' setup (points 3 & 4), means that even with many losses, nominal purchasing power will not adjust to reflect them which leads to loss of purchasing power. A 'strict' setup (points 1 & 2) will leave purchasing power unaffected as a first order effect.
|
|
|
Post by oliver on Sept 18, 2017 13:34:43 GMT
Makes sense. Though I still think it's more correct to view these as cases of one UoA. Here's why: Each bank would like its credits/notes to trade at par with dollar coins, or more generally, with strongest credits out there. Whether it is so, is decided by the public. So we should view it as a proposition by the bank, saying "Here's our ledger with records denominated in the one and only UoA, US$ -- feel free to use it when you trade". I could write a personal IOU for $10. If someone didn't accept it at face value for the goods he sells, would that mean I was referring to a personal UoA when I wrote the IOU? I wish people would accept it at face value, and the same goes for the TOMs a bank writes. Again, this is probably not important. We must remember that the modern banking system is built on the principle of each credit trading at par, and authorities have shown that they are willing to go to great lengths to ensure that it is so. Yes, for all practical purposes nowadays I'd say that't true.
|
|
|
Post by Antti Jokinen on Sept 18, 2017 13:45:37 GMT
JP and Oliver,
First I have to say that Oliver nailed it above. He's well aware of how I think.
In this case, when LETS is the only alternative to bilateral barter, the only way the balances can lose purchasing power is by a rise in the price level. "Stickiness" doesn't mean that prices won't change, but it does suggest that in practice there are no sudden, one-time leaps in prices -- leaps which are followed by again stable price level.
Oliver's option 3 ("3. The accounts of the debtors are kept on as legacy debts") makes an already unrealistic scenario even more unrealistic. As I've argued, that option is against the logic of the system (claims on goods > obligations to give up goods, at aggregate level). Nevertheless, even if assume option 3, I don't see how prices in practice would adjust immediately to a much higher level.
First, we have to remember that credit balances are not something someone uses to make purchases. I could have a zero balance and make a purchase, and in this way both total debits and total credits will increase. There's no defined amount of "balances" that are transferred from buyers to sellers. This poses a problem for calculating the "purchasing power of existing balances" in any other way than by looking at the prices of goods offered for sale. No? We don't arrive at the "right" price level by calculating the purchasing power of balances; instead, the price level defines the purchasing power of balances.
Second, the amount of gross balances relative to the total activity in the economy plays a decisive role when we consider option 3. If there was no long-term debt, so that any debts were due to short-term variation in sales and purchases, then why would merchants suddenly raise their prices just because there existed a smallish amount of credits without corresponding debits? If there was a considerable amount of balances, then I have no problem with assuming that there would be significant confusion among participants and that this would lead to a gradual, probably initially even somewhat steep, rise in prices people put on their goods.
As you see, JP, I question this claim of yours:
Why does it have to? What's the "logic of the situation" which requires this?
|
|
|
Post by Antti Jokinen on Sept 18, 2017 16:01:19 GMT
JP, let me try to remove this confusion (perhaps by adding to total confusion...).
I said earlier:
If the LETS is the only plausible alternative, then sticker price = credit entry the seller expects on his account. This is because the seller accepts that the purpose of LETS is to record the nominal amount of goods he gives to others (and takes from others), and this nominal amount is expressed in 'skilos'.
Perhaps you're confused because you expect this to differ from a system where the credit balance, say $1 in the Fed ledger (this can take the form of a $1 Fed note), is said to define the unit of account?
Well, it need not differ from that system. Instead, we might be looking at two competing descriptions (~theories) of the same system.
You said on Sep 13th:
What if monetary economists are wrong when they say that the unit of account is defined by a credit balance (which they call a 'medium of exchange'; I view coins and notes as "portable credit balances"; happy to elaborate)? I'm just asking.
How would monetary theory look like if we instead concluded that the credit balance doesn't define the unit of account, and that there is no one thing that serves both as a UoA and a MoE? Well, I know how it would look like, as I've looked at the system from that perspective for some years now.
I cannot (directly, at least) prove you wrong if you say that the UoA is defined by a credit balance. So let's not go there. The only thing I can try to prove is that it makes a lot of sense to look at the system from a perspective where the MoE doesn't serve as the UoA.
|
|
|
Post by JP (admin) on Sept 18, 2017 16:44:26 GMT
JP and Oliver, First I have to say that Oliver nailed it above. He's well aware of how I think. ... As you see, JP, I question this claim of yours: Why does it have to? What's the "logic of the situation" which requires this? Hey, no fair You said you agree with Oliver, but the claim of mine---"The purchasing power of credits has to fall by, say, half."---is what Oliver said too. How can Oliver be right and wrong? So why does the purchasing power of credits have to fall? I think Oliver explained it here, no? Something has to give: "You agree that the departure of prospective net producers tears a hole into 'net supply'? You agree that, to the extent that the remaining 'net consumers', say old people, do not or cannot change their productive capacity but insist on upholding their consumption, something has to give?" Apologies for being too obsessive on what is admittedly an unrealistic scenario, I'm only doing so because if one really wants to understand a system, investigate what happens at the extremes. We can move on if you prefer. Here's an unrelated observation. In practice, LETS don't hold any assets on their balance sheet (ok, maybe a computer or two to store entries). They are pure administrators of the system. Central banks do hold assets on their balance sheet. If a central bank is functionally like a LETS, at least according to your outlook, why does it not act like a regular LETS and hold no assets at all?
|
|
|
Post by oliver on Sept 18, 2017 17:48:02 GMT
Hey, no fair You said you agree with Oliver, but the claim of mine---"The purchasing power of credits has to fall by, say, half."---is what Oliver said too. How can Oliver be right and wrong? Ah well, I'm an illusive creature, I alternate between facts. And I pay Antti more... Seriously though, he called me on precisely that point and from the subsequent discussion I think understood and agree with his, at least to some extent. There is no linear path from counting debits and credits to supply and demand. That alone is enough to kill the quantity theory of money imo, which is what seems to be at stake here.
|
|
|
Post by oliver on Sept 18, 2017 18:15:50 GMT
Here's an unrelated observation. In practice, LETS don't hold any assets on their balance sheet (ok, maybe a computer or two to store entries). They are pure administrators of the system. Central banks do hold assets on their balance sheet. If a central bank is functionally like a LETS, at least according to your outlook, why does it not act like a regular LETS and hold no assets at all? I'll give it a jab: In contrast to central banks, LETS aren't commissioned to conduct monetary policy (irrespective of how and why such policy works or not). In contrast to commercial banks, LETS are not-for-profit. There is no reason for them (their members?) to buy financial assets. Edit: I assume by 'assets on their balance sheet' you mean something like proprietary trading / active balance sheet expansion as above. Because apart from that, LETS obviously do have assets on the balance sheets. Namely the loans to the debtors.
|
|
|
Post by Antti Jokinen on Sept 18, 2017 19:15:28 GMT
You said you agree with Oliver, but the claim of mine---"The purchasing power of credits has to fall by, say, half."---is what Oliver said too. How can Oliver be right and wrong? Hah, I knew you would catch me on that one! Luckily, Oliver admits he was wrong about the purchasing power of credits falling by a certain, calculable amount. I knew all along that he knew he was wrong, and if one knows one is wrong, then one is both right and wrong. Simple. No need to apologize -- I fully agree. Instead of moving on, I'd like to point out that there's a (weak) link between double-unrealistic option #3 and your stories about Somali shillings. Perhaps you saw it too (which could explain your obsession)? We can see some kind of "stray credits", without corresponding debits. There are differences, too. One of them is that the One Bank continues to operate, so that new balances emerge. And as balances are fungible, we cannot say that the "legacy balances" are like Somali shillings. Somali shillings were special because one could have expected their purchasing power to fall to zero, but it didn't. Right? (Yes, I'm trying to take your story and use it against you.)
|
|
|
Post by Antti Jokinen on Sept 18, 2017 20:17:39 GMT
I assume by 'assets on their balance sheet' you mean something like proprietary trading / active balance sheet expansion as above. Because apart from that, LETS obviously do have assets on the balance sheets. Namely the loans to the debtors. Careful, Oliver! The IOTs are a liability without simultaneously being an asset to anyone (yes, Nick Rowe's "red money" is back!). I think JP had real assets in mind (he mentioned computers). Let's say the LETS operator-bookkeeper bought two computers for the One Bank (the entity running the LETS system). Price SK1,000 total. How would he account for it? He would credit the account of the seller with SK1,000. Which account would he debit? That is, who bought the computers? The One Bank did. Why are the computers a liability for the One Bank? Because if the One Bank is someone, then the One Bank is the operator-bookkeeper (for full argument, see the accounting lesson I gave Nick Rowe today). And he surely doesn't own the stuff held by the One Bank. Theoretically, if we wound up the Bank, some credit-holders would end up with the computers. This shows that the One Bank had a liability to give up goods (the computers), and this liability was matched, at aggregate level, with someone's claim on goods. In addition to computers, the LETS operator, acting on behalf of the system, could buy any stuff he considered necessary. Even gold, if he was wicked enough. So any difference between our One Bank and an actual central bank is a difference in degree, not in kind.
|
|