|
Post by JP (admin) on Oct 15, 2017 18:08:46 GMT
Oliver and I might call this a "loan of credit balances", but the OB wouldn't know that I didn't buy anything from Oliver when I asked it to credit his account for $50. From the OB's perspective, we are still recording trades of goods. Alternatively, Oliver and I could call this a "forward sale", as if Oliver actually did sell me something with a delivery date in one year's time. When the date arrives, I sell the something back to Oliver at $52, with the goods (which might have served as collateral, or not) staying in Oliver's possession throughout the process. Are you following me? Yep. So an overnight market is just a way to avoid credit limits imposed by the system operator? How do open market operations work, and how do they influence the overnight rate?
|
|
|
Post by Antti Jokinen on Oct 16, 2017 12:48:44 GMT
Yep. So an overnight market is just a way to avoid credit limits imposed by the system operator? How do open market operations work, and how do they influence the overnight rate? To avoid credit limits, or to attain a lower interest rate than the one demanded by the system operator. I understand your focus on OMO, as there seems to be a clear link between the quantity of settlement balances and the overnight rate. As I said before, before we try to study OMO from the TOM perspective, we should have a model of multiple banks. And we should probably describe the relationships between these banks from the TOM perspective before we move to OMO. Perhaps we could at this point try to sum up the discussion so far? I would especially like to know whether you agree that the TOM perspective so far presented seems to describe quite accurately the banking system as a whole, or at least an imaginary One Bank world? Have you found anything novel in this approach? I'm still struggling with locating my approach accurately in relation to other explanations for money, although I feel I have more or less solved the enigma (others probably don't agree) :-)
|
|
|
Post by JP (admin) on Oct 17, 2017 16:57:03 GMT
Perhaps we could at this point try to sum up the discussion so far? I would especially like to know whether you agree that the TOM perspective so far presented seems to describe quite accurately the banking system as a whole, or at least an imaginary One Bank world? Have you found anything novel in this approach? I'm still struggling with locating my approach accurately in relation to other explanations for money, although I feel I have more or less solved the enigma (others probably don't agree) :-) Antti, I think you've done a great job explaining the TOM point of view as well as being very patient with me. As an explanation of an imaginary One Bank world, it is certainly consistent. I am still skeptical of it accurately describing the banking system, but am open to being convinced. If you want to convince people, I don't think you can just map out the structure of your system. You need to explain a real-world mystery or anomaly that the conventional view can't explain. If your TOM perspective can do this, it would be more challenging for people to continue believing the conventional view. I would describe the conventional view being that coins, banknotes, and deposits are all liquid financial instruments, or media of exchange. There is a demand for these items, since people need to hold an inventory of highly liquid instruments to engage in trade. And those instruments have certain properties inherent to them that govern their value, in the case of medieval coins their gold content, or old banknotes redeemability into metal by the issuer, or in the case of fiat money the issuer's promise to regulate the supply in order to hit some inflation target. At least one of these instruments is the unit of account, which means any alteration in the instrument creates economy-wide price changes. If I have understood you properly you are saying that coins, deposits, and banknotes are not media of exchange, but mere tokens or signs of an underlying barter transaction. The actual material that makes up the instrument, say its gold content, is not important. And there is a separate unit of account. So a different set of rules governs the system. If so, there must be some set of events that the conventional view can't explain. What are these events? Unrelated to the above, I had problems understanding the interaction between the skilo unit of account, credits, and inflation. Either I'm not smart enough to understand, or you didn't explain it properly, or it just doesn't make sense.
|
|
|
Post by Antti Jokinen on Oct 18, 2017 11:42:45 GMT
Thanks for the feedback and encouragement, JP! You ask for a real life problem which is only solvable if we adopt the TOM point of view. I think it's a fair request. That's what science is about, after all. Unfortunately I don't have any simple answer ready at this point, but perhaps it could help if I explained how I ended up thinking about the monetary system in the first place. Long story short: Since the financial crisis, I've thought that it has been, and it still is, a grave mistake to let total debt levels become this high. I have more or less shared the BIS/Bill White position (see also here). Simplified, it boils down to this (from the linked source): As far as I know, there's no theoretical backing for this that would be widely accepted. Right? Paul Krugman, for instance, has addressed this on more than one occasion, and usually talks about "one person's debt being another person's asset". All debt nets out, and so forth. Here Krugman says, somewhat related to this question, that the BIS has an attitude, not a model. Since around 2012 I've tried really hard to understand how intelligent people can be so divided on this question. I felt this is the burning question of our generation (I was really worried about this). Eventually, my search led me to money, and the disagreement on what money is. So I was initially not interested in how money works, but in how debt works at macro level. This starting point probably made it possible for me to come up with a solution where money more or less disappears (it is harder to make it disappear if you're trying to study it -- you assume it exists). The funny thing is that now that I've worked my way down to simple transactions within a monetary economy, I need to work my way backwards to debt at macro level. Anyway, it is most of all this debt problem I've been trying to solve, and I think I have a solution to it. You talk about a "conventional view". But it isn't a shared theoretical understanding, is it? People start with the conventional view -- as we all do, having learnt in our childhood that money is an object -- but end up in different places. You actually might think I'm further away from the conventional view than I really am. I'm not necessarily saying that the conventional view is wrong (in everyday life and microeconomics it seems to work pretty well). For instance, my separating of the UoA and a central bank credit balance probably doesn't have any practical consequences. A $1 Fed note will always we worth $1 (edit: either due to custom or legal tender laws). The only thing that changes is the way we look at it. By concluding that goods are priced in an abstract UoA, we establish primacy of the UoA (it exists independently of the records). This allows us to separate the records of transactions from the actual transactions. If I sell you something priced at $100 on credit and write your debt down in my ledger, the ledger entry didn't create "100 dollars". The balance is not an object we could call "100 dollars". I extend this to the central bank.
|
|
|
Post by Antti Jokinen on Oct 19, 2017 7:17:07 GMT
Unrelated to the above, I had problems understanding the interaction between the skilo unit of account, credits, and inflation. Either I'm not smart enough to understand, or you didn't explain it properly, or it just doesn't make sense. What did I say about it? Perhaps you could tell how this is conventionally viewed? I could then say if I find something wrong with the explanation.
|
|
|
Post by oliver on Oct 19, 2017 11:54:15 GMT
Sounds like you've been reading your Mike Sproul. Do we really want to get into the differences between bank money and central bank money here? Seems like that's an entirely different debate worthy its own thread. In a TOM world where people own entries that represents underlying barter transactions, the concept of money demand and reflux are irrelevant, anyways. There is no medium of exchange, just bookkeeping. Nothing actually moves between accounts. So there is no monetary instrument that can be in excess demand, and therefore nothing to reflux. That's why you and Antti had so many troubles fitting your TOM description to the actual system in which open market operations set the overnight rate... because to explain the effectiveness of OMOs you need some sort of demand for settlement balances, and a limited supply of those balances, which allows a central bank precise control over the overnight rate. Thinking about this again, now with a bigger screen to stare at, I think we're not in disagreement here. As Antti said above somewhere, there is nothing in a 'recording of barter transactions' world that forbids the record keeper to set a credit limit. This is also why I took issue with the 'ultra simplified' model in which the banker is completely passive. I think she has an active role in representing the interests of the commons, be it the people of a country in the case of a central bank, or more narrowly the stakeholders of a commercial bank.
One can then interpret the interest rate in the overnight market as the interplay between the demand by banks for higher credit limits vs. the actual credit limits set by the central bank. The prior also being a result of the availability of alternative, bilateral credits. Nevertheless, in all I think it is more precise to say that the target is the interest rate, whereas the nominal amount of credit granted is the dependent variable.
It is then a separate question whether or to what extent credit limits affect prices.
|
|
|
Post by oliver on Oct 19, 2017 12:37:23 GMT
On a completely different note, I've been thinking about why the WIR system in Switzerland is the only stable LETS. I think it may have to do with a particular debt collection procedure that seems to be unique to Switzerland. Generally, a debt collection request can be filed at the local debt collection office without having to produce proof of the validity of the claim. I.e. one does not need to produce a deed. For debt denominated in WIR, the rules are set as follows (German link). After a 30 day wating period (plus 1 week for postal delivery), the creditor can have his debt redenominated into Swiss Franc at the exchange rate of 1:1 and then may proceed to file a debt collection request as described above. The debtor is then served a summons for payment. The same applies for debts denominated in foreign currency (except for the exchange rate). So it is not so that debtors can just walk away from their WIR debt. They are effectively treated the same as any other debtor would be. Does that sound like a plausible explanation for the relative sability of the WIR system in comparison to other LETS?
|
|
|
Post by JP (admin) on Oct 20, 2017 4:13:19 GMT
So I was initially not interested in how money works, but in how debt works at macro level. This starting point probably made it possible for me to come up with a solution where money more or less disappears (it is harder to make it disappear if you're trying to study it -- you assume it exists). The funny thing is that now that I've worked my way down to simple transactions within a monetary economy, I need to work my way backwards to debt at macro level. Anyway, it is most of all this debt problem I've been trying to solve, and I think I have a solution to it. Reminds me of Michael Woodford! He took money out of macro as well. So are you just taking it out because it makes everything else easier?
|
|
|
Post by JP (admin) on Oct 20, 2017 4:18:46 GMT
Unrelated to the above, I had problems understanding the interaction between the skilo unit of account, credits, and inflation. Either I'm not smart enough to understand, or you didn't explain it properly, or it just doesn't make sense. What did I say about it? Perhaps you could tell how this is conventionally viewed? I could then say if I find something wrong with the explanation. You separate the unit of account from central bank credit balances. But the consensus view is that they are currently fused. Which means that whatever happens to the instrument gets transferred to the economy-wide price level. The consensus view also grants that there have been historical episodes when the unit of account and medium of exchange have been separated, in which case whatever happened to the instrument did not have a price level effect.
|
|
|
Post by JP (admin) on Oct 20, 2017 4:25:20 GMT
Thinking about this again, now with a bigger screen to stare at, I think we're not in disagreement here. As Antti said above somewhere, there is nothing in a 'recording of barter transactions' world that forbids the record keeper to set a credit limit. This is also why I took issue with the 'ultra simplified' model in which the banker is completely passive. I think she has an active role in representing the interests of the commons, be it the people of a country in the case of a central bank, or more narrowly the stakeholders of a commercial bank.
One can then interpret the interest rate in the overnight market as the interplay between the demand by banks for higher credit limits vs. the actual credit limits set by the central bank. The prior also being a result of the availability of alternative, bilateral credits. Nevertheless, in all I think it is more precise to say that the target is the interest rate, whereas the nominal amount of credit granted is the dependent variable. Makes sense to me. So what is happening when a central bank does an open market operation in order to affect the overnight rate? How do you translate that to a 'recording of barter transactions' world?
|
|
|
Post by oliver on Oct 20, 2017 14:03:17 GMT
I'll give it a shot, although I'm not sure this is what you or Antti have in mind. It is his theory we're discussing here, after all, and I can't claim to have one of my own. I'm just the annoying sidekick who happens to agree with him most of the time. Also, my thinking is still quite fuzzy on this matter. But maybe it'll defuzz if I put it in writing...
1. Government issues an IOU / bond in order to finance expenditure (receiving goods for nothing in Antti talk). The bond is a promise by the government to deliver goods in future in exchange for the goods it is about to receive from the public (for public or its own use). Ideally, the real return on the investments made (the goods the investments produce) will cover the commitments made (principal + interest). If not, the public will have to be debited for the real shortfall via taxation or inflation at some point.
2. Private agents agree to invest in the bonds. Thereby, they officially temporarily 'give up' their ability to redeem their credits for goods (although it is unclear whether they would have done so in a counterfactual world in which bonds could be used to purchase goods). Sellers of the investment goods are credited. Overall 0-interest credits (aka money) remain unchanged, overall bond credits increase, government indebtedness increases, sales of real goods increase. All types of credits and debits 'reside' in the books of the banking system.
3. Systemwide, such actions first lead to an increase in interest income from government (the collective) to individual agents, including banks. An increased supply of new bonds will generally decrease the demand for them, thus putting upward pressure on the nominal interest rate. That translates as an increase in real commitments for each initial $ of investment made.
4. Banks (or bank operators in our lingo) have their own, specific demand for government bonds, depending on their expectations of 'movements' of short term customer credits to and from other banks, i.e. their liquidity. But that specific demand reveals nothing about the absolute size of bank balance sheets, neither for one bank nor for the banking system as a whole. The demand only arises due to the systemic setup of having more than one bank and giving customers free choice where to bank (+ minimum reserve requirements in some countries). In normal times, each bank will want to economise on holding short term, 0-interest assets in favour of longer term, interest bearing assets.
5. Enter the CB. It purchases or sells bonds from banks and the public on the open market in order to implement the target interest rate (in the interbank market) which is revealed to it by its reading of the crystal ball (aka the model). To the extent that it purchases bonds from banks, their interest income will decrease, thus lowering the threshold at which banks are willing to take on new, risky assets (make new loans / extend new credit), should the opportunity arise.
6. In conclusion, OMOs do not directly change the size of the pie, nor do they directly affect the ability of banks to extend credit to either government or private agents. A lower interest rate will however generally lead to more investments and / or higher asset prices (as I was forced to admit, recently :-)). There is a hidden assumption that anyone holding a bond but wishing to make a purchase of goods will find a liquid market into which he can sell his bond. And, to the extent that there isnt, and an increase in the interest rate results, see 5.
7. The interest income the CB earns in the case of a purchase is credited back to government, thus further reducing the real burden of the investments made. Overall, the interest rate can be adjusted to account for the actual real returns which the investments yield. (That may be congruent to Antti's proposal to have the interest rate track inflation?)
8. The net financial effect of a CB bond purchase, all other things, including the target interest rate, equal, is an overall increase in 0-interest credits (money) for non-central bank agents and a steady expansion of the CB balance sheet in step with government indebtedness. Sloppy econ shorthand: government (and everyone else) spends by issuing new money. The real (barter) implication of debt financed (government) spending is an increased real indebtedness (promise of future goods) by the public, that is all agents present and future, to those agents, present and future, who wish to redeem their 0-interest credits (money) for goods. In comparison to tax-financed spending (or spending out of income for non-government agents), the above creates more legal commitments between agents in an economy.
9. A positive interpretation of debt finance is that more commitments allow for greater smoothing of consumption as well as better utilisation of productive capacities vs. the counterfactual in that demand and supply are less prone to fee back on one another. A negative (real) implication may be that changes in expectations about the future can lead to sudden changes in demand that cannot be met by the supply side of the economy. The availability of debt finance might also lead to unrealitsic assumptions about actual, real returns of the inverstments on behalf of borrowers (unrealistic asset prices). It is most certainly the case that increased commitments, especially nominally rigid ones and particularly among agents who are not able to absorb the downsides of the risks, add to the fragility of the financial system. And we know where that can lead us... So all credits and debits may sum to 0, but that doens't mean the size of balance sheets, whether public, private, bank or central bank, don't matter.
I hope that made some sense and isn't full of facutal errors. And to the extent that Antti agrees with any of the above, I'm sure he can fill in the intermediate steps by mapping them on to the underlying barter transactions in his own lingo.
|
|
|
Post by Antti Jokinen on Oct 20, 2017 19:21:52 GMT
Reminds me of Michael Woodford! He took money out of macro as well. So are you just taking it out because it makes everything else easier? I should have been more careful with my words. I don't take money out. I make money as we all have learned to know it disappear. Money is not an object, it doesn't flow and it cannot be transferred between accounts. A seller of goods receives absolutely nothing from a buyer. What we are used to call the seller "receiving money" is just a record, made by a third party, of the fact that the seller received nothing. The record is not nothing (it matters whether it is made or not), yet it is a record of the seller having received nothing from the buyer. Think of it this way: the money I make disappear is like the sun which travels across the sky during the day. If you choose to see the sun travel across the sky, I can't make it disappear for you. I can only ask you to consider whether the rotation of the Earth could make it look like the Sun travels across the sky. (Already Mitchell-Innes made this comparison, so don't take this as me saying I'm Copernicus! He did study money, too, as you know... Copernicus-Gresham's Law, etc.)
|
|
|
Post by Antti Jokinen on Oct 20, 2017 20:16:46 GMT
You separate the unit of account from central bank credit balances. But the consensus view is that they are currently fused. Which means that whatever happens to the instrument gets transferred to the economy-wide price level. The consensus view also grants that there have been historical episodes when the unit of account and medium of exchange have been separated, in which case whatever happened to the instrument did not have a price level effect. What I'm saying is that they should be conceptually separated. Whatever happens to the price level affects directly and proportionately the (real) value of the credit balance -- not vice versa. Prices are always raised or lowered by a seller, not by something that happens to credit balances (I don't know what can happen to them, other than that their quantity changes?). The historical episodes you are talking about are another thing. It is not what I mean by separation here. See my first blog post again. There's a UoA, skilo, in existence even before any monetary/LETS system is established. People express the prices of goods in that unit. Then they decide to start keeping skilo-denominated records of their trades (or "gifts", as in a gift economy). They don't call a SK1 credit balance 'a skilo' and they would be surprised if you suggested that they price their goods in those credit balances. They continue to price their goods in abstract skilos, just like they did before the LETS system. Yet, now that everyone, or nearly everyone, is using the LETS system, how could the sticker price differ from the amount of the credit entry the seller expects on his account? The whole point of the system being the recording of prices of goods bought and sold, how could the seller say that the price of his wares is SK10 but he demands a SK12 credit entry on his account? For all practical purposes this would mean that the price of his wares is SK12, not SK10. Of course, in case of stubborn individuals, it might help if there was a law (comparable to legal tender statutes) which stated that a seller, or a creditor, must accept a credit entry that matches the sticker price, and not more. That would guarantee that the sticker price and the credit entry won't diverge. Fast forward 7000 years and perhaps all this has been forgotten. People see that the SK1 credit entry (not balance) on the seller's account is always enough if one wants to buy goods priced at SK1, so they might start thinking that goods are somehow priced in these credit entries and not in an abstract skilo. Do you see what I mean?
|
|
|
Post by JP (admin) on Oct 23, 2017 18:23:56 GMT
Oliver and Antti, my slowness in getting back to you is due to a combination of my attention span hitting its limits and the fact that I'm loaded down with a few other duties these days that are maxing out my capacity. So I'm going to sign off for a bit. My fault, not yours. This has been an interesting topic though, and if you ever come up with some killing argument for the TOM viewpoint, don't hesitate to point them out to me. Historical examples that can only be explained by a TOM point of view are the best since it is hard to refute empirical proof. In the meantime, feel free to bring up other topics on the discussion board.
|
|
|
Post by oliver on Oct 24, 2017 8:06:37 GMT
Oliver and Antti, my slowness in getting back to you is due to a combination of my attention span hitting its limits and the fact that I'm loaded down with a few other duties these days that are maxing out my capacity. So I'm going to sign off for a bit. My fault, not yours. This has been an interesting topic though, and if you ever come up with some killing argument for the TOM viewpoint, don't hesitate to point them out to me. Historical examples that can only be explained by a TOM point of view are the best since it is hard to refute empirical proof. In the meantime, feel free to bring up other topics on the discussion board. Your slowness is probably for the better because I'm having some serious blogger's remorse about my last comment above. It's at least 50% drivel, I fear. Some time to think or do other things is good.
|
|