Post by botanic on Jun 12, 2013 10:37:49 GMT
Peterv
I think I can resolve our differences in the supply curves but let me deal with two incidental matters first.
(1) When you explain the demand curves you only refer to demand from businesses (who need to make a profit). However I think there is considerable demand for loans for what might be called 'lifetime smoothing' or 'consumption smoothing' purposes. Needing to buy a car is one reason for wanting such a loan. Also buying a house must cost tens of thousands just to cover building costs, even if there were no constraints on the number of houses that could be built. So people will want 'lifetime smoothing' loans to buy houses, even if there is no lack of houses to buy.
(2) My SillyBank example was intended to add another factor to the mix, not as a total explanation. Indeed you yourself lead me to the idea! At the start of the thread I might have assumed that a bank would need to offer savers x% interest to persuade them to put their money in a term account rather than elsewhere. But I see now that each bank must offer interest rates to savers that are at least comparable with what other banks are offering. Otherwise all the deposits would flee to other banks as in the SillyBank example. So each bank is bidding against the others to maintain their share of the term accounts. It is like a price war between supermarkets. The overall result should be that savers get higher interest rates than they would if each bank had a captive set of savers.
Now to resolve the differences in our supply curves. Start with the situation after PM has been introduced.
You are incorporating the idea of Credit Rationing in your supply and demand graph. I have looked at this idea in the 'Modernising Money' book and it is clear that they lump all the loans together in one graph, regardless of risk. Then they refer to the interest rate at the market clearing point. Then they suggest that banks should lend at less than this interest rate in order to avoid the cost of too many loan defaults. In other words they are costing the loans without taking account of any loan default costs initially. Then they are saying that banks shouldn't follow the market clearing rate slavishly because there will actually be some loan default costs!
I think it is okay to use this idea provided you realise that you are including loans of varying risks in your single supply and demand graph. I guess you do because the supply curve slopes backwards at high interest rates. Presumably you are implying that banks would supply less loans at these high rates because more defaults would occur. However if you are representing things in this way you cannot also include a set band at the bottom of the graph to cover bad debts. There can be no such band if the riskiness of the loans is increasing as one moves up the supply curve.
In contrast I have proposed a number of supply and demand graphs, each one representing loans with a different percentage risk. On my model the loans demanded and supplied to over-risky borrowers would be represented on one of the higher risk graphs. Consequently none of my supply curves would be likely to slope backwards.
So I think the difference in the upper sections of our supply curves only occurs because we are using different models. Which model is best? If banks really lump all their loans together and lend at a single interest rate regardless of risk, then the single graph model may be best. However if banks take account of each borrower's credit history and the riskiness of the project and the quality of any collateral, then my multi-graph model may represent the true situation better.
Now to deal with the shape of the supply curve in our current fractional reserve system.
I agree with your single graph model more or less, except for one thing. First you assume that the amount of money supplied by savers is essentially the same both before and after the introduction of PM. I agree with this. However your graph also suggests that only twice as much money is available to borrowers before PM in comparison with the amount that would be available after PM. This implies that banks are currently operating with about 50% reserves. In other words they are only creating £2 from 'thin air' for every £1 that is being saved with them. I don't think so! I would expect to see your 'S1' supply curve (for the current fractional reserve system) far to the right of the 'S2' supply curve (for the system after PM). I would expect to see it at least ten times to the right.
If your 'S1' supply curve was altered in this way then there is a possibility that the market clearing point would be on the lower portion of the supply curve, instead of being on the vertical portion. In other words there is a possibility that not every bank would be stretching the money it gets from savers to the very limit by creating vast numbers of loans - at least not all the time.
I would still prefer to use my multi-graph model with the current fractional reserve system. I think it deals with the riskiness of loans better, especially on the flat portions of the various supply curves. However I would change the supply curves in one way. I agree that the supply of money to banks from savers is probably fairly inelastic, especially over the lower range of interest rates that is normally paid to savers. It follows that the supply curves of loans to borrowers must turn vertical at their extreme right ends, otherwise the reserve ratio couldn't be maintained. However I still think that the largest portion of each supply curve will be almost flat and horizontal. This is because banks can pay savers high interest rates, whilst only passing on a small proportion of this particular cost to each individual borrower. For example it is quite possible for a bank to pay savers 5%, 10% or even 15% interest and still lend to borrowers at only 4% interest.
Finally I am agreeing with you that savers will save regardless of the interest rate they get, within certain limits. But My SillyBank example suggests that banks may also give them higher rates than they expect!
I think I can resolve our differences in the supply curves but let me deal with two incidental matters first.
(1) When you explain the demand curves you only refer to demand from businesses (who need to make a profit). However I think there is considerable demand for loans for what might be called 'lifetime smoothing' or 'consumption smoothing' purposes. Needing to buy a car is one reason for wanting such a loan. Also buying a house must cost tens of thousands just to cover building costs, even if there were no constraints on the number of houses that could be built. So people will want 'lifetime smoothing' loans to buy houses, even if there is no lack of houses to buy.
(2) My SillyBank example was intended to add another factor to the mix, not as a total explanation. Indeed you yourself lead me to the idea! At the start of the thread I might have assumed that a bank would need to offer savers x% interest to persuade them to put their money in a term account rather than elsewhere. But I see now that each bank must offer interest rates to savers that are at least comparable with what other banks are offering. Otherwise all the deposits would flee to other banks as in the SillyBank example. So each bank is bidding against the others to maintain their share of the term accounts. It is like a price war between supermarkets. The overall result should be that savers get higher interest rates than they would if each bank had a captive set of savers.
Now to resolve the differences in our supply curves. Start with the situation after PM has been introduced.
You are incorporating the idea of Credit Rationing in your supply and demand graph. I have looked at this idea in the 'Modernising Money' book and it is clear that they lump all the loans together in one graph, regardless of risk. Then they refer to the interest rate at the market clearing point. Then they suggest that banks should lend at less than this interest rate in order to avoid the cost of too many loan defaults. In other words they are costing the loans without taking account of any loan default costs initially. Then they are saying that banks shouldn't follow the market clearing rate slavishly because there will actually be some loan default costs!
I think it is okay to use this idea provided you realise that you are including loans of varying risks in your single supply and demand graph. I guess you do because the supply curve slopes backwards at high interest rates. Presumably you are implying that banks would supply less loans at these high rates because more defaults would occur. However if you are representing things in this way you cannot also include a set band at the bottom of the graph to cover bad debts. There can be no such band if the riskiness of the loans is increasing as one moves up the supply curve.
In contrast I have proposed a number of supply and demand graphs, each one representing loans with a different percentage risk. On my model the loans demanded and supplied to over-risky borrowers would be represented on one of the higher risk graphs. Consequently none of my supply curves would be likely to slope backwards.
So I think the difference in the upper sections of our supply curves only occurs because we are using different models. Which model is best? If banks really lump all their loans together and lend at a single interest rate regardless of risk, then the single graph model may be best. However if banks take account of each borrower's credit history and the riskiness of the project and the quality of any collateral, then my multi-graph model may represent the true situation better.
Now to deal with the shape of the supply curve in our current fractional reserve system.
I agree with your single graph model more or less, except for one thing. First you assume that the amount of money supplied by savers is essentially the same both before and after the introduction of PM. I agree with this. However your graph also suggests that only twice as much money is available to borrowers before PM in comparison with the amount that would be available after PM. This implies that banks are currently operating with about 50% reserves. In other words they are only creating £2 from 'thin air' for every £1 that is being saved with them. I don't think so! I would expect to see your 'S1' supply curve (for the current fractional reserve system) far to the right of the 'S2' supply curve (for the system after PM). I would expect to see it at least ten times to the right.
If your 'S1' supply curve was altered in this way then there is a possibility that the market clearing point would be on the lower portion of the supply curve, instead of being on the vertical portion. In other words there is a possibility that not every bank would be stretching the money it gets from savers to the very limit by creating vast numbers of loans - at least not all the time.
I would still prefer to use my multi-graph model with the current fractional reserve system. I think it deals with the riskiness of loans better, especially on the flat portions of the various supply curves. However I would change the supply curves in one way. I agree that the supply of money to banks from savers is probably fairly inelastic, especially over the lower range of interest rates that is normally paid to savers. It follows that the supply curves of loans to borrowers must turn vertical at their extreme right ends, otherwise the reserve ratio couldn't be maintained. However I still think that the largest portion of each supply curve will be almost flat and horizontal. This is because banks can pay savers high interest rates, whilst only passing on a small proportion of this particular cost to each individual borrower. For example it is quite possible for a bank to pay savers 5%, 10% or even 15% interest and still lend to borrowers at only 4% interest.
Finally I am agreeing with you that savers will save regardless of the interest rate they get, within certain limits. But My SillyBank example suggests that banks may also give them higher rates than they expect!