What will happen to interest rates with PM?

[peterv]

This question is often asked! I don’t know the answer, but please add your thoughts to this thread.

One of the objectives of monetary reform is “to reduce the burden of personal, household and government debt” (Modernising Money p25). If you are interested in the housing market, or in small businesses, then you might hope to see interest rates fall. But if you are a saver, or are interested in preserving the value of your pension, then you might hope to see them rise. When we asked Ben Dyson this question, he said that interest rates would not change dramatically in either direction.
positivemoneysheffield.pbworks.com/w/page/65864643/Bens%20answers

97% of our money supply is bank credit, and credit=debt, so we might try to analyse the supply and demand for credit, using the classical method of demand/supply curves.
See diagrams at sdrv.ms/10NwYFj

Fig 1 shows a typical curve where quantity (q) is shown on the horizontal axis and price (p) on the vertical. Since we are talking about credit, price = interest rate. The supply curve S rises to the right as more is supplied at higher price, and the demand curve D falls as more is demanded at lower price. Equilibrium is reached where the two curves cross, and the total revenue is given by the area of rectangle p*q. Total revenue in this case is the total amount paid by borrowers to lenders in our economy and if you accept that money lending transfers wealth from the poor to the rich, then the bigger the rectangle p*q, the more unequal our society becomes.

I hit a problem at this point. It doesn’t cost anything to create credit, and endogenous money theory says that banks can create any amount of credit, limited only by how much we want to borrow and can be trusted to repay. I guess that might mean the supply curve is a horizontal line – but then what sets the interest rate? I searched for “credit supply curve” and found numerous academic papers on the subject. There is a “horizontalist” school which (if I understand it) says the supply ‘curve’ is a series of horizontal lines, or a discontinuous line. There is a “structuralist” school which says the supply curve rises normally. Some even argue that the curve doubles back on itself as the only people willing to borrow at high interest rates are high-risk borrowers, so banks are reluctant to lend at high rates (search for “credit rationing”, and see Modernising Money p110). I guess it’s a problem with at least 3 dimensions – quantity/interest/risk - and impossible to show on a 2D graph.

In the absence of anything better, I’m assuming the supply curve rises normally. What happens under Positive Money? Banks will only be able to lend the money that has been invested in them, so now there is a real limit on the supply of credit. The supply curve will rise more steeply, and might become vertical at some point (Fig 2). A new equilibrium will be established at a higher interest rate p2, and a lower quantity q2. In other words, there will be less debt, but those who rely on credit will pay higher rates. Total revenue is now given by p2*q2 which might be smaller or greater than p1*q1, depending on the slope of the demand curve.

But that’s only half the story – the hole in the money supply is filled by debt-free money issued into the economy. There will be less demand for credit, so the demand curve moves to the left, and the new equilibrium is at point p3*q3 (Fig 3). The level of debt falls further to q3, and the interest rate falls to p3 – whether it ends up slightly higher or lower than its original value of p1 depends on the shape of the curves, but my conclusion is that Ben Dyson was right when he said that rates would not change dramatically.

The total revenue is now p3*q3, which must be a lot smaller than p1*q1. This implies a much lower transfer of wealth from the poor to the rich, and in my view this is a much more important conclusion than whether interest rates rise slightly or fall slightly.

(edit 4/6/13 link to attachment inserted above)

Attachments:

[botanic]

Peterv

I have just designed a representation of the supply and demand curves for credit, especially for you. I hope you will find it both persuasive and informative!

First we need to allocate all potential supplies and demands for credit into different categories, according to the riskiness of each.
Then we must construct a separate supply and demand graph for each category. Having so many graphs sounds complicated but I think they would all be similar to one another and each one would react to the introduction of PM in a similar way. So we only need to concentrate on one graph and remember the existance of the others.

The demand curve on each graph would presumably be more or less normal - both before PM is introduced and afterwards. It would start high on the left because only a small amount of credit is demanded when the interest rate is high. Then the curve would slope downwards as it goes right because more cedit is demanded when interest rates drop low.

What would the supply curve be like on each graph before the introduction of PM? I think it would be flat and horizontal, provided we make the following assumptions:
(1) All the credit loaned is created 'from thin air' by banks.
(2) There is no need for banks to hold any reserves.
(3) There is no requirement for banks to hold any capital.
These assumptions do not hold true in reality, but it is useful to make them as a first approximation.
I say that the supply curves would be horizontal with these assumptions in place because no loan would cost more than any other one to make, especially when we remember that all the loans on a single graph would have exactly the same risk, by definition.

What would the height of these horizontal supply curves be - in other words what would the interest rates be?
At first it appears that loans created 'from thin air' have no cost, but this is not really true. When a loan defaults the bank must 'pay off' the loan from its own capital. So the real cost of these loans is derived from the ones that default. Consequently the supply curves on our graphs would be at different heights, according to the risk involved in each particular category.

Now what would happen after PM is introduced?

The supply curves would each start a little higher on the left than the earlier horizontal ones, because private lenders would want some recompense and this would be added onto the bank's cost of supplying the loan. Then the supply curves would slope upwards as they go across each graph. This is because higher interest rates would be required to entice private lenders to lend more. So I would expect the supply curves to slope upwards from their starting positions on the left, in the same way that supply curves normally do.

The demand curves would also change after PM. Presumably each point on a demand curve would move leftwards, because less credit would be wanted at each interest rate. So we can shift each demand curve bodily towards the left.

This brings me to an important question. PM people predict that much less credit would be needed after the introduction of PM, whereas I think the amount would be somewhat less but not that much less, especially after the transition period.

If the PM prediction is correct then the demand curves can be moved way over to the left. In that case the interest rates would still be higher than before PM because no point on the new supply curves would fall below the 'pre-PM' supply curves. Furthermore the interest rates would only be near the earlier ones if far less loans were actually made. However if my expectation turned out to be true then the interest rates would probably be much higher than they were before PM and the amount of money actually loaned would be less than it was before PM but more than the PM prediction says it would be.

[peterv]

Thanks for that botanic, it's a good analysis; it bridges the gap between "horizontalist" and "structuralist".

I remember once learning that credit interest rate is built up of separate components -
(a) the administrative cost. This should probably include "normal profit" ie. a level of profit comparable to that achievable in other sectors
(b) the cost of insuring against default (ie the bad debt rate).
(c) an "opportunity cost" - ie. how much you need to pay me to entice me to forgo immediate spending of my money. This is the usurious element.

It is this last element which is likely to be higher if funds for lending are limited, than if they can be created "out of thin air".

So I accept your premise that, other things being equal, interest rates will be higher. I don't have a big problem with that - if you accept that the cause of our high indebtedness was cheap credit, then you have to accept that higher cost credit might be part of the solution.

I don't know if we can quantify how much rates would rise, or by how much demand for credit will reduce?

One further point - the "risk" element might be different post-reform. I have something to say about that (as I guess you do), but that's perhaps a separate discussion which we might pick up on the topic of SMEs

[botanic]

I have just uploaded a diagram to show 3 versions of the Supply and Demand graphs that I described in my last post. It is here:
imageshack.us/a/img6/7615/supplydemand.jpg

All three graphs show the situation for loans with a 2% chance of defaulting.
There would need to be more graphs to represent loans with different default risks.

The first graph shows the situation before the introduction of PM.
The second graph is after PM, but before the demand curve has been altered.
The third graph is after PM with the demand curve altered - the demand has been halved at each interest rate.

I hope this makes the idea clearer than words alone!

[botanic]

Peterv

... if you accept that the cause of our high indebtedness was cheap credit, then you have to accept that higher cost credit might be part of the solution.

I would agree with this if I thought that all debt was bad. However I think there are good loans and bad loans, and therefore there must be good and bad debts.

Here are some examples of good loans:
(1) A loan that enables someone to buy a car when the need arises, provided they weigh up the dis-benefits of the repayments.
(2) A loan to someone on benefits who needs to buy a washing machine, provided they think the considerable burden of the repayments is out-weighed by not having to wash their clothes by hand.
(3) A loan that allows a businessman to employ an extra person, provided they think the extra revenue generated would cover the interest on the loan.

And now some examples of bad loans:
(1) A loan that someone has been pressured into taking, against their better judgement.
(2) A loan on terms that will inevitably lead to grief.
(3) A loan where the borrower doesn't get the right information to make a good judgement.

Ideally I would like all good loans to have the lowest possible interest rates.
All bad loans could have high rates. However I am not really convinced that high interest rates do much to discourage bad loans because typically these loans involve a lack of proper judgement.

[peterv]

I think our earlier posts overlapped. This response is to our previous discussion about the credit supply curve.

When I saw your graph, I realised that I had made an omission in my description of the components of credit, and you have made the same omission in your graph.

Post reform, banks can't make loans without deposits. Clearly they will need to entice us to take money from our transaction accounts and put them into investment accounts, and that the more they pay the more we will invest. This is the 'investors' profit' in your graph.

Pre-reform, they don't need us to invest our money before they can make loans, but they still have to pay interest on deposit accounts. The loans create the deposits. The cost of servicing term deposit accounts has to be added to the cost of providing credit. It is the equivalent to investors' profit - call it 'depositors' profit' if you like. I don't know the shape of it - it may be another horizontal line and is currently around 2 - 2.5%.

It's more complicated to explain why the banks have to pay interest on term accounts if they don't need to attract deposits in order to make loans. All money ends up as a deposit somewhere in the banking system, so why do banks prefer it to be in a term account rather than a demand account? I guess it's about inter-bank competition - having made loans, they don't want the deposits to drain away to other banks which would drain their liquidity.

Anyway, I think this element needs to be added to the first figure in your graph.

[botanic]

So here's a thing - have I omitted something from my Pre-PM graph or have you invented an idea for something that doesn't really exist (like phlogiston)?

I believe that the supposedly missing factor would emerge naturally if the simplifying assumptions which I made were removed. I believe that the flat horizontal supply curve on the pre-PM graph would really slope gently upwards and probably more steeply at its right hand end.

My reasons are as follows:

(1) Although reserves are not legally required, banks actually need them in order to make net inter-bank transfers. Depositing money provides a relatively cheap source for such reserves, especially if it comes from another bank. So banks want people to hold deposits with them and if it is held in a time account then so much the better. This is why banks pay interest (or desist from bank charges) as I understand it.

(2) The total loans made by banks may exceed the total amount that they can create 'from thin air'. Remember that the immediate reason why Northern Rock collapsed was because it was borrowing money from the money markets in order to service its long-term loans, and those markets dried up. The requirement for extra money to lend would result in the right hand end of the supply curve to slope upwards more steeply.

In conclusion I believe my horizontal supply curve would have a somewhat different slope in reality, and this would account for the interest that banks actually pay to savers under the current system.

A correction!

I can see that your suggestion may be good - the gentle upward slope in the supply curve could be represented as an additional amount of interest that needs to be given to savers/depositors, even under the current system. Alternatively it could be represented as an extra cost which banks need to bear.

Your suggestion may be best, especially when money has to be borrowed from money markets. However I wouldn't invent a new name for it. It is still a form of 'investors profit'. There is just much less of it than would be required with PM.

[botanic]

I have thought of a better way to deal with banks paying savers some interest under the current system.

Banks cannot create an infinite amount of loans 'from thin air' without any reserves, whatever the regulations do or don't say! So banks need to obtain more reserves in order to create more loans. This gives rise to a new cost, which we could call 'the cost of reserves'.

How much would this cost be? Suppose reserves are obtained by paying savers 3%. Each £1 of new reserves might enable the bank to create £100 of new loans 'out of thin air'. So the extra cost for the new loans would only be 0.03%.

We could include this relatively small 'cost of reserves' in with the other bank 'costs and reasonable profits' category. Alternatively we could add it as an extra cost category. Either way it would result in a very slight upward slope in the banks' costs, which would then be transmitted to the supply curve.

If the figures I have chosen are anything like plausible then the change to my Pre-PM graph would be almost negligible.

[peterv]

Whatever you call it, I think it's best to treat what I have called 'investor profit' and 'depositor profit' separately, otherwise you are double counting in the post-reform case.

Banks cannot create an infinite amount of loans 'from thin air' without any reserves

They can in the long run, and when viewed as the banking system as a whole, but I agree that an individual bank in competition with others is limited in the short run by the need to conserve liquidity. Each bank wants to expand its loan book to increase its market share, but not so fast that it becomes illiquid. There is a time dimension to the supply curve as well!

So reserves ARE relevant in the pre-reform case, but I don't think your answer of 0.03% can be right. They pay effectively 0% to depositors on demand accounts, and say 2% on term accounts. It must be worth at least 2% to them not to have to hold reserves. This 2% is what I am calling 'depositors' profit' and it's pure profit to the depositor, as they bear no risk in the current system. If I could understand why it's there then I might agree that some of it would also apply post-reform, but at the moment I'm struggling to get beyond this.

You have not yet convinced me why you think 'investors' profit' would be so much greater than 'depositors' profit', except at the extreme right-hand side of the supply curve where there is a long-run limit post reform.

[peterv]

Here's an alternative description.

In the current system, the money supply is unlimited long term, but limited short term. We each keep enough in demand accounts for our immediate needs, and put any surplus into term accounts. The proportion we can invest is not really influenced by the interest rate, so our supply of savings is inelastic. Banks prefer us to keep our money in term accounts because it gives greater stability to their assets. They compete among themselves, and investors get a rate of return which is really determined by how much profit the banks can make on loans.

Post reform, the money supply will be limited both short and long term. Our banking behaviour won't change - we'll keep enough in transaction accounts for our immediate needs, and put the surplus into investment accounts. Banks will still compete for our savings, and the rate will still be determined by how much surplus the banks make on lending. There will be as much to lend post-reform so the supply curve won't change

[botanic]

Peterv

I will be offline for a few days and I doubt whether I can persuade you of much before I return. But I will try!

I read once that banks could create infinite amounts of money 'from thin air', provided they all grew exactly in step. But in practice this never happens. So according to my understanding, the main reason why individual banks need to hold reserves is to make payments to other banks when customers' money flows from one bank to another. Each bank only needs to hold enough reserves to cover its net inter-bank payments but the amounts can still be very large. In any case I assume that each bank needs to hold reserves in proportion to the total loans it has made.

I read recently that LLoyds intends to increase its reserves to 9%. It seems unlikely that they would want to do this unless it was necessary for them to hold this much reserves. See the following link; www.bbc.co.uk/news/business-22728433

How do I get to the 0.03% figure? Let's assume that all deposits and all loans are made £100 at a time, to keep the numbers simple.

Suppose a particular bank wants to make 100 loans of £100 each, using new money which it is going to create from thin air. So the bank will supply £10,000 of new loans in total.

Suppose the bank thinks it needs to hold only 1% reserves in order to deal with net payments to other banks. (Some new borrowers are likely to spend their borrowed money in such a way that it will end up in another bank.) Then the bank will need to increase its reserves by 1% of £10,000, which equals £100.

So the bank obtains the extra reserves it needs by paying 3% to its savers in order to entice one of them to deposit £100 extra. The interest on this new deposit equals £3 pa. However the bank only needs to increase the cost of its loans by 0.03%, because this percentage of £10,000 also equals £3 pa.

This is why I think the supply curve on my pre-PM graph would only slope upwards a small amount. But even if the banks need to hold 9% reserves instead of 1%, as LLoyds thinks it does, then the supply curve would slope upwards a bit more, but not much. In any case the banks don't need to pass on the whole interest rate they give their savers, to their borrowers. They only need to pass on a small fraction of it.

[peterv]

Your description of why banks need reserves is correct. I've been struggling to understand where what I called "depositors' profit" comes from in the current system, and I've come round to your view that the cost of reserves is not significant.

I guess the answer is that demand for credit is determined by how much benefit borrowers expect to get. At rate x%, borrowers will want to borrow for schemes where the benefit is greater or equal to x%. Normally, that will more than enough to cover costs and bad debt, so there is excess profit to be had. Competition between banks ensures that most of this is passed to holders of term accounts, in order to retain market share and avoid a drain on their liquidity.

That's where "depositors' profit" comes from

They only need to pass on a small fraction of it.

I suspect that they have to pass on most of it. If one bank is making excess profit, they only need to offer a slightly higher rate to depositrs and capure a large share of the market forcing other banks to follow.

[botanic]

Some replies!

In the current system ... We each keep enough in demand accounts for our immediate needs, and put any surplus into term accounts. The proportion we can invest is not really influenced by the interest rate, so our supply of savings is inelastic.

I guess this is right for most ordinary savers. An extra percent increase in interest rates makes little difference to the amount of interest actually received when one's savings are about £10,000 or less (especially after the interest has been taxed). It is unlikely that the difference would persuade most people to save extra money instead of spending it.

I have made a new version of my 'After PM' supply and demand graph to show your idea.
img809.imageshack.us/img809/944/supplydemand2.jpg
Note that the inelastic vertical portion of the supply curve can be moved left or right, depending on how much one thinks people would want to 'save' in their investment accounts after PM. Incidentally I have made the top of the supply curve more elastic because I think people would invest more instead of spending if the interest rates went high enough to make it worthwhile.

Now for a different point:

I suspect that they have to pass on most of it. If one bank is making excess profit, they only need to offer a slightly higher rate to depositrs and capure a large share of the market forcing other banks to follow.

I think you read my reference to banks passing on interest rates, the wrong way round. What I said was

In any case the banks don't need to pass on the whole interest rate they give their savers, to their borrowers. They only need to pass on a small fraction of it.

In effect I was disagreeing with some of your earlier ideas where you assumed that banks must charge their borrowers an interest rate which includes the interest rate that they give to their savers. This assumption would be true under PM because the banks are inter-mediating loans from savers to borrowers. However it is not a proper assumption to make under our current fractional reserve system. In the latter system banks 'borrow' much less from savers than they lend to borrowers. So it makes no sense to assume that the interest rates have to be carried across from the first kind of transaction to the second kind, on a one-to-one basis.

[botanic]

Here is another perspective on why banks might want to pay interest on time deposits, in the current system.

Suppose we take a bank that intends to do nothing except lend money created 'from thin air'. Call it 'SillyBank'. This bank makes no attempt to attract savers. It may charge for demand accounts and offer very low interest rates (if any) on term accounts.

On Day 1 SillyBank lends large amounts of money and at the end of the day all is fine. It has lots of debt instruments on the assets side of its balance sheet and a corresponding amount of deposits (the money lent to the borrowers) on the liabilities side.

On Day 2 many of the borrowers spend their new money (why borrow money just to leave it in the bank). Virtually all this money ends up in other banks, because SillyBanks has no customers apart from its borrowers. The remaining borrowers take their money out and deposit it in other banks that pay better rates. So SillyBank is in a bad position at the end of Day 2. All the money in its deposit accounts has flowed out to other banks. This must be paid for with reserves. Either SillyBank has enough reserves from its original capitalisation to make the payments or it will be insolvent after only two days!

In any case SillyBank has had to spend an amount of reserves equal to the amount of money it lent out. This is not a good way to make profits under the current fractional reserve system!

So we may conclude that a more sensible bank would try to attract savers, especially from other banks. This would not only bring new reserves into the bank. It would also mean that the bank wouldn't actually need to maintain such a high reserve ratio as SillyBank.

[peterv]

Yes, I did mis-read your comment about passing on interest rates, sorry.

I'm close to agreeing your 'after PM' supply curve, I have a slightly different one which curves left at the top instead of right, but that doesn't affect our discussion.

Your example of SillyBank is one way of quantifying what I called depositors profit, but is it good enough to explain depositor rates in excess of 10% in the 1990s? I think there's a better explanation.

I think there are still two differences between us. One is the shape of the pre-reform curve, the other is how to explain savers' profit. You regard savers' profit as an input to the supply function, a cost which needs to be passed down, but we've not found a good way to quantify it. I am now going to suggest that quantity supplied is not determined by interest rates, and that the profit which comes from lending gets passed up. In other words, savers are price takers not price setters.

After this lengthy thread, I now have a revised model.

********* my version 2 answer ****************

See revised graph here > sdrv.ms/14S3803

1) The credit demand curve is normal. Demand is determined by how much benefit borrowers expect to get. At rate x%, borrowers will want to borrow for schemes where the benefit is greater or equal to x%.

2) Credit supply quantity is largely independent of interest rate. It is essentially a fixed quantity, both before and after reform, and I've drawn it is a vertical line. Banks need to cover their bad debt + costs + normal profit, so there is a floor rate where quantity drops to zero. It also tails off at high interest rates because low-risk borrowers are priced out ("credit rationing"). To a first approximation, the shape of the supply curve is the same before [S1] and after reform [S2], and I've assumed the quantity halves post-reform.

Under normal economic conditions, the benefit from borrowing is greater than the floor level, so excess profits are generated at the equilibrium point. It's not like manufacturing where excess profits would increase the number of suppliers, so the excess profit is not eliminated. Competition forces the surplus to be passed on to term account holders pre-reform, investment account holders post-reform.

3) How do I justify the assumption that supply quantity is fixed before and after reform?

It's relatively easy to explain post-reform. All individuals and businesses need a certain level of transactional money, but some have a wealth surplus above their transactional needs while others have a deficit. The surplus of the first group forms the stock of loanable funds, which is borrowed by the second group. This stock is constantly turned over as old loans mature and new ones are given. The turnover rate determines the credit supply quantity at any point in time.

It's more complicated to explain pre-reform because there's no direct link between deposits and lending. Banks can create new loans faster or slower than old ones are repaid, but for now assume they are in balance, so the money supply is neither growing nor shrinking. Our entire money supply is borrowed - even the money in our 'current' accounts only exists because someone has borrowed it - and if the money supply is to remain constant then all loans must be re-lent as they mature. This is the same situation as in the previous paragraph but it's the entire money stock which is turning over.

Now factor back in the fact that the banks can create credit from nothing. If the economy is booming then new loans are created faster than they are repaid. This is best represented as a right shift of the supply curve S1, and probably a right shift of the demand curve too. The floor also drops as bad debt is low, increasing the savers' interest rate. Conversely, during a credit contraction, the supply and demand curves move left and the floor raises. The equilibrium point might then be at the floor level in which case lending severely contracts and saver rates drop to virtually nothing.

So the pre-reform supply curve is best represented as a family of vertical lines.

The post-reform curve might also move a bit, but within tighter bounds. If there is a downturn, then those with a surplus might hoard more, but this will increase the amount in their transactional accounts, inviting spending. In an upturn, there will be an upper limit to how much loanable funds can increase. The economy will be more stable post reform.

4) The demand curve will probably move left post-reform because of the injection of debt-free money. For the same level of economic activity, only half the amount of money would have to be borrowed. I haven't shown this on the diagram, but if demand is halved then the cost of credit would fall back down to its pre-reform level.