Design a better banking system [Discussion]

Flamed Diving · 01-12-2009 8:09pm #1

I thought this might be an interesting thought exercise for the Economics forum. Imagine that we are chosen to design a banking system from scratch. The powers that be have decided enough is enough and want one to be built on theoretical foundations. No more financial crises, no crony banks, no taxpayers bailing out bankers, how do we achieve this? How do we set the rules of the game, so as to avoid these problems?

I thought I would outline a few rules:

This is not a thread to have a rant about the current crisis
Posts such as "No point because bankers are evil" shall be labelled wrong, by default
All points of view are welcome, no matter how left-of-field, so long as they are valid, sensible contributions

Ok, so where do we begin? I think the first matter I would like to bring up is the matter of competition. Since we will be issuing licenses for the banks, should there be some restrictions on the number of banks? Should we simply allow licenses to be freely available to those that want them? Might we be better off having a monopolistic system? How about the Keirstu system? Perhaps different banks restricted to different sectors of the economy?

Discuss, my beauties. Discuss.

EDIT: Some people seem to have trouble getting this, so I will further explain. This is an abstraction. There is no banking system. We are starting from scratch.

nanu nanu · 01-12-2009 8:26pm

Simply put, let the banks fail.

Flamed Diving · 01-12-2009 8:36pm

Thanks for reading the OP.

silverharp · 01-12-2009 10:29pm

I'll come at it from the free market perspective. From the outset there would be no central bank or any centralised regulation. It would be left to the market to determine what is to be currency. After that I'd expect there to be the alternative of paying the bank to store your savings on the basis that they are not lent out or if the bank does lend out the savings there would be private ratings companies and information services that would rate the banks (depositors would pay for this information). The objective would be to let the market price in risk in a more effective manner as savers would have to care about the institutions they let handle their money. Even if private insurance is available for depositors, the cost of insurance would be determined by the relavive risks.

Flamed Diving · 01-12-2009 10:39pm

silverharp wrote: »

I'll come at it from the free market perspective. From the outset there would be no central bank or any centralised regulation. It would be left to the market to determine what is to be currency. After that I'd expect there to be the alternative of paying the bank to store your savings on the basis that they are not lent out or if the bank does lend out the savings there would be private ratings companies and information services that would rate the banks (depositors would pay for this information). The objective would be to let the market price in risk in a more effective manner as savers would have to care about the institutions they let handle their money. Even if private insurance is available for depositors, the cost of insurance would be determined by the relavive risks.

Would it be fair to say that this model is assuming rational agents? It is clear that there would be an optimal fee to pay banks, or interest paid based on the information provided by rating agencies, but I am very doubtful that individuals would choose this optimality, even on the average. I reckon bubbles would be very likely in such a system, as unregulated banks offer "can't go wrong" deals, (much like the debt consolidation ads you see on TV now) and offer higher rates than banks that are ranked only slightly better. I know that the whole point of this system is that failures are inevitable, and that no one will stop them from happening, and that since the information is present, the individuals cannot complain. But most people don't read the information on what they eat or drink. If they don't attain perfect information on matters of health, what hope do they have with banking?

silverharp · 01-12-2009 11:25pm

It would be more important to acknowledge uncertainty in the system then that the agents act in a rational manner. If the market creates a bubble in something then no doubt some banks will over lend and fuel the mal investment. In reality it would be a benefit to the system to have a series of mini busts as it would reinforce the idea that the future is uncertain.
As you say people dont always look after their health but the worst thing to do is to increase the moral hazard by offering free health treatment or community rating. All that can be hoped for is that a free market would pull the plug on a bubble in a more timely manner as the "smart" money leaving the system or fleeing to safety would curb lending to the riskier ventures.

Anonymous1987 · 01-12-2009 11:42pm

I think the idea of restricting a bank to certain sector could depress the sector even more. Imagine the bank for the auto sector.

Our current system isn't the worst but clearly it has flaws given very recent history so I'd like to have a few improvements in a new system. Just to note these ideas are very rough but might get the discussion going. I'd be leaning towards a regulated banking system. Some of my thoughts:

Transparency. No special purpose vehicles or other off balance sheet items everything is reported and is clear to see in the company accounts.

A systematic risk measure that would limit the size and/or inter linkages one single bank may have within the system. The limit should be expressly stated by legislation with all banks aware of the threshold which would allow banks to make informed decisions. For example one bank cannot account for more than X% of the market including exposure in credit default swaps and other financial instruments. Maybe even a tax on exposure beyond a certain threshold instead of an actual limit to penalise exposure but still allow exposure for the most successful banks. The threshold would likely be very difficult to estimate and the best approach might be a mix of historical analysis and trial and error. These approaches would hopefully avoid scenarios like AIG reoccurring.

Another problem with the banks seemed to be CEO's making decisions for short term gain which ultimately would have been detrimental for the bank in the long run. To counter this top management should recieve a % of income in the form of equity which can only be sold within a certain time frame e.g. a year after resigning. Again the problem would be what % of income should be set and for how long.

Slippers · 02-12-2009 3:19am

The system I would like to see is described here by James Robertson and Joseph Huber. It is based on this work by Huber. A possible implementation is fleshed out here by Ben Dyson.

As they introduce it in chapter one:

"It is basically simple. It is in two parts.

1. Central banks should create the amount of new non-cash money (as
well as cash) they decide is needed to increase the money supply, by
crediting it to their governments as public revenue. Governments
should then put it into circulation by spending it.

2. It should become infeasible and be made illegal for anyone else to
create new money denominated in an official currency. Commercial
banks will thus be excluded from creating new credit as they do now,
and be limited to credit-broking as financial intermediaries."

This is in contrast to the current system where the money in our accounts is essentially open ended corporate bonds with put options attached. It consists of liabilities issued by commercial banks which are then spent by them to acquire assets.

The system would work like this:

All banks would have two accounts at the central bank. One for their non-cash money, one for their customers' non-cash money. The government would have one account at the central bank. When a bank wished to make a loan it would move money from its own account to its 'customer funds' account and make a note of the borrower and how much of the money in the customer funds account belonged to that borrower. A payment from a customer of bank A to a customer of bank B would involve transferring the money from bank A's customer funds account to bank B's customer funds account.

In the event of a bank failure the money supply would not shrink as it would no longer consist of privately issued liabilities.

Open market operations would cease. All the monetary policy committee would do is decide how much money should be created that month and then type it into the government's account.

nesf · 02-12-2009 5:58am

Slippers wrote: »

The system I would like to see is described here by James Robertson and Joseph Huber. It is based on this work by Huber. A possible implementation is fleshed out here by Ben Dyson.

As they introduce it in chapter one:

"It is basically simple. It is in two parts.

1. Central banks should create the amount of new non-cash money (as
well as cash) they decide is needed to increase the money supply, by
crediting it to their governments as public revenue. Governments
should then put it into circulation by spending it.

2. It should become infeasible and be made illegal for anyone else to
create new money denominated in an official currency. Commercial
banks will thus be excluded from creating new credit as they do now,
and be limited to credit-broking as financial intermediaries."

This is in contrast to the current system where the money in our accounts is essentially open ended corporate bonds with put options attached. It consists of liabilities issued by commercial banks which are then spent by them to acquire assets.

The system would work like this:

All banks would have two accounts at the central bank. One for their non-cash money, one for their customers' non-cash money. The government would have one account at the central bank. When a bank wished to make a loan it would move money from its own account to its 'customer funds' account and make a note of the borrower and how much of the money in the customer funds account belonged to that borrower. A payment from a customer of bank A to a customer of bank B would involve transferring the money from bank A's customer funds account to bank B's customer funds account.

In the event of a bank failure the money supply would not shrink as it would no longer consist of privately issued liabilities.

Open market operations would cease. All the monetary policy committee would do is decide how much money should be created that month and then type it into the government's account.

How does the central bank know how much money to create? Will, by putting so much power in the hands of the central bank, create an entity that will be under extreme political pressure to behave counter-cyclically?

nesf · 02-12-2009 6:15am

silverharp wrote: »

It would be more important to acknowledge uncertainty in the system then that the agents act in a rational manner. If the market creates a bubble in something then no doubt some banks will over lend and fuel the mal investment. In reality it would be a benefit to the system to have a series of mini busts as it would reinforce the idea that the future is uncertain.

This assumes that there aren't long term negative effects of mini-recessions since that's what you're talking about. Are 3 mini recessions less bad that one larger one? It's not immediately obvious to me why this would be.

silverharp · 02-12-2009 7:53am

nesf wrote: »

This assumes that there aren't long term negative effects of mini-recessions since that's what you're talking about. Are 3 mini recessions less bad that one larger one? It's not immediately obvious to me why this would be.

I'd argue that credit standards would be more consistent so there would be less "fuel" to create a systemic bubble so although asset prices may get of whack in in one sector say an internet boom/bubble it would not have led to an excessive consumer credit bubble. Also as there would be more savings in the system the starting point of any recession would see the our economic actors in a stronger financial position.

Slippers · 02-12-2009 12:41pm

nesf wrote: »

How does the central bank know how much money to create? Will, by putting so much power in the hands of the central bank, create an entity that will be under extreme political pressure to behave counter-cyclically?

The same way they used to decide what interest rate to target.

Having control over the size of the money supply in a public body with the freedom to act counter-cyclically, rather than in commercial banks motivated by profit to act pro-cyclically, is one of the biggest benefits of the system. No more pushing on a string, trying to get banks to put money into the economy.

SLUSK · 02-12-2009 2:51pm

Flamed Diving wrote: »

I thought this might be an interesting thought exercise for the Economics forum. Imagine that we are chosen to design a banking system from scratch. The powers that be have decided enough is enough and want one to be built on theoretical foundations. No more financial crises, no crony banks, no taxpayers bailing out bankers, how do we achieve this? How do we set the rules of the game, so as to avoid these problems?

I thought I would outline a few rules:
This is not a thread to have a rant about the current crisis

Posts such as "No point because bankers are evil" shall be labelled wrong, by default

All points of view are welcome, no matter how left-of-field, so long as they are valid, sensible contributions

Ok, so where do we begin? I think the first matter I would like to bring up is the matter of competition. Since we will be issuing licenses for the banks, should there be some restrictions on the number of banks? Should we simply allow licenses to be freely available to those that want them? Might we be better off having a monopolistic system? How about the Keirstu system? Perhaps different banks restricted to different sectors of the economy?

Discuss, my beauties. Discuss.

EDIT: Some people seem to have trouble getting this, so I will further explain. This is an abstraction. There is no banking system. We are starting from scratch.

Bring back the free markets and let bad institutions fail. If it is good enough for Jim Rogers it is good enough for me.

jimmycrackcorm · 02-12-2009 2:54pm

Require banks to only lend on multiples of actual deposits like the good old days.

02-12-2009 3:01pm

I can't quite believe I'm saying this but I would go for a more rigid form of Glass-Steagal. Seperate and ringfence non commerical deposit bases and place rigourous limits on its uses, especially removing it from direct investement in equity and currency markets.

Essentially, I would be looking at creating a two tiers - one for speculation and the potential for high returns, the other for low risk, low returns.

However, I think consumers need to wake up to the fact that you can't have high return, low risk assets like everyone seems to want.

Flamed Diving · 02-12-2009 10:26pm

Deleted User wrote: »

I can't quite believe I'm saying this but I would go for a more rigid form of Glass-Steagal. Seperate and ringfence non commerical deposit bases and place rigourous limits on its uses, especially removing it from direct investement in equity and currency markets.

Essentially, I would be looking at creating a two tiers - one for speculation and the potential for high returns, the other for low risk, low returns.

However, I think consumers need to wake up to the fact that you can't have high return, low risk assets like everyone seems to want.

I agree, I think that the new system should have two fundamental types of bank. Consumers should have the choice of two banks:

The "traditional" bank, which is restricted to net interest income and cost/income savings for a revenue stream. None of this naughty OBS activity allowed. I would expect these banks to offer lower savings rates (perhaps even inflation-indexed low) and higher loan rates, but offer higher stability, as a result. Any group that is issued a license for this type of bank may not enter into the same markets as the high-risk sector, and may not lend to this sector, though may lend to one another.

And of course the investment bank, meaning the high risk, high return model. This market would more or less follow something closer to Silverharps model, where no central bank or government support is offered, and is left entirely to the free market. If a bank fails, it fails, and no exception shall be made on this by punishment of catapult.

Anonymous1987 · 03-12-2009 12:34am

I know we're not debating the current crisis but its still relevant to the discussion. This article might be of interest it debates whether the Feds role should include fighting asset bubbles.

http://online.wsj.com/article/SB125970281466871707.html#mod=todays_us_nonsub_page_one

I would think that it would be next to impossible to impose credible independence. There will always be interested parties that would be very critical and outspoken against the fed managing bubbles let alone the difficulty in identifying bubbles in the first place.

Anonymous1987 · 03-12-2009 12:46am

Deleted User wrote: »

I can't quite believe I'm saying this but I would go for a more rigid form of Glass-Steagal. Seperate and ringfence non commerical deposit bases and place rigourous limits on its uses, especially removing it from direct investement in equity and currency markets.

Essentially, I would be looking at creating a two tiers - one for speculation and the potential for high returns, the other for low risk, low returns.

However, I think consumers need to wake up to the fact that you can't have high return, low risk assets like everyone seems to want.

Wouldn't a system like that miss out on the benefits of a diversified risk base. Isn't the whole idea of minimising risk spreading risk as much as possible (now that I think of it it contradicts my own idea above on systematic risk). For example a bank lending to both low and high risk individuals will be less risky than the risk of one risky bank and one safe bank.

nesf · 03-12-2009 12:54am

Anonymous1987 wrote: »

Wouldn't a system like that miss out on the benefits of a diversified risk base. Isn't the whole idea of minimising risk spreading risk as much as possible (now that I think of it it contradicts my own idea above on systematic risk). For example a bank lending to both low and high risk individuals will be less risky than the risk of one risky bank and one safe bank.

In theory yes but only so long as the pool of low risk money is much greater than the pool of high risk money loaned out and only if you aren't estimating risk using models that distribute it normally which underestimates risk. This also assumes that risk is not system wide, which it can be.

nesf · 03-12-2009 12:56am

silverharp wrote: »

I'd argue that credit standards would be more consistent so there would be less "fuel" to create a systemic bubble so although asset prices may get of whack in in one sector say an internet boom/bubble it would not have led to an excessive consumer credit bubble. Also as there would be more savings in the system the starting point of any recession would see the our economic actors in a stronger financial position.

Yes, but my point is more fundamental than that. Do we have evidence that multiple small recessions are better than one larger on? If you assume recessions have only transitory effects this question is moot but that's another assumption that requires empirical backing before it can be taken for granted.

silverharp · 03-12-2009 12:57am

Anonymous1987 wrote: »

This article might be of interest it debates whether the Feds role should include fighting asset bubbles.

What about the Fed's role in creating asset bubbles? the Greenspan put.

nesf · 03-12-2009 1:02am

Slippers wrote: »

The same way they used to decide what interest rate to target.

Having control over the size of the money supply in a public body with the freedom to act counter-cyclically, rather than in commercial banks motivated by profit to act pro-cyclically, is one of the biggest benefits of the system. No more pushing on a string, trying to get banks to put money into the economy.

It should have read "not to behave counter-cyclically" sorry, was in a rush typing it!

But you still do end up pushing on a string! Just because you increase the money supply doesn't mean that banks will automatically lend it where you want unless you nationalise them as well! You'd need to control the money supply, interest rates and some amount of financial institutions in order to control credit provision which ends up in the planned economy problem, how does the planner know how much credit to create, who to give it to and in what way it should then be distributed?

nesf · 03-12-2009 1:04am

silverharp wrote: »

What about the Fed's role in creating asset bubbles? the Greenspan put.

After the experiences of the 70s and 80s where the Fed's fighting of asset bubbles led to anemic growth there was a push towards more hands off management under Greenspan, which brings its own problems, i.e. you get better growth but you'll get bigger busts and a big mess to clean up afterwards.

silverharp · 03-12-2009 1:24am

nesf wrote: »

After the experiences of the 70s and 80s where the Fed's fighting of asset bubbles led to anemic growth there was a push towards more hands off management under Greenspan, which brings its own problems, i.e. you get better growth but you'll get bigger busts and a big mess to clean up afterwards.

It depends how the growth is achieved, there is a world of difference between investment credit that finds its way into useful production versus a stepstair increase of consumer credit which is not tied to production or rising incomes. I think I am right in saying that real incomes went nowhere in the US from the 1980's onwards so there was always going to be problems if the policy aim to prop up asset prices via increased leverage when incomes were not keeping up.

Slippers · 03-12-2009 3:57am

nesf wrote: »

It should have read "not to behave counter-cyclically" sorry, was in a rush typing it!

Ah, I see what you mean. I would think they would be under the same sort of pressure as now but that there would be less risk of creating a bubble than with lowering interest rates because issuing new money wouldn't increase total debt.

But you still do end up pushing on a string! Just because you increase the money supply doesn't mean that banks will automatically lend it where you want unless you nationalise them as well!

The new money wouldn't be in the banks. It would be in the government's account at the CB. The government would spend it and then it would be circulating in the economy. There wouldn't be lending involved.

nesf · 03-12-2009 5:58am

Slippers wrote: »

The new money wouldn't be in the banks. It would be in the government's account at the CB. The government would spend it and then it would be circulating in the economy. There wouldn't be lending involved.

So the only new money is when the Government decides to print money to pay bills?

Mahatma coat · 03-12-2009 9:11am

Interest Free money Spent by the By the Government into the economy and a .5% tax on all financial transactions.

this is a system that gets touted a bit on some of the Fora I frequent.

seems like a good idea, but my question is what backs the currency, We have to link our currencies to something solid and tangible otherwise tha banks will just keep inflating the money supply and Booms and busts are the inevitable outcome, leading back to Nesf's question as to whether one big Recession or three small ones are better.

the key element is to take control of the money supply out of the hands of the banks and put it firmly in the hands of............??

well thats the question init.

What are peoples thoughts on a mass global default, a Reset so to speak, not just nations but everyone

Slippers · 03-12-2009 2:03pm

nesf wrote: »

So the only new money is when the Government decides to print money to pay bills?

It wouldn't be the government, it would be decided by the relevant monetary authority at their monthly meeting. In the Eurozone that would be the 22 members of the ECB's Governing Council. If they decided that the best thing for the people of the Eurozone was to have the money supply grow at 10% per year forever, they would have very short meetings and just increase the money supply by about 0.8% every month. If it turned out that the best thing is to have a fixed money supply then they wouldn't create any money each month.

When you said "print money" it was a figure of speech, yes? Just so everyone can follow, this money would not be physically printed, it would be held in accounts at the central bank.

When a national government received its share of the new money it could use it to reduce borrowing, to pay bills or to reduce taxes. The government would be accountable to the electorate for this decision in the same way as they are accountable for other budget decisions.

Anonymous1987 · 03-12-2009 10:42pm

nesf wrote: »

Yes, but my point is more fundamental than that. Do we have evidence that multiple small recessions are better than one larger on? If you assume recessions have only transitory effects this question is moot but that's another assumption that requires empirical backing before it can be taken for granted.

I don't have any empirical evidence to back it up but it could be the case that smaller shorter recessions are less harmful to capital. I would imagine that the rate of human capital depreciation would accelerate the longer a recession lasts. For example a carpenter loses his job at the beginning of a recession the longer he is out of work the more his skills deteriorate because he is not actively using them and the less employable he becomes to prospective employers. Of course it could also be the case that human capital depreciation is negligible unless over very long time spans.

nesf · 03-12-2009 11:01pm

Mahatma coat wrote: »

What are peoples thoughts on a mass global default, a Reset so to speak, not just nations but everyone

Highly undesirable really, it punishes all countries that were cautious about building up debt and provides a precedence for future justification for massive borrowing by some countries (i.e. if it happens once, it can happen again). Just not a good idea at all, similar to why a "cleaning of the slate" of all mortgage debt in this country would be a bad idea.

nesf · 03-12-2009 11:07pm

Anonymous1987 wrote: »

I don't have any empirical evidence to back it up but it could be the case that smaller shorter recessions are less harmful to capital. I would imagine that the rate of human capital depreciation would accelerate the longer a recession lasts. For example a carpenter loses his job at the beginning of a recession the longer he is out of work the more his skills deteriorate because he is not actively using them and the less employable he becomes to prospective employers. Of course it could also be the case that human capital depreciation is negligible unless over very long time spans.

Labour economics isn't my thing but it's definitely an interesting question. I'd think it'd be non-linear with the initial years having a much different impact than latter years of unemployment.

What you'd find interesting is the hysteresis effect in economics. It's an idea that's been around for a few decades and there's a lot of argument for and against the idea. It centres around how spikes in unemployment during recessions have long term effects on unemployment rates when the economy recovers (i.e. unemployment levels stay higher than one would expect after a recession than they would if the effect of the spike of unemployment was transitory).

Design a better banking system [Discussion]

Comments