I'm not sure whether to post this or not but seeing as there's some discussion on whether people understand banks or not I thought I'd post my simplistic veiws of the banking system.

1 - Banks take deposits and then lend money taking a margin. Very true. However:

(a) They are only required to keep a % of the money that they lend out, representing the fact that they wouldn't expect to need to repay all of their deposits in one go. So a bank takes a deposit, then lends out many times the value of that deposit. This is then essentially what your capital ratios are - how much leverage the bank has, depending on various factors. Rough rule of thumb is capital ratio = 10% = ten times lending (leverage).

(b) SIV's/ABS/structures complicated this. Using one of these a bank could set it up - put in say $500m as capital, then the SIV would leverage up by say 20 times (to say $10bn) then use that $10bn to buy loans back off the bank. So the bank lends to Mr X, sells the loan on to SIV Z (clearing it off their balance sheet and taking a margin for making the loan in the first place), then as they've had the money back off SIV Z they can lend it again! and again! and again! Even better. The $500m they have as capital in the SIV is an asset, so they can class this in their asset pool and lend even more money off it! [Now thats clever and the evolution of this is basically whats been the driver in the decade+ long credit expansion we've seen]

(c) Not all deposits are equal. An overnight deposit can't be classed as the same way as a time deposit. Especially now post Northern Rock as the regulators have seen large lines of people happy to wait and withdraw money. Thats why people like IceSave and Alliance & Leicester are now prepared to pay up to encourage you to fix your deposit with them for 1yr or more (or in the UK the new trick is to say, ok we'll pay you 10%, but you have to put in X amount every month for X months)

(d) This leverage also leverages up the issue of defaults.

2 - Banks don't just take deposits. They can also borrow money. They can then use this borrowed money to lend out. So pretty much all banks have a % of funding from deposits and a % of funding from the wholesale market. Northern Rock had too much wholesale funding, however because of the way that people took money out deposit wise the new mantra is 'balance'. You want to have deposits (preferably time deposits) but you also want to not over rely on these so you also want to have access to other borrowing from markets. People with sub 50% of their cash from deposits are all now looking to up this towards these levels (see B&B, A&L etc).

3 - All lending is not equal. You can borrow money overnight (via an instand access deposit account) then lend it on a 20yr fixed morgage. Not good - see Northern Rock. Alternatively you can borrow using a 3yr bond and lend to someone for 3yrs taking the margin. Thats called 'match funding' which is now a word thats becoming very popular. The more match funded a bank is the less of a problem there is if they can't get funding because at the end of the day they just have to wait for everything to mature. However traditionally match funding is not as profitable. Whats really profitable generally is to borrow short and lend long - mega margins - mega risk (in hindsite). This also extends to market/currency risk. You could borrow in Yen (0%) and lend in Australian dollars (8%+), but now the emphasis is that if you borrow in Yen you better bloody be lending in Yen too. The shift towards matched funding if long term reduces the long term profitability of the banking sector.

4 - All assets are not equal. 'Loan to Value' is another critical term in the current environment. Basically in the Banks loan pool what is the average mortgage % (real estate or commercial property) that is backed by an asset. So Mr X has a £400k house which is financed by a £250k mortgage gives a 62.5% LTV. Before 100% LTV was fine. Now 80% is the new target (60% shows you're in really good position). Hence various banks now upping their deposit requirements as they're trying to lower the LTV on their mortgage books.

5 - Central banks were picky but are now less so. So basically all banks can borrow from central banks - to do so they generally have to deposit 'assets' (loans) with the central bank to get cash. The ECB have started to accept 'covered bonds' and now so have the Fed. These are bonds backed by mortgages. So this is now a rapidly growing area as it is allowing certain banks to effectively deleverage. They can package a large number of mortgages from their balance sheet (with full bank support) into a covered bond and then bung that bond to the central bank getting cash back. A covered bond is different than a Mortgage Backed Security because the Bank has to provide full support if the underlying mortgages don't repay enough to cover the maturity (ie the bank has to make up the difference if there is any shortfall).

6 - Banks don't just to banking. However things like insurance are also a problem in the current environment. Insurers have large asset pools of longer dated bonds which used to be worth 100 and are now likely to be worth 80-90. AIG has to be the largest example with a $900bn asset pool. Some banks rely on income from investment banking (Barclays), some Investment Banks rely on income from Banking (Merrill Lynch).

Haha anyway I have no idea if this is helpfull or not but hopefully will provide a little help on what factors people are now looking much more closely at.

Bert

I understand the principle of banking, but the practice of borrowing short and lending long mostly leads to profits but sometimes leads to losses. (when the yield curve inverts)

This leverage issue, confuses me. Say someone deposits £100 and the leverage of that bank is 10%, does that mean a £1000 mortgage can be granted? If this is the case, where does the extra £900 come from in order to pay the home seller.

Or are you saying, only £10 of that £100 needs to be held, and £90 can be lent as a mortgage.

In both scenarios, there appears to be a pyramid selling problem which occurs when property prices are over inflated.

The assets, and liabilities are not matched. The asset is equity based, and the liability is fixed nominal. The asset is not very liquid, and the liability is extremely liquid.

In an insurer, or life assurer, you would need to hold extremely large free assets in order to mismatch in this manner. The FSA would require you to hold reserves called mismatch reserves, which would be substantial if there was more than 75% borrowed on the value of the property assets.

It would appear that banks get away with holding such large free assets or mismatch reserves, by virtue of the fact that the Bank of England acts as lendor of last resort.

In other words banks profit at the expense of the taxpayer. The taxpayer carries the risk of failure, and the banks profit.

Something needs to change here!

I understand the principle of banking, but the practice of borrowing short and lending long mostly leads to profits but sometimes leads to losses. (when the yield curve inverts)


Which is why fixed mortgages were rare. now with derivatives and (were) MBS more longer term loans are around.

This leverage issue, confuses me. Say someone deposits £100 and the leverage of that bank is 10%, does that mean a £1000 mortgage can be granted? If this is the case, where does the extra £900 come from in order to pay the home seller.


No

Or are you saying, only £10 of that £100 needs to be held, and £90 can be lent as a mortgage.


Yes


In other words banks profit at the expense of the taxpayer. The taxpayer carries the risk of failure, and the banks profit.


Ask NRK investors if that is the case.

This leverage issue, confuses me. Say someone deposits £100 and the leverage of that bank is 10%, does that mean a £1000 mortgage can be granted? If this is the case, where does the extra £900 come from in order to pay the home seller.


No

Well, actually, yes. Just not as directly as that. It's all to with fractional reserve banking and money multiplication. A bank may only be able to lend 90% of its deposits but this money may then be deposited and 90% of that lent out and so on until loans are 10x initial deposits for the system as a whole. The money for this essentially comes from the future - think about how you pay back your mortgage.

At least I think that's how it works but I'm still at the learning stage myself.

lfc

Bert,

Banks take deposits...

Deposits are loans made to the bank. Similar to what you are saying when banks borrow from the wholesale market.

In both cases the bank is obtaining funds from another source and paying interest to obtain the funds. Deposits are retail loans from individuals to the bank. A liability to the bank just like any other debt.

John Corey

Author: FunkyPinky | Date: 12/03/2008 16:46:07 | Number: 2601

he practice of borrowing short and lending long

This is what all banks do by definition. Deposits are short term loans from individuals. The money is loaned out on term loans or in the form of home loans secured by mortgagees.

In other words bankers are always trying to match up short term and long term exposures.

It does not matter much if the bank uses wholesale funds rather than deposits. It all comes down to the term mismatch.

John Corey

This leverage issue, confuses me. Say someone deposits £100 and the leverage of that bank is 10%, does that mean a £1000 mortgage can be granted? If this is the case, where does the extra £900 come from in order to pay the home seller.

Thats correct yes.

The money doesn't come from anywhere it's simply created.

Take a look at this:

http://en.wikipedia.org/wiki/Fractional-reserve_banking

The taxpayer carries the risk of failure, and the banks profit.

Thats why you have the concept of 'too big to fail' in that if a large bank (say Barclays) were to fail then it's failure would bring down the economy and thus could not be allowed to happen. You thus have regulation to try and limit the risk of failure.

B

In other words bankers are always trying to match up short term and long term exposures.

Thats not true actually, different banks are very different. Some have no mismatch at all and others have varying degrees.

Author: BertEEE | Date: 12/03/2008 18:59:34 | Number: 2605

Thats not true actually, different banks are very different. Some have no mismatch at all and others have varying degrees.

If you are sure then explain.

Banks take in deposits. Some have no restrictions and others are term deposits. The bank might or might not take in wholesale deposits and other hot money.

The same banks make loans. Loans for cars, for businesses, for credit cards, for houses.

Explain how banks do not need to manage the term of the deposits vs. the term of the loans. What banks have no mismatches at all?

John Corey

This leverage issue, confuses me. Say someone deposits £100 and the leverage of that bank is 10%, does that mean a £1000 mortgage can be granted? If this is the case, where does the extra £900 come from in order to pay the home seller.

Thats correct yes.

The money doesn't come from anywhere it's simply created.

Take a look at this:

http://en.wikipedia.org/wiki/Fractional-reserve_banking


The Wikipedia example provided says more or less the opposite. Look at the table for a simple review.

The £100 deposited turns into a cash reserve of X and a loan of Y. The loaned money gets deposited back into the banking system so there is Y more deposits.

The example does not show £100 deposited turns into a £1,000 loan.

Please explain. Either the answer above should be 'No', see Wikipedia or the above question is correct and Wikipedia is wrong.

John Corey

SIV's/ABS/structures complicated this. Using one of these a bank could set it up - put in say $500m as capital, then the SIV would leverage up by say 20 times (to say $10bn) then use that $10bn to buy loans back off the bank. So the bank lends to Mr X, sells the loan on to SIV Z (clearing it off their balance sheet and taking a margin for making the loan in the first place), then as they've had the money back off SIV Z they can lend it again! and again! and again! Even better. The $500m they have as capital in the SIV is an asset, so they can class this in their asset pool and lend even more money off it! [Now thats clever and the evolution of this is basically whats been the driver in the decade+ long credit expansion we've seen]

Yes, it's always the above that loses me completely and I usually understand balance sheets and P&L accounts (except for insurance and banks). I've seen this explanation/jargon quoted a number of times in the financial press and still have not the vaguest clue what it means.

m06een00

I've seen this explanation/jargon quoted a number of times in the financial press and still have not the vaguest clue what it means.

All I can say is you're not alone! Like watching that stuff about Hawking and superstrings the other night - I could understand each of the words individually but not when they were joined into sentences. I suspect the difference is the financial stuff is actually understandable if put into plain english and perhaps with the aid of a diagram whilst quantum mechanics is incomprehensible, period.

lfc

m06een00,

SIVs and other things are nothing but a company that someone sets up.

Special purpose has not real legal standing. It is jargon to say the business has a narrow focus but the terms are not special forms of a company.

Banks can take deposits and to protect depositors there are special rules. If you want take deposits and call yourself a bank you need to follow the rules.

SIVs that buy loans from banks and other places do not take deposits. They do not need to reserve money (fractional reserves as in banking).

Banks can reduce the capital reserves and free up capital to do new business if they sell off a loan book. Many banks make more money from originating loans than they will do from holding the loans to term. Given they have limited capital they are forced to sell off the loans to originate new business or stop all new business until they can bring in more capital.

John Corey

Hi John,

Please explain. Either the answer above should be 'No', see Wikipedia or the above question is correct and Wikipedia is wrong.

Interesting point. I wonder though, isn't the difference here just one of the number of iterations (and perhaps wording too)?

From the Wikipedia example I'd say you are right and that at the first transaction the £100 deposit is not multiplied, but after a number of iterations, the effect is that the original £100 deposit has multiplied up to £500 by the creation of commercial bank money (as per the graph to the right of the table, which can be seen in a larger version by following this link: http://en.wikipedia.org/wiki/Image:Fractional_reserve_banking_20percent_100base.gif )

So I hope I might be in the nice (and unusual situation for me) of saying that I think you are both right, broadly speaking.

One question I would add though, is that with the Wikipedia example, the illustrative Reserve Rate is 20%. Does that correspond with real rates used in the financial system? I think I have heard mention somewhere of a 10x multiplying effect - I think it was in one of BertEEE's Pub posts, which would suggest double the multiplier effect I guess and so double the contraction if the process is reversed?

Regards,

TC

TC,

The reserve requirements vary by type. It is not a flat number for all activities as different bank business decisions have different rissk.

John Corey

One of the benefits of these boards is the facility for the less well informed to listen and learn from a bunch of pros talking to each other. But it’s sometimes easier to detect that contributors know what they are talking about than to understand what they are saying. Fortunately, some of these guys occasionally throw in an explanatory piece for the unseen audience.

BertEEE did it with post 2582, which elicited appropriate appreciation. The crucial bit IMO was this tantalising para quoted by m06een00:

SIV's/ABS/structures complicated this. Using one of these a bank could set it up - put in say $500m as capital, then the SIV would leverage up by say 20 times (to say $10bn) then use that $10bn to buy loans back off the bank. So the bank lends to Mr X, sells the loan on to SIV Z (clearing it off their balance sheet and taking a margin for making the loan in the first place), then as they've had the money back off SIV Z they can lend it again! and again! and again! Even better. The $500m they have as capital in the SIV is an asset, so they can class this in their asset pool and lend even more money off it! [Now thats clever and the evolution of this is basically whats been the driver in the decade+ long credit expansion we've seen]

Do we have here in a few lines the root cause of the greatest global financial crisis since 1929? I suspect we do - but there are still a few chinks that need filling in for us plebs.

SIV's/ABS/structures complicated this. Using one of these a bank could set it up - put in say $500m as capital, then the SIV would leverage up by say 20 times (to say $10bn) then use that $10bn to buy loans back off the bank.

So a ‘SIV and other things’ is a company (thanks John Corey), set up in this case by a bank which injects $500 of capital. And the SIV can borrow up to say 20 times its capital to $10bn because it’s not a bank and subject to capital ratio rules.

At this point, is the SIV still owned by the bank that formed it (surely not????) or has it been sold/transferred to another party (Hedge Fund99)?

Assuming the latter, HF99 then uses its borrowed $10bn to buy loans (MBS1) from the same bank (in this instance). It is assumed that the yield on MBS1 is better than the interest charged by the bank for lending HF99 its $10bn and that it will stay that way!

In an either/or situation, these two transactions would make no sense for the bank. It would be better to retain the higher yielding loans and not lend at a lower rate to HF99. But if the bank can take a rake off from each transaction and then make more loans (MBS2) because MBS1 is now off its balance sheet, it does make sense because the process can continue ad infinitum.

Even better. The $500m they have as capital in the SIV is an asset, so they can class this in their asset pool and lend even more money off it!

Makes me think in this case the SIV is still owned by the bank, i.e. just an off balance sheet subsidiary, 'borrowing' money off the parent to 'buy' more loans from the parent, which the parent can provide repetitively because basic banking rules have been evaded. I bet the bank would make sure such loans were insured against default . . . . . . hmmmmm.

[Now thats clever and the evolution of this is basically what’s been the driver in the decade+ long credit expansion we've seen]

It puzzles me that what seems like a reckless ponzi scheme has survived for so long without regulation. Central banks must have been fully aware and even the odd politician.

DoY

Hi John

The Wikipedia example provided says more or less the opposite. Look at the table for a simple review.

No it says exactly what I've said except that I've been overly simple and they use a 20% capital requirement (vs 10% being normal). So look at the big green chart at the top - in their example $100 is turned into $500 because they're using 20% capital. Use 10% and it's $1000.

The £100 deposited turns into a cash reserve of X and a loan of Y. The loaned money gets deposited back into the banking system so there is Y more deposits.

The example does not show £100 deposited turns into a £1,000 loan.

What you're missing I think is that when you're saying the loan gets deposited back into the banking system, the loan still exists - what they're saying here is that the cash finds its way back in. So for example if you take a loan out to buy something someone somewhere gets money which goes back into the system.

So £100 = £20 capital + £80 loan outstanding >

£80 returns and £64 lent out again etc.

But you have to keep agregating the loan figures - so £80+£64 etc.

I hope that helps.

Bert

DoY,

So a ‘SIV and other things’ is a company (thanks John Corey), set up in this case by a bank which injects $500 of capital. And the SIV can borrow up to say 20 times its capital to $10bn because it’s not a bank and subject to capital ratio rules.


There is nothing that says the SIV is owned by the bank or under control of the bank. Frankly the SIV has to be owned by someone else or the SIV will be seen as just a division of the bank and still under the bank regulations.

So, the SIV is a separate company and any stake the bank has will be very much a minority position. If the SIV gets into trouble then the SIV can go under without the bank needing to step in (legally). There might be reputational risks if the bank's brand is associated with the SIV. Or the bank might have extended promises (guarantees) that force the bank to do something if there is a problem.

Second point...

The SIV is not borrowing from the bank directly in many cases. They raise funds from other places (other banks). The bank that wanted the SIv set up will inject capital so that the company has some assets with which to borrow.

The loan facilities plus the cash are used to borrow MBS and those MBS are pledged to secure the loans.

Assuming the latter, HF99 then uses its borrowed $10bn to buy loans (MBS1) from the same bank (in this instance). It is assumed that the yield on MBS1 is better than the interest charged by the bank for lending HF99 its $10bn and that it will stay that way!

Two things.

The SIV is issuing commercial paper so it can borrow money. The interest rates for short term money is lower than the returns being earned on the long term assets (MBS). The spreads were a bit large actually. The SIVs are borrowing short while holding assets that represent loans that lend long. The SIV is not making a loan. They bought the loans from the bank that did the origination work.

Very similar for credit card companies who pack up the payment streams from customers and then sell off the income to investors (SIVs or others). The same is done with car loans if anyone is interested. The buyers do not have to be SIVs. Legally, anyone can buy a loan and hold it for the payments.

Second thing. Historically short term rates stay below long term rates. Not all the time but much of the time. Risk of default and inflation impact is more of an issue with long term debt so the rates tend to reflect the added risk. Short term rates reflect the lower levels of risks but the rates do bounce around more.

Even better. The $500m they have as capital in the SIV is an asset, so they can class this in their asset pool and lend even more money off it!

Makes me think in this case the SIV is still owned by the bank, i.e. just an off balance sheet subsidiary, 'borrowing' money off the parent to 'buy' more loans from the parent, which the parent can provide repetitively because basic banking rules have been evaded. I bet the bank would make sure such loans were insured against default . . . . . . hmmmmm.


The idea of showing the asset on its books is not my comment. In some ways the details matter a great deal and it is rather well covered by the accounting rules.

I think it is fair to say that in most SIVs the bank does not show it as an asset in the way one might think. If a bank holds a small equity stake in a company because the bank put $500M into the company, then it is an asset. The bank could lose the full amount. It is equity and not debt so the bank's interest can be wiped out if the SIV lenders seize assets. The SIV would have to be reflected on the bank's books as a subsidiary if the bank controlled the SIV and stood behind thee SIV. Most cases this is not going to be the case. Otherwise there would be no reason to use an SIV as the loan owed by the SIV would still hit the bank capital requirements. These SIVs really do have to be separate and subject to the SIV failing without the bank needing to do a thing.

Two other issues.

Some SIVs were set up wrong or the bank made promises to guarantee or provide a backstop. One example showed a bank taking back 6 SIVs because the bank's reputation was at risk even through the bank would not have to step in legally. In another case the bank had offered a guarantee to buy back the loans if there was an issue. They were only required to do so after a set of things happened. The bank choose to step in early as the cost to do so was less than waiting until the worst case scenario happened. It cost less to step in early than to wait until much later and it was clear where things were heading.

Assume the SIVs were legal and correctly set up. If they can not obtain funding in the market they can not buy the loan books. When they issue commercial paper they are pledging assets to back the loans. Banks or institutions who provide capital on commercial paper terms have dried up. The collateral has become worth less or hard to value so the people with the cash will not take it as collateral (or they want more coverage so the numbers do not make sense).

John Corey

ponzi scheme is still my favourite analogy.

There's nothing wrong with leverage, as long as you have full disclosure of the % of assets under leverage, and who owns it.

The arbitrage of rating agencies (mortgage leverage given AAA ratings), and the hidden leverage of bank balance sheets via SPVs, VIEs and SIVs, served to obscure the amount of leverage in the system and where it was located.

Actually thats not true and is a popular misconception : leverage is obvious, has always been obvious, its just that everyone chose to ignore it.

Zak

"and the hidden leverage of bank balance sheets via SPVs, VIEs and SIVs"

There is leverage with SIVs, but I don't see why you think it's hidden.

The loans show up quite visibly on bank balance sheets.

The limits on bank lending apply as much to these loans as any other.

George

apologies - I was trying to say that in the face of popular understanding and the political story, leverage wasn't hidden. Banks and individuals made rational choices in the context of their freedoms and restrictions. Over-leverage and asset bubbles will always happen and have always happened - its part of the human condition.

(b) SIV's/ABS/structures complicated this. Using one of these a bank could set it up - put in say $500m as capital, then the SIV would leverage up by say 20 times (to say $10bn) then use that $10bn to buy loans back off the bank. So the bank lends to Mr X, sells the loan on to SIV Z (clearing it off their balance sheet and taking a margin for making the loan in the first place), then as they've had the money back off SIV Z they can lend it again! and again! and again! Even better. The $500m they have as capital in the SIV is an asset, so they can class this in their asset pool and lend even more money off it! [Now thats clever and the evolution of this is basically whats been the driver in the decade+ long credit expansion we've seen]

---------------------------------

The flaw in your argument, is that just moving things into a SIV doesn't mean you can gear at a higher ratio than you can in a bank.

The 500m put in, means you have to also reduce your lending by 500m x 10 (banks gearing)

In reality, the capital put into a SIV is very small, and the bank doesn't own it. It is invariably held in trust for a charity.

The SIV isn't geared in the case you describe. It gets the assets from the bank (loans) and raises money against the loans (same size)

Nick

Follow the cashflows.

Bank A has 1,000m in loans. It has to have 100m in capital put aside (depends on the loans too!)

The initial process

An SIV is set up with minimal capital.

The loans are sold from the Bank to the SIV.

SIV issues a bond, backed by the assets for 1,000m

SIV receives 1,000m and pays the banks for the loans it has just bought.

Now the running process.

The bank receives income, and repayments on the loans it is administering. It takes its fees, and passes the rest on to the SIV.

The SIV forwards this on to the bond holders.

The gearing on the SIV is loans = borrowings. 1:1

Nick

geebee2,

The loans show up quite visibly on bank balance sheets.

The limits on bank lending apply as much to these loans as any other.


Err no they don't. And that's the whole reason why we've been making them!

1. All of the SIVs (that I'm aware of) are off-balance sheet, they are not consolidated into the accounts of the bank sponsor. And that's probably the correct accounting treatment because as the SIV mess has begun to unwind the SIVs have not actually directly caused huge losses to the banks. They have created liquidity problems and they've stretched the solvency ratios of the bank sponsors which in itself has lead to losses (and missed opportunites) for the banks but the banks have not directly bourne the losses of their SIVs.

2. Re "limits on bank lending" SIVs are unregulated offshore investment funds, quite simply SIVs are not banks and so there are no limits whatsoever on their borrowing/lending. The only constraints on a SIV are those imposed by the various contractual documents.

Consequently the leverage that SIVs (and the numerous other leveraged vehicles as SIVs are simply a small subset of these) has been "hidden".

JakNife

haha thanks Nick but given that until recently I had around $500m in SIV exposure, one of which defaulted causing total mayhem I think my last few months of meetings, conference calls and reports that I've had to write leaves me as at least decently clued up on the subject ;o)

The 500m put into the SIV is classed as an asset, not a loan. The credit lines extended to SIV's were done so on the idea they'd never be called upon so required minimal capital exposure. The bank fully owns the capital in the SIV, I've not come across a single one thats been held in trust for a charity fwiw. The SIVs were massively geared, most of them to 15-20 times. Yes the borrowed money was then used to purchase assets from the banks but that still means they're geared.

The 500m put into the SIV is classed as an asset, not a loan.

Bertee, the 500m put in the SIV either as a loan or equity injection would be classified as an asset in the Banks books. There would be a further breakdown dependent on if it was a loan or equity investment.

For Banks -

Loans and equity investments made are assets

Deposits & money market funds borrowed are liabilities.

Loans and equity investments made are assets

Deposits & money market funds borrowed are liabilities.

It's an equity investment (effectively) so an asset. They owned capital notes.

1. All of the SIVs (that I'm aware of) are off-balance sheet, they are not consolidated into the accounts of the bank sponsor. And that's probably the correct accounting treatment because as the SIV mess has begun to unwind the SIVs have not actually directly caused huge losses to the banks. They have created liquidity problems and they've stretched the solvency ratios of the bank sponsors which in itself has lead to losses (and missed opportunites) for the banks but the banks have not directly bourne the losses of their SIVs.

2. Re "limits on bank lending" SIVs are unregulated offshore investment funds, quite simply SIVs are not banks and so there are no limits whatsoever on their borrowing/lending. The only constraints on a SIV are those imposed by the various contractual documents.

----------------

It's important do realise the difference between an SIV used by a bank to get loans off its books, and things like hedge funds.

For a SIV used to get loans off the books.

Not consolidated on the sponsor's books is correct. That is because they is no ownership relationship.

Correct. SIV's are not causing losses to the banks.

Where is the liquidity problem caused by the SIVs?

If more banks like NR had used SIVs, they wouldn't have been in the mess they are in.

SIV's don't stretch the solvency ratios of the originating banks, because they aren't part of the bank. You then confirm this by saying the banks haven't any losses from the SIVs

--------------

For a hedge fund.

2. Re "limits on bank lending" SIVs are unregulated offshore investment funds, quite simply SIVs are not banks and so there are no limits whatsoever on their borrowing/lending. The only constraints on a SIV are those imposed by the various contractual documents.


If a bank sets up a hedge fund, it does contribute capital.

If the hedge fund is geared, it will do this by collateralising its transactions to gear up. ie. It trades on margin.

Depending on the trade, and the risk taken, the gearing varies. The largest I've seen is 35 times, but that was for very low risk transactions. If you were punting direction on equities 4, maybe 5 times.

However, if shares drop 20%, you would be wiped out.

Nick

haha thanks Nick but given that until recently I had around $500m in SIV exposure, one of which defaulted causing total mayhem I think my last few months of meetings, conference calls and reports that I've had to write leaves me as at least decently clued up on the subject ;o)

The 500m put into the SIV is classed as an asset, not a loan. The credit lines extended to SIV's were done so on the idea they'd never be called upon so required minimal capital exposure. The bank fully owns the capital in the SIV, I've not come across a single one thats been held in trust for a charity fwiw. The SIVs were massively geared, most of them to 15-20 times. Yes the borrowed money was then used to purchase assets from the banks but that still means they're geared.


You are confusing an SIV used to get loans off the balance sheet with a trading entity like a hedge fund.

The 500m put into a hedge fund isn't classed as an asset, it is classed as equity.

The credit lines extended, were done on the basis that if the trading went wrong, the banks had the capital. That's why the deals were colateralised.

The bank fully owns the capital in the SIV, I've not come across a single one thats been held in trust for a charity fwiw.

Try NR and Granite.

For a bank securitising it's loans via an SIV, it will be the case.

People are confusing two separate structures.

They are taking the hedge fund SIV structure and saying that this must apply to an SIV that is used to securitise loans.

That's wrong.

Nick

It's an equity investment (effectively) so an asset. They owned capital notes.

------

That's the bank's view.

For the Hedge fund, the capital is equity, not assets

Nick

Ask NRK investors if that is the case.

When I said banks profit, and taxpayers take the risk of failure, I should of course say instead that shareholders profit due to the guarantee made by the Bank of England and taxpayers pay if the guarantee is called in (as it does as soon as the company is no longer trading as a publicly quoted company).

When people invest in a share, there is always a risk of failure. Taxpayers do not usually end up paying when a company other than a bank fails (excepting non payment of tax). So investors in NRK have lost out because the company was extremely highly leveraged. However, prior to the collapse they benefited from dividends and growth which were created from a financial bubble.

Its assets were so mismatched from its liabilities that it ran into liquidity problems.

The taxpayer is now exposed to the risk that the assets themselves are insufficient to meet the liabilities in amount as well as the fact that they are insufficient in liquidity.

I have looked at the growth in M4 money recently, and all I can say is that when this decreases as it did in 1990-1995, property prices will fall. Taxpayers will then be paying for a big black hole in NRK's accounts where the property valuation backing of the assets (mortgages) is insufficient.

Negative equity itself is not a problem, so long as the mortgage payers continue paying. However, this presumes that there will not be mass unemployment which coincides with large property devaluation.

I very much doubt that this will be avoidable (without crooked politicians manipulating and lying with statistics)

So, taxpayers will pay and investors in NRK should get nothing unless someone other than the BOE takes on the risk of failure. (failure being the risk that the assets and liabilities are not matched: in other words someone has to take on an already failed bank)

This growth in M4 appears to have been created by banks. Money which does not exist which causes a boom in consumption, and artificial increases in asset prices. M4 stops growing when the credit crunch restricts how much further artificial money can be created due to balance sheets failure to be realistic.

I think something needs to be done to change this!

You are confusing an SIV used to get loans off the balance sheet with a trading entity like a hedge fund.

The 500m put into a hedge fund isn't classed as an asset, it is classed as equity.

Oh god.... ok yeah I'm confusing SIV's with hedge funds thats right... have a good day Nick.

I wouldn't worry about Nick BertEEE, I'm still trying to get my head around his logic that an SIV with 100m capital and 1000m in borrowings isn't levered because it holds 1000m in assets!!!

Nick, in your 2628, I think you're getting a bit confused in your terminology. What you're describing doesn't appear to be a SIV - you seem to be describing a pretty standard mortgage securitisation vehicle - most definately a SPV, but most definately not a SIV. Granite's an SPV, one that securitises mortgages.

And I think you've got your gearing a little mixed up too. In your example of an SPV getting a £1bn sale of assets from an originating bank:

- the SPV initially has bugger all (call it £100) of equity. Dr Cash £100; Cr Equity £100;
- the SPV raises £1bn cash by issuing £1bn of MBS debt. Dr Cash £1bn; Cr Debt £1bn;
- The SPV uses the £1bn of cash to buy £1bn (or less) of assets. Dr investments £1bn; Cr Cash £1bn.

The balance sheet that's left after this is

Dr Investments £1bn
Dr Cash £100

Cr Debt £1bn

Cr Equity £100

The gearing would be the debt to equity ratio; or £1bn : £100 - in other words; massive gearing.

Cheers
K

Kirkie

The SPV may be highly geared, but does it really matter?

The point is that if the SPV unwinds, the CP is deposited somewhere else, and the total amount of money / gearing in the financial system as a whole doesn't change ( disregarding any loss ).

Ah well, at the end of the day it's all a squabble over who owns what, and who owes who how much.

George

The point is that if the SPV unwinds, the CP is deposited somewhere else, and the total amount of money / gearing in the financial system as a whole doesn't change ( disregarding any loss ).

You are, of course, right.

I might add, howeover, that the bit I've highlighted above is of somewhat modest relevance in these times!!! Given that there's been (I don't know how much, but guessing) $70bn of writedowns related to these assets over the last 8 months - can you really disregards any loss?

I'd suggest that disregarding any loss is a bit like the knight in Monty Python's "The Holy Grail" who insists on trying to keep on fighting despite the fact that his limbs are rapidly moving towards negative territory... ;)

Its a bit worrying when Nick who works in the Investment Banking Sector does know his SIV from SPV or how gearings calculated.

No wonder the Banking sectors up the creek with such people working in it ; )

OOOps key word missing in my earlier post

Its a bit worrying when Nick who works in the Investment Banking Sector does NOT know his SIV from SPV or how gearings calculated.

No wonder the Banking sectors up the creek with such people working in it ; )

Following the ping-pong in this thread reinforces the belief I had a year ago that buying a modest amount of gold was a good idea.

lfc

"can you really disregards any loss"

No of course not, that was "for the purposes of this discussion about gearing".

A loss is a loss, and working out where the losses have ended up is very important.

In some cases, such as highly geared hedge funds, investors lose 100%.

Much of it is spread around in ginormous pension funds, where it is inconsequential ( say less than normal daily market fluctuation ).

Some banks have taken big losses.

But that is another discussion.

Hi John / Others,

The reserve requirements vary by type. It is not a flat number for all activities as different bank business decisions have different risk.

Thanks for that. I realise you will not like the direction of the next question then, but I'll ask anyway ;-)

If I held a gun to your head and forced you to make an estimate of a weighted average (particular reserve requirements weighted by proportion of business activity 'type') of the overall reserve requirement in an 'average' banking institution, would you say it was closer to 10 or 20x - or something entirely different again?

I do realise there are a lot of averages in there. All I am trying to do is work out whether the 20% from the illustrative example you and Bert were discussing is credible in the market environment of today?

Many thanks,

TC

The SPV may be highly geared, but does it really matter?

The point is that if the SPV unwinds, the CP is deposited somewhere else, and the total amount of money / gearing in the financial system as a whole doesn't change ( disregarding any loss ).

Ah well, at the end of the day it's all a squabble over who owns what, and who owes who how much.


Yes it does matter! The financial system has seen leverage built upon leverage in a similar manner to the split-capital investment trusts. There is not enough equity to support the leverage.

FunkyPinky has already commented earlier in the thread at the enormous growth of "M4" (http://en.wikipedia.org/wiki/M4_money_supply). M4 should be expected to reduce as a consequence of a reduction in leverage, ie "the total amount of money / gearing in the financial system as a whole" should reduce.

But JakNife, I can still argue it doesn't matter all that much.

Where banks have a capital deficit as a result of losses, that capital is likely to be replenished.

By Sovereigh Wealth Funds, or whoever has more money than they know what do with ( mainly oil producers I would guess ).

I don't see how the total leverage is much different to normal once the capital losses have been mopped up.

Hmm.. I take that back.

The SIVs have been operating as unregulated banks, with less capital than a normal bank would have.

So yes, the overall leverage does decrease as they go up in smoke.

Bert,,

Thanks for the clarification.

To the group - the £100 does not get turned into 1,000 when the first bank makes the loan. They are lending £90 in Bert's example or £80 if the reserve requirement requires they hold back £20.

It is the fact that the person receiving the loan will give the money so someone else who will deposit the funds in most cases (pay a bill, etc). When the deposit comes in then the next bank can lend X% of the funds deposited.

An earlier question asked if the lender was taking a £100 deposit and then making a £1,000 loan. They wanted to know where the £900 comes from. As the bank can not make such a loan there is no need to find the £900.

John Corey

Its a bit worrying when Nick who works in the Investment Banking Sector does NOT know his SIV from SPV or how gearings calculated.


AFAIAA he has never claimed to actually work in investment banking....despite what his many posts on related topics might suggest ;-)

...whereas I'm happy to say that I don't know much about the way banks load their cash machines at Tescos.....

ee

Errr sorry John but again you're incorrect I'm afraid

An earlier question asked if the lender was taking a £100 deposit and then making a £1,000 loan. They wanted to know where the £900 comes from. As the bank can not make such a loan there is no need to find the £900.

The Bank can do exactly that. They can lend out based on their capital ratio. I mean yes I am being overly simplistic for the sake of the board, but banks tend to lend around £1000 for every £100 they have as deposits.

The money is created by the banking system - there's nothing particularly mystical in that, it's just economics. Some people don't get it and end up going lala, talking about fiat money and posting a lot on the gold board but the bottom line is that economics works.

The Bank can do exactly that. They can lend out based on their capital ratio. I mean yes I am being overly simplistic for the sake of the board, but banks tend to lend around £1000 for every £100 they have as deposits.



No this is wrong. Find a balance sheet of any bank that is like this.

Bert clearly meant to say capital, not deposits, as is clear from his previous sentence.

The money is created by the banking system - there's nothing particularly mystical in that, it's just economics

But is it sound economics? No-one would question that's how the system works but is it a good system? The problem we have at the moment is, at root cause, too much debt. FRB, coupled with low IR's, has allowed this debt bubble to inflate to the point where it is no longer reasonable or possible to finance it from any sensible projection of economic growth. Hence the inevitability of default. That's how it's always gone in the past (still simple economics) and that's how it's going now. The reason it's more 'interesting' this time is because the debt mountain grew bigger than ever before. The fact that it is bigger over here than over there means the fall-out, when it happens, has every chance of being bigger over here as well. IMO it's inevitable. The only question is how it will unravel. Mass debt deflation is one way but it's painful. Bailout, which looks on the cards over there, is only possible if govmnt finances allow. Will GB's finances allow? I doubt it. So what's left? Inflation. It really isn't surprising that people, not just kooks, are shifting into a bit of gold.

lfc

No this is wrong. Find a balance sheet of any bank that is like this.

It is possible using a net capital ratio of 9% (it's very easy to work it out on a spreadsheet). As I understood it this would be higher than the Tier 1 capital ratios of most banks in Britain.

Ben

It is possible using a net capital ratio of 9% (it's very easy to work it out on a spreadsheet). As I understood it this would be higher than the Tier 1 capital ratios of most banks in Britain.


Then find a balance sheet of a bank where the loans are 10x the deposits. Every major bank has its accounts on the web. Every bank has a line in the balance sheet that says "loans to customers" and "deposits from customers".

If you're right you should easily find a bank that proves your case.

But is it sound economics? No-one would question that's how the system works but is it a good system?

Have you got an alternative? There aren't any. Except the barter system.

Then find a balance sheet of a bank where the loans are 10x the deposits. Every major bank has its accounts on the web. Every bank has a line in the balance sheet that says "loans to customers" and "deposits from customers".

If you're right you should easily find a bank that proves your case.

My apologies I used the term capital ratio when I should have used reserve ratio.

I'm not sure you are interpreting the outcomes of fractional reserve banking correctly. With a reserve ratio of 9% under the fractional reserve system, starting with an initial deposit of £100, you would retain £9 cash and lend out £91 - which would find it's way back to you as a deposit of £91, of which you would lend £82.81 and retain £8.19 cash etc etc. I hope you will see that as the numbers tend to zero you would end up with total customer deposits of £1,111, total customer lending of £1,011 and £100 of cash.

Taking the first bank I arrive, at Barclays 2007 results, I can see cash of £33,077m and lending to customers £345,398m which is a ratio of just over 10 times. Of course banks are enormously complex businesses and this is obviously not as straightforward as this, but you asked for a real life example.

regards

Ben

Bert,

The facts do not line up based on your comments.

Let step is through.

If I deposit £100 how much more can my bank lend based only on that £100. I am talking about 1 bank and 1 deposit. Forget for a minute where the person who gets the loan deposits the money so the banking system in total and this bank will only see £100 deposited.

Is it not correct based on the Wikipedia example you offered that the bank will be able to lend £80? If we use your reserve requirement of 10% (yes, very reasonable to assume) then the bank can lend out £90. Correct so far?

Put another way the simple act of £100 being deposited does not raise 1 banks capital enough to originate £1,000 in new debt. Correct?

John Corey

PS. I am intentionally avoiding the plural form that you used (banks) or the banking system. i am talking about what happens with one bank when they receive 1 deposit for a specific amount.

Arrrrrrgh.

BertEEE said "tend to lend around £1000 for every £100 they have as deposits".

If you look at Barc's balanc sheet as you suggest, yes lending to customers is £345B + £40B to banks. Customer accounts are £294, deposits from banks are £90.5.

I hope you agree that this directly proves that Bert's statement above is wrong.

Fractional banking aside I uderstand very well having studied at university and worked for banks for 15 years.

http://www.investorrelations.barclays.co.uk/INV/A/Content/Files/2007_Barclays_Results_Announcement_web.pdf

BertEEE said "tend to lend around £1000 for every £100 they have as deposits".

If you look at Barc's balanc sheet as you suggest, yes lending to customers is £345B + £40B to banks. Customer accounts are £294, deposits from banks are £90.5.

I hope you agree that this directly proves that Bert's statement above is wrong.

I agree the use of the word deposits is wrong, it perhaps should read something like tend to lend around £1,000 from an initial deposit of £100.

But isn't this being a bit picky? The fact is that Barclays holds only £33B in reserve to settle upto £294B of customer deposits, possibly at short notice. It seems to me that this is a very good example of what the Wikipedia article and Bert were trying to elucidate.

Ben

Have you got an alternative? There aren't any.

Of course there are alternatives and many have been suggested in the past. The simplest would be a return to the gold standard, a more rigid monetary system for sure but one that at least imposes natural constraints on lending. Regardless of that, the current financial system is rooted in the pre-industrial era and is predicated on continuing exponential growth. That's not 'alternative' speak, it's fact. Look at how quickly things unravel with a sniff of recession. Look at how we have lurched from boom to bust and how the longest boom in living memory now threatens to turn into the biggest bust. It also strikes me as very pertinent, and not at all 'wacky', to ask whether exponential growth can continue as we start to push up against the limits of our planets carrying capacity.

If we are to move towards sustainability before it is forced upon us then some pretty major changes will be needed, economic and otherwise. Of course it won't be easy, replacing something as entrenched and all pervasive as 'the devil you know' is never easy. However, to believe that what has worked in the past will continue to work in the future seems little more than an act of blind faith in the powers that be.

lfc

Of course there are alternatives and many have been suggested in the past. The simplest would be a return to the gold standard, a more rigid monetary system for sure but one that at least imposes natural constraints on lending.

Ha ha. Please at a rough estimate, calculate how much gold would be required for the developed economies.

Then for places like China & India that are growing at 10% pa. How much gold should they be buying each year.

And given that the amount required is many many many times the annual production, wont the price go to $1m or so an oz/ and realistically the gold standard is completely unworkable in a modern world.

But isn't this being a bit picky?

No we're talking about two completely different things. There is a big myth around many people that banks can lend 10x their deposits. It keeps getting repeated again and again and is wrong, as you agree.

The fact is that Barclays holds only £33B in reserve to settle upto £294B of customer deposits, possibly at short notice. It seems to me that this is a very good example of what the Wikipedia article and Bert were trying to elucidate.


Works fine in practice nearly all the time. Yes if you don't know anything about fractional banking it might help but not with typos in it.

Hi Bert,

You know we are all nuts on the gold board, banking is so difficult to understand even the rating agencies have their senior moments when it comes to things like, what constitutes AAA.

http://boards.fool.co.uk/Message.asp?mid=10331509


Being a bit La La is of not much concern to a dithering old idiot like me but I still do have concerns about some of the boy racers these days.

Have things gone Mad Max enough for me to be concerned about my road safety yet or should I be making more use of my Nowcard

http://boards.fool.co.uk/Message.asp?mid=10527065

Having bought into in to gold at around $275 La La feels OK at the moment and I quite agree economics do work, I suppose the trick is going to be getting them to work for everyone.

Regards

And given that the amount required is many many many times the annual production, wont the price go to $1m or so an oz/ and realistically the gold standard is completely unworkable in a modern world.

You are confusing value with price. Growth does not require monetary inflation. Simply put, on the gold standard (or any other system in which money is real and limited) you can still have just as much growth, but in this case you can buy more goods for the same money rather than more goods for more money. What counts is spending power not the quantity of money you have to spend.

As for the price that gold would have to be set at, yes that would be higher than current but that is a simple re-evaluation excercise to match gold levels to fiat currency levels. OK, it wouldn't be simple in practice but that doesn't make it a theoretical impossibility. FWIW a back-of-the envelope calculation i did a while back gave an answer of about $50,000/oz if you were to replace all currency with gold. The fact that I would be rich if this were done is of course beside the point ;-)

lfc

You are confusing value with price. Growth does not require monetary inflation. Simply put, on the gold standard (or any other system in which money is real and limited) you can still have just as much growth, but in this case you can buy more goods for the same money rather than more goods for more money. What counts is spending power not the quantity of money you have to spend.

So the only way you can have growth is through deflation? Presumably as companies make more money their stock price goes down too as the total market cap has to stay the same.

And a growing population - does that mean everyone needs a pay cut each year or does that mean you need more gold.

& what happens what happens to the Spaniards when they discovered a rich source of gold and inflation spiralled out of control. Not exactly stable either.

So the only way you can have growth is through deflation? Presumably as companies make more money their stock price goes down too as the total market cap has to stay the same.

There was very strong growth from the onset of the industrial revolution through to WW1. This occurred whilst on the gold standard with relatively stable prices. The price of goods may go down but companies costs also fall so profits can still rise. All about the relationship between these two factors rather than the absolute size of either one.

Enterprise is about selling things for more than you produce them. Currency inflation (and deflation) merely muddy the water and, if anything, cause inefficiencies and extra expenses to enter the system.

FWIW I don't think the gold standard is a perfect system but don't be surprised if it isn't touted as an alternative before the end of the current fianncial shenanigans is over.

lfc